Title Psychological perspectives of periodontal disease

Author(s) Ng, Kwai-sang, Sam.; .

Citation
Issue Date 2005
URL http://hdl.handle.net/10722/50954
Rights
The author retains all proprietary rights, (such as patent
rights) and the right to use in future works.


Psychological Perspectives of Periodontal Disease
by
Sam Kwai Sang Ng
()
Bachelor of Dental Surgery, HKU (1987)
Post-Graduate Certificate in Psychology, HKU (1994)
Master of Social Science in Clinical Psychology, HKU (1998)









A thesis submitted in partial fulfillment of the requirements for
the Degree of Doctor of Philosophy
at The University of Hong Kong.

December 2005
Abstract of thesis entitled
Psychological Perspectives of Periodontal Disease
Submitted by
Sam Kwai Sang Ng
for the degree of Doctor of Philosophy
at The University of Hong Kong
in December 2005

Stress has been found to be directly related to various somatic diseases.
The present study explored how the stress process affects the periodontal status
and, subsequently, the impact of periodontal status on oral health-related quality
of life. The purpose of the present study was to investigate the effects of
psychosocial stress on periodontal health in a community survey study. Multiple
aspects of the stress process, with reference to the contemporary understanding of
stress and emotions, were assessed simultaneously. These included the stressors
acute and chronic ones, stress responses, moderating and mediating factors of
coping and personality traits. Potential confounding factors for periodontal disease
such as age, gender, demographic and socioeconomic factors, smoking and
general health were taken into account. The relation between dental anxiety and
periodontal status as well as the impact of periodontal health on oral-health related
quality of life was also examined.
A cross-sectional study of 1,000 predominantly Chinese Hong Kong
subjects aged 25 to 64 years was conducted. Subjects were asked to complete a set
of questionnaires measuring stressors including changes, significant life event and
daily strains, stress responses, and coping and affective dispositions. Dental
anxiety level and oral health-related quality of life were assessed. Periodontal

clinical attachment level was measured and used as the outcome variable.
The study confirmed that common risk factors like smoking, increasing
age, gender and diabetes mellitus were associated with periodontal attachment
loss while high education attainment was associated with better periodontal
condition. It was found that stressors, revealed as job strain, financial strain, and
the negative affective state of depression were associated with greater levels of
periodontal disease manifested as greater clinical attachment level which gives an
estimate of the historical amount of periodontal destruction. Individuals with more
adaptive coping behaviours (problem-focused coping) were at lower risk for
periodontal disease while those with poor coping behaviours (emotion-focused
coping) were at higher risk. Trait dispositions of anxiety and depression as well as
dental anxiety were also suggested as significant risk factors.
It was also found that those individuals who possessed adaptive coping
behaviors or lower trait disposition of anxiety or depression, even when under job
strain or financial strain, exhibited no more periodontal disease than those
individuals not under job strain or financial strain. Statistically significant
correlations were detected between clinical attachment level with dental anxiety
and oral health-related quality of life.
In conclusion, other than the common periodontal disease risk indicators
such as smoking, increasing age, gender and diabetes mellitus, chronic job and
financial strains, depression, inadequate coping, and maladaptive trait dispositions
including dental anxiety are significant risk indicators for periodontal attachment
loss. High education level and adequate/adaptive coping behaviours as evidence
by high level of problem-focused coping or low level of emotion-focused coping,
and favourable trait dispositions as evidenced by low anxiety or depression traits

were found to reduce the stress-associated risk for periodontal disease.
Periodontal attachment loss is significantly associated with reduced oral
health-related quality of life indicating compromised general well being in the
affected individuals. (497)







Declaration

I declare that this thesis represents my own work, except where due
acknowledgement is made, and that it has not been previously included in a thesis,
dissertation or report submitted to this University or to any other institution for a
degree, diploma or other qualifications.



Signed ...
Sam Kwai Sang Ng




1
i

Acknowledgments



I would like to specially thank Dr. W. K. Leung, my project supervisor, for
his support, advice and guidance through the course of my study. The support of
Periodontology, Faculty of Dentistry, the University of Hong Kong was much
appreciated.
I would like to express my sincere thanks to the three dental clinics, the
seven dental surgeons, the six psychologists, the statistician, the four dental
surgery assistants and the seven interviewers who kindly offered help and support
in the validation of the Dental Anxiety Inventory and in conducting the
community study. I also appreciate the discussions with Dr. Calaise Chan, Dr. K.
O. Ng and the research team of Professor R. J . Genco regarding research design,
planning and preliminary analysis of a subset of the thesis data.
The work described in this thesis was substantially supported by a grant
from the Research Grants Council of the Hong Kong Special Administrative
Region (HKU 7331/00M) and was approved by Ethic Committee, The University
of Hong Kong.




2
ii

Contents
Declaration ..... i
Acknowledgments ...... ii
Table of Contents ... iii
List of Tables ..... ix
List of Figures xiii
List of Abbreviations .. xiv

Chapter 1 Overview and Summary .. 1
1.1 Introduction .... 1
1.2 Periodontitis a complex disease .. 2
1.2.1 Periodontal pathogenic bacteria . 4
1.2.2 Genetics .. 4
1.2.3 Cigarette smoking .. 5
1.2.4 Hormones and systemic factors ..... 6
1.2.5 Age, gender and education . 9
1.2.6 Stress and psychosocial factors .. 9
1.3 Stress and periodontal disease: Biological basis 10
1.3.1 Animal studies 11
1.3.2 Physiological studies of stress in human 13
1.3.3 Psychoneuroimmunology (PNI) studies .... 17
1.3.4 Concluding remarks ... 21
1.4 Stress .. 22

iii
1.4.1 Components of stress process 23
1.4.2 Types of stressors .. 26
1.4.3 Responding to stress .. 27
1.4.3.1 Psychological responses 28
1.4.3.2 Physiological responses . 28
1.4.3.3 Behavioral responses . 30
1.4.4 Mediating and moderating factors of the stress process ... 31
1.4.4.1 Appraisal ... 31
1.4.4.2 Coping ... 32
1.4.4.3 Personality traits .... 33
1.4.4.4 Dental anxiety .... 35
1.4.5 Stress and periodontal disease: Human psychosocial studies. 38
1.4.5.1 Review of studies ... 38
1.4.5.1.1 Necrotizing ulcerative gingivitis (NUG)........ 45
1.4.5.1.2 Psychiatric patients .... 46
1.4.5.1.3 Dental patients 47
1.4.5.1.4 Community population groups .. 53
1.4.5.2 Conceptual and methodological problems in
human studies......................... 55
1.4.5.3 Erie County study .... 57
1.4.6 Concluding remarks ... 60
1.5 Oral health-related quality of life ... 63
1.6 J ustification for the study and statement of the problem ... 67
1.6.1 Hypothesis . 73

iv
1.6.2 Aim .... 73
1.7 Addendum .. 75

Chapter 2 Materials and Methods .... 76
2.1 Validation of Dental Anxiety Inventory (DAxI) 79
2.1.1 DAxI translation . 79
2.1.2 Chinese DAxI validation 80
2.1.3 Chinese SDAxI derivation and validation . 85
2.1.4 Data analysis .. 85
2.2 The cross-sectional community study 86
2.2.1 Subjects .. 86
2.2.1.1 Recruitment of subjects .. 86
2.2.1.2 Study sample ... 87
2.2.2 Design of study .. 89
2.2.2.1 The research team .. 89
2.2.2.2 Procedure ... 90
2.2.3 Data collection ... 91
2.2.3.1 Demographic and medical history .. 91
2.2.3.2 Periodontal examination . 92
2.2.3.3 Psychological questionnaires .. 98
2.2.3.3.1 Evaluation of stressors ... 100
2.2.3.3.2 Evaluation of stress responses. 104
2.2.3.3.3 Evaluation of stress coping and
stress-related trait dispositions 105

v
2.2.3.3.4 Measurement of dental anxiety .. 107
2.2.3.4 Measurement of oral health-related quality of life.. 108
2.2.4 Data analysis 109
2.2.4.1 Descriptive analysis and data processing ... 110
2.2.4.2 Reliability and validity ... 111
2.2.4.3 Inferential statistics: evaluation of associations . 111
2.3 Ethics .. 114

Chapter 3 Results 115
3.1 Demographic characteristics .. 116
3.1.1 Validation of DAxI ..... 116
3.1.2 The cross sectional study ... 118
3.2 Validation of DAxI and its Chinese short form . 123
3.3 The cross sectional study ... 127
3.3.1 Systemic diseases, smoking habits, occupational hazards
and alcohol drinking 127
3.3.2 Dental history and oral hygiene habits .. 136
3.3.3 Dental and periodontal profile of the sample 138
3.3.4 Stressors and periodontal status .... 146
3.3.5 Stress responses and periodontal status . 150
3.3.6 Dispositional factors, coping behaviours and periodontal
status... 154
3.3.7 Dental anxiety and periodontal status..... 160


vi

3.3.8 Risk evaluation and interaction of the significant
demographic, social and psychosocial variables on
periodontal status ... 160
3.3.9 Oral-health related quality of life and periodontal status .. 168
3.3.10 Oral-health related quality of life and concurrent
psychosocial factors 174

Chapter 4 Discussion .. 175
4.1 Current results and implications 175
4.1.1 J ob strain and financial strain ........ 176
4.1.2 Coping 179
4.1.3 Traits of anxiety and depression 182
4.1.4 Impact of dental anxiety on periodontal health ..... 184
4.1.5 Interaction of strains, coping, and traits . 186
4.1.6 Depression .. 188
4.1.7 Life events and changes, and other chronic stressors .... 190
4.1.8 Systemic conditions, smoking, age, gender and education. 192
4.1.9 Impact of periodontitis on oral health-related quality of life.. 193
4.1.10 Oral health-related quality of life and dental anxiety .. 196
4.2 Methodology and limitations .. 197
4.2.1 Representativeness ........ 198
4.2.1.1 Validation of DAxI . 198
4.2.1.2 The cross sectional study ... 198

vii
4.2.2 Validity of measurements ... 201
4.2.2.1 Clinical examination .. 201
4.2.2.2 Psychological questionnaires . 201
4.2.2.3 Measurement of dental anxiety .. 204
4.2.2.4 Measurement of oral health-related quality of life.. 207
4.2.3 Cross-sectional community study .. 208
4.3 General remarks..... 209

Chapter 5 Conclusion . 211
5.1 Conclusion . 211
5.2 Recommendation ... 214

Appendix 1 Brockprobe
TM
Periodontal Probe . 219
Appendix 2 List of questionnaires used in this study, copy of recruitment
notice, information sheet, consent form and Chinese version
of questionnaires used 220
Appendix 3 Table extracted from: Department of Health, HKSAR
(2002) . 241
Appendix 4 Table extracted from: Holmgren et al., (1994) 245
Appendix 5 Publications . 246
References .. 247



viii
List of Tables
Table Page
1.1 Common symptoms of periodontal disease... 3
1.2 Common systemic conditions associated with periodontal
disease 8
1.3 Summary of human studies regarding relationship between
psychosocial factors and periodontal disease.... 40
1.4 Health problems that may be linked to stress 72
2.1 The psychological questionnaires used in the cross-sectional
community study 99
3.1 Demographic background of the Hong Kong population and
subjects recruited for validation of Chinese DAxI and Chinese
SDAxI 117
3.2 Demographic characteristics of subjects in the cross sectional
study... 120
3.3 Demographic characteristics of subjects (n=727) taking part in
the study of oral-health related quality of life 122
3.4 Psychometric characteristics of the Chinese versions of DAxI
and SDAxI. 125
3.5

Population norms, convenient sample means (n=500) of
SCL-90, DASS, and STAI, and their expected and observed
correlations with DAxI.. 126
3.6 Prevalence of systemic diseases in the study sample, n=1,000. 130


ix
Table Page
3.7 Distribution of cigarette smoking (packyear) by categories of
full mouth mean clinical attachment level, n=1,000..... 131
3.8 Mean (SD) cigarette smoking in packyear by categories of full
mouth mean clinical attachment level. 132
3.9

Distribution of cigarette smoking in packyear by age cohorts,
n=1,000.. 133
3.10 Distribution of subjects with occupational hazards by categories
of full mouth mean clinical attachment level, n=1,000. 134
3.11 Distribution of subjects by categories of mean full mouth
clinical attachment level and drinking frequency, n=1,000... 135
3.12 Percentage distribution of dental care habits according to age,
n=1,000.. 137
3.13 Intra-examiner reproducibility for periodontal variables
according to age cohorts 140
3.14 Distribution of subjects by number of teeth present.. 141
3.15 Probing pocket depth, recession and clinical attachment level -
prevalence and extent (mean number of teeth/sites) according to
severity... 142
3.16 Distribution of subjects by categories of full mouth mean
clinical attachment level and age, n=1,000 143
3.17 Mean proportions of teeth and tooth sites with calculus and with
bleeding on probing... 144


x
Table Page
3.18 Mean, SD, internal consistency, item-scale correlation and
inter-scale correlations between the individual subscales of
Measure of Chronic Stress (Daily Strains) ... 148
3.19 Adjusted scores of measures of psychosocial stressors by
categories of clinical attachment level (CAL), n=1,000 149
3.20 Mean, SD, internal consistency, item-scale correlation and
inter-scale correlations between the individual subscales of
SCL-90 and DASS-S. 152
3.21 Adjusted Scores by levels of clinical attachment level,
n=1,000.. 153
3.22 Mean, SD, internal consistency, item-scale correlation and
inter-scale correlations between the individual subscales of
DASS-T and SDAxI.. 156
3.23 Statistical results of factor analysis using the COPE subscales as
predictors...
157
3.24 Mean and SD of COPE scales, n=1,000....... 158
3.25 Adjusted scores of DASS-T, COPE and SDAxI by categories of
clinical attachment level, n=1,000. 159
3.26 Stepwise ordinal logistic regression analysis of potential
indicator for clinical attachment levels.. 164
3.27 Statistics of subjects stratified according to anxiety and
depression dispositions, and coping styles 165


xi
Table Page
3.28 Interaction of trait dispositions and coping styles with daily
strains in risk evaluation of periodontal attachment level. 166
3.29 Distribution of OHIP-14S individual items response.... 170
3.30 Mean scores and internal consistency for OHIP-14S and
individual subscales.. 171
3.31 Comparison of OHIP-14S scores of individual self-reported
symptoms of periodontal disease... 172
3.32 Unadjusted scores (Mean SD) and adjusted scores (Mean
SE) of OHIP-14S and individual subscales of subjects in the two
levels of CAL severity... 173


xii
List of Figures
Figure Page
1.1 A model to evaluate the neural and endocrine mechanisms of
psychosocial stress and coping behaviors in periodontal
disease. 16
1.2 A model of impact of psychological and behavioural responses
of stress on periodontal disease... 25
1.3 A simplified model of stress process... 62
1.4 A diagrammatic conceptual representation of the present study. 74
2.1 The sequence of activities in validation of Chinese DAxI and
SDAxI. 77
2.2 The sequence of activities in the cross-sectional community
study 78
2.3 Measurement of pocket depth using Brockprobe 96
2.4 Measurements of periodontal supports... 97
3.1 Proportion of tooth sites with various severities of calculus
deposit. 145
3.2 Analysis of risk differential between minimal versus
established periodontal disease and (A) J ob strain, or (B)
Financial strain in subjects with high/low psychosocial stress
modulating and mediating abilities. 167


xiii
List of Abbreviations

AAP American Academy of Periodontology
ACH alveolar crestal height
ACTH adrenocorticotropic hormone
AIDS acquired immune deficiency syndrome
ANOVA analysis of variance
ANS autonomic nervous system
APA American Psychiatric Association
BAI Beck Anxiety Inventory
BDI Beck Depression Scale
BOP bleeding on probing
CAL clinical attachment level
CEJ cemento-enamal junction
CHQ-12 12-item Chinese Health Questionnaire
CI confidence interval
Cl calculus
CNS central nervous system
COPE COPE inventory
CP chronic periodontitis
CRH corticotropic-releasing hormone
DAS dental anxiety scale
DASS Depression Anxiety Stress Scales
DASS-S state version of Depression Anxiety Stress Scales
DASS-T trait version of Depression Anxiety Stress Scales
DAxI Dental Anxiety Inventory
DMFS index of decayed, missing and filled surfaces
DMFT index of decayed, missing and filled teeth
DSM-IV Diagnostic and Statistical Manual of Mental Disorder, 4th Edition
EPI Eysenck Personality Inventory
EPQ Eysenck Personality Questionnaire
F F statistic
GAS general adaptation syndrome
GHQ General Health Questionnaire
GHQ-12 12-item General Health Questionnaire
HIV human immunodeficiency virus
HKSAR Hong Kong Special Administrative Region

xiv
HPA hypothalamus-pituitary-adrenal
IgA immunoglobulin A
IgG immunoglobulin G
IL interleukin
IMD International Institute for Management Development
LEQ Life Event Questionnaire
MMP matrix metalloproteinase
NK natural killer cell
NR-group non-responding group
NUG necrotizing ulcerative gingivitis
OHIP Oral Health Impact Profile
OHIP-14S 14-item short form of Oral Health Impact Profile
OHQoL oral health-related quality of life
OHQoL-UK 16-item UK Oral Health-Related Quality of Life
OR odds ratio
P significance level
PESS periodontal emotional stress syndrome
PHA phytohemagglutinin
PMN polymorphonuclear
PNI psychoneuroimmunology
PPD probing pocket depth
R
2
R square in multiple regression
REC recession
R-group responding group
RPP rapidly progressive periodontitis
r
xy
correlation coefficient
SCL-90 Symptom Checklist 90
SD standard deviation
SDAxI Dental Anxiety Inventory Short Form
sIgA secretory immunoglobulin A
SRRS Social Readjustment Rating Scale
STAI State-Trait Anxiety Inventory
STAI-S State Anxiety scale of State-Trait Anxiety Inventory
STAI-T Trait Anxiety scale of State-Trait Anxiety Inventory
t t statistic
TMD temperomandibular joint disorders
Chi square statistic

xv
1
Chapter 1
Overview and Summary

1.1 Introduction
Oral health problems, which range from tooth decay to severe dental
infections and abscesses, affect hundreds of millions of people around the world,
have been a major preoccupation of mankind since time immemorial. Untreated
oral diseases cause not only pain and distress, but may result in disabilities and
secondary health problems. Of all the oral health problems, periodontal disease is
one of the most common oral diseases nowadays. Depending on the measure of
disease applied, 5% to 36% of the adult population of the United States is reported
to be affected by the problem, and locally in Hong Kong 15% to 17% of the adult
population are said to have been affected (Brown et al., 1989; Holmgren et al.,
1994; Oliver et al., 1998; Owens et al., 2003; Petersen & Ogawa, 2005). On the
basis of these epidemiological figures, periodontal disease as a clinical entity is
obviously of significance.
Periodontal disease involves a complicated interplay between plaque
aetiological agents and various genetic and environmental risk factors, and its
occurrence is often unpredictable (Page & Kornman, 1997). Although there
have been genuine advances in our understanding of the pathogenesis, prevention
and treatment of periodontal disease in recent years, these advances have not been
accompanied by a significant reduction in the prevalence and severity of
periodontal disease (Downer, 1998; Albandar et al., 1999). For this reason,
studies to further elucidate the nature of the disease would be of significant value,
2
not only in our understanding of its etiological process, but also in the prevention
of the disease.

1.2 Periodontitis a complex disease
Periodontal disease is an insidious and destructive disease which usually
runs a slow and chronic course. Specific symptoms of periodontal disease may not
appear until the problem has reached an advanced stage, resulting in the loosening
of the affected tooth, which may have to be removed as a consequence.
Non-specific symptoms include those of chronic inflammation and destruction of
the tooth-supporting tissues, ranges from redness, bleeding on brushing, persistent
bad breath, and loosening of affected teeth (Table 1.1).
Periodontal diseases refer to chronic inflammatory conditions caused by
subgingival bacteria. The aetiopathogenesis of chronic inflammatory periodontal
diseases is complex. There are many processes at work, and probably no single
one is the causative factor in all cases (Moulton et al., 1952; Gupta, 1966; Meyer,
1989; Genco et al., 1998 & 1999).
3
Table 1.1 Common symptoms of periodontal disease (American Academy
of Periodontology, 2003).

1. Red, swollen or tender gums
2. Bleeding while brushing or flossing
3. Gums that pull away from the teeth
4. Loose or separating teeth
5. Pus between the gum and the tooth
6. Persistent bad breath
7. A change in the way your teeth fit together when you bite
8. A change in the fit of partial dentures




4
1.2.1 Periodontal pathogenic bacteria
Periodontal diseases result from colonization and subsequent unfavourable
interaction with host by the pathogenic bacteria at the subgingival area. For
instance, two of these microorganisms, Porphyromonas gingivalis and
Actinobacillus actinomycetemcomitans, express potent virulence factors and can
invade human gingival tissues (Grossi et al., 1990, 1994 & 1995; Beck et al.,
1992). However, other non-microbial factors may also play a role in the
pathogenesis of the diseases. Similar to many other chronic infections, the onset
and progression of periodontal infections are modified by local and systemic host
conditions, as well as risk factors that significantly affect the resistance of the host
to the infecting pathogens.

1.2.2 Genetics
It has been found that despite adopting a habit of regular oral care, certain
individuals are still prone to periodontal disease. Research has suggested that up
to 30% of the population may be genetically susceptible to gum disease; and,
some of these individuals are as much as 6 times more likely to experience
periodontal disease (Michalowicz, et al., 1991; Michalowicz, 1994). Recent
studies have provided further evidence that a person's genetic makeup plays a
major role in the onset and severity of periodontal disease. A study examined
periodontal health in 64 pairs of identical and 53 pairs of fraternal twins and
concluded that approximately half of the variance in periodontal disease, such as
attachment loss and probing depth, in the population can be attributed to genetic
differences (Michalowicz, et al., 2000). A consistent racial difference was not
established in epidemiological studies on prevalence of periodontal disease
5
(Brown & Loe, 1993).
Seymour (1991), based on clinical observations, proposed a model of
chronic inflammatory periodontal diseases in which genetic factors divide the
population into susceptible and non-susceptible individuals. For susceptible
individuals, the periodontal tissues are normally in a balance with their
microbial flora and any lesion of the periodontium is present in its stable form.
Progressive pathology appears should this balance be disrupted, resulting in a
depression of immune responsiveness caused by factors such as physical and
mental stress (Ballieux, 1991). What seems evident from these studies is that
genetic influence plays an important role in predisposing certain individuals
vulnerable to periodontal disease, but one thing which remains unclear is under
what condition the development of the disease or the destruction process would be
activated.

1.2.3 Cigarette smoking
Cigarette smoking and tobacco use, apart from its association with many
serious illnesses such as cancer, lung disease and heart disease, as well as
numerous other health problems (Newcomb & Carbone, 1992), are also linked
with an increased risk of periodontal disease (Schei et al., 1959a; Miller et al.,
1986; Grossi et al., 1994; Genco, 1996; Genco et al., 1999; Tomar & Asma, 2000).
Recent studies have shown that smoking or tobacco use may be one of the most
significant risk factors in the development and progression of periodontal disease
(Johnson & Hill, 2004). Genco et al. (1999) in the Erie County study reported a
statistical significant odds ratio of 4.74 (95% CI = 3.34 to 6.73) for heavy smokers
to experience periodontal disease. There is mounting evidence that smokers may
6
be colonized by different periodontal microflora and they recover less favourably
after periodontal therapy than non smokers (Grossi et al., 1996; Zambon et al.,
1996).

1.2.4 Hormones and systemic factors
It has also been shown that certain physiological changes such as
hormonal fluctuation may also increase ones vulnerability to the development of
periodontal disease. For example, during puberty, an increased level of sex
hormones, such as progesterone and possibly estrogen, causes increased blood
circulation to the gingiva. During this time, the gingiva may become swollen, turn
red and feel tender. During the period of pregnancy, women also experience
bodily hormonal changes. These changes can affect many of the tissues in her
body, including her periodontal tissues. The gingiva can become sensitive, and at
times react strongly to the hormonal fluctuations. This may make an individual
more susceptible to periodontal disease (Kinane, 2000). All these hormonal
changes may lead to an increase in the gum's sensitivity with greater reaction to
any irritation, including food particles and plaque; and, hence a greater risk for
periodontal infection (Michelberger & Matthews, 1996).
Thus, it would appear that factors which affect ones basic response or
reaction to plaque may also increase the risk of periodontal disease (Paquette,
2002). Indeed, altered systemic health conditions, such as cardiovascular disease,
respiratory diseases and diabetes were found to increase not only the odds for
periodontal disease, but may also aggravate the periodontal disease condition
further (Hujoel et al., 2000; Scannapieco & Ho, 2001; Soskolne & Klinger, 2001).
Likewise, Grossi et al. (1994) conducted a study of risk assessment for
7
periodontal disease and found that diabetes, allergy (negatively associated) and
anaemia were significant risk indicators for clinical attachment loss and
periodontal disease. Review of previous studies suggested a higher incidence of
periodontitis in diabetics compared to healthy controls and that this difference was
also statistically significant. A direct causal or modifying relationship has been
proposed in which the hyperglycemia and hyperlipidemia of diabetes result in
metabolic alterations that may then exacerbate bacteria-induced inflammatory
periodontitis (Soskolne & Klinger, 2001; Dietrich & Garcia, 2005; Genco et al.,
2005). Table 1.2 summarizes the influence of some of the systemic conditions
which have been investigated.
Such periodontal-systemic diseases relationship is not only limited to the
fact that these systemic diseases risk factor such as diabetes and osteoporosis
(Soskolne & Klinger, 2001; Wactawski-Wende et al., 2001) could increase the
odds for the severity of periodontal disease, but also periodontal disease was
found to increase the risk for systemic problems i.e. cardiovascular problem,
ischemic stroke or adverse pregnancy outcomes (Offenbacher et al., 1996;
Scannapieco et al., 2003).
8
Table 1.2 Common systemic conditions associated with periodontal
disease
a
.

1. Genetic conditions
e.g. Downs Syndrome, Papillon Lefevre Syndrome, Chediak Higashi
Syndrome, and Cyclic Neutropenia
2. Diabetes Mellitus
3. Leukemia
4. Anemia (negatively associated)
5. Allergy
6. Pregnancy
7. Use of medicines
e.g. oral contraceptives, epileptic medications (Dilantin),
immunosuppressants (cyclosporin)


a
Modified from Grossi et al. (1994) & American Academy of Periodontology (2003).
9
1.2.5 Age, gender and education
Among the many factors suggested, age, male gender and low education
level have been shown to be related to increased risk in periodontal disease (Schei,
1959b; Miller et al., 1986; Beck et al., 1990; Grossi et al., 1994; Kocher, 2005).
Early studies suggested that older populations were at greater risk for periodontal
disease (Marshall-Day, 1955; Barmes, 1977; Garcia & Cutress, 1986; Hugoson et
al., 1986; Albandar & Brunelle, 1999; Albandar & Kingman, 1999). Similarly,
studies have also suggested gender as a risk indicator with males experiencing
more periodontal disease (Russell, 1960; Milleret al., 1986; Grossi, 1994; Grossi
et al., 1995). Genco et al. (1999), in a population study of 1,426 subjects in Erie
County, New York, reported increased risk with statistical significant odds ratio
of 11.21 (95% CI = 7.39 to 17.02; c.f. aged 25 to 34) for people aged 65 to 74 and
1.33 (95% CI = 1.03 to 1.70; c.f. female) for male subjects to experience
periodontal disease. The precise reason for this relationship is not exactly clear;
one explanation is that men and elderly subjects tended to pay less attention to
their oral care. This has been demonstrated in a study that a patient's oral hygiene
status is of more predictive significance than chronological age for the
development of periodontal disease (Abdellatif & Burt, 1987). The prevalence of
severe periodontitis was higher in less educated persons (Kocher et al., 2005;
Senna et al., 2005; Torrungruang et al., 2005). It is generally believed that less
educated persons have less knowledge about significance of periodontal health
and hence compromised maintenance care; longitudinal studies, however, are
necessary to explore whether this is a true risk factor (Torrungruang et al., 2005).

1.2.6 Stress and psychosocial factors
10
Although the significance of periodontal disease risk factors such as
genetic predisposition, smoking, hormonal changes, systemic conditions, age,
gender and education have been demonstrated, there is however a significant
portion of the variation in disease severity (variance in statistical term) which
cannot be explained by these factors alone (Marcenes & Sheiham, 1992). The
possible association between psychological factors and inflammatory periodontal
diseases has also become the subject of many studies (De Marco, 1976; Lowental,
1981; Davies et al., 1985; Genco et al., 1999). Psychosocial factors such as stress
have been linked to many serious conditions such as hypertension, cancer, and
numerous other health problems (Lazarus, 1999; Weiten, 2004) including dental
disease (Krasner, 1978). Stress has also been suggested as a risk factor for
periodontal disease (Monteiro da Silva, 1995; Genco et al., 1999). Research
demonstrates that stress, especially in chronic form, can influence host defenses in
terms of immunosuppression hence increasing ones vulnerability to disease
(Rogers, 1979). Stressful demands can affect the homeostasis of ones human
body and immune responses (Selye, 1963; Ballieux, 1991).

1.3 Stress and periodontal disease: Biological basis
Studies directed towards the biochemical and physiological mechanisms
by which psychosocial stress contributes to periodontal disease are needed to
establish the biological basis for this relationship. Such studies may include
assessment of bodily biochemical, neurological, immunological and
endocrinological alterations in response to psychological and behavioural changes.
Evaluation and/or investigation into biological basis of these mechanisms using
animal models is deemed necessary and instructive.
11

1.3.1 Animal studies
Laboratory animal studies employing a variety of stressors have been
conducted to investigate the causal relationship between stress and inflammatory
periodontal disease. Most of these experiments were based on the concept of the
general adaptation syndrome (GAS) (Selye, 1946, 1948, 1975). In accordance
with this concept, laboratory animals were exposed to prolonged stress, similar to
the chronic stress often endured by humans. As stress continues, the body's
resources for fighting stress may be depleted and corresponding physiological
arousal decreases. The organism's resistance declines and it may eventually
collapse from exhaustion.
Stressors in these animal studies, mainly conducted in 1950s and1960s,
included prolonged immobilization, fasting, cold and hot bath, violent physical
exercise, exposure to extreme sensory stimulations (e.g. ringing of bells, bright
light), and injection of chemicals (Shklar & Glickman, 1953; Ratcliff, 1956; Fedi,
1958; Gupta et at., 1960; Karren & Ingle, 1964; Cohen et al., 1969). Findings of
these experimental studies revealed consistent features associated with
degeneration of connective tissue of the periodontal tissues and reductions in
remodelling activity.
In a more recent study, Breivik et al. (2001) demonstrated a positive
feedback loop between the hypothalamus-pituitary-adrenal (HPA) axis and
periodontal disease: the disease activates the HPA axis, and a genetically
determined high HPA responsivity further increases disease susceptibility. In this
study, the periodontal tissue destruction was more severe in Fischer 344 rats (than
the Lewis strain) that responded to high HPA axis reactivity following
12
subcutaneous corticosterone delivery, and that adrenalectomy reduced the severity
of periodontitis.
Gaspersic et al. (2002) investigated the influence of stress on periodontal
breakdown in Wistar rats. The rats were divided into three groups of 10 animals in
each; the first group was exposed to restraint stress for 12 hours every day for a
period of 4 weeks; the second group was exposed to restraint stress for 2.5 hours
every day for a period of 4 weeks; the third group served as control. Periodontitis
was induced by placing silk ligatures around the maxillary right second molar
teeth in the experimental groups. Combined stress and ligation lead to a
significantly greater attachment loss and alveolar bone resorption than either
treatment alone, while no differences were seen between the two stress regimens.
A reduced body weight was found in both restrained groups of rats after 4 weeks
and a reduced weight of the thymus in the rats restrained for 12 hours every day,
while no changes were observed in the weight or composition of the suprarenal
glands. It was thus suggested that stress alone does not result in periodontal
disease but may modulate the pathophysiological processes of already present
periodontal inflammation, resulting in accelerated destruction of periodontal
tissues.
Takada et al. (2004) investigated the effect of restraint stress on the
progression of periodontitis in rats. A nylon ligature was placed around the second
right maxillary molars of 100 male Wistar rats. The rats were then divided into
group S, subjected to restraint stress for 12 hours per day for up to 10 days, and
group N, controls. Ten animals were sacrificed on days 2, 4, 6, 8, and 10. Blood
biochemical parameters of stress include concentration of blood glucose,
adrenocorticotropic hormone (ACTH), corticosterone, and adrenaline were
13
measured. The size and weight of the thymus and the spleen were evaluated; and
the furcation area of the second maxillary molars were examined histologically
and histomorphometrically. In group S, all values of stress parameters were
increased, and the thymus and the spleen were atrophied. In group N, only slight
alveolar bone resorption was detected. A marked alveolar bone resorption
occurred in group S between days 8 and 10. The results of this study supported
that the restraint stress modulates the progression of periodontal inflammation.
Animal studies have provided impressive evidence to substantiate the
association between stress and periodontal disease. It also allowed opportunities
to evaluate further the biological basis such as biochemical and immunological
mechanisms involved.

1.3.2 Physiological studies of stress in human
The interaction between the immune system and central nervous system
(CNS), that mediates the effects of stress in maintaining immunologic and
behavioural homeostasis, has been the subject of intense study. Analyses of the
role of stress in periodontal disease begin with the overall effect of stressors, be it
physical or psychological, as appraised by the brain. Physiological responses of
the stress process have been shown to modulate the immune system through the
neural and endocrine systems in at least three different pathways (Brevik et al.,
1996; Seiffert et al., 2002).
The first pathway of stress-induced response is routed through the
autonomic nervous system (ANS). In response to stress, particularly if acute in
nature, the hypothalamus activates the sympathetic division of the ANS, resulting
in secretion of catecholamines such as epinephrine and norepinephne from the
14
adrenal medulla into the bloodstream and produces the physiological changes
seen in the fight-or-flight response. It is reported that the sympathetic nervous
system modulates immune function at a number of levels and that epinephrine
exerts an overall immunosuppressive effect (Anisman et al., 1996; Seiffert et al.,
2002). Catecholamines also affect prostaglandins and proteases (e.g. matrix
metalloproteinases or MMPs), which in turn, could enhance periodontal
destruction (Dimsdale & Moss, 1980; Fibiger et al., 1984) (Figure 1.1).
Under stress, the release of adrenaline and noradrenaline may not only
cause a decrease in blood flow supplying oral cavity, but possibly also in those
blood elements necessary for maintaining resistance against microbial infection
and/or destruction. Glucocorticoids released during stress may also prolong this
vascular response (Meyer, 1989).
The second pathway of stress-induced response, primarily in stress of less
acute nature, is transmitted to the HPA, to promote the release of
corticotropic-releasing hormone (CRH) from the hypothalamus to the hy-
pophyseal portal system, in turn ACTH from the pituitary, and then glucocorticoid
hormones from the adrenal cortex (Ader et al., 1990; Blalock, 1994). A model to
evaluate the role of psychosocial stress and coping behaviours in periodontal
disease is shown in Figure 1.1.
At the molecular level, glucocorticoids significantly inhibit essential
functions of inflammatory cells including macrophages, neutrophils, eosinophils,
and mast cells such as chemotaxis, secretion, and degranulation (Schleimer et al.,
1989; Schleimer, 1990; Davis et al., 1991). Glucocorticoids also inhibit
macrophage medicated antigen presentation, lymphocyte proliferation and
differentiation to effector cell types such as helper lymphocytes, cytotoxic
15
lymphocytes, natural killer cells, and antibody-forming B cells (Snyder & Unanue,
1982). Thus, the cascade process of the immune response is compromised.
Corticosteroids also inhibit production of cytokines (including interleukins IL-1,
IL-2, IL-3 and IL-6), tumor necrosis factor, interferon gamma, granulocyte and
monocyte colony stimulating factors, and pro-inflammatory mediators such as
prostaglandins and leukotrienes. Production of endogenous anti-inflammatory
proteins and lipocortins was also enhanced by glucocorticoids (Schleimer et al.,
1989). Glucocorticosteroids have major suppressive effects on immune functions
through highly specific mechanisms at multiple levels, including those involving
humoral immunity like IgA, lgG, and innate cellular immunity like neutrophils.
All these immune responses play important roles in protection against infection
by periodontal pathogenic organisms.
The third pathway is the sensonic peptidergic nervous pathway, named
neurogenic inflammation, in which neuropeptides are released from sensory
nerve fibers upon stimulation by external stimuli (Farber et al., 1990; Bartold et
al., 1994). Studies have suggested that peripheral release of neuropeptides may
promote various inflammatory processes (Helke, 1991; Bartold et al., 1994).
Neuropeptides also modulate the activity of the immune system and inhibit the
release of cytokines.
16

Figure 1.1 A model to evaluate the neural and endocrine mechanisms of
psychosocial stress and coping behaviors in periodontal disease.
Pathway of neurogenic inflammation not included. (Modified from
Genco et al., 1998)
+

-
+
+ Acute
+ -
Chronic
+
Adequate Coping - + Inadequate Coping
Psychosocial Stress
Activation of CNS
Hypothalamus (CRH)
Autonomic Nervous
System
Pituitary (ACTH)
Adrenal Cortex
(Cortisol)
Immunie Responses (sIgA, IgG,
production & PMN cells functions;
i.e. localized immne response)
Periodontal
immunoinflammatory
response (IL-1, MMP etc.)
Periodontal Disease
+
Adrenal medulla
(norepinephrine,
epinephrine)
Prostaglandins &
Proteases (e.g. MMPs)
17
1.3.3 Psychoneuroimmunology (PNI) studies
Interdisciplinary psychoneuroimmunology (PNI) studies have provided
further support to the argument that excessive psychosocial stress or some clinical
psychological conditions (e.g. anxiety disorder, depression) can alter host
defences and increase vulnerability to certain illnesses, especially those intimately
associated with immunologic mechanisms, such as chronic infection, autoimmune
disease and malignancy (Kiecolt-Glaser et al., 1984a, 1984b, 1987, 1988 & 1991;
Linn et al., 1984; Schleifer et al., 1984; Dorian et al., 1985; Irwin et al., 1987;
Arnetz et al., 1987; Kemeny et al., 1989; Kovar et al., 1989; Kiecolt-Glaser &
Glaser, 1992; Fredrikson et at., 1993; Andersen et al, 1994). Earlier studies also
suggested the presence of a relationship between mental illness and
immunosuppression (Liedemann & Prilipiko, 1978; Solomon, 1981).
Researchers have consistently detected an association between stress and
suppressed immune function, Sheldon Cohen and his colleagues commented, It
is unclear whether the immune changes related to stress in these studies are of the
type or magnitude that would influence susceptibility to infection (Cohen et al,
1993, p.131). Methodological shortcomings have plagued the studies that have
linked stress to actual infections (Cohen & Williamson, 1991). Cohen et al. (1993)
conducted an outstanding study with enormous effort for control of possible
confounding variables and highly unusual quarantine conditions for excluding the
likelihood that subjects would develop colds from other sources besides the virus
tested. Volunteer subjects, including 154 men and 266 women, aged from 18 to 54,
participated the study at the Medical Research Councils Common Cold Unit in
Salisbury, England. Subjects were given free accommodation and quarantined for
nine days. Medical examinations were conducted in the first two days to verify
18
that they were in good physical health and were administered a series of
questionnaires to assess their recent stress, personality and health habits. They
were then given, with informed consent, nasal drops that contained either a
respiratory virus or a harmless saline solution. They were then followed for the
subsequent six days to see whether they developed cold symptoms or viral
infection based on clinical assessment and results of laboratory nasal secretion
cultures. The results suggested that the associations between high stress and an
increased incidence of colds were still significant after statistical adjustment for
variations in subjects personality and health practices.
Vulnerability studies have been conducted to investigate the correlations
of immune function with psychological vulnerability which primarily refers to
the inability to forestall, or assuage the distress accompanying adverse life
circumstances (Kaplan 1991). Sieber et al. (1992) in a laboratory study
investigated the changes in natural killer cell (NK) activity and proportions of
circulating T and NK lymphocyte in adult males immediately after exposure to
controllable or uncontrollable stress (noise) as well as the lasting effects of such
stress in 24 and 72 hours later. The relationship between the personality
disposition of control-perception and stress-immunosuppression was evaluated.
No reduction in NK activity was detected in subjects who perceived they had
control over the noise as well as control (no-noise) subjects. On the other hand,
subjects who perceived that they had no control over the stressor showed reduced
NK activity immediately after the stress session, and for as long as 72 hours
afterwards. The results suggested the importance of perceived control in
moderating the short- and long-term effects of stress on NK cell activity. Lack of
social supports, including those manifested as loneliness, disruption of marital
19
relationships due to death or separation, rendering people vulnerable to stressors
have been shown to have a positive association with immunosuppression
(Kiecolt-glaser, 1984a; Bartrop, 1977; Schleifer, 1983; Levy, 1985).
Evidence regarding the association between various psychosocial factors
and immunosuppression has led researchers to explore the possibility of
psychological interventions to allow positive effects or influence for the immune
system activity. PNI intervention studies have employed a number of intervention
strategies including relaxation training, hypnosis, physical exercise, classical
conditioning, and cognitive-behavioural therapy to modulate individuals negative
affective responses and vulnerability in stressful conditions. The most common
interventions are relaxation and hypnosis. Walker et al. (1993) reviewed
extensively studies on the modulation of immune responsiveness to stress by
hypnosis and relaxation. Some studies (Black 1963a, 1963b; Black & Friedman,
1965; Zachariae & Bjerring, 1990) involved the investigation of immediate and
delayed hypersensitivity responses in subjects following injection of the same
antigen in both arms (e.g. purified tuberculin protein derivative). Subjects were
then hypnotized and suggested different reactions on each arm. Significant
differences between the two arms were usually detected. It is however not clear
whether such differences were due to change in skin or modulation of immune
response. Relaxation techniques have also been shown consistently to enhance
immune function with increase in NK cell activity and decrease in Herpes
Simplex Virus antibody titres (Kiecolt-Glaser et al., 1985).
Classical conditioning in animal studies provided impressive evidence for
its effects on the modulation of immunity (Ader et al., 1991). Classical
conditioning studies were limited, however, in humans. Bovbjerg et al. (1990)
20
investigated nausea and immune function in 20 cancer patients in the hospital
prior to chemotherapy and compared with assessments conducted at home.
Proliferative responses to T-cell mitogens were lower for cells isolated from
hospital blood samples than from home samples obtained several days earlier.
Patients also experienced increased nausea in the hospital. Multiple regression
analyses revealed that decreased immune function in the hospital was not related
to increased anxiety. The observed anticipatory immune suppression is consistent
with the hypothesis that chemotherapy patients may develop conditioned immune
suppression as well as conditioned nausea after repeated pairings of hospital
stimuli with the emetic and immunosuppressive effects of chemotherapy.
LaPerriere et al. (1990) studied the impact of aerobic exercise training as a
buffer to the affective distress and immune decrements which accompanied the
notification of HIV-1 antibody status in AIDS risk subjects. Fifty asymptomatic
gay males with a pretraining fitness level of average or below (determined by
predicted VO2 max, i.e. the maximum amount of oxygen in milliliters, one can
take in and assimilate in one minute per kilogram of body weight; this figure is
commonly used as a measure of physical fitness) were randomly assigned to
either an aerobic exercise training program or a no-contact control condition, for a
period of five weeks. At 72 hours prior to notification of HIV serological status,
psychometric, fitness and immunologic data were collected from all subjects.
Psychometric and immunologic measures were repeated one week after
notification. Seropositive controls showed significant increases in anxiety and
depression, and decrease in natural killer cell number following notification.
Seropositive exercisers did not show such changes and a statistical significance
difference was not evident compared with both seronegative groups. These
21
findings suggest that an experimentally manipulated aerobic exercise training
intervention might attenuate the concurrent changes in some affective and
immunologic measures in response to an acute stressor. However, a study
involving an intensive stress reduction intervention (included relaxation, stress
management skills and health behaviours) failed to report any positive
immunological alterations among HIV-seropositive subjects as compared with
their controls (Coates et al., 1989). Pennebaker et al. (1988) found that
self-disclosure, a standard counseling technique in psychotherapy, could lead to a
positive alteration of immune functioning with higher blastogenic response to
mitogens.
There have been very few reports with commonly used PNI measures in
large community samples. Evans et al. (2000) reported such data for secretion
rates of secretory immunoglobulin A (sIgA) from saliva samples taken from 1971
subjects in the West Central Scotland. Lower sIgA and salivary flow were found
to be significantly associated with poorer social class, increased age, and being
female. Smokers also had lower sIgA but not lower salivary flow. Multiple linear
regression showed that demographic variables were significant predictors of sIgA
independently of each other and assay variation. This gave probably the first
results from a community study that significant associations exist between stress
and sIgA levels which serve as a first line of mucosal defense against infection.
PNI study focused specifically on periodontal disease, however, has not been
reported.

1.3.4 Concluding remarks
The suppressive impact of stress on the immune system has been well
22
researched and reasonably established. Animal studies provided evidence linking
stress and periodontal disease. PNI studies served to provide evidence on a
molecular and cellular base regarding the association between immunologic
functioning and stressful life events, negative affective states (e.g. anxiety,
depression, anger), and psychological vulnerability. PNI intervention studies
approached the issue from an evaluative perspective assessing the outcome
immune responses following manipulation of these factors and provided further
supportive evidence.

1.4 Stress
Reports on the impacts of psychosocial factors on general health status of
an individual were available years ago (Cohen et al., 1986; Cooper & Paragher,
1993). The relationship between psychological stress, distress, and periodontal
disease has received some attention in the past few decades. Studies have reported
a higher prevalence of chronic destructive periodontal disease in persons with
certain psychological disorders (Belting & Gupta, 1961; Gupta, 1968; De Marco,
1976). Several studies have suggested that psychological stress may be associated
with acute necrotizing periodontal diseases (Giddon et al., 1964; Cohen-Cole et al.,
1986; Johnson & Engel, 1986; Schoor & Havrilla, 1986; Melnick et al., 1988;
Murayama et al., 1994; Horning & Cohen, 1995).
Psychological factors, which were suspected of increasing the risk for
periodontal diseases, have been investigated in several studies in the last decade
(Ballieux, 1991; Monteiro da Silvia et al., 1995; Beck, 1996; Brevik et al., 1996;
Linden et al., 1996; Monteiro da Silvia et al., 1996; Axtelius et al., 1997a, 1997b
& 1998; Page & Beck, 1997; Salvi et al., 1997; Genco et al., 1998; Albandar,
23
2000; Hildebrand et al., 2000). Population and epidemiological studies strongly
suggest that stress and distress, among other things, are important risk indicators
for periodontal disease (Green et al., 1986; Marcenes et al., 1992; Machtei et al.,
1992; Freeman & Goss, 1993; Minneman et al., 1995; Moss et al., 1996; Croucher
et al., 1997; Genco et al., 1999; Wimmer et al., 2002). Results of investigations
also suggest that psychological stress may contribute to an unfavorable response
to periodontal therapy (Axtelius et al., 1997a, 1997b & 1998).
These studies, however, failed to address some major conceptual and
methodological issues such as assessment and elucidation of contribution and
interaction of various components of the stress process including the role of dental
anxiety, as well the as the issue of how oral health related quality of life was
affected.

1.4.1 Components of stress process
The word stress has been used in different ways by different theorists.
Stress, in the broadest sense, can be defined as any circumstances that threaten or
are perceived to threaten ones well-being and that thereby tax ones coping
abilities. The threat may be to ones immediate safety, long-term well-being, or, in
more abstract perspectives, reputation, self-esteem, peace of mind, or things that
one values.
People struggle with stress every day. Most of the time, stress comes and
goes without leaving any enduring effect. However, when stress is severe or
prolonged, or when many stressful demands pile up, ones physiological and
psychological functioning may be affected. Stress can contribute to a host of
common problems related to emotions and cognitive functions, including poor
24
academic or work performance, poor temper control or emotional outburst,
insomnia, and irritability.
A convenient way to understand the stress process is in terms of stressors
and stress responses. Stressors are events that threaten or challenge people. They
are the sources of stress, such as having to make decisions, attending examination,
and natural disasters. Stress responses include psychological, physiological, and
behavioural reactions to stressors. Examples include emotions of annoyance,
agitation and anger, muscle tension, physiological arousal, and concentration
difficulties. Mediating processes and moderating factors determine how we react
to an external stressor. These include appraisal process, coping, and personality
traits (Figure 1.2).
25


Model of Stress Process
Stressors

Frustration,
Conflict, Change &
Pressure existing in
acute or chronic
stressor, live events,
or daily hassles
(strains)




Mediating &
Moderating Factors

Appraisal

Coping Skills

Personality






Stress Repsonses

Psychological

Physiological

Behavioural



Figure 1.2 A simplified model of stress process.
26
1.4.2 Types of stressors
An enormous variety of events can be stressful for an individual. The four
principal types of stress are frustration, conflict, change, and pressure. Frustration
arises in any situation in which the pursuit of some goal is thwarted. Failures and
losses are common examples of frustration that are often quite stressful. Conflict
occurs when two or more incompatible motivations or behavioural impulses
compete for expression. Examples include conflicts in playing tennis or football,
taking a demeaning job or going on welfare, and dating with an attractive person
or accepting the possibility of being rejected. Change refers to noticeable
alterations or losses in ones living circumstances that one has to readjust oneself.
There are plenty of aversive events (e.g. being fired from work, death of a friend
or relative), as expected, are stressful. There exists however many seemingly
positive events (e.g. getting married, having a baby, or getting promoted), that
people can perceive as stressful even though the changes are welcomed (Holmes
& Rahe 1967; Rahe & Arthur, 1978). Pressure involves expectations or demands
that one behaves in certain way. Examples are salespersons under pressure to meet
specific target sales revenue, teenagers under parents pressure to adhere to certain
rules and values and parents struggling to sustain family finance.
These four principal types of stress, either alone or more often in
combinations, appear in peoples life in one of the following kinds of stressors,
namely a chronic stressor, major life event or acute stress, and daily hassles. A
chronic stressor refers to one that is recurrent or continues indefinitely. Examples
include frustration and conflict in marital relation, chronic illnesses such as cancer
or diabetes, poverty, living or working in crowded conditions, and too many work
responsibilities. A major life event or acute stress is bad for a period of time and
27
may leave some prolonged effects as it goes away. It usually comes from major
and significant changes, frustration, and pressures of the recent past or the
anticipated near future. Acute stress tends to be thrilling, exciting, and, not
uncommonly, exhausting. Examples include death of a loved one, severe illness
like life-threatening cancer, job loss, divorce, natural disasters such as hurricanes
or earthquakes, and terrorists attack. Daily hassles are the minor nuisances that
are unpleasant and often temporary. Examples include grocery shopping,
unpleasant salespeople, extreme weather conditions, forgetting your keys, and a
broken down personal computer system.
People might guess that minor stresses would lead to minor effects, yet
that is not necessarily true. In recent years, studies have indicated that everyday
problems and the chronic minor nuisances in daily living are important form of
stress and may have significant harmful effects on mental and physical health
(Kohn et al., 1991; Pillow et al., 1996). Why would minor hassles affect ones
mental and physical well being is not yet clear. It may be due to the cumulative
nature of stress (Seta et al., 1991). Routine and minor stresses at workplace, at
home, in studying, in looking after children, and in a relationship may be fairly
benign individually; however, as stress adds up, collectively they could lead to
great strain.

1.4.3 Responding to stress
Peoples response to stress is complex and multidimentional. A persons
reactions to stress can be analysed at three levels: psychological responses
(affective and cognitive), physiological responses, and behavioural responses
(Lazarus & Folkman, 1984; Lazarus, 1993b, 1999; Lazarus & Lazarus, 1994;
28
Weiten, 2004).

1.4.3.1 Psychological responses
Psychological responses can be evaluated at two levels: affective and
cognitive. When people are under stress, they tend to react emotionally; stress,
more often than not, lead to unpleasant emotions rather than pleasurable ones
(Lazarus, 1993b). Common affective responses to stress include emotions of anger,
anxiety and sadness. Emotions of anger may range from mild annoyance to
irritability, hostility to uncontrollable rage. Emotions of anxiety include
nervousness, tension, mild apprehension, panic/phobia, guilt, and shame.
Emotions of sadness encompass sense of dejection, moodiness, loneliness, to
clinical depression or grief (Woolfolk & Richardson, 1978; Lazarus & Folkman,
1984; Lazarus, 1993b & 1999; Lazarus & Lazarus, 1994). Examples of cognitive
responses are concentration problems, indecisiveness, rumination/intruding
thoughts, forgetfulness, sensitivity to criticism, and not uncommonly negative
thoughts and evaluations including the depressive cognitive triad of worthlessness,
helplessness and hopelessness (Lazarus & Lazarus, 1994; Lazarus, 1999).

1.4.3.2 Physiological responses
Physiological arousal in response to stress was originally called the
fight-or-flight response by Canon (Cannon 1932). This is mediated by the
sympathetic division of the autonomic nervous system in reaction to threat so as
to mobilize the organism for attacking (fight) or fleeing (flight) an enemy. The
immediate physiological reactions in response to the increased sympathetic
discharge include pupils dilation, salivation inhibition, bronchial dilation with
29
increased respiration, increased heart rate, slow down of digestive processes,
peripherial vasoconstriction in viscera redirecting more blood to voluntary
muscles, heart, brain and other important organs. This fight-or-flight response
can backfire if a person remains in a chronic state of stress and autonomic arousal.
Concern about the effects of prolonged physical arousal was first
described by Seyle (1936). Seyle (1936, 1956 & 1982) formulated an influential
theory of stress reactions called the general adaptation syndrome (GAS). The
GAS has three stages: alarm, resistance, and exhaustion.
The first stage of the GAS, alarm, is basically the fight-or-flight
response, the various physiological changes preparing the body to attack or to flee
a threatening situation. The sympathetic branch of the autonomic nervous system
is activated and prompts the release of two catecholamines: epinephrine (also
called adrenaline) and norepinephrine (also called noradrenaline), from the
medulla of the adrenal glands. The net result of catecholamine elevation is that
your body is mobilized for immediate action. Heart rate and blood flow increase
and more blood is pumped to your brain and muscles. Respiration and oxygen
consumption speed up, which facilitates alertness and response. Digestive
processes are inhibited to conserve your energy. The pupils of your eyes dilate
increasing visual sensitivity and acuity. Glucocorticoids (stress hormones) like
cortisol are also released from the adrenal cortex resulting from the stimulation of
CRH from the hypothalamus following the release of ACTH from the pituitary, i.e.
the HPA axis.
In the second stage of the GAS, resistance, the body tries to calm itself
and restrain the fight-or-flight response from the alarm stage. During this phase,
physiological changes stabilize as coping efforts get under way. These changes
30
allow people to deal with stressors more effectively over a longer period of time.
If the stress continues over a substantial period of time, the organism may
enter into the exhaustion stage. According to Seyle (1982), the bodys resources
for fighting stress are limited. If the stress cannot be overcome, the bodys
resources may be depleted, and physiological arousal will decrease. At this stage,
the body admits defeat and suffers the negative consequences of the stressors,
such as a decreased capacity to function correctly, less sleep, or even collapse
from exhaustion. Prolonged autonomic arousal can eventually contribute to the
development of various physical diseases (Kaplan, 1989; Weiten, 2004)

1.4.3.3 Behavioural responses
People react differently in stressful situations. Sometimes, the behaviours
are somewhat subtle, such as strained facial expressions, shaky voice, tremors or
spasms, accident proneness, difficulty sleeping, over-eating or loss of appetite
(Weiten, 2004; Lazarus & Lazarus, 1994). Behavioural responses are more
obvious when people take advantage of the preparatory physiological responses of
the fight-or-flight response. People sometimes may behave aggressively toward
other people. Avoidance behaviours in which people try to escape from
threatening situations are not uncommon, examples include quiting jobs, dropping
out from school, abusing alcohol or other drugs, or attempting suicide.
Most, if not all, behavioural responses to stress involve coping. This refers
to active efforts to master, reduce, or tolerate the demands created by stress
(Weiten, 2004). This definition is neutral as to whether the coping behaviours are
adaptive or not. One of the not-so-healthy coping behaviours is striking out at
others with aggression, a typical response in frustration. Giving up (withdrawal
31
from the battle) and self-indulgence (engaging in excessive consummatory
behaviours such as eating, drinking, smoking, spending) are coping patterns that
tend to be of limited value. Defensive coping with subconscious/unconscious
reactions is rather common and allows a person to stay away from unpleasant
emotions such as anxiety and guilt; denial of reality, fantasy, isolation,
overcompensation (making up for frustration in one area by overgratification in
another) are common examples of defense behaviours (Freud, 1940; Laughlin,
1979; Vaillant. 1994; Erickson et al., 1996).
Thus far, we have probed the nature of stress and described how people
typically respond to it in stressful situations. We turn next to the issue of why
stress affects everybody differently and examine the factors mediating and
moderating the impact of stress.

1.4.4 Mediating and moderating factors of the stress process
1.4.4.1 Appraisal
Appraisal is a mediating processes that determines how we react to an
external stressor. Lazarus and Folkman (1984) distinguished between primary and
secondary appraisal. Primary appraisal is an initial evaluation of whether an event
is irrelevant to oneself, relevant yet not threatening, or stressful. When one views
an event as stressful, one is likely to make a secondary appraisal, that is an
evaluation of ones coping resources and options for dealing with the stress.
People are not very objective in appraisals of potentially stressful events.
Stressors can be interpreted in different ways, such as harm or loss, as threats or as
challenges. A study of hospitalized patients awaiting surgical operations showed
only a slight correlation between the objective seriousness of a persons upcoming
32
surgery and the level of fear experience (Janis, 1958). Studies also showed that
anxious and neurotic people report more stress than others (Brett et al, 1990).
Some people are more prone than others in perceiving being threatened by lifes
difficulties. When appraising the situation, aspects such as how predictable and
controllable a stressor is, whether is stable or unstable, global or specific, and
internal or external, affect how the individual will react to the stressor (Allred &
Smith, 1989). If the event is judged to be uncontrollable, it will be more stressful;
if it's more stable and global, people will react in a helpless manner; if it's more
internal, people will feel worse about themselves. These perceptions are
influenced by familiarity of the event, personal experience or reference, and
personality traits. Thus, stress lies in the eyes or, more precisely, the mind of the
beholder.

1.4.4.2 Coping
Another mediating process is coping. There are in general two main
groups of coping strategies: problem-focused coping and emotion-focused coping.
Problem-focused coping strategies involve confronting problem directly, are
usually task relevant and action oriented. Problem-focused coping tries to manage
and alter stressors and is more adaptive in exploring a constructive solution
(Allred & Smith, 1989; Carver, 1989). Emotion-focused coping refers to
regulating the emotional responses to stressors and is more useful in situations in
which the problem must be accepted. Some of these coping strategies include
self-controlling, distancing, positively reframing (exploring positive meaning in
an otherwise stressful experience such as by focusing on personal growth),
accepting responsibility, and escaping/avoiding (e.g overeating, drinking, using
33
drugs) (Allred & Smith, 1989; Carver, 1989). Same as the appraisal process,
people exhibit persistent patterns of coping strategies across stressful events, and
these can be considered as a dispositional construct.

1.4.4.3 Personality traits
Moderating factors, as well as mediating processes, influence individuals'
stress responses. The main moderating factor is the personality traits of an
individual. Since the 1950s, people have become much more interested in the idea
of personality and have witnessed an upsurge of interest in how personality affects
the stress process (Hjelle & Ziegler, 1992). Since the 1980s, conceptualization of
personality has evolved from controversial concepts with various measurement
difficulties; the consensus of psychologists gradually has come to adopt the Big
Five model as the best comprehensive system of basic, independent personality
factors (Briggs, 1992; McDonald, 1995).
Personality affects the stress process in various aspects. However, certain
personality traits relate to how an individual reacts to the stressors. Kobasa (1979)
studied the personality construct hardiness and compared high-stress executives
who exhibited the expected high incidence of illness against high-stress
executives who stayed healthy. She found that the hardier executives were more
committed, felt more in control, and had bigger appetites for challenge (Kobasa,
1984, p.70). Hardiness is composed of a set of three related personality traits:
commitment, control, and challenge (Cohen & Edwards, 1989; Florian et al.,
1995). Commitment refers to the tendency for people to confront and involve
themselves in what they encounter. Control refers to the belief in people that they
can influence their internal states and behaviour, influence their environment, and
34
bring about desired outcomes. Challenge refers to the willingness to face change
and try new activities, and allows opportunities for personal growth (Taylor &
Aspinwall, 1996; Powell, 1997). Hardiness is associated with stress resistance
(Kobasa, 1979, 1984; Kobasa & Puccetti, 1983), favorable appraisals of potential
stressors (Allred & Smith, 1989; Wiebe, 1991) and effective use of coping
strategies (William et al., 1992).
There are several other personality traits which are relevant to stress
resistance and reactions. Extroversion is associated with more enthusiasm and
energy, leading to eustress while neuroticism is associated with anxiety and
depression, leading to distress (Watson & Clark, 1984; Watson & Tellegen, 1985;
Bolger & Schilling, 1991). Optimism is associated with stress resistance and lack
of unfavourable stress responses, such as depression (Scheier & Carver, 1985;
Scheier & Carver, 1987; Scheier et al., 1994). An individual's self-esteem and
power motivation help determine how much stress the stressor will cause
(McClelland & Burnham, 1976; Scherwitz et al., 1978; Foder, 1984; DeLongis et
al, 1988). A study by Ng et al. (2004) has demonstrated that individuals of
different anxiety trait levels would give different severities of stress response
towards various formats of presentation of pre-operative procedural information
prior to oral surgery.
Friedman and Rosenman (1974) divided people into two basic types
Type A and Type B who exhibit different characteristics. The Type A personality
includes 3 elements, namely strong competitive orientation, impatience and time
urgency, and anger and hostility. These people are ambitious, hard-driving
perfectionists and extremely time-conscious. They routinely try to do several
things at a time. The Type B personality is marked by relatively relaxed, patient,
35
easygoing, amicable behaviour. These people are less competitive, less hurried,
and less readily angered than Type As. Type As appear physiologically
over-responsive to stress and the risk for Type As suffering coronary heart
problem is perhaps double that for Type Bs (Rosenman, 1993).

1.4.4.4 Dental anxiety
Dental anxiety is a significant health issue for many people. A remarkable
proportion of the population in the United States, United Kingdom, Netherlands,
Denmark, Norway, Sweden, Hong Kong and Canada reported certain degrees of
anxiety about dental visits and treatment (Green, 1985; Stouthard & Hoogstraten,
1990a; Hakeberg et al., 1992; Moore et al., 1993; Vassend, 1993; Skaret et al.,
1998; Schwarz & Birn, 1995; Maggirias & Locker, 2002; Smith & Heaton, 2003).
The general term dental anxiety might have diverse meanings (Edelmann, 1992).
Different meanings or definitions have been given in the dental literature,
covering a rather wide range of emotions from a relatively mild feeling of
apprehension, to extreme anxiety and to dental phobia (Lautch, 1971). Dental
anxiety in the present study is defined as a situation-specific trait anxiety and as
the disposition to experience anxiety in dental situations (Stouthard et al., 1993 &
1995).
Research in dental anxiety has explored its causes, prevalence and
consequences. Some have also investigated its impacts on the use of dental
services and oral health status. The general perception behind this concern is that
dental anxiety can lead to avoidance behaviours resulting in lack of regular dental
care and delay in seeking necessary treatment (Gale & Ayer, 1969; Schuurs et al.,
1980; Cohen, 1985; Woolgrove et al., 1987). It is generally assumed that
36
avoidance behavioural patterns of dental care and treatment have a detrimental
effect on dental health (Berggren & Meynert, 1984). While these propositions
appeared logical and intuitively appealing, most studies of the relationship
between dental anxiety and oral health status have used subjective rather than
clinical indicators of oral health. For instance, Locker & Liddell (1991) showed in
a random sample of 580 people aged between 50 and 89 year that dentally anxious
subjects were more likely than non-anxious ones in perceiving a need for dental
care, rating their oral health as poor and reporting problems in chewing. Similarly,
Milgrom et al. (1988) reported in a survey of dental fears, dental experiences and
perceived oral health status that dentally anxious people were less likely to be
satisfied with the appearance of their teeth, more likely to report dental problems
such as toothache or bleeding gum, and report a need for dental care. There are
relatively few data which have quantified the nature and magnitude of these
effects.
A few studies have attempted to examine the association between dental
anxiety and clinical indicators of oral health. For instances, two studies on
self-reported groups of fearful dental patients reported that their dental and
periodontal health were poorer than people of the same age. However,
non-anxious comparison groups were not included in these studies (Molin &
Seeman, 1970; Berggren & Meynert, 1984).
Shimura et al. (1983) investigated anxiety and personality of children
based on the assumption that emotional stress causes various negative effects on
physiological changes. A statistically significant relationship between dental
caries and anxiety was reported. This study explored, however, the general anxiety
of children rather than the more specific construct of dental anxiety.
37
Cohen (1985) reported a possible association between scores on Corahs
Dental Anxiety Scale (DAS) (Corah, 1969) and DMFS (Index of decayed, missing
and filled surfaces) status in a rather restricted population of naval recruits
consisting primarily of young men with relatively low rates of dental caries.
Several other studies compared dental anxiety in edentulous and dentate subjects.
Schuurs et al. (1985) and Locker et al. (1991) found that mean DAS scores were
higher in edentulous subjects, whereas Stouthard and Hoogstraten (1990b) found
no differences in dental anxiety measured by the 36-item Dental Anxiety
Inventory (Stouthard et al., 1993 & 1995).
Bedi et al. (1992) examined the clinical indicators (DMFT-index of
decayed, missing and filled teeth) of high self-reported dental anxiety in a group
of Scottish secondary schoolchildren (N = 1103). All DMFT components were
higher compared with their contemporaries but only the mean MT (missing teeth)
reached statistical significance after adjusting for gender and social class. These
dentally anxious children appeared more accurately in perceiving their treatment
need and were more likely to defer, cancel or not turn up for dental appointments.
Locker & Liddell (1992) examined differences between older adults who
were and were not dentally anxious. The data showed that dentally anxious
individuals were more likely to be edentulous. Among the dentate subjects,
dentally anxious individuals had more missing and fewer filled teeth. Dentally
anxious dentate subjects were more likely to need prosthodontic treatment,
immediate treatment for the relief of pain and infection and periodontal care. It
was further suggested that patterns of dental treatment were different between
those who were and were not dentally anxious.
A radiographic study (Hakeberg et al., 1993) of dental health in a group of
38
90 patients with severe dental fear showed that dental fear patients have a
deteriorated dental health compared to ordinary dental patients. The dental care
habit appeared to be different as well. The generalizability of this study however
was limited due to its small sample size and inclusion of only extreme cases
requiring specialist treatment for severe dental fear.
Besides showing in general compromised dental health resulting from
avoidance of dental care, studies reported significant yet small correlations of
dental anxiety with depression, state anxiety and trait anxiety (Ng et al., 2005).
Studies also suggested that dentally anxious subjects tended to experience more
psychological or social distress, and report stronger negative social consequences
(Abrahamsson et al., 2000 & 2002; Berggren et al., 2000; Locker, 2003). This
appears compatible with the general picture of co-morbidity of anxiety and
depressive disorders as well as the hypothesis of personality predisposition that
individuals who are fearful or anxious about dental treatment have a constitutional
vulnerability to anxiety or related disorders as evidenced by the presence of
multiple fears, generalized anxiety or depressive disorders (Stouthard et al., 1995;
Locker et al., 2001).
Periodontal disease is a major oral health problem (Holmgren et al., 1994;
Oliver et al, 1998) and, with its nature of chronicity, appears possible to be
affected by anxiety and avoidance behaviours. However, the impact of dental
anxiety on periodontal condition, being a major human oral health problem, has
not been reported.

1.4.5 Stress and periodontal disease: Human psychosocial studies
1.4.5.1 Review of studies
39
Table 1.3 summarizes some of the human studies exploring the
relationship between psychosocial factors and periodontal disease. A detailed
description of these studies are as follow.

40
Table 1.3 Summary of human studies regarding relationship between psychosocial factors and periodontal disease.


Stress process
evaluated
Conceptual and
methodological issues
Study/Country Sample (n) Major findings SS MF SR 1 2 3
Baker et al.
(1961)/USA
Psychiatric patients
(n=81)
Significant correlations between periodontal
status and such factors as age, broken home,
marital adjustment, hysteria scores and
somatization

* * * *
Belting & Gupta
(1961)/Sweden
Psychiatric patients
(n=141)
Severity of inflammatory periodontal disease
increased significantly as the degree of anxiety
increased
* * *
Davis & Jenkins
(1962)/USA
Psychiatric patients
(n=60)
Anxiety significantly correlated with
periodontal index
* * *
De Marco
(1976)/USA
Patients with severe
periodontal bone loss
(n=11)
Significant association between periodotnal
disease and emotional stress associated with
active duty in Vietnam
* * * *
Vogel et at.
(1977)/USA
Patient of dental
school clinic
(n=55)
Significant correlations between introversion
and inflammatory periodontal disease
* * *

41
Table 1.3 Continued.


Stress process
evaluated
Conceptual and
methodological issues
Study/Country Sample (n) Major findings SS MF SR 1 2 3
Ludenia & Donham
(1983)/USA
Dental out patients
(n=101)
No significant correlation between periodontal
disease and the personality disposition of
locus of control
* * *
Green et al.
(1986)/USA
Male patients of
hospital dental clinic
(n=50)
Significant correlation between life events
stress and periodontal status
* * *
Marcenes et at.
(1992 & 1993)/UK
Male workers
(n=164)
Significant association between poor
periodontal status and high work mental
demand and low marital quality;
Significant relationship between marital or
family problems with self-reported acute and
chronic oral symptoms
* * *
Minneman et al.
(1995)/USA
Military recruits
(n= 241)
Significant association between introversion
and gingival inflammation
* * *



42
Table 1.3 Continued.


Stress process
evaluated
Conceptual and
methodological issues
Study/Country Sample (n) Major findings SS MF SR 1 2 3
Monteiro da Silva et
al.
(1996)/UK
Dental patients
(n=150)
Significant correlation between rapid
progressive periodontal disease with
depression and loneliness
* * *
Linden et al.
(1996)/Northern
Ireland
Regular dental
attenders
(n=23)
Increase in loss of periodontal attachment
significantly associated with increasing age,
lower socio-economic status, lower job
satisfaction, and type A personality
* * * *
Axtelius et al.
(1997a; 1997b;
1998)/Sweden
Patient with
therapy-resistant
periodontitis
(n=22)

Non-responding patients showing more
psychosocial strain and more
passive-dependent personality
* * *
Croucher et al.
(1997)/UK
Dental patients
(n=100)
Periodontal disease significantly associated
with negative life-events
* *
Genco et al.
(1999)
a
/USA
General population
(n=1426)
Significant association between periodontitis
with financial strains, coping and depression
* * *

43
Table 1.3 Continued.


Stress process
evaluated
Conceptual and
methodological issues
Study/Country Sample (n) Major findings SS MF SR 1 2 3
Wimmer et al.
(2002)/Austria
Patients of chronic
periodontitis
(n=89)
Periodontitis patients with defensive coping at
greater risk for severe periodontal disease
* * *
Elter et al.
(2002)/USA
Dental patient with
depression
(n=697)
Significant association of clinical depression
with worse clinical periodontal treatment
outcome
* *
Hugoson et al.
(2002)/Sweden
General population
(n=298)
Significant association between periodontal
disease with loss of spouse and trait of
external locus of control
* * *
Pistorius et al.
(2002)/Germany
Patients of chronic
periodontitis
(n=120)
Significant association between periodontal
disease and life event stress
* * *
Aleksejuniene et al.
(2002)/Norway
General population
(n=571)
Significant correlation between levels of
remaining periodontal support and life style
* *


44
Table 1.3 Continued.


Stress process
evaluated
Conceptual and
methodological issues
Study/Country Sample (n) Major findings SS MF SR 1 2 3
Vettore et al.
(2003)/Brazil
Patients of chronic
periodontitis
(n=79)
Significant correlation between anxious trait
and clinical attachement loss
* * *
Teng et al.
(2003)/Taiwan
Patients of chronic
periodontitis
(n=250)
Significant association between mental illness
and periodontitis
* * * *
Torabi-Gaarden et al.
(2004)/Norway
General population
(n=96)
Statistically significant association not evident
between periodontal disease and negative life
events, anxiety, depression and coping ability
* * * *


Stress process evaluated: SS = stressors, MF = moderating factors, SR = stress responses.
Conceptual and methodological issues: 1 = failing to address stress as a process, 2 = problem in operationalization of stress components, 3 = sample
bias/small sample size.
a
Genco et al., (1999) addressed all the various aspects of the stress process; however, they did not attend to the moderating and mediating factors of
personality traits and dental anxiety.
45
1.4.5.1.1 Necrotizing ulcerative gingivitis (NUG)
Human studies have been conducted and yielded positive correlative
findings between psychosocial factors and acute necrotizing ulcerative gingivitis
or what is currently called necrotizing ulcerative gingivitis (NUG). Moulton et al.
in an early study back to 1952 reported that severe cases of NUG were preceded
by acute anxiety arising from a conflict on issues about dependency or sexual
needs. Severe chronic periodontitis cases were also documented, including one
with a background of longstanding conflicts around dependency needs, and one
with significant marital conflict and psychosomatic symptoms. Possibly because
of the very nature of NUG (acute painful onset, short-lived infection, ease of
diagnosis, and multiple predisposing factors), it is perhaps the single most studied
periodontal disorder in relation to psychosocial predisposing factors (Pindborg,
1951; Grupe & Wilder, 1956; Dahlstrom & Welsh, 1960; Giddon et al., 1963 &
1964; Goldhaber & Giddon, 1964; Shannon et al., 1969; Formicola et at., 1970;
Maupin & Bell, 1975; Comstock & Helsing, 1976; Shields, 1977; Goldberg, 1978;
Cohen-Cole et at., 1983; Cogon et al., 1983; Johnson & Engel, 1986; Schoor &
Havrilla, 1986; Melnick et al., 1988). Horning & Cohen (1995) evaluated a total
of 68 consecutive NUG patients over a period of 5 years. Ten of these 68 patients
were HIV (Human Immunodeficiency Virus)-positive. Besides HIV infection, it
was found that unusual life stress, inadequate sleep, and recent illness were also
significant predisposing factors to this form of periodontal disease. All these
findings provided support for the hypothesis that stress is a predisposing factor for
NUG, and that the relationship between psychosocial factors and NUG may be
mediated by endocrine and immune changes.

46
1.4.5.1.2 Psychiatric patients
Besides the studies on NUG, the relationship between psychological stress,
distress, coping behaviours, personality, and periodontal diseases has also
attracted much attention in the past few decades (Gupta, 1968; Machtei et al, 1992;
Freeman & Goss, 1993; Monteiro da Silva, 1995; Salvi et al., 1997; Hildebrand et
al., 2000). Baker et al. (1961) investigated the relationship between psychosocial
factors and periodontal disease by assessment of psychiatric patients and normal
subjects (N=81). Significant correlations were observed between periodontal
status and such factors as age, broken home, marital adjustment, hysteria scores
and somatization (defined in this study as the tendency to develop psychogenic
physical complaints or psychosomatic disorders).
Belting and Gupta (1961) studied the influence of psychiatric disturbances
on severity of periodontal disease (N=141). It was reported that psychiatric
patients presented significantly higher periodontal scores than their controls when
brushing frequency, calculus, bruxism and clenching were held constant. In the
experimental group the severity of inflammatory periodontal disease increased
significantly as the degree of anxiety increased. The authors concluded that the
periodontal changes in the psychiatric patients were mediated through one or
more processes associated with anxiety and under the control of the autonomic
nervous system.
Davis and Jenkins (1962), without employing controls, investigated
possible associations in psychiatric patients (N=60) between periodontal disease
and what they called psychological measures of stress. These measures were
actually assessing the affective signs and symptoms of anxiety and depression.
They found that anxiety was significantly correlated with periodontal index. The
47
authors postulated that anxiety altered concentrations of adrenal corticoids and
other hormones. Thus the level of inflammatory periodontal disease is presumably
related to levels of circulating corticosteroids.
Elter et al. (2002) reviewed the change of periodontal status for 3 years
(January 1996 to December 1998) in 697 patients with a clinical diagnosis of
depression in accordance with Diagnostic and Statistical Manual of Mental
Disorder (4
th
Ediction) (American Psychiatric Association, 1994) who were
receiving continuous dental care. Depression has a strong negative impact on
individuals physical and psychological well-being, leading to withdrawal from
daily activities and compromised working performance (Broadhead et al., 1990;
Rost et al., 1992; Weiten, 2004). This study demonstrated the association of
clinical depression with worse clinical periodontal treatment outcome. This study
however did not elucidate mechanisms for these worse treatment outcomes; the
authors speculated that depressed patients may have inadequate oral hygiene
practices and other associated behaviours, or the immunological process may be
compromised resulting in worse periodontal health.

1.4.5.1.3 Dental patients
De Marco (1976) coined the term periodontal emotional stress syndrome
(PESS) for periodontal disease due to significant emotional stress. The author
evaluated eleven cases, men aged 22 to 32, with severe periodontal bone loss
especially in the posterior segments unexplained by any other local aetiological
factors. The only factor in common in all these cases was severe emotional stress
associated with active duty in Vietnam.
Vogel et at. (1977) evaluated possible relationships between personality
48
traits and periodontal disease in fifty subjects registering for treatment at a dental
school clinic. Significant correlations were found between introversion and plaque,
and between introversion and inflammatory periodontal disease as measured both
clinically and radiographically. There was also a significant correlation between
neuroticism and radiographic measures of inflammatory periodontal disease.
Ludenia and Donham (1983) studied the relationship between periodontal
disease and the personality disposition of locus of control (n = 101). Locus of
control refers to generalized expectancy about the degree to which individuals
perceive that they are in control of the outcomes of the events in their lives (Rotter
1966, 1975, 1990; Lefcourt 1976;). Individuals with an external locus of control
believe that their successes and failures are governed by external factors such as
fate, luck, and chance. In contrast, individuals with an internal locus of control
believe that their successes and failures are determined by their actions and
abilities (internal, or personal, factors). The Multidimensional Health Locus of
Control Scale (Wallston & Wallston, 1978) was used in the study to examine the
relationship between health locus of control and the following variables: age, de-
pression, trait anger, trait anxiety, and dental ratings of oral hygiene, and in-
flammatory periodontal disease. This psychological trait variable refers to an in-
herited or acquired characteristic which is relatively consistent, persistent and
stable (Wolman 1989), while the state variable is conceptualized as transitory
state or condition which may vary in intensity and fluctuate over time. Trait anger
and trait anxiety were measured by trait subscales of the State-Trait Personality
Inventory (Spielberger et al. 1979). The Beck depression inventory (Beck 1967)
was used to measure depression. They found that trait anxiety, depression and trait
anger were correlated negatively and significantly with health internality. Trait
49
anxiety was related positively and significantly to both health externality and
powerful others externality (that is, the belief that powerful others are to control,
e.g., health care professionals). Contrary to the authors prediction, however, oral
hygiene status and degree of inflammatory periodontal disease were not correlated
to either health internality or health externality.
Besides the earlier studies which focused on the relationship between
stressful life situations and NUG scores, the first study involving the systematic
evaluation of life events stress with self-reported measures and periodontal dis-
ease generally (gingivitis and periodontitis) in humans appears to have been
reported by Green et al. (1986). Gingival and periodontal pathology, stressful life
events and somatic symptoms were investigated in 50 male subjects, aged 23 to
74, attending a veteran hospital dental clinic. Somatic symptoms included distress
arising from perceptions of bodily dysfunction, e.g., headaches, faintness and
pains in the heart or chest. A significant correlation was evident between life
events stress and periodontal status. For individuals who scored high on somatic
symptoms, an especially conspicuous relationship between life stress and
periodontal disease was detected.
Monteiro da Silva et al. (1996) investigated the possible associations
between a number of relevant psychosocial factors and the then named rapidly
progressive periodontitis (RPP). One hundred and fifty patients were equally
divided into 3 groups (n = 50, in each group) - RPP, chronic periodontitis (CP),
and without significant periodontal destruction (controls). The RPP group
presented significantly increased depression and loneliness compared to the CP
and control groups.
Linden et al. (1996) examined the association between occupational stress
50
and the progression of periodontitis in adult workers. A total of 23 regular dental
attenders, with a mean age of 41.1 years, enrolled in a longitudinal study on
periodontal disease. They were examined on 2 occasions at an interval of 5.5
years on average. Clinical measurements of periodontal status were made; and, an
occupational stress indicator questionnaire was used to assess stress. In the final
regression model, an increase in loss of periodontal attachment was significantly
associated with increasing age, lower socio-economic status, lower job
satisfaction, and type A personality. The personality construct of locus of control
was a significant predictor variable and explained 65% of the variance in the loss
of periodontal attachment.
Axtelius et al. (1997a, 1997b & 1998) attempted to explore, from the
perspective of a stress system disorder, the pathogenesis of therapy-resistant
periodontitis for indications that the stress-behaviour-immune system model holds
as an explanatory model for the understanding of periodontal disease. From
retrospective comparisons of probing pocket depth (PPD) charts, 22 patients were
classified as either non-responding (NR-group, n = 11) or responding (R-group, n
= 11) to periodontal treatment according to the profile of PPD reduction over time
in response to periodontal treatment. Somatic and psychological factors were
described as obtained by interviews and psychological testing. The results of
logistic regression analysis indicated that the NR-group patients displayed more
psychosocial strain and a more passive-dependent personality. The R-group
patients displayed a more rigid personality and possibly a less stressful
psychosocial situation in the past. The result supported the possible contribution
of stress factors in the context of therapy for resistant periodontal disease.
Croucher et al. (1997), in a case-control study (n = 100; dental patients
51
matched for age and sex), investigated the role of life-events in periodontitis.
Logistic regression analysis showed that periodontal disease was associated with
the negative impact of life-events, the number of negative life-events, high levels
of dental plaque, tobacco smoking and being unemployed. Marital status became
statistically significant after adjusting for the other variables (P < 0.05). A model
was suggested that life events may affect periodontal health. It was concluded that
psychosocial factors and oral health risk behaviours may cluster together and
serve as important determinants of periodontal disease.
Wimmer et al. (2002) studied the influence of stress coping in periodontal
disease. Eighty-nine patients with different forms of chronic periodontitis were
included in this retrospective case-control study. The control group consisted of
63 persons employed in health care services. All subjects completed a 114-item
stress coping questionnaire and 19 actional and intrapsychic stress coping modes
for assessment of stress coping strategies. Clinical attachment loss was examined.
Five factors of coping strategies were extracted in factor analysis with Varimax
rotation. Significant differences were found between the study and control groups
in active coping, distractive coping, defensive coping, coping through aggression,
and coping with the use of pharmaceutical drugs. In consideration of the severity
of periodontal disease, the patients were divided into 2 groups - mild to moderate
and severe. Statistical analysis (t-test) showed significance for defensive coping in
that patients with a defensive coping style had greater attachment loss. The data
suggested that periodontitis patients with inadequate stress management
behaviours (defensive coping) were at greater risk of severe periodontal disease.
Pistorius et al. (2002) studied stress factors comparing 120 patients with
chronic periodontitis against a control group of 122 patients matched for age and
52
gender. The results suggested that life event stress may exert an unfavourable
effect on the course of periodontal disease. It should be noted that only loss of a
significant other in the evaluation of private life and family was found to be
statistically significantly different between the study and control groups. Even
though the questions on personal information and occupation/retirement were
found to be significantly different, the validity of the assessment tools used
remained inconclusive. A study with statistical method of path analysis
(Aleksejuniene et al., 2002) in a sample of 571 subjects in Norway supported the
link between levels of remaining periodontal support and life style, yet failed to
confirm the path between psychosocial stress and periodontal status.
Vettore et al. (2003) investigated the relationship of stress and anxiety with
periodontal characteristics. The study detected, after adjusting for smoking and
socioeconomic data, only a significant correlation between subjects with high
anxious trait and moderate clinical attachment loss or probing pocket-depth
(4-6mm). Other measures of stress and anxiety with the Stress Symptoms
Inventory (Lipp & Guevara, 1994) and Social Readjustment Rating Scale
(Holmes & Rahe, 1967) failed to reveal any significant association with
periodontal status. The study sample, however, was small, with 79 subjects
divided into three groups in accordance with their levels of probing pocket
depth and clinical attachment loss.
A study was conducted by Teng et al. in 2003 to explore the lifestyle and
psychosocial factors in patients with chronic periodontitis and to estimate the odds
due to these factors. A case-control study of 250 cases of chronic periodontitis
patients and 250 age and sex matched controls. Structured questionnaires were
used to collect lifestyle and psychosocial data. Multivariate logistic regression
53
showed that mental illness (OR: 5.32, if Chinese Health Questionnaire scores 6),
and smoking (OR: 3.93, if pack years smoked > 21) were significantly associated
with chronic periodontitis. These findings however appeared rather non-specific
in the assessment of psychosocial factors which, in this study, was measured by
the 12-item Chinese Health Questionnaire (CHQ-12). This instrument is primarily
a screening instrument originally developed by Goldberg in 1972 for detection of
psychiatric disorders in community settings and non-psychiatric clinical settings,
such as primary care or general practice. The original version is known as General
Health Questionnaire (GHQ). The GHQ-12 is the shortest version and gives only
a uni-dimensional score referring to the degree of distress. It is usually used in
assessment of changes in psychological distress with time by repeated testing and
there is good evidences that scores rise and fall as clinical psychological status
changes (Goldberg & William, 1988).

1.4.5.1.4 Community population groups
Marcenes & Sheiham (1992) carried out a correlational study in Belo
Horizonte, Brazil to evaluate whether oral health status is associated with work
stress. Clinical examination recorded the number of decayed teeth, missing teeth
and filled tooth surfaces, periodontal pockets and presence of gingival bleeding on
probing. Questionnaires were used to measure psychosocial factors (work demand,
work variety, work control, and marital quality) and risk-related behaviours
including frequency of dental visits, toothbrushing frequency, sugar consumption
and type of toothpaste used. A total of 164 male workers aged from 35 to 44 years,
equally distributed over four socio-economic groups were recruited in the study.
Age, socio-economic status, the behavioural data, years of residence in Belo
54
Horizonte and marital quality were considered in the data analysis. The results
showed a significant association between poor periodontal status and high
work-related mental demand and low marital quality. In addition, the relationship
between work-related mental demand and periodontal status was independent of
risk-related behaviours. The authors postulated that psychosocial factors may
affect periodontal tissues including alterations in saliva flow and suppression of
immune system.
Marcenes et at. (1993) investigated possible relationships between eight
specific negative life events and self-reported oral symptoms of toothache or
trouble with the gums. Marital or family problems were significantly associated
with self-reported acute and chronic oral symptoms, after adjustment for other
variables. Similarly, occupational stress and type-A personality were suggested to
be associated with increase pocket depth and as predictors of periodontal disease
(Freeman & Goss, 1993).
Minneman et al. (1995) conducted a study to examine the relationship of
personality traits and stress with gingival inflammation and with soft-tissue oral
pathology in a sample of military recruits (N=241). Eysenck's Personality
Questionnaire was used to measure personality traits of psychoticism,
extroversion-introversion, and neuroticism. A modified version of the
Organizational and Individual Assessment Survey developed by Hendrix (1985)
was employed in assessment of stress. Significant correlations were detected
between personality traits and various measures of stress tolerance. Physical stress
was found to affect soft-tissue pathology, while gingival inflammation correlated
significantly with extroversion-introversion scores, tolerance to change, and
anxiety. This finding supported a possible relationship among certain personality
55
traits or stress variables, and gingival inflammation or soft-tissue pathology in
recruits with extreme personality characteristics and perception of high physical
stress levels in basic combat training.
Genco et al. (1996) studied 1,426 subjects between the ages of 25 and 74
years in Erie County, New York found that financial strain, depression are
significant risk indicators for more severe periodontal disease after adjusting for
age, gender, smoking. High level of problem-based coping was also found to
reduce the stress-associated risk.
Hugoson et al. (2002) studied the prevalence of some negative events and
psychological factors and their relation to periodontal disease in 298 subjects in
Sweden. The results revealed, in addition to the well-documented risk factors of
age and smoking and oral hygiene, the loss of a spouse and the personality trait of
external locus of control were associated with severe periodontal disease.
Torabi-Gaarden et al. (2004) studied in a Norwegian adult population the
relationships between periodontal disease and negative life events, anxiety,
depression and coping ability. The results failed to reveal any statistically
significant association when smoking was introduced in the analysis within a
multiple logistic regression model. These negative findings may be related to the
rather small sample size of 96 subjects (42 as study and 54 as controls), the use of
The Hospital Anxiety Depression Scale which has been developed primarily for
use in a clinical setting (Zigmond & Snaith, 1983), as well as the inclusion of only
the psychological construct of neuroticism as personality measure.

1.4.5.2 Conceptual and methodological problems in human studies
Going back to research studies concerning stress and periodontal disease,
56
while there seems a wealth of evidence linking stress and the disease, failing to
appreciate stress as a dynamic and interactive process based on a contemporary
understanding of stress process with precise operationalization of various stress
components remains the common flaw in many of them. Some of the studies
attempted to evaluate only some of the components in the stress process without
taking into consideration the interaction with other components. The relative
contribution of various stress components has not been clearly understood. For
example Ludenia & Donham (1983) focused on locus of control and traits which
were the only moderating factors in the stress process, without attending to the
stressors or stress reactions in the stress process. Marcenes & Sheiham (1992)
evaluated only the role of stressors (within work and marital relation) contributing
to periodontal disease. A similar conceptual issue appeared also in many studies
including Marcenes et al. (1993), Minneman et al. (1995), Croucher et al. (1997),
Wimmer et al. (2002), Hugoson et al. (2002), Vettore et al. (2003), and
Torabi-Baarden et al. (2004).
Another common problem has been that stress, stressors, stress response or
stress mediating factors are used interchangeably in many studies. For example,
Davis & Jenkins (1962), investigated possible associations in psychiatric patients
between periodontal disease and what they called psychological measures of
stress. These measures were actually assessing the affective signs and symptoms
of anxiety and depression. A study by Teng et al. (2003) also displayed the
problem in operationalization of the stress components. The scores in the GHQ-12
which, in the broadest sense, measures only a general state of distress, was stated
as an indicator of mental illness in coming up with the conclusion of a positive
association between psychosocial stress and periodontitis. Other examples
57
included Baker et al. (1961), De Marco (1976), and Vettore et al. (2003).
Sampling bias appears to be another common flaw in the studies of stress
and periodontal disease. The strategy of observing a limited sample in order to
generalize about a much larger population rests on the assumption that the sample
is reasonably representative of the population. Studies carried out in clinical
populations, for example Belting & Gupta (1961) with psychiatric patients, and
Elter et al. (2002) with clinical depression patients, apparently limited the
generalizability of their findings and rendered the conclusion speculative. The
methodological issue of small sample size also existed in some of the studies. The
study by Axtelius et al. (1997a; 1997b; 1998) represented an extreme case with a
sample size of 11 (another 11 as control) to evaluate the relationship between
periodontal disease and a series of somatic and psychological variables. Other
examples of sampling issues include Vogel et al. (1977), Monteiro da Silva et al.
(1996), Linden et al. (1996), and Torabi-Gaarden et al. (2004).
A review of the literature highlighted several conceptual and
methodological problems which might have confounded the findings of studies in
this area. A study of adequate sample size representative of the general population
with inclusion of measurements of the various components of stress process
appears a rational systematic approach to explore the relation between stress and
periodontal disease.

1.4.5.3 Erie County Study
The Erie County Risk Factor Study (Grossi et al., 1994 & 1995; Moss et
al., 1996; Genco et al., 1998 & 1999) was one of the most extensive and
systematic studies of the relationship of stress, distress, and coping behavirours
58
with periodontal disease conducted in the early 1990s.
A cross-sectional study of 1,426 subjects between the ages of 25 and 74
years, mostly metropolitan dwellers, residing in Erie County, New York, and
surrounding areas, was carried out to assess these relationships. Subjects were
requested to complete a set of 5 psychosocial questionnaires which measured life
events and their impact, chronic stress or daily strains, distress, coping styles and
strategies, and hassles and uplifts. Clinical examinations of supragingival plaque,
gingival bleeding, subgingival calculus, probing depth, clinical attachment level
(CAL) and radiographic alveolar crestal height (ACH) were performed, and 8
putative bacterial pathogens from the subgingival flora measured. It was found
that stress associated with financial strain and distress manifest as depression, are
significant risk indicators for more severe periodontal disease in adults in an
age-adjusted model in which gender (male), smoking, diabetes mellitus,
Tannerella forsythia (previously known as Bacteroides forsythus), and
Porphyromonas gingivalis are also significant risk indicators (Grossi et al., 1994
& 1995).
Logistic regression analysis indicated that, of all the daily strains
investigated, only financial strain was significantly associated with greater
attachment loss and alveolar bone loss (odds ratio, OR = 1.70, 95% CI = 1.09 to
2.65 and OR = 1.68, 95% CI = 1.20 to 2.37, respectively) adjusted for age, gender,
and cigarette smoking. For those with more financial strain, high emotion-focused
copers (a form of inadequate coping) had a higher risk of having more severe
attachment loss (OR = 2.24, 95% CI = 1.15 to 4.38) and alveolar bone loss (OR =
1.91, 95% CI = 1.15 to 3.17) than those with low levels of financial strain, after
adjustment for age, gender, and cigarette smoking. Similar results were found
59
among the low problem-focused copers for alveolar bone loss (OR = 2.21, 95%
CI = 1.11 to 4.38) and ACH (OR = 2.12, 95% CI = 1.28 to 3.51). Subjects with
high levels of financial strain reporting high levels of problem-focused coping
(considered adequate or good coping), however, had no more periodontal disease
than those with low levels of financial strain. It was further suggested that the
effects of stress on periodontal disease can be moderated by adequate coping
behaviours (Genco et al., 1998 & 1999).
Subjects were also selected from among the 1,426 participants in the Erie
County Risk Factor Study for an exploratory case-control study (Moss et al.,
1996). The association between social factors and adult periodontal disease was
assessed by comparing self-reported information for daily strains and symptoms
of depression in 71 case subjects and 77 control subjects. Baseline data were
collected and the subjects were examined again after 1 year. It was found that
baseline smoking status and Anti-T. forsythia IgG titre were associated with more
extensive disease in subjects scoring high on depression at baseline, with 8.1% or
more of the periodontal sites showing further breakdown,. This exploratory
analysis has served to establish specific lines of inquiry concerning psychosocial
factors as important environmental variables in adult periodontal disease.
The Erie County Study gave a far more systematic approach than previous
research studying stress and periodontal disease. The study attempted to tease out
the different roles of stressor, stress response and mediating factor. Personality
traits which have been suggested to have significant contribution towards stress
process, however, were not included in the study. The role of trait dispositions,
including the ones associated with negative affectivities of stress, were not
evaluated. In measurement of stress response, the psychological constructs of
60
anxiety and depression could also be measured with more specific instruments. At
the same time, dental anxiety and quality of life, being the apparent psychosocial
correlates of periodontal disease were not included or investigated.

1.4.6 Concluding remarks
Mechanisms have been proposed that psychological or behavioural factors
could be putative mediators for plaque induced inflammatory periodontal diseases
(Corby, 1947; Gilbert, 1947; Miller & Firestone, 1947; Moulton et al., 1952;
Briken, 1953; Zaidens, 1954; Moulton, 1955; Davis & Jenkins, 1962; OLeary et
al., 1962; Rubin, 1963; Gupta, 1966; Ringsdorf & Cheraskin, 1969; Manhold,
1970; Regenbaum, 1970; Manhold et al., 1971; Krasner, 1978; Clarke et al.,
1981; Meyer, 1989; Merchant et al., 2003). The main effects occur primarily
through behavioural changes which affect at-risk health behaviours (Figure 1.3),
such as neglect of oral hygiene, changes in diet, increase in smoking and other
pathogenic oral behaviours, and bruxism. For example, one may have oral
hygiene care practice decreased during the stressful period, and at-risk oral health
behaviour such as bruxism, smoking, drinking, use of drug, and adverse dietary
practice may be increased at the same time (McGlynn et al., 1990). Attendance for
maintenance therapy after periodontal treatment was shown to be significantly
influenced by personality factors and life events (Becker et al., 1988). There are
certainly many other possible behaviours that could be affected by stress and
inadequate coping and distress, such as depression, which would have significant
effects on periodontal disease (Genco et al., 1999).
On that basis, lack of adequate coping skills to stress, in addition to
compromised or maladaptive dental care habits, such as reduced oral hygiene or
61
teeth grinding, could also lead to undesirable biochemical changes in salivary
secretion and weakening of the body's immune functions. Genco (1999) reported
that high levels of financial stress with poor coping abilities increase about
twofolds the likelihood of developing periodontal disease compared with those
with low levels of financial strain, as observed with higher levels of attachment
loss and alveolar bone loss, after accounting for other risk factors such as age,
gender, smoking, poor dental care and diabetes.
62




Figure 1.3 A model of impact of psychological and behavioural
responses of stress on periodontal disease. (Modified from Genco et al., 1998)

Depressed
Immunity
+
+
+
+
+
+
+
+
+
Behavioral Change
Poor Oral Hygiene,
Poor Compliance
Overeating
(High-fat diet)
Bacterial Infection Cortisol
Smoking & other
at-risk behaviours
(e.g. drinking, use of
drug, bruxism)
Periodontal disease
Psychosocial Stress
63
1.5 Oral health-related quality of life
Dentists have been trained to recognize and treat oral diseases such as
caries, periodontal disease and malocclusions. Various parameters and indices
have been used to describe the prevalence of these diseases in the population. The
severity of periodontal disease is usually documented by researchers/clinicians
using clinical parameters such as bleeding on probing (BOP), probing pocket
depth (PPD) and clinical attachment level (CAL). However, the symptoms of
periodontal disease experienced by the patients include the consequences of
chronic inflammation and destruction of tooth supporting tissues, such as redness,
bleeding on brushing, loosening of affected teeth, progressive loss of fit or
difficulties wearing removable partial prostheses, and persistent bad breath; and
their impact on individuals daily functioning and psychological well-being were
not normally documented in a research report. Such symptoms, however, are
highly relevant from the patients point of views and often have a considerable
impact on the patients daily living and quality of life (Locker, 1988). This area
deserves further exploration.
Locker (1995) found that there was only a weak association between
scores on the Oral Health Impact Profile (OHIP) and clinical indicators of oral
disease. He also suggested that health problems may affect quality of life but such
a consequence is not inevitable (Locker, 1997; Locker, 2004). The implication of
this is that poor health or presence of disease does not necessarily mean poor
quality of life. Allison et al. (1997) further explained that quality of life was a
"dynamic construct", and very likely subjects to change over time. Individual
beliefs, values, attitudes and perception are not constant, and vary with time and
experience; and, they are modified by psychological and behavioural processes
64
such as coping, expectancy and adaptation. An example is that individuals with
eating problems due to common oral diseases that caused pain and discomfort
would have rated this problem as extremely important at one point in time. Should
the same problem, however, be diagnosed as oral cancer, and treated with
radiotherapy and/or surgery, the same individual may perceive the original oral
discomfort as relatively benign and relatively not so important.
There has been considerable debate on the use of traditional outcome
indicators in periodontal therapy. Hujoel et al. (1997) commented that these are no
more than just surrogate markers. Such indications are also therapist-centered.
Studies have recently begun to explore in a boarder perspective the relationship
between various satisfaction factors and periodontal treatment, that is,
patient-centered outcomes (Whitehead & Watts, 1987; Kalkwarf et al., 1992;
Mathews & McCullock, 1993; Fardal et al., 2002; Lee et al., 2002). This emphasis
on quality of life is consistent with the concept that health is a resource and not
simply the absence of disease (US Department of Health and Human Services,
2000). Interest in the idea of quality of life is growing rapidly. More than 1,000
new articles are indexed each year under this heading (Muldoon et al., 1998).
Quality of life is increasingly acknowledged as a valid and appropriate
indicator of service need and intervention outcomes in public health research and
services. Measures of health-related quality of life, along with the objective and
subjective assessments of health, are especially useful for evaluating efforts in
prevention and treatment of chronic diseases. All these information can give rise
to significant implications on areas from health policy planning to provision of
service to patients (Hennessy et al., 1994).
The psycho-social consequences of oral conditions have received little
65
attention. This may be due to the fact that they are rarely life threatening.
Furthermore, the oral cavity used to be dissociated from the rest of the body when
considering general health status (Allen, 2003). Recent research has served some
evidence that oral disorders may give rise to appreciable emotional and
psycho-social consequences. Reisine (1984) and Gift et al. (1989) have indicated
that approximately 160 million work hours are lost in a year in USA due to oral
disorders. Cushing et al. (1986) found that pain, difficulty with eating, and
communication problems were frequently reported in a study involving employed
adults in the United Kingdom. Reisine et al. (1989) compared quality of life
scores of patients with temporomandibular joint disorders (TMD) against a group
of patients with cardiac disorders. It was found that TMD patients were disabled
to a greater extent in the areas of sleep and rest, social interaction, intellectual
functioning and communication.
A few studies on oral diseases with quality of life as outcome measures
have been reported in the last few years. Cunningham and Hunt (2001) suggested
that a comprehensive exploration and understanding of the effects of orthodontic
treatment on health-related quality of life is essential to ensure that improvements
in aesthetics and subsequently psychosocial well-being are derived from treatment.
McGrath et al. (2003) evaluated patients perceptions of changes in oral
health-related quality of life (OHQoL) in one hundred patients over a period of six
months after third molar surgery. Two specific OHQoL measures, the 14-item
Oral Health Impact Profile (OHIP-14S) and the 16-item UK Oral Health-Related
Quality of Life measure (OHQoL-UK), were used to measure changes in life
quality. A significant deterioration in OHQoL in the immediate postoperative
period was followed by an improvement in OHQoL compared with preoperative
66
status. Patients perceived physical, social, and psychologic improvements in life
quality after surgery. The study concluded that third molar surgery is associated
with favourable changes in OHQoL from patients' perspectives. There is also
increasing agreement among dentists that patients perceptions should be included
in the decision-making process to provide a more comprehensive evaluation of the
value and effectiveness of third molar surgery (McGrath et al., 2003).
Needleman et al. (2004) explored the impact of oral health on quality of
life in a group of periodontal patients referred for specialist periodontal therapy
and reported that periodontal status affects life quality. Little has been reported,
however, about the impact on oral health related quality of life associated with
periodontal health or disease in the general population.
A better understanding of the consequences of periodontal disease and its
treatment on patients perceptions of how their oral health affects their daily lives
can help to ensure that the planning and evaluation of periodontal care and
treatment adequately addresses patients needs and concerns (McGrath & Bedi,
1999; Allen, 2003). The use of patient-centered measures in dentistry is increasing.
A number of instruments have emerged with promising psychometric properties
(Slade 1997a; Allen, 2003), such as Oral Health Impact Profile (OHIP) (Slade &
Spencer, 1994), Subjective Oral Health Status Indicators (Locker & Miller, 1994),
UK Oral Health-Related Quality of Life Instrument (OHQoL-UK) (McGrath &
Bedi, 2000).
Besides, a cross-sectional study of a random sample of 300 residents in
Britain revealed that dental anxiety is associated with the impact of oral health on
quality of life (McGrath & Bedi, 2004). Those giving high score in the
psychological construct of dental anxiety are among those with poorest oral
67
health-related quality of life.

1.6 Justification for the study and statement of the problem
Stress has been related to development of various psychological disorders,
such as depression, anxiety disorder, eating disorders, and schizophrenia.
Similarly, the assertion that stress can contribute to physical disease is nothing
new. Research data on stress and physical illness began to accumulate back in the
1930s and 1940s. The concept of psychosomatic disease has widely been accepted
since 1950s; that refers to actual physical disorders with genuine organic basis
caused in part by psychological factors, especially emotional distress (Schultz,
1990; Weiten, 2004). Common psychosomatic diseases (Table 1.4) include
hypertension, asthma, peptic ulcers, skin allergy, tension headache and migraine
(Kaplan, 1989).
Health psychology is a subject focused on how psychological factors relate
to the promotion and maintenance of health, and relate with the causation,
prevention, and treatment of illness. The growing recognition of psychological
factors provides an even more complex illustration of multifactorial etiological
factors in human diseases. Studies thus far point to an association of human
diseases with psychosocial stress and distress.
Contemporary conceptualization of the stress process supports the
evaluation of stress at three levels: stressors, moderating factors, and stress
reactions. Stress has to be evaluated as a dynamic and interactional process of
intricate systems with valid formulations and operationalization of the various
components at these three levels; it would be even more complex if we take into
account of the spontaneous environmental factors (Lazarus, 1999). Affective
68
responses (emotions) would be determined primarily by the appraisal process that
makes personalized perceptions of a stressor or threat, which in turn is influenced
significantly by factors including personality trait, experience and reference
information. These responses are further affected by the existing affective and
physical condition of an individual at that particular moment (Lazarus & Lazarus,
1994). The factor of personality trait is generally considered as a major
moderating factor. Physiological response including autonomic arousal, hormonal
fluctuations, and neurochemical changes so aroused would interact with emotional
response. Behavioural response in coping with the stressor such as lashing out on
others or seeking help may somehow modulate the emotions and physiological
status, and lead to different reciprocal response from the outside world, making it
more stressful or less. This spontaneously affects the impact of the stressor, and
subsequent appraisal, coping, emotional and physiological responses. This
spontaneity and automaticity adds further to the complexity of the array of static
as well as dynamic variables that are involved in the evaluation of the stress
process.
The relationships between certain systemic diseases (e.g. heart disease,
gastric ulers) and stress, personality or coping strategies have been studied at
various scales over many years; the findings were splendid. Similar studies on the
effects of these psychosocial factors on periodontal disease are limited, especially
in Chinese populations. The earlier studies were predominantly focused on the
relationship between stressful life situations and necrotizing periodontal diseases
(Monteiro de Silva et al., 1995). Most, if not all, of the studies have been
conducted to investigate the relationship separately between periodontal disease
and particular stressors (e.g. life events, marital problems), between that and
69
moderating factors (e.g. coping strategies, trait), or between that and stress
responses (e.g. emotions, immunologic contingency). The study in Erie County
(Genco et al., 1999) appeared as recent and rational attempt to examine the
relationships, on a population scale, between periodontal disease and psychosocial
stress with the inclusion of some of the major components of different levels of
the stress process. The study, however, did not investigate the relationship
between periodontal disease and the other major stress response moderating
factors such as personality traits/dispositions, nor the specific relevant disposition
of dental anxiety. Personality traits/dispositions have been considered to be
important factors regarding stress response moderation (Lazarus, 1999).
Studies reviewed in previous sections supported primarily the existence of
a positive correlation between psychological stress and periodontal disease. Many
of these studies however attempted to investigate and evaluate individual
psychological variables in the stress process. Limited studies have addressed the
issue of personality trait dispositions. Furthermore the sample size in some of
these studies was limited and rendered the results of investigations inconclusive
when making generalization conclusion. That may also account for some of the
inconsistencies in the findings. Thus it would be desirable to select and include
parameters across the stress process based on the contemporary understanding of
stress. A study of the relationship between stress and periodontal disease in a large
population is thus indicated so as to allow for inclusion of concurrent evaluation
of the variance due to the factors at different levels of the stress process, and at the
same time for analysis of possible confounding factors (such as age, gender,
education, smoking, systemic diseases) which have been suggested as significant
risk indicators for periodontal disease (Clarke & Hirsch, 1995). The study sample
70
size should be big enough to reflect broad variation in periodontal disease severity
and in potential risk indicators for assessing the relationship between explanatory
and outcome variables (Genco et al., 1988 & 1999). The impact of dental anxiety,
being a specific personality trait about ones propensity of developing anxiety
towards dental care, on periodontal condition in a general population has never
been reported.
In Dentistry as well as in other areas of Medicine, it has been recognized
that objective measures of disease provide little insight into the impact of
disorders on daily living and quality of life. There is a substantial body of
knowledge regarding patterns of dental disease in both adult and child populations,
there is yet little information concerning the impacts of oral diseases for
well-being and the quality of life.
Studies assessing the association between objective measures of dental
disease (such as presence of dental caries or periodontal attachment loss) and
patient based opinions of oral status reported only a weak association and
suggested that these objective measures failed to reflect accurately patients'
perceptions (Gooch et al., 1989; Locker, 1992; Locker & Slade, 1994).
Limitations of the traditional "biomedical" paradigm of health have been
recognised principally in that this model only deals with disease. Consequently,
any measure of health should address the social and emotional aspects of health in
addition to the mere presence or absence of disease.
A number of studies have attempted to investigate the contributory role of
stress in periodontal disease (Monteiro da Silva, 1996; Genco 1996 & 1999).
However, study of psychological perspectives of periodontal disease probably
cannot be considered as completed without exploring the impact of periodontal
71
disease on quality of life. Studies with quality of life as outcome measure have
been reported in areas such as oral surgery, orthodontics, and referred periodontal
patients (Cunningham & Hunt, 2001; McGrath et al., 2003; Needleman et al.,
2004). Besides, little has been reported about the impact on oral health-related
quality of life associated with periodontal health or disease in general population.
The aim of this study was to investigate the relationship of periodontal
disease to psychosocial stress, making reference to the major components of the
stress process including stressors, mediating and moderating factors (coping
strategies and traits), and stress responses (psychological and somatic responses)
based on the contemporary understanding of the stress process. Periodontal
disease was assessed by status of clinical attachment level. Stressor were assessed
by questionnaires on major life events and daily strains. Coping and traits were
measured by a coping scale and a trait scale. The construct of dental anxiety was
measured by a comprehensive dental anxiety scale. Psychological and
physiological responses were measured with negative affectivity scales and
symptoms checklist. Details of demographic and socioeconomic data, systemic
disease, smoking, and dental health care habit were also collected for analysis of
confounding factors. The study also assessed the impact of periodontal status on
oral health-related quality of life.
72
Table 1.4 Health problems that may be linked to stress.

Health Problem Representative Evidence

Hernias Rahe & Holmes (1965)
Tuberculosis Wolf & Goodell (1968)
Leukemia Greene & Swisher (1969)
Glaucoma Cohen & Hajioff (1972)
Female reproductive problems Fries et al. (1974)
Asthma Plutchik et al. (1978)
Hyperthyroidism Weiner (1978)
Menstrual discomfort Siegel et al. (1979)
Hemophilia Buxton et al. (1981)
Vaginal infections Williams & Deffenbacher (1983)
Hypertension Egan et al. (1983)
Headaches Featherstone & Beitman (1984)
Cancer Cooper (1984)
Genital herpes VanderPlate et al. (1988)
Skin Disorders Fava et al. (1989)
Appendicitis Creed (1989)
Multiple sclerosis Grant et al. (1989)
Ulcers Ellard et al. (1990)
Chronic back pain Craufurd et al. (1990)
Diabetes Gonder-Frederick et al. (1990)
Complications of Pregnancy Pagel et al. (1990)
Stroke Harmsen et al. (1990)
Coronary heart disease Rosengren et al. (1991)
Inflammatory bowel disease Garrett et al. (1991)
Common Cold Stone et al. (1992)
Rheumatoid arthritis Thomason et al. (1992)
Periodontal disease Genco et al. (1999)


73
1.6.1 Hypothesis
1. An association exists between the various components of the stress
process stressors, stress responses, mediating and moderating
factors of coping and traits and periodontal disease.
2. An association exists between the construct of dental anxiety and
periodontal status.
3. An association exists between the periodontal status and oral
health-related quality of life.

1.6.2 Aim
A diagrammatic conceptualization of the present study is shown in Figure
1.4. We aimed at exploring how the stress process affects periodontal status and,
subsequently, the impact of periodontal status on oral health-related quality of life.
1. To investigate the relationships between variables of the stress
process and periodontal status;
2. To identify and evaluate, with respect to the stress process, risk
indicators in periodontal attachment loss;
3. To examine the interactions of the identified risk indicators;
4. To assess the impact of dental anxiety on periodontal status;
5. To assess the impact of periodontal status on oral health-related
quality of life.
6. To conduct a preliminary investigation of the relationships between
psychosocial factors and oral health-related quality of life.
74




















Figure 1.4 A diagrammatic conceptual representation of the present study.
Stress Process
Stressors
Life Events/Changes
Daily Strains

Mediating & Moderating Factors
Coping Skills
Personality traits
Dental anxiety

Stress Repsonses
Psychological
Physiological
Behavioural


Periodontal Disease
Clinical Attachment Level

Oral Health-related
Qaulity of Life

75
1.7 Addendum
Parts of the present study have already been published (Ng et al., 2005; Ng
& Leung, 2006a & 2006b). Portions of the Overview and Statement of Problem,
Materials and Methods, Results, Discussion, Tables and Figures are taken directly
(verbatim) from the work published during the course of this study.
76

Chapter 2
Materials and Methods

This chapter presents the materials and methods relating to validation of
Dental Anxiety Inventory (DAxI) for assessment of dental anxiety in the local
Hong Kong adult population as well as a cross-sectional study evaluating the
relationship between psychosocial variables, periodontal status and oral
health-related quality of life. An overview of the sequence of activities is
presented in Figures 2.1 and 2.2.
77
Validation of Dental Anxiety Inventory (DAxI)

Figure 2.1 The sequence of activities in validation of Chinese DAxI and
SDAxI.
DAxI translation
Chinese DAxI validation
A convenient sample of 500 subjects aged between 18 and 64 years were
recruited from 10 railway stations in Hong Kong.
Questionnaires issued by trained interviewers:
1. Chinese DAxI
2. Symptom Checklist 90 (SCL-90) (Derogatis, 1994)
3. Depression Anxiety Stress Scales (DASS) (Lovibond &
Lovibond, 1995a & 1995b)
4. State-Trait Anxiety Inventory (State Anxiety STAI-S, Trait
Anxiety STAI-T) (Kendall, 1976).
One month later, selected subjects attended a dental check-up and completed:
1. Chinese DAxI
2. Beck Anxiety Inventory (BAI) (Beck & Steer, 1990).
Two months after initial questionnaire interview, all 500 subjects were asked by
mail to complete the Chinese DAxI again for evaluation of test-retest
reliability.
Chinese SDAxI derivation and validation
The same set of items was selected as the original DAxI to construct the
SDAxI.
A convenient sample of 300 subjects aged between 18 and 64 years were
recruited from 10 railway stations.
Questionnaires issued by trained interviewers:
1. Chinese DAxI
2. Chinese SDAxI.
A random sample of 150 individuals were selected asked by mail to complete
the SDAxI a second time, two months after the initial questionnaire
interview, for evaluation of test-retest reliability.
78
The cross-sectional community study


Figure 2.2 The sequence of activities in the cross-sectional community
study.
Procedure
Trained interviewer first explained the details of the research project to subject
individually.
Patients who agreed to participate would sign a consent form.
Subjects were asked to complete a questionnaire including:
1. demographic and socioeconomic details
2. medical history reporting symptoms and diagnosed systemic
diseases
3. dental habits and dental care utilization
4. history of cigarette smoking and exposure to occupational hazards.
The investigator dentist confirmed before the clinical examination that subject
had no relevant medical history requiring prophylactic antibiotic cover, nor
any positive psychiatric history.
Periodontal examination were then carried out and mean CAL was calculated.
Questionnaires issued by trained interviewers:
1. Evaluation of stressors
2. Evaluation of stress responses
3. Evaluation of stress coping and stress-related trait dispositions
4. Measurement of dental anxiety
5. Measurement of oral health-related quality of life
Recruitment of subjects
Three general dental practices were selected in each of the three main
geographic districts of Hong Kong.
A convenient sample of 1,000 subjects aged between 25 and 64 years were
recruited.
79
2.1 Validation of Dental Anxiety Inventory (DAxI)
The Dental Anxiety Inventory Short Form (SDAxI) was planned to be used
to measure the specific trait disposition of dental anxiety in the present study. The
DAxI appeared promising in allowing an appreciable coverage of the concept of
dental anxiety and its empirical data also appeared to justify a positive assessment
(Schuurs & Hoogstraten, 1993). The DAxI was originally developed in Dutch and
subsequently translated into English (Skaret et al., 1998; Aartman et al., 2000). In
order to make use of this instrument to measure the dental anxiety of Chinese in
Hong Kong, translation and validation was necessary.

2.1.1 DAxI translation
Since the majority of the adult population in Hong Kong are literate (Census
& Statistics Department of the HKSAR, 2004), the Chinese DAxI was planned, the
same as for the original version of DAxI, to be a self-administered questionnaire. A
panel was set up for translation of DAxI, including three dentists, three psychologists
and one statistician specializing in survey studies. Advice was solicited from this
panel and DAxI was translated into Chinese by the author. The draft Chinese version
of the DAxI was back-translated into English by another two independent individuals,
one dentist and one psychologist, fluent in both Chinese and English, who were not
involved in the study. The backward translated English version was assessed and
evaluated by the panel to check whether the questions were properly translated.
Feedbacks from the panel were employed in further modification of the translated
version. The translated DAxI was then pilot-tested on a convenient sample of 50
adults studying or working in a university. Modifications were then made according
to the comments made by this sample.
80

2.1.2 Chinese DAxI validation
For the validation of the Chinese DAxI, a convenient sample of subjects aged
between 18 and 64 years were recruited at 10 railway stations in Hong Kong. The
target sample size was planned to include 500 subjects (that is, 50 subjects from each
station). The questionnaires were issued by 5 trained interviewers who were not
involved in any future assessment and analysis. These interviewers were all
university undergraduates majoring in psychology; they received training in subject
selection, introducing the research project, soliciting consent for participating in the
study and giving instructions on how to complete the questionnaires (Sommer &
Sommer, 2002).
The questionnaires used in the first interview included the Chinese DAxI, the
Symptom Checklist 90 (SCL-90) (Derogatis, 1994), the Depression Anxiety Stress
Scales (DASS) (Lovibond & Lovibond, 1995a & 1995b), and the State-Trait Anxiety
Inventory (State Anxiety STAI-S, Trait Anxiety STAI-T) (Kendall, 1976).
Symptom Checklist 90 (SCL-90) is a multidimensional self-report inventory
designed to screen for a broad range of psychological problems and symptoms of
psychopathology, including somatization (12 items), obsessive-compulsive (10
items), interpersonal sensitivity (9 items), depression (13 items), anxiety (10 items),
hostility (6 items), phobic sensitivity (7 items), paranoid ideation (6 items) and
psychoticism (10 items). The purpose of the scale is to screen for psychological
problems in normal populations, to evaluate symptoms changes over time, and to
support managed care decisions (Derogatis, 1996).
When completing the SCL-90 (Derogatis, 1994), the participant was asked to
rate on a 5-point Likert scale, with values ranging from 0 (not at all distressing) to
81
4 (extremely distressing) the degree of distress experienced due to each symptom

over the past seven days. Raw scores for each of the primary symptoms were
converted into standardized scores. Total scores for each factor were computed.
Scores for each of the nine factors were the average rating given to the symptoms of
that factor. The remaining seven items did not measure any particular factor, but
were evaluated qualitatively in clinical application setting.
The SCL-90 has been used extensively in research studies (Derogatis, 1996).
Studies of the SCL-90 have demonstrated satisfactory levels of concurrent,
convergent, discriminant, and construct validity comparable to other self-report
inventories. The SCL-90 instrument is used by clinical psychologists, psychiatrists,
and counseling professionals in mental health, medical, and educational settings as
well as for research purposes. It can be useful in both the initial evaluation of patients
and for measuring patient progress during treatment. The SCL-90 is a
well-researched instrument with numerous research studies supporting its reliability,
validity, and utility (China Association of Mental Health, 1993; Schmitz et al., 2000;
Schmitz et al., 2002; Hardt & Gerbershagen, 2001; Pearson Assessments, 2004). The
internal consistency coefficients (Cronbachs alpha) for the nine symptom
dimensions ranged from 0.77 for Psychoticism, to a high of 0.90 for
Depression. Test-retest reliability coefficients range between 0.80 and 0.90 after one
weeks interval.
The Depression Anxiety Stress Scales State (DASS-S) Chinese short
version was used to allow a relatively pure measure of the affective responses of an
individual in face of stress (Chan & Lovibond, 1996; Wong, 1996). Among these,
stress probably evokes anxiety, fear and depression more frequently than any other
82
emotions (Woolfok & Richardson, 1978, Weiten, 2004). The DASS (Depression
Anxiety Stress Scales) was designed to cover the full range symptoms of depression,
anxiety and stress, while at the same time providing maximum discrimination among
these affective states. It was constructed not only as another set of scales to measure
conventionally defined emotional states, but also to further the process of defining,
understanding, and measuring the ubiquitous and clinically significant affective
states of depression, anxiety and stress (Lovibond & Lovibond, 1995b). The short
form DASS was employed in this study for its promising psychometric properties
and substantial correlations with the standard version (Brown et al., 1997; Antony et
al., 1998). It is composed of three scales: anxiety, depression, and stress, each
consisting of 7 items. The depression scale assesses dysphoria, hopelessness,
devaluation of life, lack of interest or involvement in daily living, social withdrawal,
anhedonia, and inertia. The Anxiety scale assesses autonomic arousal,
musculo-skeletal effects, situational anxiety, and subjective experience of anxious
affect. The Stress scale sorts to detect the levels of chronic non-specific arousal. It
assesses difficulty relaxing, nervous arousal, and being easily upset, agitation,
irritability, over-reactivity and impatience.
Participants were asked to read each statement and use 4-point
severity/frequency scales to rate the extent to which they had experienced each state
over the past week, from 0 (Did not apply to me at all) to 3 (Applied to me very
much, or most of the time). Scores for Depression, Anxiety and Stress were
calculated by summing the scores for the relevant items. Subject was reminded that
there are no right or wrong answers and not to spend too much time on any
statement.
Internal consistency of the DASS subscales was high,

with Cronbachs alpha
83
above 0.88 for all three scales. Factor analysis suggested a

three-factor solution,
which corresponded well with the three

subscales of the DASS. Exploratory and
confirmatory factor analyses have substantiated the proposition of the three factors
(Brown et al., 1997). The DASS Anxiety scale correlates 0.81 with the Beck Anxiety
Inventory (BAI), and the DASS Depression scale correlates 0.74 with the Beck
Depression Scale (BDI) (Lovibond & Lovibond, 1995b). Reliability of the three
scales is considered adequate; and, test-retest reliability is likewise considered
adequate with 0.71 for depression, 0.79 for anxiety and 0.81 for stress (Brown et al.,
1997). Studies in Chinese and German also supported the validity of DASS (Wong,
1996; Nieuwenhuijsen et. al, 2003). The three-factor structure of DASS discriminates
between anxiety and depression better than other commonly used measures (Brown
et al., 1997). Groth-Marnat (1990) added further that the DASS allows a more
specific evaluation of depression as compared to the BDI which has also been
criticized for its ambiguity in measuring state or trait variables.
The State-Trait Anxiety Inventory (STAI) differentiates between the
temporary condition of "state anxiety" (STAI-S) and the more general and
long-standing quality of "trait anxiety" (STAI-T); this measure is commonly used in
research study for measurement of an individuals state and trait anxiety (Condon,
1993; Zheng et al., 1993; Hishinuma et al., 2000). Questions regarding the subjects
demographic data, educational level, income and brief dental history were also asked.
Subjects were asked to complete and return the questionnaires during the interview.
These instruments had been translated and validated for Chinese populations
(Condon, 1993; The Mental Health Association of China, 1993; Zheng et al., 1993;
Wong, 1996; Tang et al., 1999; Wang et al., 2000; Cheng et al., 2002).
One month later, individuals were selected from the pool of subjects surveyed
84
based on preliminary analysis of the DAxI scores and were invited by letter (with
telephone follow-up) to attend a free dental check-up by an uninvolved dentist at his
dental clinic. This subset of subjects ensured inclusion of individuals from upper,
middle and lower portions of the spectrum of the DAxI scores. In brief, all subjects
with initial DAxI score one standard deviation (SD) above (upper portion) or below
(lower portion) the mean were recruited. These formed two-thirds of the dentally
examined group. The remaining one-third were subjects with initial DAxI scores
within the range of the one standard deviation of the mean (middle portion) selected
as described below. The questionnaires with middle portion scores were arranged in
the order of the original coding and then subjects were selected by an interval
method. These selected subjects were asked to complete the Beck Anxiety Inventory
(BAI) (Beck & Steer, 1990) and to repeat the Chinese DAxI on the dental chair
immediately before the clinical examination commenced. The BAI is a
self-administered screening test, with good convergent and discriminant validity, for
measuring anxiety levels of clinical and non-clinical subjects by the response to 21
items rated on a scale from 0 to 3 (Beck et al., 1988; Beck & Steer, 1990). Each item
is descriptive of subjective, somatic, or panic-related symptoms of anxiety.
The dentist and the dental surgery assistants involved had also received
training in introducing the research project, and in providing instructions on how to
complete the questionnaires. Oral hygiene instructions and advice on their individual
treatment needs were given upon conclusion of the examination.
For evaluation of test-retest reliability, two months after initial questionnaire
interview, all 500 subjects were asked by mail to complete the Chinese DAxI again
and to return by mail using an enclosed stamped return envelop.

85
2.1.3 Chinese SDAxI derivation and validation
In the derivation of the short-form Chinese version of DAxI, the methodology
used by Stouthard (Abrahamsson et al., 2000) was followed and hence the same set
of items was selected. For the validation of the Chinese version of the SDAxI,
another convenient sample of subjects aged between 18 and 64 years were recruited
following the same method as for the validation of the full version of DAxI, i.e.
subjects were recruited from 10 railway stations by the same trained interviewers.
The target sample size was planned to include 300 subjects, i.e. 30 subjects from
each station. The questionnaires used included the Chinese DAxI and SDAxI. For
evaluation of test-retest reliability, all subjects who initially completed a
questionnaire with an even code number, target size of 150 individuals, were asked
by mail to complete the SDAxI a second time, and to return this by mail using an
enclosed stamped return envelop, two months after the initial questionnaire
interview.

2.1.4 Data analysis
The total DAxI score was calculated by summing the scores of the responses
to the 36 items. Item-scale correlation coefficients were used to assess the correlation
between the individual items and the DAxI scores. Validation of the Chinese DAxI
was achieved at three levels, namely construct validity, discriminant validity, and
reliability.
Construct validity was verified by examining the predictive ability of DAxI
and its provision of measuring the stable dental anxiety proneness, independent of
the situation. Predictive ability was measured by the correlation between the BAI
scores recorded for those subjects during the clinical dental examination and the
86
scores of the Chinese DAxI which they had completed at the first interview.
Measurement of stable dental anxiety proneness was assessed by repeated measures
ANOVA. This was studied by comparing the DAxI scores of those subjects who
attended the clinical dental examination, which were administered at three different
time points: namely at first interview, one month later upon dental examination and
two months after first interview by mail. The discriminant validity was evaluated by
assessing the correlations of the Chinese DAxI with related and unrelated variables.
Correlation coefficients indicating good construct validity and predictive
validity should be substantial. Good discriminant validity should be reflected by a
relatively small effect (r
xy
between 0.10 and 0.3) or zero correlations. The validity of
the Chinese version of the short-form DAxI was assessed by evaluating the
associations between the SDAxI scores and DAxI scores. The associations were
studied by the Pearson correlation and the linear regression of SDAxI on DAxI.
Test-retest reliability and internal consistency were obtained to assess the
reliability of the Chinese DAxI and SDAxI. Pearson correlation and intraclass
correlation were used to measure the test-retest reliability. Cronbach's Alpha was
used to measure the internal consistency. The level of significance was set at 0.05 for
all tests.

2.2 The cross-sectional community study
2.2.1 Subjects
2.2.1.1 Recruitment of subjects
Recruitment of subjects was designed to accomplish two objectives. Firstly,
the cross-sectional sample should be able to display broad variation in periodontal
condition and in potential risk factors to allow an objective assessment of the
87
relationship between explanatory and outcome variables. Secondly, efforts should be
made to ascertain the generalizability of the findings of this study to the broader
population.
As only limited emergency dental treatments, mainly extraction and
intervention for acute pain or infections, are provided by the HKSAR government in
a few public clinics or hospitals and people receive their dental care primarily from
private settings in Hong Kong. Three general dental practices were selected for
conducting the study, one in each of the three main geographic districts of Hong
Kong (namely, Hong Kong Island, Kowloon peninsula and the New Territories).
Patients who presented themselves for treatment in these three clinics were invited to
participate in the study.
Subjects were also invited through recruitment advertisement posted in these
three clinics. In this case, individuals (patients, relatives or others) interested in
participating in this project were asked to respond to the receptionists in these clinic
with their name, age and telephone number for further contact regarding their
eligibility for this study.
The selection criteria for subjects recruitment included:
1. within the age range of 25 to 64 years,
2. dentated,
3. not requiring antibiotic prophylaxis against bacterial endocarditits
for clinical periodontal examination.
4. without psychiatric history.

2.2.1.2 Study sample
A pilot study of 50 subjects was conducted before the main study. Based on
88
the result from the pilot, assuming an 80% power and a statistical significance level
at 5%, a sample of 820 subjects was sufficient to detect the minimal effect size in
ANOVA was 0.11 and to detect the minimal odd ratio at 1.7 for univariate regression
analysis. The target sample size was thus decided to include 1,000 subjects in
accordance with a balance between the above mentioned objectives and the
limitations of budget, resources and logistics.
A total of 589 subjects were approached as patients attending for dental
check-up or treatment in the three dental clinics. A verbal explanation and invitation
was given by a trained dental surgery assistant to these patients. Of these, 52 refused
to participate for personal reasons and 20 were rejected for incompatibility with
selection criteria. A total of 517 subjects were recruited to participate in the study
A total of 677 enrollments were received in response to the recruitment
posters in these clinics. The subjects were randomly assigned a sequence number. A
trained interviewer then approached the subjects by phone following the assigned
sequence number screening for compatibility with selection criteria. Eligible subjects
were invited to attend an interview and free dental check-up at the respective dental
clinics. A total of 503 subjects were successfully contacted and 20 were excluded for
incompatibility with selection criteria; that is, a total of 483 subjects were recruited
from this channel.
Recruitment of subjects was carried out in both channels simultaneously and
stopped as the total number of participants reached the target sample size of 1,000.
The study sample finally consisted of 531 females and 469 males (total 1,000)
between the ages of 25 and 64. The subjects were all permanent residents of Hong
Kong. Subjects from both recruitment channels, i.e. patients attending the three
dental clinics and respondents to poster advertisement, were offered an appointment
89
in one of the three dental clinics for interview and dental examination.

2.2.2 Design of study
2.2.2.1 The research team
The team for the community study consisted of the author as principal
investigator, assisted by two trained dental surgery assistants and two interviewers. A
panel was set up for supervising the research project, including two dentists, two
psychologists and one statistician specializing in health surveys.
Training was provided to the two dental surgery assistants in introducing the
research project and recording the clinical data. They received hands-on training in
recording the data items directly into a notebook computer as the investigator dental
surgeon conducting the examination. Briefing sessions with written instruction
materials were provided to these two dental surgery assistants regarding the theoretical
background, the design, and the aims and objectives of this study.
The questionnaires were issued by two trained interviewers who were not
involved in any future assessment and analysis. Two psychology undergraduates
from The University of Hong Kong were recruited as interviewers and trained to
assist in administration of the psychological questionnaires. They were all in the final
year of their study, fluent in both Chinese and English. They received training on
introducing this research study and explaining its purpose to subjects, as well as data
collection procedures. Training also included explanation of confidentiality and
information protection issues, and soliciting informed consent with signature to
participate the study. Regarding the various psychological instruments used in this
study, interviewers received training seminars on basic features of individual
questionnaires, the guidelines in administration and answers to frequently asked
90
questions, and the scoring and coding of raw data (Sommer & Sommer, 2002). They
all had the chance to experience filling out the whole set of questionnaires used in
this research. Scoring, coding and digital input of data from questionnaires were
completed by the interviewers. Data set from each subject was input twice by the two
interviewers and any discrepancy was then clarified.

2.2.2.2 Procedure
During the appointment, the trained interviewer first explained the details of
the research project to subject individually. Patients who agreed to participate were
asked to read and sign a consent form which summarized the study procedures and
highlighted their right to terminate their participation at any time without any
obligation.
Subjects were asked to complete a questionnaire including the following
sections: 1) demographic and socioeconomic details; 2) medical history reporting
symptom and diagnosed systemic diseases; 3) dental habits and dental care
utilization; and 4) history of cigarette smoking and exposure to occupational hazards.
All the questionnaires were designed in self-administered format for the subject. For
those illiterate or marginally literate subjects, mainly from the older age groups, they
received the questionnaire in an interview format with the research assistants and the
answers were recorded accordingly.
All participants were checked and confirmed in a clinical interview by the
investigator dentist who is also a qualified clinical psychologist before the clinical
examination that they had no relevant medical history requiring prophylactic antibiotic
cover, nor any positive psychiatric history (to exclude the confounding due to the
extreme or unpredictable stress response in psychiatric patients). Periodontal
91
examination was then carried out. Mean CAL was calculated for each subject and
this value represented the mean attachment loss status and was used as the outcome
variable. When clinical examinations were completed, a brief verbal report of dental
status was given to the subject including indications for treatment in accordance with
the standard professional ethical requirements; oral hygiene instructions were given
by a trained dental surgery assistant.
Subjects were then given a set of self-administered psychological
questionnaires in a face-to-face interview with a trained interviewer. Instructions for
completing the questionnaires were explained. Subjects were arranged to complete
the questionnaires in a private setting in the clinic and the interviewer was standby to
clarify any queries arising during completion of the questionnaires. Same as before
for the illiterate or marginally literate subjects, interviewers helped to complete the
questionnaires in an interview format in a private and closed environment.
Upon completion of the psychological questionnaires, participants were
invited to describe their feelings and comment on what they had experienced through
the course of the study procedures, including clinical examination and completing
questionnaires.

2.2.3 Data collection
2.2.3.1 Demographic and medical history
Data of demographic characteristics collected in this study included gender,
marital status, education, ethnicity, age, household income and occupation. The
Hong Kong Census 2001 (Census and Statistics Department of the HKSAR, 2001)
track data were referred in design of the format for collection of demographic data in
the present study.
92
A structured questionnaire was used to record the presence of systemic
diseases with special reference to the following categories: 1) cardiovascular, 2)
pulmonary, 3) allergy, 4) metabolic, 5) infectious, 6) neoplastic, 7) sexually
transmitted disease, followed by open ended question regarding other medical history.
The medical history details were then reviewed, further explored, if indicated, before
the dental examination by the investigator.
Data on personal dental habits and usage of oral health service were collected
through structured questions. History of smoking was reported by the number of
cigarettes smoked per day and the number of years of smoking. Exposure to
occupational hazards was put in an open question and also confirmed later by the
investigator dental surgeon. History of drinking was explored by frequency of
alcoholic beverage or wine consumption. These data would be categorized before
further analysis (details in Section 2.2.4 Data Analysis).

2.2.3.2 Periodontal examination
The author conducted all the dental examinations for the subjects of this
study in the three dental clinics. Training and calibration was completed in Prince
Philip Dental Hospital before the commencement of the research study. Intra-rater
reliability was ensured by repeating the examination of a random quadrant in every
tenth subjects and analysed with Kappa statistics (Guggenmoos-Hozmann, 1996).
Calibration was repeated back in Prince Philip Dental Hospital, The University of
Hong Kong after every one hundred subjects were examined.
Six to eight subjects were scheduled and examined in each clinic session.
Audio tape recording was used for counterchecking all the data input after each
session.
93
Clinical examination included recording the number of standing teeth;
measurement of the following parameters at six sites on each tooth (mesio-buccal,
mid-buccal, disto-buccal, mesio-lingual, mid-lingual, and disto-lungual): calculus (Cl,
visible or detectable through tactile sense using a periodontal probe), bleeding on
probing (BOP), followed by recession (REC) and probing pocket depth (PPD) after
dental prophylaxis (Pilgram et al. 2000). Tooth-sites excluded from the examination
were impacted teeth, retained roots, grossly broken down teeth or teeth which were
difficult to examine due to inaccessibility of the sites or had the cemento-enamel
junction (CEJ) indeterminable on clinical examination. The Brockprobe periodontal
probe (Figure 2.3) was used, which is an autoclavable metal probe with markings
identical to the standard Williams probe and gives approximately a calibrated 20
gram force for measurement of BOP and PPD. The measurement of REC, PPD and
clinical attachment level (CAL) was done according to Pilgram et al. (2000) with
modification: REC was measured from the CEJ to the gingival margin, with a
positive value if there was recession and a negative value in the absence of recession;
CAL was calculated by summation of PPD and REC.
The clinical criteria for the various parameters were:
Calculus (Cl) (Loe, 1967)
0 No calculus detected, neither supragingival nor subgingival
1 Only supragingival calculus detected
2 Detected subgingival calculus, with or without supragingival
calculus
3 Abundant amount of calculus detected.

Bleeding on probing (BOP)
94
0 No bleeding was detected from the gingival sulcus after probing
1 Bleeding was detected from the gingival sulcus after probing

Recession (REC) (Figure 2.4)
Recession was measured to the nearest millimeter from the
cemento-enamel junction (CEJ) to the gingival margin using the
Brockprobe probe, with a positive value if there was recession and a
negative value in the absence of recession but gingival margin coronal to
CEJ. Recession greater than 10mm was given a score 11.

Probing pocket depth (PPD)
Probing depths was measured to the nearest millimeter from the gingival
margin using the Brockprobe probe. Probing depth greater than 10mm
was given a score 11.

Clinical attachment level (CAL)
This equals to the sum of recession and probing pocket depth.

The sequence of examination began with charting of teeth present, caries,
calculus, bleeding on probing, prophylaxis, recession, and finally probing depth. The
recording of a single parameter for all six sites of each tooth was completed before
proceeding to the next. The recording of the same parameter was completed for all
teeth before proceeding to next parameter.
This sequence of examination was designed for three reasons. Firstly, it saved
time of examination and appeared more comfortable for patients. This was supported
95
by pilot trials and proved to be the most convenient order. Patients in pilot trials also
reported more comfortable for less manipulative movement of the cheek and lips.
The average time required for examination of a subject was approximately 20
minutes; less time was needed for subjects with healthy gingival condition.
All these data would be categorized before further analysis (details in Section
2.2.4 Data Analysis).
96








Figure 2.3 Measurement of pocket depth using Brockprobe.
The manufacturers instruction was followed: inserting the probe into sulcus/pocket
until the shaft flexes to meet the stop, at this point the proper pressure of 20 grams
(+/- 2 grams) has been reached. The probing depth is recorded.

97




Figure 2.4 Measurements of periodontal supports.
CEJ: cemento-enamel junction; GM: gingival margin; REC: recession; PPD: probing
pocket depth; CAL: clinical attachment level (sum of REC and PPD)



REC
PPD
CAL
CEJ
GM
98
2.2.3.3 Psychological questionnaires
Psychological instruments were used in measuring stressors including
changes, significant life event and daily strains, stress responses, coping, affective
dispositions, dental anxiety level and oral health-related quality of life (Table 2.1).
99
Table 2.1 The psychological questionnaires used in the cross-sectional
community study.

Evaluation of stressors
Life Event Questionnaire (LEQ) (Brugha & Cragg, 1990)
Social Readjustment Rating Scale (SRRS) (Holmes & Rahe, 1967;
McGrath & Burkhart, 1983)
Measure of Chronic Stress (Pearlin & Schooler, 1978)

Evaluation of stress responses
Symptom Checklist 90 (SCL-90) (Derogatis, 1994)
Depression Anxiety Stress Scales State (DASS-S) (Lovibond &
Lovibond, 1995a & 1995b)

Evaluation of stress coping and stress-related trait dispositions
COPE Inventory (COPE) (Carver, 1989)
Depression Anxiety Stress Scales Trait (DASS-T) (Lovibond &
Lovibond, 1995a & 1995b; Chan & Lovibond, 1996)

Measurement of dental anxiety
Dental Anxiety Inventory Short form (SDAxI) Chinese version

Measurement of oral health-related quality of life

The Chinese short-form version of the Oral Health Impact Profile
(OHIP-14S) (Wong et al, 2002)


100
2.2.3.3.1 Evaluation of stressors
Three psychological instruments were used in assessment of stressors in the
subjects daily living: 1) Life Event Questionnaire (Brugha & Cragg, 1990); 2)
Social Readjustment Rating Scale (Holmes & Rahe, 1967; McGrath & Burkhart,
1983); and 3) Measure of Chronic Stress (Pearlin & Schooler, 1978). These
psychological assessment instruments have been used widely by researchers all over
the world in evaluation of stressors in daily life; and, the empirical validity have been
well established (Sandler & Guenther, 1985; Weiten, 2004; Lynch et al., 2005).
The Life Event Questionnaire (LEQ) is a 12-item instrument measuring
common life events that tend to be perceived as threatening. The instrument was
designed to overcome labour-intensive and lengthy exploration interviews. The LEQ
is useful in making an assessment for psychosocial and environmental problems, that
is an Axis IV diagnosis in accordance with the Diagnostic and Statistical Manual of
Mental Disorders Fourth Edition (DSM-IV) (American Psychiatric Association
(APA), 1994). The problems include those associated with primary support group
(e.g. death of a family member, divorce), those related to social environment (e.g.
death or loss of a friend, adjustment to life-cycle transition such as retirement),
educational, occupational, housing, and economic problems (e.g. discord with
teacher or classmates, job dissatisfaction, homelessness, inadequate finance), and
problems related to interaction with the legal system (e.g. arrest, litigation, victim of
crime) (Brugha & Cragg, 1990).
Each subject was asked to consider whether any of the twelve categories of
life events or problems, as listed in the questionnaire, had happened to him/her
during the last 6 months. He then needs to check the box or boxes corresponding to
the month or months in which the event/events happened or began. Once a subject
101
has successfully identified the stressful events, the specific LEQ items may serve as a
good point of departure in developing a self-anchored rating scale (Brugha & Cragg,
1990). Each item of the LEQ is scored 1 if it is checked and 0 if not. A total
score would be the sum of all items; yet the scores will tend to be skewed due to the
nature of the instrument.
Concurrent validity estimates were derived from the concordance between
inpatient psychotic patients identification of stressful events and those identified by
a significant other; there was a 90% agreement when assessed at a three-month
period and a 70% agreement when assessed at six months. The study involving
extensive interview of stressful events as a base rate showed that the LEQ is sensitive
to the identification of stressful events. The test-retest reliability of the LEQ was
reported as 0.84 for a three-month period and 0.66 for a six-month period. The
instrument is thus acceptable to the general population and psychiatric patients. It has
high sensitivity and is more likely to produce false positives (identify as a stressful
event one that may not be) than false negatives (failing to identify a stressful event
that is present). (Rijsdijk et al., 2001). The psychological instrument had been
validated for use in a Chinese population (Chan, 1998; China Association of Mental
Health, 1993).
The Social Readjustment Rating Scale (SRRS) was the first ever instrument
focusing on identifying and quantifying potential stressors, or stressful life events.
Holmes and Rahe (1967) found that life changes, such as changes in personal
relationships, changes at work, can be stressful even when the changes are welcome.
Based on this analysis, the SRRS was developed and numerous improvements were
made on the initial checklist (Holmes & Rahe, 1967; McGrath & Burkhart, 1983).
The SRRS assesses a wide range of stressful experience in addition to life change
102
(Weiten, 2004).
The scale assigns numerical values to 43 major life events. These values are
supposed to reflect the magnitude of the readjustment required by each change. In
using the scale, subjects are asked to indicate how often they experience any of these
43 events during a certain time period, typically in the past year. The numbers
associated with each event checked are then summed as an index of the amount of
change-related stress that the subject has recently experienced (Holmes & Rahe,
1967; McGrath & Burkhart, 1983).
The SRRS was first developed in 1967 and the weight of stress of life event
was originally done objectively by a panel of judges. There have been revised
versions that followed the initial scale and dealt with the issue of how much weight
each life event should be given. Many critics have argued that the SRRS fails to
address the individuals perception of life events. For examples, different people may
experience with different perspectives on the specific life event of divorce; some
may consider divorce to be a great relief, like given a second chance in life to start
over again, others view it as the end of the world. In addition, the duration of each
stressful life event is not measured in the SRRS (Turner & Wheaton, 1995).
The advantages of the SRRS are that it is a relatively short survey consisting of
43 items in a checklist format. This life events measure is flexible enough to be
adapted to address specific research questions or needs of different research designs.
Validity of the SRRS is satisfactory (Creed, 1993) and numerous studies have
demonstrated that people with higher scores on the SRRS tend to be more vulnerable
to many kinds of physical illness and to many types of psychological problems as
well (Creed, 1993). It is a rather robust instrument used extensively in life event
research and is the basis for other life event inventories (Derogatis & Coons, 1993).
103
The psychological instrument had been validated for use in Chinese population
(Chan, 1998; China Association of Mental Health, 1993).
Measures of Chronic Stress (Daily Strains), adapted from the Problems of
Everyday Living Scale of Pearlin and Schooler (Pearlin & Schooler, 1978), assesses
chronic stressors associated with the central roles of people in daily life. These
include worker, financial manager, spouse and parent. The scale was developed with
the appraisal of stress from sociological perspective (Pearlin, 1975; Pearlin &
schooler, 1976). Pearlin et al. (1975, 1976) stated that many of the difficult problems
that people have to deal with in their daily living are not unusual problems impinging
on exceptional people in rare situations. People are persistently experiencing
challenges in daily life. Most of these are the persistent life-strains that people
encounter as they act as worker, breadwinners, husbands and wives, and parents
(Lazarus, 1993a, 1999; Lazarus & Lazarus, 1994; Weiten, 2004). By strains, it refers
to those enduring problems with the potential for arousing threat. They are the
chronic stress and within the themes repeatedly surfaced in exploration of stress. The
strains identified in this measurement are by no mean exhaustive, yet represent
problems that are frequently outstanding in the experiences of people in their roles as
workers, financial manager for self and family, spouses and parents.
In completing this instrument, the subject was asked to indicate on a Likert
scales how often or how serious he was affected in various conditions as stated in the
structured questions capturing the strains that exit within each of the roles. The strain
score in each role area is calculated and a role strain composite score is computed
from the total strain scores.
Because the structured questions about life-strains were so closely developed
from progressively focused exploratory interviews, these measures have been
104
successfully applied in numerous studies either for measuring the stress being in a
specific role or the overall strain as a whole (Link et al., 1990; Aneshensel et al.,
1993). The psychological instrument had been validated for use in Chinese
population (China Association of Mental Health, 1993).

2.2.3.3.2 Evaluation of stress responses
Two psychological instruments were used in assessment of the stress
response in the subjects daily living: 1) Symptom Checklist 90 (SCL-90) (Derogatis,
1994) and 2) Depression Anxiety Stress Scales - State (DASS-S) (Lovibond &
Lovibond, 1995a & 1995b). These are comprehensive, well-researched and popular
measures of the social, psychophysiological and affective responses of human in face
stressful condition (Sandler & Guenther, 1985; Vines et al. 2004; Weiten, 2004).
The Symptom Checklist 90 (SCL-90) is a multidimensional self-report
inventory designed to screen for a broad range of psychological problems and
symptoms of psychopathology.
The Depression Anxiety Stress Scales State (DASS-S) Chinese short
version was used to allow a relatively pure measure of the affective responses of an
individual in face of stress (Chan & Lovibond, 1996; Wong, 1996). Among these,
stress probably evokes anxiety, fear and depression more frequently than any other
emotions (Woolfok & Richardson, 1978, Weiten, 2004). The DASS (Depression
Anxiety Stress Scales) was designed to cover the full range symptoms of depression,
anxiety and stress, while at the same time providing maximum discrimination among
these affective states.
Both these scales give also measures of the symptoms reflecting the
physiological response of the subjects in face of stress. Details description of these
105
scales can be referred in previous section on validation of DAxI (Section 2.1.2).

2.2.3.3.3 Evaluation of stress coping and stress-related trait dispositions
Two psychological instruments were used in assessment of coping and trait
dispositions in the subjects daily living: 1) COPE Inventory (COPE) (Carver, 1989);
and, 2) Depression Anxiety Stress Scales Trait (DASS-T) (Lovibond & Lovibond,
1995a & 1995b; Chan & Lovibond, 1996). These scales were developed on a strong
theoretical basis addressing the stress-related traits and coping behaviours of people
in face of stress. Empirical evidence also supported their application in research and
clinical practice (China Association of Mental Health, 1993; Clark et al., 1995;
Wong, 1996; Ng et al., 2004).
The COPE Inventory (COPE) is used to measure the coping styles and
strategies. The COPE was developed to assess a broad range of coping responses. It
is believed that people may engage in a wide range of coping during a given period
(Kohlmann, 1993). Some responses in COPE are based on specific theoretical
arguments about functional and potentially less functional properties of coping
strategies. Other responses are included because research have indicated that the
coping tendencies they reflect either may be of value or may impede adaptive coping
(Carver, 1989). This coping measure was found to have adequate convergent and
discriminant validity, correlate with a variety of external criteria, including hassles
and uplifts, physical symptoms, satisfaction with life, positive affectively and
negative affectivity (Clark et al., 1995; Smari et al., 1997). The original factor
structures were reasonably replicated in subsequent studies (Smari et al., 1997).
The "dispositional" or trait-like brief version is used in this study. This COPE
version consists of 28 items measuring 14 different coping behaviours. The COPE
106
begins with an explanation about the purpose of the instrument in exploring how
people respond when they confront difficult or stressful events in their lives. It also
reminds the subject that there are lots of ways to try to deal with stress, and that
different events bring out somewhat different responses; subject is advised to think
about what he/she usually does when he is under a lot of stress. Subject is then asked
to indicate what he/she generally does and feels on a 4-point Likert scale, with values
ranging from 1 (I usually don't do this at all) to 4 ( I usually do this a lot), when
he/she experiences stressful events as listed in the COPE items (Carver, 1989 &
1997).
Despite the rather abstract nature of coping styles and strategies, supportive
evidence of both the convergent and discriminant validity of the COPE has been
established. The COPE scales also proved to be relatively free of strong association
with the social desirability scale. Even though the scales were developed with the
conceptual distinct aspects of "problem focused" and "emotion focused" coping, use
of aggregate scores or an "overall" coping index is not recommended (Carver, 1989
& 1997). An alternative, as suggested by the developer of the COPE, is to create
second-order factors from a particular data set and using the factors as predictors
because different samples or populations may exhibit different patterns of behaviours
(Lazarus, 1981, 1993a, 1999; Somerfield et al., 1997). This is one of the most
commonly used instruments in assessment of coping behaviours in the local
population (China Association of Mental Health, 1993).
The Depression Anxiety Stress Scales Trait (DASS-T) Personality
dispositions of anxiety and depression traits were included in the present study for
that these negative affectivities are very sensitive to stress (Spielberger, 1985;
Lovibond and Lovibond, 1995b; Lazarus, 1999). The Chinese version of the
107
DASS-T is used to assess the trait dispositions of depression, anxiety and stress of
the subjects (Chan & Lovibond, 1996; Wong, 1996; Ng et al., 2004). It contains the
same question items as the DASS standard version mentioned above. The instruction
is manipulated to tap the relatively stable individual proneness towards negative
affectivities; the respondents report the extent to which they usually experience in the
past year or a longer period.
It consists of 42 items, with 14 items for each scale of depression, anxiety and
stress. This dispositional version has been employed in several studies with
satisfactory validity (Chan & Lovibond, 1996; Brown et al., 1997; Antony et al.,
1998).

2.2.3.3.4 Measurement of dental anxiety
Dental Anxiety Inventory Short form (SDAxI) Chinese version was
developed to measure the situation-specific trait anxiety disposition to experience
anxiety in dental situations (Stouthard et al., 1993, 1995; Sections 2.1, 3.1.1 and 3.2
of this thesis). This instrument was included in the present study to explore and
evaluate if the dental trait anxiety would have an impact on an individuals
periodontal status. The original version of the Dental Anxiety Inventory (DAxI) was
developed by Stouthard in 1989 (Stouthard, 1989). The questionnaire was developed
with the aid of a facet design (Canter, 1985). All relevant facets, namely time,
situation and reaction, of the construct dental anxiety were distinguished and
combined in order to give a systematic description as completely as possible.
The original DAxI, however, was too long and proved inconvenient for use in
general dental practice (Skaret et al., 1998). A short form of DAxI (SDAxI,
Stouthard, 1993) with 9 items was derived by Stouthard in 1993, which has
108
demonstrated satisfactory psychometric characteristics comparable with the original
version. Answers are given on a 5-point Likert scale, ranging from complete
disagreement [1] to complete agreement [5]. Dental anxiety score is computed by
the total score of all items.
Since its development, the DAxI has been translated into and/or validated in
several languages (Dutch, English, German, French, Spanish, Italian and Norwegian)
(Stouthard et al., 1995), and adopted in various population studies in Europe
(Schuurs & Hoogstraten, 1993; Stouthard et al., 1993 & 1995;). The psychometric
characteristics of DAxI and SDAxI appeared promising (Schuurs & Hoogstraten,
1993; Stouthard, 1993). It allows an appreciable coverage of the concept of dental
anxiety and it empirical data also appeared to justify a positive remark (Schuurs &
Hoogstraten, 1993; Aartman et al., 2000). The SDAxI offers a short, easy to
complete, valid, reliable and interpretable scale for measuring dental anxiety.

2.2.3.4 Measurement of oral health-related quality of life
To focus attention on the influence of periodontal conditions on quality of life,
we sought to compare individuals at upper and lower ends of the spectrum of
periodontal attachment loss severity. From the cross-sectional sample, subjects with
either a mean full mouth CAL of 2mm (healthy/low periodontal attachment level
group), or > 3mm (high/severe periodontal attachment level group) (Section 2.2.3.2)
were selected. The Chinese short-form version of the Oral Health Impact Profile
(OHIP-14S) and a checklist of self-reported periodontal symptoms were sent by mail
to the subjects. A covering letter explaining the purpose and procedures of the study
and an informed consent were attached. Demographic data would be retrieved
directly from the database of the community study.
109
The impact of oral health on the patients' quality of life was assessed using
the Chinese version of OHIP-14S (Slade & Spencer, 1994; Wong et al, 2002). This is
a patient-centered outcome measure based on the World Health Organizations
disease-impairment-disability-handicap model. OHIP-14S is one of the most
comprehensive instruments available. It is a self-completed questionnaire consisting
of 14 items subdivided into seven domains (subscales): functional limitation,
physical discomfort, psychological discomfort, physical disability, psychological
disability, social disability, and handicap. These seven conceptual domains were
derived from the oral health model described by Locker (1988). The instrument's
psychometric properties, validity and reliability have been assessed and good results
were obtained (Wong et al., 2002; Slade, 1997a). Subjects were asked how frequently
they had experienced negative impacts in these respects in the preceding 12 months.
Responses to the items were recorded in a 5-point Likert scale: 0 = never; 1 = hardly
ever; 2 = occasionally; 3 = fairly often; 4 = very often.
Subjects were asked to complete a simple yes/no checklist of symptoms
relating to their periodontal health (self-reported periodontal status) in the past year.
This would allow further evaluation of the associations between subjects
self-reported periodontal status and oral health-related quality of life. They were
asked if they had experienced either swollen gums, sore gums, receding gums, loose
teeth, drifting teeth, bad breath and toothache (American Academy of Periodontology,
2004; Needleman et al., 2004).

2.2.4 Data analysis
All analyses were conducted using SPSS (Version 12.0) (SPSS, 2004) for
Windows. The level of significance was set at 0.05 for all tests.
110

2.2.4.1 Descriptive analysis and data processing
Descriptive analysis was conducted to describe the demographic
characteristics of subjects. Comparison was made of the characteristics of the sample
and those from local census data to evaluate the generalizability of the findings of
this study to broader populations. Dental habits and dental service utilization were
also evaluated with descriptive analysis.
History of smoking reported as the number of cigarettes smoked per day and
the number of years of smoking was then quantified as a composite value of the
number of packs of cigarettes smoked per day number of years smoked; that is, the
number of packyears. Tobacco consumption history was categorized as Grossi et al.
(1994) into 5 ordered categories: non-smoker; very light smoker more than 0 to 5.2
packyears; light smoker 5.3 to 15.0 packyears; moderate smoker 15.1 to 30.0
packyears; and heavy smokers above 30.0 packyears. Occupational hazards
assessed included chemicals, asbestos, radiation, and other (Grossi et al., 1994). The
data were coded in a dichotomous manner; that is, positive or negative. Drinking
history was categorized as Grossi et al. (1994) into 5 ordered categories:
non-drinker/ex-drinker; drink less than once a month, drink 1-3 days a month; drink
1-3 days a week; and daily drinker.
Descriptive statistics was used to report the periodontal status of the sample,
including the number of teeth present, Cl, BOP, PPD, REC, and CAL. After being
used to calculate CAL, negative REC values were transformed to 0 before further
relevant data analysis. A preliminary step in the analysis involved stratifying mean
CAL into 5 ordered categories as described by Genco et al. (1999): healthy 0 to 1.0
mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and
111
severe above 4.0 mm. The range of mean CAL observed in this sample was from
0.3 to 7.1 mm. The measurement of clinical attachment level, even though not
intended as an indication for treatment, allows an estimate of the historical amount of
periodontal destruction in a given patient (Loe et al., 1978; Locker & Leake, 1993;
Greenstein, 1997).

2.2.4.2 Reliability and validity
Kappa statistics was employed to examine the reliability of measurements
during periodontal examination - the examination was repeated in a randomly
selected quadrant in every tenth subjects.
To ascertain the validity of applying the various psychological measurements
in the present study sample, the collected psychological data was examined by
assessing the internal consistency of items within each subscale or individual
psychological instrument, the item-scale correlation and the correlation between
other subscales. Cronbachs alpha and correlation coefficient were utilized
accordingly for these purposes. In analyses of coping styles and strategies, as
suggested by the developer of the COPE scales (Carver, 1989), factor analysis with
Varimax rotation technique was conducted to extract a set of second-order factors
serving as predictor variables in subsequent analysis. This, according to Carver (1989,
1997), gives a more precise description of coping styles in a particular population.

2.2.4.3 Inferential statistics: evaluation of associations
Analysis of variance was employed to evaluate the differences in all
psychological scores of each subscale and dental anxiety between the various ordered
levels of CAL after adjusting for age, gender and cigarette smoking. Contingency
112
tables, Spearman rank correlations and analysis of variance were used to examine the
association of the ordered classification of CAL with age, demographic data,
systemic diseases, smoking, occupational hazards, and psychological variables.
Among the systemic diseases, only those with sufficient prevalence (at least 15
subjects with a positive history) to support approximately normal estimates for
regression analysis were considered. Significance level of 0.05 was adopted and
post-hoc comparisons were performed using the Tukeys HSD test (Altman, 1991;
Bland, 2000).
Clinical attachment levels were dichotomized into 2 groups for odds
assessment: combining healthy and low CAL categories as group 0 (minimal
disease) and combining high and severe CAL categories as group 1 (High/Severe
CAL). Ordinal logistic regression models were then used to evaluate the association
of the outcome variables, namely clinical attachment level (CAL) and other
explanatory variables including age, gender, education, smoking, systemic diseases,
occupational hazards and psychological variables. Age (in decades of 25-34, 35-44,
45-54 and 55-64) was first entered into the regression model because of its known
strong association with periodontal attachment loss. Systemic diseases such as
diabetes (yes/no), allergy (yes/no) and anemia (yes/no), were also entered
independently into the logistic model. Those variables of significance level of 0.10 or
less were then entered into the regression model in a stepwise approach. Odds ratios
(OR) and the corresponding 95% confidence intervals (CI) were calculated.
To further examine the detail picture of the odds of periodontal attachment
loss due to the interaction of the stressors on one hand and the dispositional
constructs of coping behaviours and personality traits on the other, median split of
relevant scores (MacCallum et al., 2002) was conducted to stratify subjects into
113
high and low groups of problem-focused copers, emotion-focused copers, trait
anxious subjects and trait depressive subjects respectively. Ordinal logistic
regression models were then used to measure respectively the association between
severity of attachment loss and various stressors in individuals grouped according to
high and low levels in these dispositional constructs of coping behaviours and
personality traits. OR and the corresponding 95% CI of individual subject group
were calculated.
Scores from the OHIP-14S and the self-reported checklist of symptoms of
periodontal health in the past year were used to assess the impact of periodontal
status on oral health-related quality of life. Scores were derived from the OHIP-14S
by summating the responses on the Likert scales to each of the individual questions.
Possible OHIP-14S scores ranged from 0 (no problems at all) to 56 (all problems
experienced very often). The unweighed OHIP-14S and subscales scores were used
in this study, as the weighed and unweighed OHIP scores in both the long and the
short form of the OHIP had similar psychometric performance (Allen & Locker,
1997).
Variations in mean OHIP-14S and subscales scores against self-reported
periodontal health (symptoms of periodontal disease) were explored through
bivariate analysis employing t-tests for independent samples. Associations between
OHIP-14S and the socio-demographic data of gender, income and education was
evaluated with correlation analysis. These items have been shown to correlate
significantly with oral-health related quality of life (McGrath & Bedi, 2004). The
correlations between various psychosocial variables of the stress process, the number
of teeth present, the number of occluding pairs and number of anterior teeth present
with oral-health related quality of life were also examined as these may affect
114
subjects perception and ability in chewing and personal appearance. Analysis of
variance was employed to examine the differences in OHIP-14S and subscales scores
between the different periodontal status (healthy/low periodontal attachment level
group versus high/severe periodontal attachment level group) after adjustment for
possible confounding factors.

2.3 Ethics
The Ethics Committee of the Faculty of Dentistry, the University of Hong
Kong approved the study. All participants volunteered themselves to participate in
form of written consent and they all received comprehensive information on the
study.

115

Chapter 3
Results

This study was designed to evaluate the relationship between psychosocial
variables and periodontal status as well as the impact of periodontal disease, as
reflected by concurrent mean CAL or periodontal attachment loss, on quality of life.
This chapter presents the data and research findings in the following order:
1. This chapter begins with presentation of the demographic characteristics of
the samples in the validation of DAxI and the cross sectional study.
2. The validation of the DAxI and its Chinese short form will be presented,
which will to be used later in the cross-sectional study.
3. The profiles of various independent variables and their relationships with
the dependent variable full mouth mean CAL will then be presented. The
independent variables include systemic diseases, smoking, occupational
hazards, alcohol drinking, stressors, stress responses, personality traits,
dental anxiety and coping. To ascertain the appropriateness of application of
the various psychological measures in the present study sample, the validity
of the data will be examined. The profiles of dental care and periodontal
status of the sample will also be presented.
4. Risk evaluation and interaction of the various significant demographic,
social and psychosocial variables on periodontal status will be assessed.
5. The impact of periodontal status on oral health-related quality of life of the
cohorts surveyed will be evaluated.
116
6. Results regarding preliminary investigation of the relationships between
psychosocial factors and oral health-related quality of life will be presented
in the last section.

3.1 Demographic characteristics
This section presents the demographic characteristics of the subjects,
including gender, age, ethnicity, marital status, education, household income and
occupation.

3.1.1 Validation of DAxI
For validation of the Chinese DAxI, 500 subjects with mean age 38.3 11.8
years were interviewed and their demographic data are shown in Table 3.1. All the
subjects had received at least primary education, approximately half of the subjects
were male and less than 30% received regular dental care including check-up and
scaling at least once a year.
For validation of the Chinese SDAxI, 300 subjects with mean age of 38.1
11.2 years were interviewed and the demographic data is shown in Table 3.1. These
data were essentially the same as those of the 500 subjects surveyed for the DAxI
earlier.
117
Table 3.1 Demographic background of the Hong Kong population and
subjects recruited for validation of Chinese DAxI and Chinese SDAxI.


Population
reference
Chinese DAxI
validation
Chinese SDAxI
validation
(%)
a
n % n %
Subjects 500 300
Gender
Male 48.6 240 48.0 146 49.0
Female 51.4 260 52.0 154 51.0

Educational level
Primary or less 22.8 108 21.6 67 22.3
Secondary 48.0 251 50.2 145 48.3
Tertiary or above 29.2 141 28.2 88 29.3

Individual income (in HKD)
<10,000 NA 369 73.8 211 70.3
10,001 20,000 NA 71 14.2 48 16.0
20,001 30,000 NA 37 7.4 24 8.0
>30,000 NA 23 4.6 17 5.7

Dental care
Regular 26.3 138 27.6 86 28.7
Irregular 73.7 362 72.4 214 71.3


NA = data not available for direct comparison.

a
Population reference for gender and educational level are from Census and Statistic
Department of the HKSAR (2001); population reference for dental care patterns is from Oral
Health Survey 2001 (Department of Health of the HKSAR, 2002).

118
3.1.2 The cross sectional study
A total of 1,266 subjects were approached. Of these 98 did not consent to
participate, 40 were excluded due to incompatibility with selection criteria after
confirmation by the investigator, and 128 quitted during the phase of questionnaire
completion and failed to complete all the psychological measurements. The
recruitment period lasted for nine months. A total of 1,000 subjects between the age
of 25 and 64 years (mean = 41.3 10.5 years) participated in this study. The sample
surveyed was predominantly Chinese and were all Chinese speaking permanent
residents in Hong Kong. Table 3.2 shows the summary of their demographic
characteristics. Statistical analysis failed to detect any significant difference between
the included and excluded subjects with respect to age and sex.
All the subjects completed the self-administered questionnaire of
demographic data before the clinical examination. The interviewer helped to clarify
queries, if any, arising during completion of the questionnaires. Almost all subjects
could complete the questionnaires themselves without any problem; 38 illiterate
subjects required assistance from the interviewer to read out the questions in
colloquial Chinese and record their answers accordingly.
As a group, the subjects were moderately well educated. Over two-thirds
(74.9%) of the respondents had secondary or above education with the majority
(57.6%) having secondary education (Table 3.2).
Most subjects reported a moderate income as a group. 82 subjects refused
to answer the question on household income. The largest groups had monthly
household incomes of $5,000 to $9,999 (30.2%) and $10,000 to $14,999 (25.7%).
About one-tenth of the respondents reported a monthly household income of less
than HK$5,000.
119
To focus attention on the influence of periodontal conditions on quality of
life, we sought to compare individuals at the upper and lower ends of the spectrum
of periodontal attachment level. A total of 767 subjects were selected from the
cross-sectional sample, with either a mean full mouth CAL of 2mm (healthy/low
periodontal attachment level group), or > 3mm (high/severe periodontal attachment
level group). Of the 767 subjects selected from the cross-sectional sample, 727
subjects responded and completed the questionnaires. The overall response rate was
94.7%. The demographic characteristics of this sub-sample of subjects are shown in
Table 3.3. More than two-third of the surveyed subjects had at least secondary
education. More than half of the sample reported a monthly household income
above HK$10,000. About a quarter of the sample had regular annual dental
check-up and preventive care. The subjects had a mean of 26 teeth (range 5-32)
comprising healthy/low CAL subjects with mean 28 teeth (range 18-32) and
high/severe CAL subjects with mean 17 teeth (range 5-26) (t = 2.39, P = 0.009).
120

Table 3.2 Demographic characteristics of subjects in the cross sectional study.


Demographic Characteristics n
Percentage
(%)
Population
(%)
a

Gender (n=1000)
Male 469 46.9 48.5
Female 531 53.1 51.5
Ethnicity (n=1000)
Chinese 955 95.5 94.9
Others 45 4.5 5.1
Age in Years (n=1000)
25 to 34 292 29.2 28.2
35 to 44 355 35.5 34.6
45 to 54 233 23.3 24.4
55 to 64 120 12.0 12.8
Marital Status (n=1000)
Never Married 350 35.0 31.9
Married 550 55.0 59.4
Separated/Divorced 65 6.5 2.7
Widowed 35 3.5 6.0
Education (n=1000)
None/Pre-school 38 3.8 3.8
Primary 213 21.3 21.4
Secondary 576 57.6 48.0
Tertiary (Non-Degree) 45 4.5 12.7
University Degree or above 128 12.8 14.1
Household Income (in HKD) (n=918)
b

Less Than $ 4,999 100 10.9 14.9
$ 5,000 - $ 9,999 277 30.2 29.5
$ 10,000 - $ 14,999 236 25.7 23.6
$ 15,000 - $ 19,999 128 13.9 11.8
$ 20,000 - $ 24,999 73 8.0 8.2
$ 25,000 - $ 29,999 32 3.5 3.8
More Than $ 30,000 72 7.8 8.2

121


Table 3.2 Continued.


Demographic Characteristics n
Percentage
(%)
Population
(%)
a

Occupation (n=1000)
Managers and administrators 88 8.8 NA
Professionals 30 3.0 NA
Associate professionals 101 10.1 NA
Clerks 51 5.1 NA
Service workers and shop sales workers 7 0.7 NA
Craft and related workers 45 4.5 NA
Plant/machine operators/assemblers 52 5.2 NA
Elementary occupations 125 12.5 NA
Skilled agricultural and fishery workers 127 12.7 NA
Occupations not classifiable/self-employed 126 12.6 NA
Housewife 62 6.2 NA
Retired 55 5.5 NA
Students 130 13.0 NA
Unemployed 1 0.1 NA
Dental Care
Regular 249 24.9 26.3
Irregular 751 75.1 73.7


NA = data not available for direct comparison.

a
Population reference data were retrieved from Census and Statistic Department of the
HKSAR (2001); population reference data for dental care pattern were retrieved from Oral
Health Survey 2001 (Department of Health of HKSAR, 2002).
b
82 subjects refused to disclosed income details.


122
Table 3.3 Demographic characteristics of subjects (n=727) taking part
in the study of oral-health related quality of life.

Demographic Characteristics n Percentage (%)
Gender
Male 342 47.0
Female 385 53.0
Age in Years
25 to 34 229 31.5
35 to 44 256 35.2
45 to 54 161 22.1
55 to 64 81 11.2
Marital Status
Never Married 254 35.0
Married 400 55.0
Separated/Divorced 48 6.6
Widowed 25 3.4
Education
None/Pre-school 23 3.2
Primary 148 20.3
Secondary 431 59.3
Tertiary (Non-Degree) 35 4.8
University Degree or above 90 12.4
Household Income (in Hong Kong Dollars)
a

Less Than $4,999 73 10.9
$ 5,000 - $ 9,999 202 30.2
$ 10,000 - $ 14,999 173 25.9
$ 15,000 - $ 19,999 93 13.9
$ 20,000 - $ 24,999 53 7.9
$ 25,000 - $ 29,999 23 3.4
More Than $ 30,000 52 7.8
Dental Care
Regular 178 24.5
Irregular 549 75.5


a
Total no. = 669; 58 subjects refused to disclose income details.
123
3.2 Validation of DAxI and its Chinese short form
The mean Chinese DAxI score, the Cronbachs alpha of the Chinese DAxI,
the item-scale correlation coefficients are shown in Table 3.4. No statistically
significant correlation was detected between DAxI scores and age, gender, education
level, income and dental care attendance profile.
There were a total of 90 subjects with DAxI scores greater than one standard
deviation above (upper portion) or less than one standard deviation below (lower
portion) the mean DAxI score. According to the pre-determined protocol, 45 subjects
were then selected from among those individuals who had initial DAxI scores within
one standard deviation from the mean (middle portion, n = 410). One out of every
nine subjects was selected. This subgroup of 135 subjects was then invited to attend a
free dental examination one month after the first interview. Eight subjects (6.0%),
however, refused to participate further in this study. The mean BAI score of the
remaining subjects (n = 127), measuring the state anxiety level at the time of the
clinical examination, was 16.0 15.2 (range 0 50). The correlation of the Chinese
DAxI with the BAI as measured by the Pearsons correlation coefficient was 0.97 (P
= 0.007).
For those subjects receiving a clinical dental examination, the Chinese DAxI
was administered thrice: at first interview, one month later upon clinical dental
examination, and two months after the first interview by mail; repeated measure
ANOVA showed no effect of the time of administration of the Chinese DAxI, F(2,
125) = 0.94, P = 0.393. (DAxI scores at first interview, one month later, upon clinical
examination, and by mail two months after first interview were 69.8 32.8, 69.6
32.4, and 68.3 31.9 respectively).
Correlations between the Chinese DAxI and the other tests are presented in
124
Table 3.5. The somatization, depression, anxiety and phobic sensitivity scales of the
SCL-90, the depression and anxiety scales of DASS, and the state and trait scales
from STAI, were all shown to be significantly correlated to the DAxI. In terms of
effect size (Cohen, 1977), a small effect (r
xy
around 0.3) was found for these scales.
No effect was found with other psychological problems as measured in SCL-90,
namely the obsessive-compulsive, interpersonal sensitivity, hostility, paranoid
ideation and psychoticism.
While 26 subjects refused to complete the Chinese DAxI a second time, the
test-retest reliability measured by the Pearsons correlation was 0.90 and intraclass
correlation was 0.90 (n = 474). Cronbach's Alpha was calculated at 0.77 which
indicated that good internal consistency was achieved.
The mean and range of the SDAxI score, the Cronbachs alpha, and the
item-scale correlation coefficients of the sample used in validation are shown in
Table 3.4. No statistically significant correlation was detected between Chinese
SDAxI scores and age, gender, education level, income and dental care profile.
The correlation coefficient between the Chinese SDAxI and the Chinese
DAxI was 0.93. The reported R
2
was 0.86 (P < 0.001) in the regression line of
SDAxI on DAxI, with a regression coefficient of 0.26. Thus, the score on the
Chinese SDAxI is approximately one fourth of the score on the DAxI which
consisted of four times as many items as compared to SDAxI. Out of the 150
subjects selected for SDAxI test-retest reliability, 7 refused to participate further. The
test-retest correlations and Cronbach's Alpha are presented in Table 3.4.
125
Table 3.4 Psychometric characteristics of the Chinese versions of DAxI and
SDAxI.

DAxI
(n = 500)
SDAxI
(n = 300)
Mean SD 60.3 20.5 15.2 6.0
Range 36 146 9 40
Internal consistency Cronbachs alpha 0.77 0.80
Item-scale correlation coefficient 0.72 0.81 0.91 -0.94
Test-retest Pearson correlation
a

Test-retest intraclass correlation
a

0.90
0.90
0.84
0.85

a
26 subjects dropped-out from DAxI and 7 dropped-out from SDAxI validation
respectively, i.e. n = 476 for DAxI, n = 143 for SDAxI.


126
Table 3.5 Population norms, convenient sample means (n=500) of SCL-90, DASS, and STAI, and their expected and observed
correlations with DAxI.

Population Norm Sample Correlation with DAxI
Test Variable (mean SD)
a
(mean SD) Expected
b
Observed (r
xy
) P - value
SCL-90 Somatization 8.9 7.6

8.1 7.7 Positive 0.15 0.032
Obsessive-Compulsive 11.9 6.8

11.4 6.4 Positive 0.22 0.341
Interpersonal Sensitivity 6.8 5.3

7.3 6.3 Positive 0.16 0.585
Depression 11.1 8.2

10.5 7.5 Positive 0.15 0.021
Anxiety 4.3 4.1

4.5 4.1 Positive 0.28 0.010
Hostility 4.5 3.9

4.9 4.1 None/negative 0.03 0.643
Phobic Sensitivity 2.6 3.0

2.3 2.9 Positive 0.10 0.042
Paranoid Ideation 8.4 5.9

8.9 5.3 None/negative 0.20 0.165
Psychoticism 5.6 5.6

6.0 6.2 None/negative 0.20 0.632

DASS Depression 5.7 6.4 5.6 6.0 Positive 0.10 0.009
Anxiety 6.6 5.3 6.7 5.3 Positive 0.22 0.005
Stress 12.0 7.0 12.3 7.1 Positive 0.01 0.756

STAI State anxiety 39.3 8.7 39.0 8.5 Positive 0.11 0.029
Trait anxiety 41.2 7.6 39.8 8.2 Positive 0.22 0.009

a
Population norms for SCL-90 and STAI are from China Association of Mental Health (1993); population norms for DASS are from Wong (1996).
b
Positive = positive relationship; none = no relationship; negative = negative relationship.
127
3.3 The cross sectional study
3.3.1 Systemic diseases, smoking habits, occupational hazards and alcohol
drinking
This section presents the profile of the subjects with respect to systemic
diseases, smoking habits, occupational hazards and alcohol drinking. The
association with periodontal status is examined.
The presence of a self-reported systemic medical condition or symptoms
suggestive of systemic disease reported in medical history questionnaire was
followed up by the investigator before the clinical examination. The diseases were
grouped under the following categories: 1) Allergy, 2) Anaemia 3) Asthma, 4)
Cardiovascular disease, 5) Diabetes, 6) Hypertension, 7) others (frequency less
than 5 in the whole sample) including arthritis, cancer, renal disease, thyroid
disease, gout, hepatitis, cirrhosis, angina, cataracts and emphysema; known cases
of hepatitis B virus carrier were also recorded. A summary of the frequency of
systemic diseases is shown in Table 3.6.
The most prevalent self-reported diseases in this sample were allergies,
including skin allergy, allergic rhinitis, allergic to medicine, food and other
substance; followed by hypertension, diabetes, asthma, anemia and cardiovascular
disease. (Table 3.6). Of the 62 diabetic subjects surveyed, 18 (29.0%) reported
using insulin to control their diabetic status.
The relationship between smoking habit and the various clinical
attachment level categories is shown in Table 3.7. Those never having smoked
accounted for 86.0% of the sample; moderate and heavy smokers accounted for
less than one-tenth of the subjects (3.5% and 5.5%). There were no periodontally
healthy subjects who were, or had been, heavy smokers. Analysis with
128
contingency table gave a significant Chi-square statistics, = 87.45, P < 0.001. A
summary of cigarette smoking in packyears and its relationship with the full
mouth mean clinical attachment levels was also evaluated using the Spearman
rank correlation coefficient (Table 3.8). The smokers in this sample exhibited a
substantial range of packyears from 0.5 to 140. Despite the different packyear
levels observed for each category of CAL, there was a trend towards increasing
packyears and increasing mean severity of clinical attachment level, Spearman
Correlation Coefficient = 0.369, P = 0.029. The healthy, low, moderate, high, and
severe CAL groups smoked 0.9, 2.2, 5.5, 7.2 and 15.6 mean packyears,
respectively (Table 3.8). The distribution of subjects smoking habit by age
cohorts is shown in Table 3.9; Chi-square tests suggested statistical significant
association between smoking habit and age groups, = 126.10, P < 0.001. The
correlation between age cohorts and packyear levels was statistically significant,
Spearman Correlation Coefficient = 0.176, P < 0.001. Analysis of variance
revealed that, after adjusting for age, there were statistically significant
differences between the mean full mouth CAL across different smoking habit
groups, F(4, 994) = 13.54, P < 0.001. Statistical significant correlations were also
detected between smoking with full mouth mean PPD and number of teeth present,
Spearman Correlation Coefficients of 0.251 (P = 0.032) and -0.150 (P = 0.027)
respectively; a statistical significant relationship, however, was not found between
smoking and other periodontal parameters including BOP, Cl and REC.
The relationship between exposure to occupational hazards and categories
of full mouth CAL is shown in Table 3.10. A total of 125 subjects (12.5%) of the
entire sample reported that they were positive for exposure to at least one of the
occupational hazards that were investigated. The most common exposure was
129
chemical (8.0%), followed by radiation (2.7%). The least number of positive
subjects was seen in the severe group (4 subjects), with various numbers of
positive subjects across the other four categories of clinical attachment levels.
Chi-Square tests failed to detect any statistically significant association between
the various categories of clinical attachment level and the exposure to various
occupational hazards.
The distribution of subjects by their mean full mouth CAL categories and
drinking frequency is shown in Table 3.11. Chi-Square tests failed to detect any
statistically significant association between the various categories of clinical
attachment level and the drinking habits. No statistical significant correlation was
observed between drinking habit and mean full mouth CAL, Spearman
Correlation Coefficient = 0.024, P = 0.455, nor any significant difference in mean
CAL between subjects of different profile of drinking habits after adjusting for
age, F(4, 994) = 1.39, P = 0.236.









130

Table 3.6 Prevalence of systemic diseases in the study sample,
n=1,000.

Disease Prevalence Precentage
n %
Allergy
skin 18
nasal 24
medicine 5
food 19
other 8
skin and nasal 36
Total 110 11.0%


Anaemia 27 2.7%

Asthma 51 5.1%

Cardiovascular 26 2.6%

Diabetes 62 6.2%

Hypertension 77 7.7%


Others


Arthritis 3

Cancer 2

Renal disease 2

Thyroid Disease 2

Gout 3

Hepatitis 2

Cirrhosis 2

Angina 4

Cataracts 2

Emphysema 1

Total 23 2.3%


Hepatitis B carrier
98 9.8%





131
Table 3.7 Distribution of cigarette smoking (packyear) by categories of full mouth mean clinical attachment level,
n=1,000.

Clinical Attachment Level
a
Total CAL
Smoking Habit
b
Healthy Low Moderate High Severe n % Mean SD

None 81 454 199 84 42 860 86.0 2.0 0.9
Very light 3 7 1 0 0 11 1.1 1.5 0.6
Light 5 32 2 0 0 39 3.9 1.4 0.4
Moderate 1 24 7 3 0 35 3.5 1.8 0.7
Heavy 0 8 24 14 9 55 5.5 2.9 1.0

Total 90 525 233 101 51 1,000 100.0


Chi-square statistics, = 87.45, P < 0.001

a
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above 4.0 mm
(Genco et al., 1999).

b
Smoker categories: very light smoker > 0 to 5.2 packyears; light smoker 5.3 to 15.0 packyears; moderate smoker 15.1 to 30.0
packyears; and heavy smokers >30.0 packyears; classified as described by Grossi et al. (1994).
132

Table 3.8 Mean ( SD) cigarette smoking in packyear by categories of full mouth mean clinical attachment level.

Clinical Attachment Level
a

Age (years) Healthy Low Moderate High Severe Total

25-34 0.1 0.7 1.0 3.0 1.0 3.6 7.2 11.1 11.2 8.2 1.2 3.6
35-44 0.9 2.3 1.7 5.5 2.1 7.8 8.1 12.7 14.0 26.5 2.8 7.6
45-54 2.7 8.5 4.2 10.5 6.8 15.2 7.0 15.8 14.7 37.9 6.2 16.4
55-64 2.4 4.5 4.7 8.0 18.2 24.9 20.9 26.6 23.3 48.5 13.5 25.5

Overall 0.9 3.4 2.2 6.6 5.5 14.5 7.2 16.9 15.6 37.1 4.0 9.0


Spearman Correlation Coefficient = 0.369, P =0.029

a
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above 4.0 mm
(Genco et al., 1999).

133

Table 3.9 Distribution of cigarette smoking in packyear by age cohorts, n=1,000.

Smoking Habit
a

Age (years) None Very Light Light Moderate Heavy Total

25-34 261 10 15 6 0 292
35-44 322 1 14 12 6 355
45-54 196 0 0 11 26 233
55-64 81 0 10 6 23 120

Overall 860 11 39 35 55 1,000


Chi-square statistics, = 126.10, P < 0.001








a
Smoker categories: very light smoker > 0 to 5.2 packyears; light smoker 5.3 to 15.0 packyears; moderate smoker 15.1 to
30.0 packyears; and heavy smokers >30.0 packyears; classified as described by Grossi et al. (1994).


134

Table 3.10 Distribution of subjects with occupational hazards by categories of full mouth mean clinical
attachment level, n=1,000.

Clinical Attachment Level
a
Total Chi-square statistics
Hazard
b
Healthy Low
c
Moderate
d
High Severe n % P - value
Asbestos 0 0 2 0 0 2 0.2 6.597 0.159
Chemical 9 42 20 6 3 80 8.0 1.490 0.829
Radiation 0 17 7 2 1 27 2.7 3.463 0.484
Others 0 12 3 3 0 18 1.8 4.414 0.353
None 81 455 202 90 47 875 87.5


a
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above
4.0 mm (Genco et al., 1999).
b
Occupational hazards are classified as described by Grossi et al. (1994).




c
One subject in low CAL category reported exposure to both chemical and radiation hazards.




d
One subject in moderate CAL category reported exposure to both chemical and radiation hazards.





135

Table 3.11 Distribution of subjects by categories of mean full mouth clinical attachment level and drinking frequency,
n=1,000.

Clinical Attachment Level
a
Total Mean CAL
Drinking Frequency Healthy Low Moderate High Severe n % (SD)
Non-drinker/Ex-drinker 47 258 115 49 18 487 48.7 2.0 0.9
Drink less than once a month 30 147 78 33 22 310 31.0 2.1 1.0
Drink 1-3 days a month 6 45 20 7 2 80 8.0 1.9 0.8
Drink 1-3 days a week 5 60 16 6 5 92 9.2 2.0 1.0
Daily drinkers 2 15 4 6 4 31 3.1 2.3 1.3

Total 90 525 233 101 51 1,000 100.0


Chi-Square statistic, = 23.32, P = 0.105
Spearman Correlation Coefficient = 0.024, P = 0.455


a
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above 4.0 mm
(Genco et al., 1999).


136
3.3.2 Dental history and oral hygiene habits
This section describes, with descriptive analysis, the dental history and
oral hygiene habits of the sample. The details regarding time of last dental visit,
dental care service utilization, dental care habits including brushing and flossing
were evaluated.
Table 3.12 summaries the findings on dental care habits. A total of less
than six percent of the sample reported that they had never visited a dentist before.
Among the subjects who had visited a dentist within the past year (36.1%), about
one-third of them went because of dental problems (e.g. pain, caries) while
two-third of them did so for check-up examination or professional cleaning. That
means about a quarter of the sample were regular dental attenders. This regular
maintenance practice appeared more common in the age group of 25-34 and least
common in the age group of 55 to 64 (Table 3.12).
Most of the subjects reported a tooth brushing habit of once or twice daily
(26.3% and 70.7% respectively). Less than one percent of the subjects reported
brushing occasionally or had never brushed before. These undesirable oral
hygiene practices were more common in the age group of 55-64 years. Flossing
was a far less common dental habit than brushing; only about one-eighth of the
sample reported regular use of dental floss on a daily base and less than
one-fourth of the sample claimed occasional flossing. The rest of the sample
reported that they had never used dental floss (Table 3.12).




137
Table 3.12 Percentage distribution of dental care habits according to age,
n=1,000.

Age
25-34 35-44
a
45-54 55-64 Overall
n=292 n=355 n=233 n=120 n=1000
Time of last dental visit
1 year or less
for check-up and professional cleaning 27.4 24.2 24.9 20.8 24.9
for dental problem 9.3 9.9 10.7 20.8 11.2
1 to 3 years 34.9 33.2 30.0 22.5 31.7
More than 3 years 23.6 26.2 24.5 27.5 25.2
Never visited dentist 3.8 5.9 8.2 6.7 5.9
Could not remember 1.0 0.6 1.7 1.7 1.1

Toothbrushing habit
Three times daily 5.1 0.0 0.0 0.0 1.5
Twice daily 67.8 71.3 71.7 74.2 70.7
Once daily 26.4 28.7 25.7 20.0 26.3
Brushed occassionally 0.7 0.0 0.9 2.5 0.7
Never brushed 0.0 0.0 1.7 3.3 0.8

Use of dental floss habit
Flossed everyday 14.4 12.4 11.6 9.2 12.4
Flossed occassionally 27.4 24.5 21.0 19.2 23.9
Never flossed 58.2 63.1 67.4 71.6 63.7


a
Consistent with previous Hong Kong findings regarding corresponding age group,
i.e. 35-44 years (Department of Health of HKSAR, 2002, Appendix 3).


138
3.3.3 Dental and periodontal profile of the sample
This section describes the intra-rater reliability assessed with weighted
Kappa statistics (Cohen, 1968; Fleiss, 1981; Hafkenscheid, 1993; Grossi et al.,
1994; Guggenmoos-Hozmann, 1996) and the dental and periodontal profile of the
sample. Number of teeth present, PPD, REC, CAL, presence of Cl and BOP were
reported with descriptive statistics.
Intra-rater reliability All clinical examinations were carried out by the
author from 2003 to 2004. Intra-rater reliability was assessed on one randomly
selected quadrant in every 10
th
patient. The intra-examiner reproducibility of
clinical periodontal examination results expressed as proportion of agreement was
never lower than 85%. The weighted kappa statistics were between 0.72 and 0.78
for CAL, PPD and REC (Table 3.13). That is, substantial agreement in scoring
of the various periodontal variables (Landis & Koch, 1977; Byrt et al., 1993;
Guggenmoos-Holzmann, 1996; Cook, 1998).
Teeth present It can be seen from Table 3.14 that about ninety percent
of the subjects had 20 teeth or more. Among the subjects in the age cohort of 25
to 34, 98% had 20 teeth or more, and none of them had less than 10 teeth. The
mean number of teeth present was less in older age cohorts. In the age cohort of
55 to 64, one-tenth had 9 teeth or less.
Probing Pocket Depth, Recession and Loss of attachment Table 3.15
summarizes the prevalence and extent of involvement, measured by number of
teeth and by number of tooth sites affected, the PPD, REC and CAL data
according to severity thresholds. In general, lower prevalence of PPD, REC and
CAL, and smaller extent of involvement were observed in higher severity
thresholds in various age cohorts and in the sample as a whole. On the other hand,
139
a higher proportion of older age subjects experienced more severe PPD, REC and
CAL. The highest prevalence of PPD was recorded in the 45-54-year-old cohort
in all PPD severity thresholds; the highest prevalence of REC or CAL was
recorded in the eldest age cohort of 55-64 years in all severity thresholds.
The distribution of subjects according to categories of full mouth mean
CAL as defined by Genco et al. (1999) and age is shown in Table 3.16.
Considering the sample as a whole, more than half of the subjects were in the
healthy and low CAL categories compared to approximately 15% in the high and
severe CAL categories. Of the 90 healthy CAL subjects, 72 (80%) were from the
two younger age cohorts of 25-34 and 35-44 years; of the 525 low CAL subjects,
377 (72%) were in the two younger age cohorts. A greater proportion of subjects
showed severe clinical attachment loss in the older age cohorts; 55 (54%) of the
101 high CAL subjects and 34 (67%) of the 51 severe CAL subjects were in the
two older age cohorts of 45-54 and 55-64 years.
Calculus Almost all the subjects had one or more teeth with calculus.
The mean proportions of teeth and tooth sites with calculus in different age
cohorts are shown in Table 3.17. Figure 3.1 shows the proportions of tooth sites
with calculus according to various severities of calculus deposit.
Bleeding The distribution of the proportion of teeth and tooth sites with
bleeding on probing for various cohorts are shown in Table 3.17.






140

Table 3.13 Intra-examiner reproducibility for periodontal variables
according to age cohorts.

Age
cohort
Clinical Attachment
Level (CAL)
Pocket Probing Depth
(PPD)
Recession
(REC)
(years) % agree weighted % agree weighted % agree weighted
25-34 91 0.71 90 0.78 94 0.70
35-44 92 0.75 99 0.79 90 0.81
45-54 89 0.68 88 0.76 98 0.76
55-64 85 0.67 85 0.74 91 0.70
Overall 87 0.72 99 0.78 85 0.75



141
Table 3.14 Distribution of subjects by number of teeth present.

No. of teeth present
a

1-9 10-19 20-32
Age n % n % n % Mean no. of teeth present (SD)
25-34 (N=292) 0 0.0% 7 2.4% 285 97.6% 27.7 3.4
35-44 (N=355) 0 0.0% 16 4.5% 339 95.5% 26.3 4.0
45-54 (N=233) 7 3.0% 36 15.5% 190 81.5% 23.8 5.6
55-64 (N=120) 12 10.0% 33 27.5% 75 62.5% 21.4 7.8

Overall 19 1.9% 92 9.2% 889 88.9% 25.5 5.2


a
Consistent with previous Hong Kong findings regarding corresponding age group i.e. 35-44 years : 1-9 teeth 0%, 10-19 teeth
4%, 20-32 teeth 96% (Holmgren et al., 1994, Appendix 4).


142

Table 3.15 Probing pocket depth, recession and clinical attachment level - prevalence and extent (mean number of teeth/sites)
according to severity.


4mm 6mm 9mm
Extent Extent Extent
Clinical variable
Age
(year) n
Prevalence
(% persons)
No. of
teeth
No. of
site
Prevalence
(% persons)
No. of
teeth
No. of
site
Prevalence
(% persons)
No. of
teeth
No. of
site
Probing pocket depth 25-34 292 58.9 2.1 3.5 12.7 1.2 2.0 1.7 1.6 1.6
(PPD) 35-44
a
355 61.7 4.6 8.4 17.5 2.1 3.3 2.3 1.8 2.3
45-54 233 68.2 4.7 9.8 28.8 1.8 3.3 3.4 1.1 1.4
55-64 120 59.2 4.5 8.7 20.0 2.0 2.8 1.7 1.5 1.5
Overall 1000 62.1 3.9 7.4 19.0 1.8 3.0 2.3 1.5 1.7
Recession 25-34 292 15.4 2.1 5.2 3.8 2.6 3.9 0.0 0.0 0.0
(REC) 35-44
a
355 49.0 3.1 5.6 12.4 1.8 3.0 0.6 1.0 1.0
45-54 233 57.1 3.0 6.3 15.5 1.7 3.7 2.6 1.2 1.8
55-64 120 60.8 3.4 7.3 25.0 2.1 3.6 4.2 1.2 1.4
Overall 1000 42.5 3.0 6.1 12.1 1.9 3.5 1.3 1.2 1.5
Clinical attachment level 25-34 292 61.6 4.8 13.0 19.5 1.8 7.3 2.1 6.0 9.8
(CAL) 35-44
a
355 71.8 8.0 26.7 33.8 3.2 11.2 6.8 2.5 5.2
45-54 233 79.8 8.2 29.6 45.1 3.2 12.8 14.2 1.9 5.0
55-64 120 85.8 8.8 31.9 50.8 4.1 13.5 16.7 2.2 3.9
Overall 1000 72.4 7.4 24.8 34.3 3.1 11.4 8.3 2.4 5.1

a
Similar to corresponding data from a Hong Kong periodontal health survey (Holmgren et al., 1994, Appendix 4); 35-44 age group i) 4 mm
(PPD/REC/CAL): 81/22/74% persons, 7.3/4.1/8.0 teeth; ii) 6 mm (PPD/REC/CAL): 20/3/33% persons, 2.8/2.2/3.3 teeth; iii) 9 mm
(PPD/REC/CAL): 2/0/7% persons, 1.7/1.2/2.2 teeth.

143

Table 3.16 Distribution of subjects by categories of full mouth mean clinical attachment level and age, n=1,000.

Clinical Attachment Level
a

Healthy Low Moderate High Severe Overall Age
(years) n mean (SD) n mean (SD) n mean (SD) n mean (SD) n mean (SD) n mean (SD)
25-34 44 0.70 0.21 183 1.81 0.12 51 2.12 0.09 9 3.22 0.28 5 4.64 0.45 292 1.79 0.66
35-44 28 0.73 0.10 194 1.57 0.21 84 2.19 0.08 37 3.45 0.22 12 4.55 0.39 355 1.95 0.85
45-54 10 0.98 0.01 106 1.30 0.08 63 2.19 0.08 35 3.47 0.26 19 4.83 0.83 233 2.14 1.16
55-64 8 0.96 0.03 42 1.49 0.27 35 2.47 0.24 20 3.53 0.32 15 4.46 0.40 120 2.45 1.13

Total 90 0.81 0.29 525 1.57 0.22 233 2.20 0.15 101 3.40 0.33 51 4.63 0.60 1000 2.01 0.94
% 9.0 52.5 23.3 10.1 5.1 100.0
















a
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above 4.0 mm
(Genco et al., 1999).








144

Table 3.17 Mean proportions of teeth and tooth sites with calculus and with bleeding on probing.

Calculus Bleeding on probing
% teeth % tooth sites % teeth % tooth sites
Age (SD) (SD) (SD) (SD)
25-34 (n=292) 99.2 2.3 70.7 10.4 87.3 21.2 39.7 15.9
35-44 (n=355) 98.9 7.8 76.0 13.1 82.9 24.9 38.2 19.6
45-54 (n=233) 98.3 7.5 83.5 11.0 93.5 11.0 46.3 12.2
55-64 (n=120) 99.2 2.6 74.1 11.7 74.7 22.4 41.0 34.2
Overall 98.9 6.1 76.0 12.6 85.7 21.7 40.9 19.8


145
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
25-34 35-44 45-54 55-64 Overall
Age cohorts
SITES WITH ABUNDANT CALCULUS
SITES WITH SUBGINGIVAL OR SUPRA- & SUB-
GINGIVAL CALCULUS
SITES WITH SUPRAGINGIVAL CALCULUS ONLY
SITES WITHOUT CALCULUS

Figure 3.1 Proportion of tooth sites with various severities of calculus deposit.
146
3.3.4 Stressors and periodontal status
Three sets of questionnaires were completed for measuring the stressors in
daily living of the subjects, namely Life Events Questionnaire (LEQ), Social
Readjustment Rating Scale (SRRS) and Measure of Chronic Stress (Daily
Strains).
Table 3.18 shows the results of evaluation of the validity of the Measure of
Chronic Stress (Daily Strains) questionnaire in this study. The Cronbachs alpha
ranged from 0.79 to 0.94 for the five subscales and was 0.91 for the overall
composite scale. The item-scale correlation coefficients ranged from 0.65 to 0.88
for various subscales, and ranged from 0.51 to 0.79 for the composite scale. The
discriminant validity was measured by the correlation with other subscales and
ranged from 0.07 to 0.22 for various subscales, with the exception that the
correlation coefficient between job and financial strain scores was 0.41 (P =
0.016). The correlation coefficients ranged from 0.08 to 0.23 for the composite
scale with various subscales. These results supported the reliability, and the
convergent and discriminant validity of the collected psychosocial data in this
study sample.
The mean scores of the LEQ, SRRS and Measure of Chronic Stress after
adjusting for the effects of age, gender and smoking are shown for subjects with
different categories of clinical attachment level in Table 3.19. The mean number
of life events in the last six months as measured by the LEQ ranged from 0.26 to
0.34 in different categories of clinical attachment level; statistically significant
difference was not detected. The mean scores of life changes in last one year, after
adjusting for age and gender and smoking, as measured by SRRS ranged from
80.57 to 96.81; this is within the range of low stress (Holmes & Rahe, 1967)
147
approximately equivalent to facing retirement and change in financial state, or
suffering a personal injury. Statistical analysis failed to detect any significant
difference in SRRS scores across the different categories of clinical attachment
level.
In assessment of chronic daily strains with the Measure of Chronic Stress,
statistically significant differences were detected in job, financial and role
composite strain scores across the various categories of clinical attachment level.
Subjects suffering more severe clinical attachment loss had higher job, financial
and role composite strain scores than the periodontally healthy subjects (Table
3.19). Post hoc tests revealed that the scores in the severe clinical attachment level
group were significantly higher than the other severity groups, while the
differences among the other severity groups were statistically insignificant.
Statistical analysis failed to detect any significant correlation between
scores of the LEQ and SRRS with CAL, nor between number of teeth present, Cl,
BOP, REC, PPD, and the various parameters of psychological stressors.
148

Table 3.18 Mean, SD, internal consistency, item-scale correlation and inter-scale correlations between the individual subscales of Measure
of Chronic Stress (Daily Strains).

Psychological Variables
No. of
items Mean SD
Internal consistency
(Cronbach's )
Item-scale correlation
coefficients
Correlation coefficients
with other subscales
Daily Strains
Job 19 2.38 0.81 0.94 0.78-0.82 0.13-0.20
a

Financial 9 2.13 0.70 0.93 0.80-0.88 0.12-0.21
a

Spouse 16 2.37 0.66 0.87 0.65-0.81 0.07-0.17
Being single 7 2.61 1.20 0.79 0.78-0.85 0.12-0.22
Children 33 2.54 1.08 0.89 0.81-0.85 0.17-0.22
Role strain composite 84 2.54 1.04 0.91 0.51-0.79 0.08-0.23


a
With the exception that the correction coefficient between Job and Financial subscales is 0.41, P = 0.016.


149

Table 3.19 Adjusted scores
a
of measures of psychosocial stressors by categories of clinical attachment level (CAL), n=1,000.



Clinical Attachment Level (Mean SE)
Post hoc

Psychological Scale Healthy Low Moderate High Severe
Statistics Significance analysis
b





LEQ 0.26 0.07 0.33 0.04 0.34 0.05 0.30 0.07 0.28 0.09 F = 0.41 P = 0.799





SRRS
c
80.57 7.42 88.23 4.44 92.15 5.36 96.81 7.11 91.91 9.19 F = 0.95 P = 0.434





Daily strains


Job
d

2.03 0.30 2.04 0.13 2.19 0.20 2.76 0.25 2.95 0.14 F = 4.69 P = 0.001
1, 2, 3, 4<5;

Financial
d

1.77 0.26 1.90 0.11 2.01 0.17 2.48 0.22 2.69 0.12 F = 5.04 P = 0.001
1, 2, 3, 4<5;

Spouse
2.68 0.22 2.34 0.11 2.70 0.11 2.64 0.18 2.51 0.21 F = 1.85 P = 0.117


Being single
2.74 0.28 2.68 0.15 2.60 0.19 2.73 0.27 2.80 0.31 F = 1.87 P = 0.113


children
2.99 0.44 2.81 0.25 3.48 0.40 2.73 0.52 3.12 0.34 F = 1.13 P = 0.340


Role strain composite
d

2.39 0.25 2.19 0.10 2.43 0.16 2.70 0.21 2.85 0.12 F = 4.17 P = 0.002
1, 2, 3, 4<5;





a
Adjusted for age (decades, please refer to Table 3.16), gender and levels of smoking (please refer to Table 3.7).
b
Post hoc analysis by Turkey HSD tests, groups 1, 2, 3, 4 and 5 refer to levels of clinical attachment loss from health to severe.
c
SRRS scores within the range of "low" stress (Holmes & Rahe, 1967).
d
Statisitcally significant differences in mean scores between the various severities of clinical attachment loss, P < 0.05, ANOVA.
150
3.3.5 Stress responses and periodontal status
Two sets of psychosocial questionnaires were used in measuring the stress
responses of the subjects; that is measuring the state reactions in face of stressors.
These included the Symptoms Checklist (SCL-90) Chinese version and
Depression Anxiety Stress Scale-State (DASS-S) Chinese Short version.
Table 3.20 shows the number of psychosocial items in subscales
comprising the SCL-90 and the DASS-S, the results of internal consistency of
items within each subscale, item-scale correlation and correlation with other
subscales. Besides the psychoticism subscale which had a Cronbachs alpha of
0.85, all other subscales of SCL-90 had coefficients over 0.93. Similarly, high
Cronbachs coefficients of 0.88 to 0.94 were recorded for all the subscales of
DASS-S. Item-scale correlation coefficients ranged from 0.77 to 0.89 and 0.75 to
0.93 for SCL-90 and DASS-S respectively. The inter-subscale correlations
between the subscales of SCL-90 and DASS-S ranged from 0.11 to 0.24 and 0.29
to 0.34 respectively. These results supported the reliability, and the convergent and
discriminant validity of the collected psychosocial data in this study sample. The
mean scores and SDs of the subscales of both the SCL-90 and DASS-S were
comparable with local norms (Wong, 1996; Tang et al., 1999; Wang et al., 2000).
The mean scores of the various subscales of SCL-90 and DASS-S in
subjects with different categories of clinical attachment level are shown in Table
3.21 after adjusting for age, gender and smoking. For the SCL-90, a statistically
significant difference was detected in the means scores of the Depression
subscale. Post hoc testes revealed that, for the Depression subscale, the mean
score of the severe CAL category was significantly higher than that of the other
severity groups and no significant difference was detected between the other
151
severity groups.
For DASS-S, statistical analysis detected significant difference only in the
mean scores of the Depression subscale. Post hoc tests revealed that the mean
scores of the Depression in the moderate, high and severe CAL categories were
significantly higher than that of the healthy and low CAL categories. No
difference was observed between the moderate, high and severe levels, or between
healthy and low CAL categories.

152

Table 3.20 Mean, SD, internal consistency, item-scale correlation and inter-scale correlations between the individual
subscales of SCL-90 and DASS-S.

Psychological Scales No. of items Mean SD
Internal Consistency
(Cronbach's a)
Item-scale correlation
coefficients
Correlation coefficients
with other subscales
Population
Norm
a

(Mean SD)
SCL-90
Somatization 12 8.40 6.50 0.93 0.79-0.82 0.11-0.16 8.9 7.6


Obsessive-compulsive 10 11.91 6.42 0.95 0.84-0.82 0.15-0.17 11.9 6.8


Interpersonal sensitivity 9 7.10 5.42 0.93 0.78-0.83 0.11-0.18 6.8 5.3


Depression 13 11.34 8.05 0.94 0.82-0.89 0.14-0.19 11.1 8.2


Anxiety 10 4.81 3.52 0.93 0.83-0.86 0.11-0.17 4.3 4.1


Hostility 6 3.72 3.24 0.93 0.84-0.84 0.19-0.23 4.5 3.9


Phobic sensitivity 7 3.21 3.37 0.97 0.84-0.83 0.20-0.24 2.6 3.0


Paranoid ideation 6 7.30 6.32 0.95 0.81-0.83 0.15-0.19 8.4 5.9


Psychoticism 10 5.81 5.42 0.85 0.77-0.87 0.11-0.19 5.6 5.6



DASS-S
Depression 7 6.34 6.18 0.94 0.75-0.91 0.31-0.33 5.7 6.4
Anxiety 7 6.88 5.56 0.88 0.85-0.93 0.29-0.34 6.6 5.3
Stress 7 13.04 7.00 0.92 0.84-0.92 0.32-0.34 12.0 7.0


a
Population norms for SCL-90 and STAI are from China Association of Mental Health (1993); population norms for DASS are from Wong (1996).

153

Table 3.21 Adjusted Scores
a
by levels of clinical attachment level, n=1,000.

Clinical Attachment Level (Mean SE)
Psychological Scale Healthy Low Moderate High Severe Statistics Significance Post hoc analysis
b

SCL-90
Somatization 3.64 2.48 7.37 1.05 7.14 1.63 7.10 2.09 9.58 1.17 F = 0.81 P = 0.521
Obsessive-compulsive 7.52 2.44 11.27 1.03 9.93 1.60 12.60 2.05 12.56 1.16 F = 2.08 P = 0.081
Interpersonal sensitivity 5.05 2.06 7.10 0.87 5.18 1.35 6.65 1.73 9.20 0.98 F = 0.83 P = 0.507
Depression
c
6.50 3.07 10.67 1.29 8.88 2.01 12.10 2.58 13.99 1.45 F = 2.43 P = 0.046 1, 2, 3, 4<5;
Anxiety 2.99 2.41 7.14 1.01 6.40 1.58 6.33 2.02 8.77 1.14 F = 0.67 P = 0.613
Hostility 3.02 1.24 4.35 0.52 4.00 0.81 3.15 1.04 4.81 0.59 F = 0.82 P = 0.515
Phobic sensitivity 1.85 1.28 3.17 0.54 2.07 0.84 3.94 1.08 4.59 0.61 F = 1.04 P = 0.384
Paranoid ideation 4.06 1.34 5.12 0.56 4.04 0.88 4.82 1.13 5.81 0.63 F = 1.03 P = 0.392
Psychoticism 3.32 2.07 5.76 0.87 4.78 1.36 4.75 1.74 7.77 0.98 F = 0.85 P = 0.494

DASS-S
Depression
c
4.51 2.27 4.64 0.96 5.57 1.49 6.13 1.91 10.00 1.08 F = 2.50 P = 0.046 1,2<3, 4, 5;
Anxiety 5.07 2.06 5.41 0.87 9.09 1.35 7.77 1.74 8.46 0.98 F = 2.13 P = 0.075
Stress 14.81 2.64 13.92 1.11 15.45 1.73 16.15 2.22 15.31 1.25 F = 0.64 P = 0.634


a
Adjusted for age (decades, please refer to Table 3.16), gender and levels of smoking (please refer to Table 3.7).
b
Post hoc analysis by Turkey HSD tests, groups 1, 2, 3, 4 and 5 refer to levels of clinical attachment loss from health to severe.
c
Statisitcally significant differences in mean scores between the various severities of clinical attachment loss, P < 0.05, ANOVA.

154
3.3.6 Dispositional factors, coping behaviours and periodontal status
Depression Anxiety Stress Scale-Trait (DASS-T) Chinese version was
used to measure the personality dispositions of the negative affectivities of
depression, anxiety and stress. The COPE inventory was employed to analyse the
pattern of coping behaviours in face of stressful experiences.
Table 3.22 shows the number of psychosocial items in subscales
comprising the DASS-T, the results of internal consistency of items within each
subscale, item-scale correlation and correlation between the DASS-T subscales.
High Cronbachs coefficients were recorded for all the subscales of DASS-T.
Similarly high item-scale correlation coefficients ranged from 0.76 to 0.90 were
observed for various sub-scales. Inter-subscale correlations between the subscales
of DASS-T ranged from 0.29 to 0.36. These results supported the reliability, and
the convergent and discriminant validity of the collected psychosocial data in this
study sample. The mean scores and SDs of both DASS-T were comparable with
local norms (Wong, 1996; Tang et al., 1999; Wang et al., 2000).
Factor analysis using oblimin rotation was carried out to extract the
second-order factors from among the COPE scales as suggested by Carver et al.
(1989) so as to determine the composition of the higher-order factors in this study
sample. The statistical results of factor analysis are shown in Table 3.23. A total of
three factors were obtained accounting for 73.2% of the total variance. The factor
loadings on factors 1 and 2 were all above 0.7 while that on factor 3 were above
0.4. These patterns of relationships suggested that the items in individual factors
clustered together with reasonably high correlation. The three factors so derived
included one focusing on problem solving, one with the theme of soliciting
emotional support and focusing on re-interpretation of event, and one concerning
155
emotion ventilation and avoidance (distraction and behavioural disengagement).
These findings were compatible with the contemporary conceptualization of
coping strategies (Lazarus, 1999, 2000). Three items of COPE were dropped after
factor analysis, namely religious coping, substance abuse and self-blame. The
final factor scores are shown in Table 3.24 comprising three factors, namely
Problem-focus coping, Emotion-focused coping, and Less adaptive coping.
The mean scores of the scale and various subscales of DASS-T and COPE
of subjects in different categories of clinical attachment level are shown in Table
3.25 after adjusting for age, gender and smoking. Among all the psychosocial
instruments measuring trait disposition and coping behaviours, statistical
significant difference (P < 0.05) were detected in the Depression trait and
Anxiety trait subscales of DASS-T, Problem-focused coping and
Emotion-focused coping of COPE. For the Depression trait subscale of
DASS-T, post hoc analysis revealed that scores of the subjects with healthy and
low categories of clinical attachment level were smaller than those of the
moderate category, which in turn was smaller than that of the high and severe
categories; while for the Anxiety trait subscale of DASS-T, mean scores of the
subjects with healthy and low categories of clinical attachment level were smaller
than that of the subjects with severe category. For Problem-focused coping
factor, means scores of the subjects with healthy, low and moderate categories of
clinical attachment level were greater than that of the subject with severe category,
while for Emotion-focused coping factor, means scores of the subjects with
healthy and low categories of clinical attachment level were smaller than that of
the subjects with high and severe categories.

156

Table 3.22 Mean, SD, internal consistency, item-scale correlation and inter-scale correlations between the individual
subscales of DASS-T and SDAxI.


Psychological Scales
No. of
items Mean SD
Internal Consistency
(Cronbach's a)
Item-scale correlation
coefficients
Correlation coefficients
with other subscales
Population Norm
a

(Mean SD)
DASS-T
Depression 14 5.76 6.05 0.95 0.76-0.88 0.31-0.35 5.6 6.0
Anxiety 14 6.41 5.26 0.94 0.84-0.89 0.33-0.36 6.2 4.8
Stress 14 12.46 6.70 0.94 0.86-0.90 0.29-0.32 10.8 6.1

SDAxI 9 15.81 5.51 0.81 0.91-0.94 15.2 6.0


a
Population norms for DASS are from Wong (1996); local reference data for SDAxI please refer to Table 3.4.

157
Table 3.23 Statistical results of factor analysis using the COPE subscales
as predictors.

Component
a

1 2 3
Initial Eigenvalues
Total Variance Explained 4.75853 3.12907 2.35601
% of Variance 34.0 22.4 16.8
Cumulative % 34.0 56.3 73.2

Rotated Component Matrix
b,c

Factor 1
Planning .88721
Active coping .87491
Use of instrumental social support .82931
Humour .82707 -.41571

Factor 2
Use of emotional support -.44937 .94813
Positive re-interpretation .93851
Acceptance .81115
Denial .79416

Factor 3
Focus on venting of emotions .66383
Distraction .44738
Behavioral disengagement .40206

Religious coping
Substance abuse
Self-blame

Correlation Matrix Factor 1 Factor 2 Factor 3
Factor 1 1.00000
Factor 2 -.18163 1.00000
Factor 3 .03963 -0.07416 1.00000


a
Extraction Method: Principal Component Analysis.
b
Rotation Method: Oblimin with Kaiser Normalization.
c
Suppress absolute value less than 0.3
158

Table 3.24 Mean and SD of COPE scales, n=1,000.

Psychological Scales Mean SD
COPE
Active coping 5.20 2.80
Planning 5.57 2.41
Use of instrumental social support 5.51 2.23
Humour 5.91 2.03
Factor 1 - Problem-focused coping 22.19 4.67

Use of emotional support 4.99 2.02
Positive re-interpretation 5.00 2.12
Acceptance 5.93 2.01
Denial 4.95 2.94
Factor 2 - Emotion-focused coping 20.87 4.15

Distraction 4.32 2.23
Focus on venting of emotions 2.10 0.52
Behavioral disengagement 3.42 1.82
Factor 3 - Less adaptive coping 8.16 3.05

Religious coping 2.64 1.33
Substance abuse 3.55 1.75
Self-blame 3.09 1.01








159
Table 3.25 Adjusted scores
a
of DASS-T, COPE and SDAxI by categories of clinical attachment level, n=1,000.

Clinical Attachment Level
b
(Mean SE)
Psychological Scale Healthy Low Moderate High Severe Statistics Significance Post hoc analysis
c

DASS-T
Depression
d
3.80 2.20 3.80 0.93 5.22 1.44 5.28 1.85 9.55 1.04 F = 2.53 P = 0.039 1, 2<3 <4, 5;
Anxiety
d
4.69 2.00 5.64 0.84 8.11 1.31 6.37 1.69 6.85 0.95 F = 2.43 P = 0.047 1, 2<5;
Stress 13.28 2.57 14.33 1.08 14.49 1.68 14.30 2.16 15.10 1.21 F = 0.13 P = 0.970

COPE
Problem-focused coping
d
23.58 1.76 23.29 0.74 24.88 1.15 21.59 1.48 19.45 0.83 F = 2.55 P = 0.038 1, 2, 3>5;
Emotion-focused coping
d
20.20 1.33 20.53 0.75 19.33 1.04 22.97 1.58 22.22 0.67 F = 2.53 P = 0.039 1, 2<4, 5;
Less adaptive coping 7.32 1.14 7.97 0.48 6.80 0.75 8.04 0.96 10.24 0.54 F = 1.72 P = 0.143

SDAxI
d
13.52 2.13 14.94 0.90 14.91 1.39 16.38 1.79 15.85 1.01 F = 2.81 P = 0.025 1<4;


a
Adjusted for age (decades, please refer to Table 3.16), gender and levels of smoking (please refer to Table 3.7).
b
CAL categories: healthy 0 to 1.0 mm; low 1.1 to 2.0 mm; moderate 2.1 to 3.0 mm; high 3.1 to 4.0 mm; and severe above 4.0 mm
(Genco et al., 1999).
c
Post hoc analysis by Turkey HSD tests, groups 1, 2, 3, 4 and 5 refer to levels of clinical attachment loss from health to severe.
d
Statisitcally significant differences in mean scores between the various severities of clinical attachment loss, P < 0.05, ANOVA.
160
3.3.7 Dental anxiety and periodontal status
Dental anxiety, being a specific personality trait about ones propensity to
developing anxiety towards dental care, is a complicated phenomenon and its
multifactorial nature is very often undermined in its measuring instruments
(Stouthard et al., 1993; Schuurs & Hoogstraten, 1993). The construct of dental
anxiety was measured by a comprehensive dental anxiety scale.
The Chinese Dental Anxiety Inventory Short version (SDAxI) was used to
measure the trait of dental anxiety. High Cronbachs coefficient of 0.81 and high
item-scale correlation coefficient ranged from 0.91 to 0.94 were observed for
SDAxI (Table 3.22). The mean scores and SDs of SDAxI (15.81 5.51), with a
range from 9 to 39, were comparable with local reference data (Table 3.4). The
mean scores of SDAxI in different severities of clinical attachment level are
shown in Table 3.25 after adjusting for age, gender and smoking. The mean score
was highest in subjects of high category of CAL, and in decending order severe,
low, and moderate, with the lowest score in the healthy group. A statistically
significant difference (P < 0.05) however was only detected between subjects of
the high and healthy CAL categories. There was statistically significant negative
correlation between dental anxiety and number of teeth present (r
xy
= - 0.21, P
= 0.027).

3.3.8 Risk evaluation and interaction of the significant demographic, social
and psychosocial variables on periodontal status
The degree of association between the outcome variable of clinical
attachment level and potential explanatory variables, including demographic
characteristics, systemic diseases, smoking, occupational hazards, drinking habits
161
and psychosocial variables, were measured using the logistic regression model.
Age (in decades, Table 3.16), gender and smoking levels (Table 3.7) and three
systemic diseases including diabetes, allergy and anaemia were entered into the
regression model. Variables with P < 0.10 were then included for analysis in the
model in a stepwise manner.
The results of the logistic regression are shown in Table 3.26. Males had
higher odds of 1.27 (95% CI: 1.05-1.65) for high/severe CAL than females. Age
was positively associated with CAL, the odds ratios ranged from 2.24 (95% CI:
1.05-3.87) for those aged 35 to 44, to 4.07 (95% CI: 2.89-5.81) for those aged 55
to 64 years as compared to the younger age group of 25 to 34 years old. Education
was inversely associated with CAL giving an odds ratio of 0.75 (95% CI:
0.59-0.91).
For subjects with a history of diabetes, the odds for high/severe CAL was
more than twice that of subjects who did not have diabetes, with an odds ratio
2.15 (95% CI: 1.31-2.87). The odds for high/severe CAL in smokers increased
with increasing amounts of smoking. The light smoker, an individual with a
smoking history of 5.3 to 15 packyears, had an odds ratio of 2.33 (95% CI:
1.32-3.52) compared to non-smokers. The odds ratio for a moderate smoker, an
individual with a smoking history of 15.1 to 30.0 packyears, increased to 3.50
(95% CI: 2.50-4.92) while the heavy smoker, those who had smoked more than 30
packyears had an odds ratio of 4.61 (95% CI: 2.88-5.68), almost double that of the
light smoker. No association was detected between high/severe CAL and
non-smokers or very light smokers. Other systemic diseases, occupational hazards
and drinking habits were not significant variables in the model (Table 3.26).
High/severe CAL was significantly associated with job strain, financial
162
strain and depression, with odds ratios of 1.47 (95% CI: 1.21-2.01), 1.38 (95% CI:
1.13-1.71) and 1.41 (95% CI: 1.17-2.78) respectively. For dental anxiety, there
was a significant association with clinical attachement level, with odds ratios of
1.20 ( 95% CI = 1.10-1.50) (Table 3.26). Depression trait and anxiety trait were
found associated with high/severe CAL, with odds ratios 1.62 (95% CI: 1.15-2.35)
and 1.51 (95% CI: 1.09-2.72), respectively. Problem-focused coping was
significantly and inversely associated with high/severe CAL, with odds ratio of
0.85 (95% CI: 0.71-0.90) whereas emotion-focused coping was significantly
associated with clinical attachment loss, with odds ratio of 1.21 (95% CI:
1.09-1.73).
Subjects were stratified by median-split in accordance with their coping
styles and trait dispositions to further assess the risk differential for minimal
disease versus high/severe CAL between subjects with high and low
problem-focused coping, emotion-focused coping, depression disposition and
anxiety disposition (Table 3.27). Statistically significant differences between the
high and low level groups were detected in the respective disposition and
coping variables after the median-split stratification.
Result of analysis of logistic regression according to the various
dichotomized variables, controlling for age and gender and smoking, is shown on
Table 3.28. Subjects with high trait depression and more job strain or financial
strain had a greater odds for high/severe CAL, with OR of 2.12 (95% CI:
1.36-3.06) and 1.97 (95% CI: 1.19-3.21), respectively. Similar findings were
detected for high anxiety trait in subjects with more job strain or financial strain,
with OR of 2.27 (95% CI: 1.65-2.98) and 2.03 (95% CI: 1.69-2.96), respectively.
Odds for high/severe CAL was also observed in subjects with more job strain or
163
financial strain and low problem-focused coping, with OR of 2.94 (95% CI:
2.21-3.88) and 2.33 (95% CI: 1.68-2.93), respectively. The picture was similar in
subjects with high emotion-focused coping and more job strain or financial strain,
with odds ratios of 2.96 (95% CI: 2.12-3.96) and 2.42 (95% CI: 1.64-2.93),
respectively.
The interaction of trait dispositions and coping styles, with job and
financial strains in risk evaluation of periodontal attachment loss was graphically
depicted in Figure 3.2. It can be seen that the odds for high/severe CAL for all
subjects is greater in those with high levels of job strain or financial strain. Those
scoring high on depression trait, anxiety trait or emotional-focused coping (poor
coping), or those scoring low on problem-focused coping (good coping) are at
even greater odds for periodontal destruction. On the contrary, those subjects with
high job or financial stain scoring low on depression trait, anxiety trait or
emotion-focused coping (poor coping), or those scoring high on problem-focused
coping (good coping) are at no more odds for periodontal attachment loss than
those who report little or no job or financial strains, respectively.

164

Table 3.26 Stepwise logistic regression analysis of potential indicators for
clinical attachment levels
a
.


Estimated
Odds Ratio
b
95% Confidence
Interval
Heavy Smoker
c
4.61 2.88 5.68
Age 55-64 4.07 2.89 5.81
Age 45-54 3.50 2.50 4.92
Moderate smoker
c
2.69 1.39 4.31
Light smoker
c
2.33 1.32 3.52
Age 35-44 2.24 1.05 3.87
Diabetes 2.15 1.31 2.87
Depression trait 1.62 1.15 2.35
Anxiety trait 1.51 1.09 2.72
Job strain 1.47 1.21 2.01
Depression (SCL-90) 1.41 1.17 2.78
Finacial strain 1.38 1.13 1.71
Gender (male) 1.27 1.05 1.65
Emotion-focused coping 1.21 1.09 1.73
Dental Anxiety (SDAxI) 1.20 1.10 1.50
Problem-focused coping 0.85 0.71 0.90
Allergy 0.77 0.58 0.96
Education 0.75 0.59 0.91


a
n = 767; dichotomized clinical attachment levels: 0= healthy/low mean CAL categories,
1 = high/severe mean CAL categories; please refer to Table 3.16 for CAL categories
classification (Genco et al., 1999).
b
Statistically significant (P < 0.05).
c
Light smoker: 5.3 to 15.0 pack-years, moderate smoker: 15.1 to 30.0 pack-years, heavy
smoker: more than 30.0 pack-years (Grossi et al., 1994).


165
Table 3.27 Statistics of subjects stratified according to anxiety and depression dispositions, and coping styles
a
.

High
a
Low
a

Mean SD Mean SD t-statistics P - value
Depression - Trait 9.48 5.07 2.17 1.09 31.96 < 0.001
Anxiety - Trait 9.99 4.32 2.19 1.91 32.28 < 0.001
Problem-focused coping 26.19 1.10 18.20 3.22 45.89 < 0.001
Emotion-focused coping 17.51 2.52 24.24 2.60 36.46 < 0.001


a
n = 767; dichotomized clinical attachment levels: 0 = healthy/low mean CAL categories, 1 = high/severe mean CAL categories,
please refer to Table 3.16 for CAL categories classification (Genco et al., 1999); subjects were stratified into "High" and "Low"
trait depression, trait anxiety, problem-focused coping, or emotion-focused coping groups by median-split (MacCallum et al.,
2002); trait dispositions measured by The Depression Anxiety Stress Scale - Trait (Lovibond & Lovibond, 1995a & 1995b);
coping styles measured by The COPE Inventory (Carver et al., 1989).




166

Table 3.28 Interaction of trait dispositions and coping styles with daily strains in risk evaluation of periodontal attachment level
a
.

Depression (Trait) Anxiety (Trait) Problem-focused coping Emotion -focused coping
Low High Low High Low High Low High

Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
Odds
Ratio 95% C.I.
LEQ
1.18 0.59 - 1.79 1.32 0.81 - 2.17 1.07 0.63 - 1.24 1.38 0.98 - 1.77 2.27 0.66 - 2.83 1.45 0.76 - 2.17 2.39 0.68 - 2.90 1.45 0.75 - 2.25


SRRS
0.88 0.82 - 3.11 1.56 0.52 - 2.03 1.12 0.88 - 2.24 2.30 0.85 - 2.89 1.60 0.99 - 3.02 1.07 0.47 - 1.93 1.77 0.16 - 3.09 1.22 0.43 - 1.92


Daily strains

Job
1.77 0.35 - 3.78 2.12 b 1.36 - 3.06 1.95 0.14 - 3.18 2.27 b 1.65 - 2.98 2.94 b 2.21 - 3.88 2.15 0.28 - 3.09 2.10 0.23 - 3.06 2.96 b 2.12 - 3.96
Financial
1.62 0.22 - 2.87 1.97 b 1.19 - 3.21 1.83 0.24 - 3.06 2.03 b 1.69 - 2.96 2.33 b 1.68 - 2.93 2.01 0.23 - 3.27 2.05 0.32 - 3.24 2.42 b 1.64 - 2.93
Spouse
1.28 0.78 - 2.50 1.65 0.88 - 2.78 0.81 0.67 - 1.99 1.02 0.80 - 1.96 1.81 0.80 - 2.83 1.35 0.87 - 2.56 1.41 0.93 - 2.63 1.75 0.72 - 2.80
Being single
0.98 0.80 - 1.90 1.28 0.97 - 2.81 1.51 0.15 - 1.84 1.57 0.92 - 2.94 1.15 0.31 - 2.76 0.89 0.79 - 1.89 1.03 0.80 - 1.82 1.28 0.42 - 2.67
children
0.87 0.54 - 1.75 1.07 0.78 - 3.11 1.81 0.92 - 2.91 1.99 0.59 - 3.97 1.01 0.72 - 3.05 0.69 0.64 - 1.84 0.86 0.72 - 1.91 1.15 0.69 - 2.95
Role strain
1.33 0.88 - 2.17 1.51 0.98 - 2.33 1.63 0.84 - 3.29 1.83 0.56 - 2.75 1.42 0.96 - 2.17 1.31 0.94 - 2.36 1.25 0.87 - 2.24 1.34 b 0.88 - 2.38



a
n = 767; dichotomized clinical attachment levels: 0 = healthy/low mean CAL categories, 1 = high/severe mean CAL categories, please refer to
Table 3.16 for CAL categories classification (Genco et al., 1999); subjects were stratified into "High" and "Low" trait depression, trait anxiety,
problem-focused coping, emotion-focused coping groups by median-split (MacCallum et al., 2002); all models were adjusted for age (decades,
please refer to Table 3.16), gender and levels of smoking (please refer to Table 3.7).


b
Statistically significant (P < 0.05) difference between "High" and "Low" groups .
167

Figure 3.2 Analysis of risk differential between minimal versus established
periodontal disease and (A) Job strain, or (B) Financial strain in subjects
with high/low psychosocial stress modulating and mediating abilities [Odds
ratio (y-axis) and 95% confident intervals (bars)].
Included are subjects with minimal periodontal disease, i.e. healthy/low mean
CAL categories (n = 615) or established periodontal disease, i.e. high/severe mean
CAL categories (n = 152). Subjects (n = 767, all subjects) were stratified into high
or low trait depression/anxiety, problem-/emotion-focused copers by median-split
(MacCallum et al., 2002). All models were adjusted for age (decades, please refer
to Table 3.16), gender and smoking levels (please refer to Table 3.7). Black circles
indicate statistically significant risk for established periodontal disease. Please
refer to Table 7 for mean full mouth CAL categories classification.
168
3.3.9 Oral-health related quality of life and periodontal status
Of the 767 subjects selected from the cross-sectional sample, 727 subjects
responded and completed the questionnaires (Table 3.3). The distribution of
responses according to items of OHIP-14S is shown in Table 3.29. The impact of
oral health on the life quality of the patients was considerable in terms of causing
functional limitation, physical pain, and physical disability. More than one-tenth
of the subjects perceived that they had functional limitation, physical pain or
disability fairly or very often. In other words, they had difficulty chewing, found it
uncomfortable to eat, or could not taste their food properly, because of problems
with their teeth, mouth or dentures. The prevalence of negative impact on
psychological domains (discomfort and disability) varied between 4.0% to 6.3%.
The impact on the domains of social disability and handicap was less prevalent.
Distributions of subjects with respect to the OHIP-14S and individual
subscales were skewed with more subjects having lower score (Table 3.29). The
mean scores and internal consistency for the OHIP-14S and individual subscales
are shown in Table 3.30. Cronbachs alpha varied from 0.73 to 0.94.
Subjects oral health-related quality of life was associated with the
self-reported periodontal symptoms over the past 12 months. About one-sixth of
the subjects reported having symptoms of sore gums and receding gums. Only a
small number of subjects reported drifting teeth (less than 10%) (Table 3.31). The
OHIP-14S and subscales scores were significantly associated with the occurences
of swollen gums, sore gums, receding gums, loose teeth, bad breath and toothache
in the previous year (Table 3.31). The experience of drifting teeth was not
significant.
A statistically significant correlation was detected between the OHIP-14S
169
score with education (Pearson Correlation Coefficient = -0.23, P < 0.001) and
number of teeth present (Pearson Correlation Coefficient = -0.45, P < 0.001), but
no significant correlation was detected in respect of gender, income, number of
anterior teeth or occluding pairs. A comparison of the mean scores of the
OHIP-14S and individual subscales between the subjects of healthy/low
periodontal attachment level group versus high/severe periodontal attachment
level group after adjustment for the effects of education and number of teeth
present is shown in Table 3.32. The differences were significant in the total score
and the domains of functional limitation, physical pain, psychological discomfort,
physical disability, and psychological disability. The differences in social
disability and handicap subscales were not significant.


170
Table 3.29 Distribution of OHIP-14S individual items response
a
.

never
hardly
ever occasionally
Fairly/very
often
b

n % n % n % n %
Functional limitation
Difficulty chewing 253 34.8 252 34.7 98 13.5 124 17.1
Trouble pronouncing words 276 38.0 339 46.6 65 8.9 47 6.5
Physical pain
Uncomfortable to eat 252 34.7 262 36.0 130 17.9 83 11.4
Sore spots 376 51.7 203 27.9 116 16.0 32 4.4
Psychological discomfort
Worried 470 64.6 197 27.1 26 3.6 34 4.7
Miserable 505 69.5 105 14.4 71 9.8 46 6.3
Physical disability
Less flavour in food 406 55.8 151 20.8 63 8.7 107 14.7
Interrupt meals 409 56.3 183 25.2 77 10.6 58 8.0
Psychological disability
Upset 432 59.4 172 23.7 91 12.5 32 4.4
Been embarrassed 426 58.6 179 24.6 93 12.8 29 4.0
Social disability
Avoid going out 620 85.3 80 11.0 20 2.8 7 1.0

Trouble getting on with
others
632 86.9 70 9.6 16 2.2 9 1.2
Handicap
Unable to function 657 90.4 50 6.9 16 2.2 4 0.6
Unable to work 658 90.5 37 5.1 18 2.5 14 1.9

a
n=727; 767 subjects were selected from upper and lower ends of the spectrum of
periodontal attachment level, with either a mean full mouth CAL of 2mm (healthy/low
periodontal attachment level group), or > 3mm (high/severe periodontal attachment level
group), please refer to Table 3.16 for CAL categorization (Genco et al., 1999), 727
subjects responded and completed the questionnaires.
b
subjects with negative impacts.


171
Table 3.30 Mean scores and internal consistency for OHIP-14S and
individual subscales
a
.


Mean Scores
(SD) Range
Internal Consistency
(Cronbach's )
Functional limitation 1.99 1.92 0 - 8 0.77
Physical pain 1.82 1.88 0 - 8 0.88
Psychological discomfort 1.02 1.69 0 - 8 0.88
Physical disability 1.54 2.08 0 - 8 0.79
Psychological disability 1.25 1.76 0 - 8 0.85
Social disability 0.38 1.09 0 - 8 0.81
Handicap 0.29 1.01 0 - 7 0.73
OHIP-14S 8.31 10.76 0 - 55 0.94

a
n=727; 767 subjects were selected from upper and lower ends of the spectrum of
periodontal attachment level, with either a mean full mouth CAL of 2mm (healthy/low
periodontal attachment level group), or > 3mm (high/severe periodontal attachment level
group), please refer to Table 3.16 for CAL categorization (Genco et al., 1999), 727
subjects responded and completed the questionnaires.


172
Table 3.31 Comparison of OHIP-14S scores of individual self-reported
symptoms of periodontal disease
a
.

n
OHIP-14S scores
(Mean SD) P value
b
Swollen gums
yes 67 12.82 5.09 0.008
no 660 7.85 7.07
Sore gums
yes 118 12.28 4.97 0.007
no 609 7.54 6.97
Receding gums
yes 125 12.72 6.66 0.007
no 602 7.39 6.49
Loose teeth
yes 84 13.77 5.77 0.009
no 643 7.60 6.54
Drifting teeth
yes 44 14.57 6.18 0.083
no 683 7.91 6.31
Bad breath
yes 67 15.52 5.40 0.009
no 660 7.58 6.75
Toothache
yes 95 9.83 5.36 0.007
no 632 8.08 6.83

a
n=727; 767 subjects were selected from upper and lower ends of the spectrum of
periodontal attachment level, with either a mean full mouth CAL of 2mm (healthy/low
periodontal attachment level group), or > 3mm (high/severe periodontal attachment level
group), please refer to Table 3.16 for CAL categorization (Genco et al., 1999), 727
subjects responded and completed the questionnaires.
b
t test.


173
Table 3.32 Unadjusted scores (Mean SD) and adjusted scores (Mean SE) of OHIP-14S and individual subscales of subjects in
the two levels of CAL severity
a
.

Periodontal attachment level
Unadjusted Scores
(Mean SD)
Adjusted Scores
(Mean SE)

Healthy/Low High/Severe Healthy/Low High/Severe Statistics
(n = 584) (n = 143) (n = 584) (n = 143) F
b
P- value
Functional limitation 1.22 0.51 5.14 1.42 1.11 0.44 5.41 1.24 6.72 0.010
Physical pain 1.04 0.53 4.76 1.27 0.96 0.43 4.86 1.26 6.13 0.014
Psychological discomfort 0.35 0.39 3.78 1.23 0.32 0.37 3.95 1.12 4.36 0.037
Physical disability 0.64 0.57 5.23 1.28 0.59 0.43 5.33 1.23 5.43 0.020
Psychological disability 0.50 0.51 4.32 1.20 0.45 0.40 4.38 1.09 4.28 0.039
Social disability 0.37 0.74 0.42 0.59 0.35 0.73 0.45 0.57 2.22 0.137
Handicap 0.28 0.60 0.33 0.52 0.26 0.56 0.34 0.51 2.32 0.128
OHIP-14S 4.41 2.74 24.19 7.04 3.78 2.25 25.09 5.94 4.24 0.036


a
n=727; 767 subjects were selected from upper and lower ends of the spectrum of periodontal attachment level, with either a mean full mouth CAL of
2mm (healthy/low periodontal attachment level group), or > 3mm (high/severe periodontal attachment level group), please refer to Table 3.16 for CAL
categorization (Genco et al., 1999), 727 subjects responded and completed the questionnaires.
b
Adjusted for age(decades, please refer to Table 3.16), education (please refer to Table 3.2) and number of teeth; ANOVA.
174
3.3.10 Oral-health related quality of life and concurrent psychosocial factors
Preliminary investigation of the relationships between oral-health related
quality of life and concurrent psychosocial factors revealed statistical significant
correlation between OHIP-14S scores with DAxI scores (r
xy
= 0.33, P = 0.008).
OHIP-14S however did not give any significant association with other
psychosocial stress variables including stressors, stress responses, coping, and
personality traits.
175
Chapter 4
Discussion

4.1 Current results and implications
The present study aimed to explore how the stress process affects
periodontal status and, subsequently, the impact of periodontal status on oral
health-related quality of life (Figure 1.4). This study investigated the effects of
acute and chronic psychosocial stressors and the various stress responses, and the
moderating effects of personality traits and coping behaviours on periodontal
health in a community survey study in which potential confounding factors such
as age, gender, demographic and socioeconomic factors, smoking, and general
health were controlled. It is known that the stress process is a dynamic and phasic
phenomenon regulated by complex interactions among different component
aspects. Adequate stress mediation, for example, may decrease physiological and
psychological stress response and, at the same time, help reduce the impact of a
stressor through effective management of the problem or successful
accomplishment of the stressful task. Multiple aspects of the stress process, with
reference to the contemporary understanding of stress and emotions, were
assessed simultaneously as there is no meaningful single composite index to study
stress. These included the stressors acute and chonic ones, stress response
physiological and psychological ones, moderating and mediating factors of coping
and personality traits. Psychological instruments with valid and precise
operationalization of constructs, such as coping, depression and anxiety, were
used.
176
The relation between dental anxiety, being a specific dentally relevant
personality disposition, and periodontal status as well as the impact of periodontal
health on oral-health related quality of life were also examined.

4.1.1 Job strain and financial strain
Results from the multivariate analysis demonstrated the significant
association of job strain and financial strain with periodontal disease, which
remained significant after controlling for the various established risk factors of
age, gender, and smoking (Table 3.19). The two scales in measuring job strain and
financial strain are parts of the Measures of Chronic Stress (Daily Strains),
adapted from the Problems of Everyday Living Scale of Pearlin and Schooler
(1978), assessing chronic stressors associated with the central roles that people
engage in daily life. These findings are consistent with the chronic stress model
that chronic stressors were associated with suppression of immune system as a
whole predisposing individuals towards chronic inflammatory reactions
(Segerstrom & Miller, 2004).
Subjects with more severe attachment loss showed significantly more job
strain and financial strain scores than healthy individuals (P = 0.001). Subjects
with health, low, moderate, and high levels of CAL/attachment loss had mean job
strain scores of 2.03, 2.04, 2.19, and 2.76, respectively, compared to 2.95 recorded
from the severe CAL/attachment subjects. Subjects with health, low, moderate,
and high levels of CAL/attachment loss had mean financial strain scores of 1.77,
1.90, 2.01, and 2.48, respectively, compared to 2.69 in the severe subjects. (Table
3.19).
These two particular measures evaluate the role of an individual as worker
177
and as financial manager. The questions asked assess chronic and long-term status
rather than transient and acute stress. Examples of these questions are: Do you
have more work than you can handle? Do you work too many hours? Is the
income I earn just about right for the job I have? Can I count on a steady income?
Is there always a chance I may be out of a job? Do people act toward you as if you
are a person without real feelings? Are you told that you're doing a good job? At
the present time are you able to afford a home that is large enough? In general,
how do your (you/your family's) finances usually work out at the end of the month?
How often does it happen that you don't have enough money to afford the leisure
activities that you/your family want(s)?
In summary, these questions likely elicit a response representative of
chronic, persistent and recurrent daily strain with the concomitant of long lasting
and chronic stress. This chronic stress may lead to adverse effects on immune
response. It is also possible that those under job strain or financial strain may visit
the dentist less frequently or only when symptom arises, and thus have poorer
dental care and more disease (Genco, 1999). A longitudinal study relating the
temporal relationship of stressors such as job and financial strains to periodontal
disease onset and progression, however, is necessary to establish whether these
are true risk factors for periodontal disease.
Previous studies (Gardell, 1971, 1982a, 1982b) suggested that important
job stressors include high mental demands, excessive work and time pressure,
under-stimulation, under-utilization of skills, and lack of novelty; these were
included in the job-related questions employed in the present study. A previous
study also showed that an increase in probing depth was significantly predicted by
job stress (Freeman & Goss, 1993). Dorian et al. (1985) demonstrated an increase
178
in immunological defense at the time of peak stress in a study of chronic work
stress in accountants, followed by suppression during the post-stress period as
reflected in the immunologic parameters of interleukin generation, interleukin
responsiveness, NK cell activity, and lymphocyte reactivity to
phytohemagglutinin (PHA). A study in Belo Horizonte, Brazil (Marcenes &
Sheiham, 1992) showed a significant association between poor periodontal status
and high work mental demand. Genco et al. (1999), on the other hand, reported
that the likelihood of developing periodontal disease was significantly increased
in subjects with high levels of financial stress with poor coping abilities after
accounting for other risk factors such as age, gender, smoking, poor dental care
and diabetes. The present findings appear consistent with all these studies.
In the present study, both job strain and financial strain were found to be
significantly associated with periodontal disease. At the same time, there existed a
rather appreciable correlation between job strain and financial strain (r
xy
= 0.41, P
= 0.016) (Table 3.18). This statistical co-linearity may be explained by the job
attitude and the social characteristics of the local population. Surveys in 2004
revealed that Hong Kong, well-known for its capitalistic context, had retained the
highest rating worldwide for economic freedom, and was the 5
th
most expensive
city with respect to cost of living and at the 6
th
position on the world
competitiveness scoreboard (IMD, 2004; Gwartney & Lawson, 2004; Mercer
Human Resource Consulting, 2004). The pressure and stress of maintaining an
adequate standard of living in such society is tremendous; and, people are used to
considering the utility purpose of their job as of paramount importance in their life.
Financial and material rewards from the job are usually carefully evaluated, while
issues of interest and aptitude are usually assigned a less significant rating. Issues
179
of job and finance hence very often associated closely together. Thus the author
was not surprised finding that job and finance strains were closely associated.
Stress as measured by LEQ and SRRS, for stressors of less chronic nature,
was not found to have any significant correlation with CAL. These observations
appeared consistent with the nature of periodontal disease as being a chronic and
usually slow progressing inflammatory disease. In contrast to some of the
previous studies (Ballieux, 1991; Monteiro da Silva et al., 1995 & 1996; Beck,
1996; Brevik et al., 1996; Linden et al., 1996; Axtelius et al., 1997a, 1997b &
1998; Page & Beck, 1997; Salvi et al., 1997; Genco et al., 1998; Albandar, 2000;
Hildebrand et al., 2000) which had attempted to investigate individual
psychological variables in the stress process, and/or with limited sample size
suggesting a positive association between acute stressor(s) and periodontal status
(Table 1.3), the present findings remained consistent with an earlier population
study with the inclusion of the systematic variables of the stress process and
adjustment for possible confounding factors (Genco, 1999).

4.1.2 Coping
Both Problem-focused coping and Emotion-focused coping were
found to have a significant association with the severity of CAL.
Problem-focused coping appeared to reduce the odds of periodontal disease
(OR = 0.85, 95% CI = 0.71 to 0.90) while Emotion-focused coping increase the
odds (OR = 1.21, 95% CI = 1.09 to 1.73) (Table 3.26). This suggested that
adequate coping strategies (problem-focused) exert a favourable effect on the
periodontal status while inadequate or less desirable ones (emotion-focused) give
an adverse effect.
180
Coping has to do with the way people manage life conditions that are
stressful. It is what we think and do in an effort to manage stress and the emotions
associated with it, whether or not these efforts are successful. Emotion-focused
coping (as reflected in the items in Tables 3.24 and Appendix 2) is aimed at
managing or regulating the emotions tied to the stressful situation without
changing it. These emotion-focused coping strategies include strategies such as
admitting that one cannot deal with it and quit trying; and pretending that the
problem hasnt really happened. The theme of emotion-focused coping is more on
avoidance by accepting the problem with re-interpretation or denial and such
strategies are likely to be less effective and maladaptive in dealing with stress. On
the other hand, problem-focused coping is directed at mounting an action to
change a troubling situation. This action entails problem solving. One checks out
what is going on, obtains information and appraise the situation as amenable to
preventive or corrective actions, solicits resources and mobilizes actions for the
purpose of changing the reality of the troubled situation.
Emotion-focused coping, which included seeking social and emotional
support, has some positive effects in relieving the stress perception. For example,
worries may appear less daunting for a time, it however remains a weak and
temporary coping strategy as the stress does not go away until the problems are
actually confronted or resolved. This may also explain why those with high
problem-focused coping behaviour are more effective in stress management as the
coping actions are directed at either the self or the environment aiming at
resolving the trouble (Carver et al., 1989; Lazarus, 1993a, 1999; Carver, 1997).
To illustrate with one who has found a tumour mass suspected of malignancy.
Instead of merely sitting down, soliciting emotional support, ventilation,
181
comforting or re-interpreting the event as less threatening, a problem-focused
coper actively strives to address the problem, manage and prepare for the
uncertainties and complications. He seeks the opinions of different medical
specialists about the nature of it and what treatment to select and which surgeon is
the best available. He not only makes a plan of action, but actually carries out the
planned actions to try to get rid of the problem. People may argue that
emotion-focused coping is more desirable, say, when something cannot be
changed and has to be accepted. However, as advocated by the
cognitive-behavioural paradigm in contemporary psychotherapy, one still has to
confront the unchangeable and derive personal meaning about that before one is
able to survive the challenge adaptively with the unchangeable reality and without
significant psychological distress (Beck, 1995; McMullin, 2000; Simos, 2002).
Dispositional maladaptive and ineffective coping strategies usually result
in frequent or chronic state of hardship and tension (Master & Gershman, 1983;
Haaga & Davison, 1993; Lazarus, 1993a, 1999; Lazarus & Lazarus, 1994; Ellis,
1997). As the distinguished stress physiologist, Hans Selye, showed (Selye, 1956
& 1982) and plenty of study findings that followed (Weiten, 2004), the stress
responses of sympathetic nervous system and hormonal components may lead to
compromised functioning of the immune system and hence the defense against
virulent or opportunistic pathogens. Extensive research by James Pennebaker and
his associates (1989, 1990) also strongly suggested that coping with stress is
facilitated by confronting and working through the threats they produce. This may
also explain why problem-focused coping is often associated with high levels of
wellbeing (Hynes et al., 1992).

182
4.1.3 Traits of anxiety and depression
Subjests high in either anxiety trait or depression trait had higher odds for
periodontal disease in the present study. Anxiety and depression personality
predispositions appeared to play unfavourable roles towards periodontal health.
In other words, subjects who were trait anxious or trait depressive were more
vulnerable to periodontal disease as measured by clinical attachment level.
It has long been recognized that people differ widely in their susceptibility
to anxiety or depression states. In an attempt to clarify and stimulate research in
this area, Spielberger (1966, 1972) advocated the well known distinction between
state and trait anxiety (Cattell & Scheir, 1958; 1961). State anxiety is viewed as a
transient condition of subjective feelings of tension, apprehension and increased
autonomic activity, while trait anxiety is viewed as a relatively stable individual
proneness to anxiety, or a tendency to respond to situations with characteristic
levels of state anxiety. Traits of anxiety and depression, as personality variables,
are therefore not directly manifested in behaviour, but may be inferred from the
frequency and intensity that a person experiences the respective affective state
over time (Hjelle & Ziegler, 1992; Weiten, 2004). This dispositional factor
characterizes affective responses over long periods (months or years), is
associated with increased vulnerability to stress (Costa & McCrae, 1985). For
example, some individuals may have such high levels of trait anxiety that they are
chronically in a generalized state of anxiety.
An issue that has emerged as of central importance in stress research is
that of vulnerability. The major factor associated with the onset of most cases of
anxiety or depressive disorders is the occurrence of stressful life events (cf.
Brown & Harris, 1978; Dohrenwend & Dohrenwend, 1974; Endler & Edward,
183
1982; McKeon et al., 1984; Miller, 1989). However, the fact that only a minority
of those who experience such events become clinically anxious suggests that the
events only have severe consequences for individuals who are in some way
vulnerable. Traditionally, the personality dimension most often considered to be a
vulnerability factor for psychological problem is almost certainly neuroticism,
which is said to reflect individual differences in the strength and reactivity of
emotional responses (Eysenck, 1967, 1980, 1989; Eysenck & Eysenck, 1985;
Eysenck & Mathews, 1987). Neuroticism (N) scores as measured by the Eysenck
Personality Inventory (EPI; Eysenck & Eysenck, 1965) or the Eysenck
Personality Questionnaire (EPQ; Eysenck & Eysenck, 1975) have been found to
be elevated in a variety of psychological conditions including both anxiety and
depression (Eysenck, 1967; Eysenck & Mathews, 1987). Recently, it was
suggested that trait anxiety could possibly be a vulnerability factor which
predisposes individuals to develop clinical anxiety (e.g. Turner & Michelson,
1984; Eysenck & Mathews 1987; Chan & Lovibond, 1996). According to
Spielberger (1985), people who are high in trait anxiety as measured by the
State-Trait Anxiety Inventory (STAI; Spielberger et al., 1970) are more
vulnerable to stress and respond to a wider range of situations as dangerous or
threatening. Individuals high in trait negative affect are more prone to
psychological symptoms of distress than persons scoring low on this dimension.
Trait negative affect is thought to play a role in the onset and progression
of physical disease and compromised immune function (Herbert & Cohen, 1993;
Watson & Pennebaker, 1989). Findings in the present study appear compatible
with existing evidence that high anxiety or depression trait render these subjects
more susceptible to stressful status, more vulnerable in developing stress reactions
184
and in turn adverse effects on the immune response resulting in reduced resistance
to periodontal disease.

4.1.4 Impact of dental anxiety on periodontal health
This is probably the first ever study focused on the relation between dental
anxiety and periodontal status in general population. Results of the present study
reflect a statistically significant association between dental anxiety and
periodontal status (OR 1.20, 95% CI: 1.10-1.50) (Table 3.26). Dental anxiety was
found to be statistically significant correlated with number of teeth present (r
xy
= -
0.21, P = 0.027). There exists no clinical norm or cut-off score of DAxI
established for the Hong Kong population. When one of the more stringent
statistical practice (2SD above mean score) was used, about 8.1% of the surveyed
subjects were classified as experiencing high levels of dental anxiety (c.f. 7% in
Ng et al., 2005).
In spite of research in the last few decades, several cardinal issues
concerning dental anxiety have not been answered. Little is known about the
etiological pathways which lead to the development of dental anxiety, its
incidence at different points in the life span, and its course over time. Some of the
studies investigated the relationship between dental anxiety and oral health status
with subjective indicators (e.g. Milgrom et al., 1988; Locker & Liddell, 1991).
Studies concerning the clinical impacts of dental anxiety on dental problems, such
as caries and periodontitis, as well as the treatment need were limited. Where
clinical indicators have been used, investigator have either employed specific
population of limited sample size (e.g. Hakeberg et al., 1993), failed to include
comparison group (e.g. Molin & Seeman, 1970; Berggren & Meynert, 1984), or
185
limited the study to differences in DMF status among subjects too young for
noticeable differences yet to have emerged (e.g. Cohen, 1985). Studies have
produced rather contradictory results, some reporting differences between those
who are and are not dentally anxious (e.g. Stouthard & Hoogstraten, 1990b; Bedi
et al., 1992; Locker & Liddell, 1992) and some noting few if any differences
between the two groups (e.g. Stabholz et al., 1999).
The present community study served admissible evidence to several issues
about dental anxiety in spite of the differences or contradictions mentioned above,
Firstly, dental anxiety contributes to poor periodontal health. This finding is
intuitively supported and consistent with human psychology in that the disposition
to experience anxiety in dental situations leads to avoidant behaviours and, in turn,
delays in receiving adequate preventive care or treatment. Furthermore, neglect of
oral health leads to increased pain, more stress and further avoidance. The vicious
cycle so established may be enhanced by the memories of past distressing
experiences (Eli, 1992). Secondly, the number of teeth present is inversely
correlated with dental anxiety as measured by different scales. Individuals with
high dental anxiety may suffer a compromised dental condition due to avoidance.
They may resort to more definitive and less complex treatment than root treatment,
crowns and bridge works when they have to seek treatment for pain, swelling, or
other condition. This is possibly due to the fact that they try to avoid staying too
long in the dental chair/clinic necessary for these treatments, and/or that the
dentists may not be prepared to treat these patients with high levels of dental
anxiety. This is consistent with the finding that individuals with high dental
anxiety not only have more missing teeth than those who are not anxious but also
have their dentitions largely unrestored (Locker & Liddell, 1992).
186
In summary, dental anxiety contributes to poor periodontal and dental
health as well as compromised oral-health related quality of life (McGrath & Bedi,
2004). In treatment of periodontal disease, especially refractory cases, dental
anxiety and subsequent avoidance and neglect behaviours should be properly
addressed so as to facilitate and enhance the intervention process and outcome. It
has also been generally accepted that dentally anxious individuals are not a
homogenous group and they differ in the origins and manifestations of their
anxiety about dental treatment (Milgrom, 1985; Thomson et al., 2000). Adequate
understanding of dental anxiety is necessary to allow dentists to provide
appropriate and effective treatment (Moore et al., 1993; Aartman et al., 2000;
Zuniga, 2000).

4.1.5 Interaction of strains, coping, and traits
Interesting relationships were found between the severity of periodontal
disease, job strain and financial strain, coping behaviours, and trait dispositions of
anxiety and depression (Table 3.28). Subjects with job strain or financial strain
who used more emotion-focused coping strategies had even more periodontal
disease. For those who used less emotion-focused coping strategies, despite the
high chronic strain in job or finance, had no more odds of periodontal disease than
those with low strain in job or finance. A similar yet reverse relation was detected
between job strain or financial strain, and periodontal disease for problem-focused
coping strategies. Job strain or financial strain added no more odds to periodontal
disease in subjects using more problem-focused coping strategies, while subjects
using less problem-focused coping had more periodontal disease.
In other words, adequate and adaptive coping behaviours, either low
187
emotion-focused coping or high problem-focused coping, with the chronic
stress resulted in little or no effect of the stress on periodontal status. Inadequate
coping, evidenced as either high emotion-focused coping or low
problem-focused coping, with the chronic stress lead to more severe periodontal
disease.
The present study also showed that individuals with more favourable
personality dispositions, that is, those with low scores of anxiety trait or
depression trait, had no more periodontal tissue destruction, even though they
reported high levels of job strain or financial strain. Conversely, those with high
levels of job strain or financial strain with less favourable personality dispositions,
evidenced as high scores of anxiety trait or depression trait, were found to have
even more severe periodontal destruction.
The Erie County study (Genco et al., 1999) demonstrated the same pattern
of interaction between financial strain and coping behaviours, and the effects of
stress on periodontal disease can be moderated by adequate coping behaviours.
The findings of present study added further the role of personality traits in
modifying the stress reaction. These interactions echo the contemporary
theoretical concept of coping strategies being the mediating processes and
personality traits being the moderating factors that determine how people react to
stressor (Lazarus, 1999, 2000). Mediating processes and moderating factors
appear to modulate impact of stressors and in turn individuals' stress responses.
To these ends, the possibility of employing psychological intervention as
adjunctive measure in treatment of periodontal disease would probably deserve
further evaluation.

188
4.1.6 Depression
Compared to the healthy subjects, there were trends towards more severe
psychological symptoms of depression in those with more severe clinical
attachment loss as measured by the SCL-90. Higher scores were observed on each
of the somatization, interpersonal sensitivity, depression, anxiety, hostility, phobic
sensitivity, paranoid ideation, and psychoticism sub-scales for those with more
periodontal disease. These 8 dimensions encompassed a wide range of symptoms
which may reflect mood disorders, negative affectivities, life events encountered
and impacts on them, and possible role changes of subjects. However, only
depression was statistically significantly related to periodontal disease after
adjusting for age, gender and smoking with an odds ratio of 1.41 (95% CI = 1.17
to 2.78).
Depression is often conceptualized as the result of a sense of hopelessness
regarding the issue of loss (Lazarus & Lazarus, 1994). A friend or loved one dies
and we grieve. A loved one leaves us and we hurt, we miss them and want them
back. We fail to reach some important goal and we cry. The loss could be a major
one such as the death of spouse, a less traumatic one as finding oneself stuck in a
job without any further career development, or an existential issue as feeling that
life as disorder and meaningless. The exact meaning of the loss varies from person
to person, varies with the time period following the loss and how the loss is
interpreted. While being emotional, depression is not a single emotion but a
complex emotional response to loss. It is a mixture of several emotions that come
and go depending on where one is in the process of grieving after the loss and
what has happened to produce the loss. The emotions of depression may consist of
anxiety, anger, guilt and shame.
189
The Depression scale of the SCL-90 assesses dysphoria, hopelessness,
devaluation of life, self-deprecation, lack of interest/involvement, anhedonia, and
inertia. Depression thus can be very distressing and, if not managed, may extend
its effect in establishing a vicious cycle making itself a role of stressor enhancing
the whole stress process adding further to the resulting psychological stress and
distress. Furthermore, the behavioural concomitants of depression can also be
maladaptive, predispose a person to compromised physical health, and contribute
indirectly to the susceptibility of various diseases, including periodontal disease.
Examples of these behaviours include withdrawal from social activities, poor self
care, increase tobacco smoking and/or alcohol consumption, decreased physical
activity, drug abuse, and eating disorder.
Depression is associated with increased risk for a variety of illnesses,
including viral illnesses, chronic and malignant diseases (Rimon et al., 1971;
Lycke et al., 1974; Cappel et al., 1978; Whitlock & Siskin, 1979). Studies of
patients with major affective disorders have shown defects in lymphocyte function
which may be indicative of an impairment in cell-mediated immunity (Kronfol et
al., 1983; Schleifer et al., 1984). Clinical depressive disorder is the affective
disorder which consistently demonstrated immunologic changes (Dorian et al.,
1982 & 1985; Dorian & Garfinkel, 1987; Gerra et al., 2003; Atanackovic et al.,
2004). This provided a possible explanation of depression as a significant risk
indicator in periodontal disease. Longitudinal studies on the temporal relationship
of negative affectivity, such as depression, with the onset and progression of
periodontal disease are necessary to evaluate the underlying biochemical
mechanism and whether depression is truly a risk factor for periodontal disease, or
whether depression is merely a complication of poor health associated with
190
periodontal disease.
The DASS-S scores showed also a significant correlation of depression
with periodontal status. This parameter, however, was excluded from the logistic
regression model for risk analysis of periodontal disease after adjusting for age,
gender and smoking. The odds ratio with respect to depression as measured by
DASS-S was 1.39 (95% CI = 1.09 to 1.98) with a marginally insignificant P of
0.057.
A possible explanation of the discrepancy between the results of the two
measures may be the slight difference in exploration of the depression signs and
symptoms. The SCL-90 is relatively a more clinical instrument than the DASS
which in turn aims to assess negative affectivities in a general population. After
all, management of depressive signs and symptoms may need to be assessed and
considered in relation to treatment of periodontal disease.

4.1.7 Life events and changes, and other chronic stressors
Major life events and changes, as measured by the LEQ and the SRRS,
made little impact on periodontal health in our population. These measures were
supposed to tap the experiences of changes. Many experts however have criticized
this group of assessments for their problems in methods and interpretation of
findings (Watson & Pennebaker, 1989; Brett et al., 1990; Raphael et al., 1991;
Rabkin, 1993). Undoubtedly, some life changes may be quite challenging, but
others may be quite benign. At present, there is little reason to believe that change
is inherently or inevitably stressful. These measures failed to address the issues
such as the desirability of a change and whether it is the frustration or the change
itself causes stress to a person. This inaccurate assessment of stressful impact may
191
account to some extent for these negative findings in the present study.
Other chronic stressors as measured by the Daily Strains scale included
those associated with the roles as spouse, parent, and being single. The daily
strains are considered as those minor but repeated or chronic daily stressors which,
in some studies, had been defined as significant and harmful to ones well-being.
However, the impact on somatic health may depend on how an individual adjusts
oneself to cope with and control these daily stressors.
Study has found that subjects who were single had less risk for periodontal
disease (Genco et al., 1999). The results of the present study failed to support this
finding, nor the spouse strain as significant factor in periodontal disease. Being
single may help to foster independence and hardiness, which will increase and
strengthen in the face of challenge. Those who were married may exhibit some
forms of stress which could have positive or negative impacts on their physical
health. It may depend on the quality of marriage, i.e., whether or not an individual
is satisfied with his/her marriage. Marcenes and Sheiham (1992) postulated that
those who were dissatisfied with their marriages were more likely to be in poorer
physical and psychological health than those who were single, widowed, or
divorced.
Weinberger et al. (1987) suggested that a subjects cultural, economic, and
social background had to be considered when measuring effects of chronic
stressors. The issue of cultural specificity had been cautioned by the development
of the Daily Strains scale (Pearlin & Schooler, 1978). Some studies also suggested
that daily hassles should be assessed along with life events in order to have a full
understanding and practical prediction of health outcomes (DeLongis et al., 1982,
1988). Further studies with more elaborate evaluation of the cultural and social
192
background may be necessary to delineate the significance of the impact of these
chronic stressors on physical health.

4.1.8 Systemic conditions, smoking, age, gender and education
The most prevalent self-reported diseases in this sample were allergies;
followed by hypertension, diabetes, asthma, anemia and cardiovascular disease.
(Table 3.6). Diabetic individuals were about twice as likely to exhibit periodontal
attachment loss compared with non-diabetics (Table 3.26). These findings appear
primarily consistent with previous studies (Davies & Davies, 2005; Pihlstrom et
al., 2005). Allergies were negatively associated with full mouth mean CAL; when
allergies were present, periodontal attachment loss was less likely to be seen. The
negative association between periodontal attachment loss and allergies has only
been reported in the community study in Erie County (Genco et al., 1999). The
role of medication, usually antihistamine, in the treatment of allergies may have a
non-specific modulating effect on inflammation and the host response to
periodontal pathogens (Grossi et al., 1994). This interpretation regarding the role
of medications in allergic individual is worthy of further study to explain the
apparent protective effect of history of allergy against periodontal disease.
The strong association between amount of smoking and periodontal
attachment loss (Table 3.26) appear substantially consistent with literature data.
This could be explained by the biological phenomena in response to nicotine and
its byproducts (Palmer, 1988).
Periodontal attachment loss was positively correlated with increasing age
and negatively correlated with education, and the odds was higher for male
subjects to experience more severe periodontal attachment loss than females
193
(Table 3.26). The precise reasons for these relationships are not exactly clear
besides the speculations that elderly, less educated subjects or men tended to pay
less attention to their oral care.

4.1.9 Impact of periodontitis on oral health-related quality of life
This study represents one of the first attempts to explore the impact of
periodontal disesase on oral health-related quality of life. Quality of life is
increasingly acknowledged as a valid, appropriate and significant indicator of
service need and intervention outcomes in contemporary public health research
and practice. Health-related quality of life measures, including objective and
subjective assessments, are especially useful for evaluating efforts to prevent
disabling chronic diseases and their effectiveness (Hennessy et al., 1994).
Assessing the consequences of impaired oral health from the patient's perspective
has emerged as an important research area (Buck & Newton, 2001). This has lead
to an increase in the use of patient-centered oral health status measures, primarily
attempting to measure the impact of oral health on quality of life (Birch & Ismail,
2002).
A study by Needleman et al. (2004) attempted to explore the impact of oral
health on quality of life in periodontal patients. However, this sample was
confined to referred periodontal patients attending a private periodontal clinic.
Accordingly, periodontal status was found to have significant impact on quality of
life. The lack of a control sample of subjects limited the extent to which these
findings could be generalized to a larger general population.
The present study attempted to explore the difference in quality of life in
subjects with various periodontal conditions. The criterion variable of full mouth
194
mean clinical attachment level (CAL) was not intended as an indication for
treatment, nor a direct and specific parameter in measurement of disease severity.
Nevertheless, it provided a valid estimate of the historical amount of periodontal
destruction in a given patient (Locker & Leake, 1993, Greenstein, 1997). Making
use of the sub-sample in this community study of psychological factors
contributing to periodontal disease, it allowed a broad variation in periodontal
condition to be studied. The potential for difference in CAL at various vulnerable
sites between individuals might theoretically influence the QoL of the
corresponding subject in varying extend. A bigger sample size, however, would be
needed for such purpose considering the present results from functional limitation,
physical pain, physical disability, handicap items of OHIP-14S.
The impact of oral health on the quality of life of the subjects was
appreciable, 22% (157/727) reported that their oral health status impacted on their
quality of life in one or more ways (i.e. scores of fairly often or very often in
one or more of the OHIP-14S items). Oral health status was frequently perceived
as impacting on quality of life because it affected function (by making food taste
worse), led to physical pain (by making food uncomfortable to eat), and resulted
in physical disability (by interrupting meals). This draws attention to the influence
of periodontal condition on daily life and its significance for overall quality of
life.
Clinical periodontal status was significantly associated with oral
health-related quality of life. Those with full mouth mean CAL above 3mm (i.e.
high/severe CAL group) scored significantly higher on the impact of oral health
on their quality of life in the OHIP-14S and various subscales, except on the
social disability and handicap subscales (Table 3.32). That is, people might
195
perceive that their social functions and overall satisfaction with life would not be
significantly affected because of their oral health status. The low prevalence of
negative impact in the latter two subscales with respect to the overall study
sample might help account for the insignificant results. Many local Chinese, as
reflected in the utilization of dental care pattern (Table 3.3), tend to pay little
attention to dental care and fail to anticipate the need for treatment and
maintenance care (Department of Health of the HKSAR, 2002). The overall
OHIP-14S score demonstrated a significant difference in subjects with different
periodontal status. After all, the cohort studied was derived from a community
sample. The generalizability of present findings would be considered as
satisfactory.
In comparison with the study conducted by Needleman et al. (2004) in a
private specialist periodontal clinic, findings of the present study differed in that
the social disability and handicap domains of the quality of life were not
associated with periodontal attachment loss. There are several possible
explanations for the discrepancy. The periodontal attachment loss in the subjects
of the present study, based on a community sample, can be expected to be less
severe than the sample of patients attending a referral periodontal clinic. It
appeared to be the case that even the low use of dental services among the study
population would potentially increase their disability and handicap. Furthermore,
the scores of the subscales of these two particular domains were relatively small
(Table 3.30) and probably failed to register the difference. Cultural specificity
may also be one of the reasons accounting for the difference. The relatively low
utilization of preventive and maintenance dental health care in the local
population (Table 3.3) probably reflects the perceived lack of importance of the
196
oral health condition in its social context.
In summary, there exists a significant association between oral
health-related quality of life and periodontal status in the present population of
predominantly non-regular dental attenders as assessed using the OHIP as
quality-of-life measure. Those with better periodontal condition, i.e. with minimal
history of periodontal destruction, are more likely to have a better quality of life,
and vice versa. This is the first ever scientific study in general population to
demonstrate that periodontal destruction can directly affect the quality of life.

4.1.10 Oral health-related quality of life and dental anxiety
This study evaluated the conceptualized relationships between the various
psychosocial variables of the stress process and periodontal disease, and in turn
the impact of periodontal disease on oral health-related quality of life. Besides the
finding of a significant correlation of periodontal disease with oral health-related
quality of life, dental anxiety was also found to correlate with oral health-related
quality of life (r
xy
= 0.33, P = 0.008) after adjustment for the effects of education,
number of teeth present, and full mouth mean CAL.
An explanation why dental anxiety and poor oral health-related quality of
life correlated may be that dentally anxious people avoid dental care to such a
degree that the oral health condition was significantly compromised beyond
merely the periodontal condition and affected their daily living to a considerable
extent. This is plausible given the evidence that dentally anxious people have poor
oral health, tend to delay in treatment and resort to more definitive and less
conservative care than non-anxious people. A longitudinal study, however, is
required to substantiate the above hypothesis. Further study investigating the
197
facets of dental anxiety and periodontal health would be necessary to give a more
precise picture; and, if this is the case, attention and resources should be allocated
for dentally anxious people to enable them to maintain better oral health condition
and hence enjoy better quality of life.

4.2 Methodology and limitations
Reports on the impacts of psychosocial factors on general health status of
an individual were available some years ago (DeLongis et al., 1982, 1988; Cohen
et al., 1986; Cooper & Faragher, 1993). Most of those studies involved a small
sample of subjects, and only a few reported the relationship between psychosocial
factors and periodontal health (Table 1.3) (Green et al., 1986; Marcenes &
Sheiham, 1992; Freeman & Goss, 1993). The population study conducted in the
Erie County (Genco et al., 1999) initiated an attempt to include a population
sample of broad variation in periodontal disease severity and potential
psychosocial risk factors. The findings were impressive especially regarding the
evaluation of coping behaviours in moderating the stress impact on periodontal
disease.
The present community study made an original attempt to address the
issues of personality traits, dental anxiety and oral health-related quality of life,
which had not been properly considered in previous studies. Personality traits are
significant moderating factors in contemporary conceptualization of the stress
process while dental anxiety is theoretically as well as intuitively an influencing
factor in the development and progress of periodontal disease. Oral health-related
quality of life, on the other hand, reflects the psychological well-being of people
with reference to their periodontal status and oral health condition as a whole.
198

4.2.1 Representativeness
4.2.1.1 Validation of DAxI
This part of the present study sought to translate and validate DAxI as an
instrument for measuring dental trait anxiety among Chinese in Hong Kong.
According to the Hong Kong Census 2001 (Census & Statistic Department of the
HKSAR, 2004), the demographic data of the subjects conveniently recruited
appeared comparable with the general population of Hong Kong in respect of age,
education, income and dental habit (Schwarz & Lo, 1994; Department of Health
of the HKSAR, 2002;) (Table 3.1). Furthermore, the corresponding psychometric
inventories/scales scores of the surveyed subjects did not differ a lot from the
Chinese population norms (Table 3.5). This suggests that despite the limitations of
the sampling protocol, and hence cautions needed in data interpretation, the
similarities of the observed data to population norms indicate that the sample
surveyed might represent a broader population.
The numbers of dropout subjects in the validation process of Chinese
DAxI and SDAxI were small and acceptable (24 of 500 for the DAxI and 7 of 150
for the SDAxI). A concern was whether such sample attritions would confound
the study. Analyses of their DAxI scores, SDAxI scores and demographic data
suggested no statistically significant difference between the dropout subjects and
those who completed the study.

4.2.1.2 The cross sectional study
This part of the present study was designed to include a wide range of
periodontal disease severity defined by different categories of clinical attachment
199
level. Rather than an epidemiological study, the main objective of this
cross-sectional sampling was to ensure an adequate representation of the local
population with sufficient variances in periodontal disease, demographic data,
psychosocial stress, coping and disposition, for assessment of the relationships
between the explanatory and outcome variables.
The sample in the present study, within the limitation of available
resources, achieved a reasonable size comparable with similar studies involving
evaluation of periodontal status and/or psychosocial variables in the general
population (c.f. Holmgren et al. 1994; Genco et al., 1999). To ascertain the
generalizability of the findings of this study to broader populations, a
comparison was made on the social demographic characteristics of the sample
and those of local populations in Hong Kong. The local population data from the
Hong Kong Census 2001 (Census and Statistics Department of the HKSAR,
2001) was employed to examine the representativeness of the study sample. A
summary of the comparison is shown in Table 3.2. The age distribution of the
subjects across the four decades was similar to that of the population. The
distributions of gender, ethnicity, marital status and education level of the
sample and that of the population were similar. In terms of household income,
the profile of the sample subjects were also similar to the general population.
Overall, the general demographic characteristics of the sampled group, as
compared with local data, appear satisfactory to suggest that this sample likely
represent the broader population of Hong Kong.
The profile of dental habits and dental care utilization was comparable
with the results of the recent territory wide dental health survey conducted by the
Department of Health of the HKSAR (Department of Health of the HKSAR, 2002)
200
(Appendix 3). Three-quarters of the subjects reported that they had not visited a
dentist for at least 1 year, except to seek treatment for a specific dental problem.
This indicated that most of the surveyed individuals were non-regular attenders,
and it was in line with earlier reports in Hong Kong population (Schwarz & Lo,
1994; Department of Health of the HKSAR, 2002). It was assumed that roughly
the same proportion of the individuals surveyed with periodontal attachment loss
had not had their periodontal disease properly treated or controlled. The
demographic data of the sample as a whole and for the subsample employed in
QoL analysis, were comparable with that of the local population. Furthermore, the
profiles of medical disease and smoking habits of the surveyed sample were
comparable with local population data (Department of Health of the HKSAR,
2005). The samples used in the present study would be considered valid and
appropriate as a community-based study.
CAL was employed as estimation of the historical amount of periodontal
destruction in a given patient in the present study (Loe et al., 1978; Locker &
Leake, 1993; Greenstein, 1997). The periodontal data of this sample, including
number of teeth present, Cl, BOP, REC, PPD and CAL appeared comparable with
previous studies of the local population (Holmgren et al., 1994; Lo & Schwarz,
1994; Appendix 4). This is especially true if the general improvement in dental
health awareness is taken into account.
The size of various sub-samples, number of subjects in categorized or
dichotomized sub-groups, remained adequate and sufficient for further statistical
analysis (Myers & Hansen, 2002; Cohen & Lea, 2004). Full mouth mean CAL
was employed as an estimation of the historical amount of periodontal destruction
in a given patient in the present study (Greenstein, 1997). Similar to many other
201
studies, high/severe full mouth mean CAL was associated with smoking,
increasing age, diabetes mellitus, and gender, while higher education status was
associated with better periodontal status (Monteiro da Silva et al., 1995; Genco et
al., 1999).

4.2.2 Validity of measurements
4.2.2.1 Clinical examination
The intra-examiner reproducibility results (Table 3.13), based on
proportion of agreement between scores of clinical periodontal examination, were
generally high; the lowest agreement being for the scoring of recession. The lower
value for this clinical variable could probably be attributed to the difficulty in
identifying the cementoenamel junction which was often obscured by restoration
margins or abrasion cavities, especially in subjects of the older age group. The
kappa statistic values were generally high, between 0.72 and 0.78 for CAL, PPD
and REC (Table 3.13). Based on Landis and Kochs benchmarks (1977),
substantial agreement was achieved in scoring of the various periodontal
variables.

4.2.2.2 Psychological questionnaires
Since analyses of predictor variables and subsequent interpretations and
conclusions are based on self-reported psychosocial traits, the goodness of fit of
the collected data of our study population to the hypothetical factor structures of
the various psychosocial instruments used was of crucial importance. However, it
was difficult, and more often unsuccessful, to reproduce the exact factor structures
of the original instruments. Some instruments, the SCL-90 for example, was a
202
90-item self-report system inventory developed in the 1980s by Derogatis and
designed primarily to reflect the psychological symptom patterns of medical and
psychiatric respondents as compared with the normal community population.
This instrument has been validated in a number of studies in clinical settings. In
the special application of cerebral acceleration and deceleration (whiplash) trauma
and its aftermath, the SCL-90-R is particularly useful (Wallis & Bogduk, 1996;
Wallis et al., 1996, 1997, 1998). Despite the validation of the psychological
instruments in a population, the cultural, socio-economic and religious
backgrounds still have some influence on their applications. These may also
explain why the exact factor structures of psychological instruments are not
always fully recovered in subsequent studies, even in the population where they
were developed, when the scores are subjected to sophisticated and stringent
evaluation of factor analysis (Myers & Hansen, 2002; Cohen & Lea, 2004; Giri,
2004; Morrison, 2004; Weiten, 2004).
A factor analysis by Angst et al.(1984) on the Symptom Checklist (SCL)
results of 1,162 non-clinical young men reflected similar difficulties in replicating
some of the factor structures. The factor anxiety has also been suggested to be
included in the factor phobic anxiety (Clark and Friedman, 1983; Holcomb et al.,
1983; Hoffmann and Overall, 1987). Studies even suggested that the individual
dimensions are highly correlated with one another and that the results are
primarily a measure of a General Distress Index. (Cyr et al., 1985; Alvir et al.,
1988).
Nevertheless, Cronbachs coefficients of all scales and individual
sub-scales of the psychological instruments in the present study were high (Tables
3.18, 3.20 and 3.22). In fact, the lowest Cronbachs coefficient recorded was 0.79,
203
from the Being Single subscale of the Measure of Chronic Stress (Daily Strains)
(Table 3.18). The high Cronbachs alphas obtained from the 9 subscales of the
SCL-90 were all above 0.85 (Table 3.20) indicating that the data collected from
the dimensions used were quite reliable. The high item-scale correlations within
the individual subscales or scales, and low inter-scale correlations between the
subscales supported the discriminant validity of the instruments used. The validity
of the instruments used was also empirically supported as the results were
comparable with the local norms (Table 3.5) (The Mental Health Association of
China, 1993; Wong, 1996; Chan, 1998; Ng et al., 2004).
Coping strategies, by nature of these constructs, are very sensitive to
cultural and social background (Lazarus & Folkman, 1988; Carver et al., 1989;
Lazarus, 1993a, 1999; Lazarus & Lazarus, 1994; Carver, 1997). Factor analysis,
as recommended by Carver (1989, 1997) was conducted to evaluate the factor
structures of the study population. Factor structures so recovered were not
identical with that reflected when the COPE was developed in 1989 by Carver in
American population (Carver et al., 1989). Component items of problem-focused
coping and emotion-focused coping however were comparable with that
explored by Carver et al. (1989). Another different factor found in the present
study population referred to less adaptive coping. This fact may in return
confirm the claim of cultural specificity of coping behaviours (Carver et al., 1989;
Carver, 1997; Lazarus, 1999, 2000).
When a potential threat comes up, the brain goes through an appraisal
process, be it conscious or unconscious (Lazarus and Folkman, 1984; Lazarus,
1999). Lazarus and Folkman (1984) coined the term cognitive appraisal
referring to the process by which an individual evaluates relevance of a potentially
204
stressful event and, if so, judges its meaning and significance to his/her own
well-being. The concept of cognitive appraisal has been widely accepted in the
study of stress and emotion, as well as in psychotherapy; however, cognitive
appraisal has not been well studied (Kessler 1998). One reason is the difficulty of
quantifying the process of the mental operations of thinking and reasoning; and,
appraisal is apparently a concept of multidimensional complexity (Kessler 1998).
Measurement of appraisal is considered a relatively new area of study. Appraisal
scales were developed for specific population groups, such as Appraisal of Illness
Scale (AIS) for stress appraisal in cancer patients (Oberst et al., 1991), Emotion
Appraisal of Nursing Home Placement Tool (EANH) (Kammer, 1994). Mishel
and Sorenson (1991) used the 15-item appraisal scale that was actually part of the
Ways of Coping Checklist (Folkman 1984). After all, how one interprets potential
threats as stressful or not usually follows a pattern or style and is more a personal
disposition; and, appraisal and coping are conjoined in nature (Lazarus, 1999). In
the present study, appraisal process was not measured by a specific instrument but
rather inferred from the stress related traits and coping styles (Lazarus, 1999).
A shortcoming in the present research is the lack of a direct measure of
biochemical parameter in the stress process. This is primarily a rational
compromise between resources, logistics and aims of this project as the scales
measuring stress responses (SCL-90 and DASS-S) give measures of the
symptoms highly correlated with the various physiological responses of the
subjects in face of stress (Derogatis, 1994; Lovibond & Lovibond, 1995a &
1995b).

4.2.2.3 Measurement of dental anxiety
205
The Dental Anxiety Inventory (DAxI) was originally developed in Dutch
and subsequently translated into English (Skaret et al., 1998; Aartman et al., 2000).
In order to make use of this instrument to measure the dental anxiety of Chinese
in Hong Kong, translation and validation was necessary. Cross-cultural
adaptations of health-related, self-administered instruments have been discussed
by several researchers (Corless et al., 2001; Locker & Allen, 2002). Studies have
suggested that instruments considered for cross-culture research should be
reviewed by considering issues with respect to language, ethnic, cultural and
socioeconomic differences, while at the same time instruments should be easy to
complete and should demonstrate acceptable psychometric properties (Allen &
Walsh, 2000; Corless et al., 2001; Kojima et al., 2002). It is thus essential to carry
out strict and rigorous translation and validation procedures prior to the final
application of an instrument in another population or culture.
An instrument measuring dental anxiety should be able to measure the
anxiety proneness and predict the state anxiety of an individual when he or she is
actually facing a dental situation. At the same time, if the claim is that DAxI
should be measuring specific dental trait anxiety, the time of administration of the
DAxI with regard to the dental visit should not influence the results.
In the present validation study, the predictive validity of the Chinese
DAxI was supported and the DAxI was shown to measure a rather stable construct
of anxiety proneness to the dental visit situation independent of the time of
administration with regard to the dental visit. The construct validity, the
discriminant validity, reliability and internal consistency of the Chinese DAxI
were supported.
Dental anxiety is a complicated phenomenon and its multifactorial nature
206
is very often undermined in its measuring instruments (Stouthard et al., 1993;
Schuurs & Hoogstraten, 1993). Dentally anxious subjects are not a homogenous
group of people and they differ in various aspects including the etiology of fear,
and its manifestation in terms of affective, behavioural and cognitive reactions
(Locker et al., 1999). A review has suggested that different aspects of dental
anxiety should be included in a measurement instrument, namely the situation to
which it pertains, the reactions it evokes and its duration (Schuurs & Hoogstraten,
1993). The Dental Anxiety Scale (DAS) (Corah, 1969) and the Dental Fear
Survey (DFS) (Kleinknecht et al., 1973), being two of the most well known
instruments, fail to address this multifactorial dimension (Stouthard, 1989). The
DAxI employs the facet theory (Canter, 1985; Roskam & Broers, 1996) which
offers a useful heuristic in the construction of measurement instruments for
multifactorial concepts, enabling the DAxI to specify an exhaustive and
systematic description and definition of the complex phenomenon of dental
anxiety. The facet approach obviously offers added value on the level of
operationalization for measurement of dental anxiety.
A few years after DAxI was formulated, Stouthard constructed a practical,
shortened version of the tool, named SDAxI (Stouthard et al., 1993). In the
present validation study, the empirical data justified a positive remark for the
Chinese SDAxI.
The aim of the original DAxI, according to Stouthard (Stouthard, 1989;
Stouthard et al., 1993, 1995), was not only to identify extremely anxious dental
patients, but also to assess anxiety proneness in regular dental patients and the
prevalence of dental anxiety in the general population. The validity and reliability
of DAxI in this respect seemed accomplished (Schuurs & Hoogstraten, 1993). In
207
conclusion, the translated Chinese DAxI is as valid and reliable as the original
version of DAxI and is a suitable instrument for measuring dental anxiety both in
research and in dental practice in Chinese populations. It offers a reliable
measurement of dental trait anxiety and satisfactory prediction of state anxiety in
dental situations.

4.2.2.4 Measurement of oral health-related quality of life
The OHIP-14S was used in this study to examine the impact of periodontal
condition on oral-health related quality of life. Since its development in 1994
(Slade & Spencer, 1994), the applications of the OHIP in research and public
health care practice, have empirically substantiated its appropriate validity and
sensitivity to the disease-related attributes. Establishment of goodness of fit of the
collected data of the studied population to the hypothetical structure of instrument
used is important. The Cronbachs coefficients of the OHIP-14S and subscales
were high (Table 3.30). In fact, the lowest Cronbachs alpha recorded (0.73) was
from the handicap subscale. These high correlations indicated that items being
used and constructed from the hypothetical constellation of items of each subscale
measured a common factor and had reasonably satisfactory convergent validity
when applied in the present sample of subjects.
Variations in oral health impact on quality of life in relation to
self-reported symptoms of periodontal diseases were apparent (Table 3.31).
Experiences of swollen gums, sore gums, receding gums, loose teeth, bad breath,
and toothache were associated with increased impact. This also added further to
the discriminant validity of the instrument in differentiating subjects with different
self-reported periodontal status. Discriminative ability is an important issue of
208
patient-centered measures so as to ensure that they are sensitive and responsive in
assessing the consequence of periodontal disease, identifying treatment needs, and
evaluation of treatment result (Weintraub, 1998).
Besides validity and reliability, issues concerning the cultural specificity of
health-related self-reported measures have been discussed by various researchers
(Slade, 1997b; Allison et al., 1999; Corless et al., 2001). Impact due to the
socio-demographic parameters of age, gender and social class are culture-sensitive.
The impacts of gender and social class on oral-health related quality of life have
been demonstrated in a study of its association with dental anxiety in the United
Kingdom, accounting for about 18% of the variance of the total score (McGrath &
Bedi, 2004). The variance due to age, gender and income were adjusted in the
present study to control for the possible confounding effects.

4.2.3 Cross-sectional community study
The present study examined systematically the component factors at
various aspects of the stress process and their relationship with periodontal
disease. Despite the explicit results from the sophisticated statistical analysis, the
most we can state about the variables is correlation; that is, how one correlates to
another. Results of the present study are based on cross-sectional data. Whether
stress-associated odds of periodontal disease is related to behavioural and/or
physiological changes is yet to be determined. It is by no means a test of
cause-effect. Even with the rather powerful multiple logistic regression statistical
analysis, it tells only the probability (odds) of observing the dependent variable
which in the present study is periodontal attachment level. This is one limitation
to this method.
209
Correlations can be deceiving. Finding a significant correlation with 2
variables does not guarantee that they are the only 2 explanatory variables. There
may be an intervening variable that wasn't measured. Further study, probably of
bigger scale and with more sophisticated design and methodology, may be
necessary to explain the variance remained. Longitudinal studies relating the
temporal relationships of components of stress process, such as depression, job or
financial strain, to periodontal disease onset and progression are necessary to
establish whether these are truly a risk factor for periodontal disease, or whether
these are only a concomitant of poor health associated with periodontal disease.

4.3 General remarks
Almost all of the participants in the present study expressed during the
debriefing time upon completion of psychological assessments that the
questionnaires were very long and they felt rather tired completing them. On
average, participants took 25 to 30 minutes to complete all the instruments.
Acknowledging the subjects workload and burden in completing the
questionnaires, it has to be admitted that exploration of psychological components
and contribution in physical disease inevitably involves evaluation of a certain
number of psychological constructs. These constructs are usually inferred from
ones feelings, behaviours and physical responses; the diversity of that needs no
further explanation. Thus, it is not uncommon to see that multiple questionnaires
or inventory measures are used in these psychosomatic studies; and furthermore,
evaluation of a construct, depending on its complexity, may require the respondent
to answer a list of question items to ensure valid and reliable assessment. Despite
these comments from the subjects, the results in the present study remained
210
reliable and valid as discussed earlier.

211
Chapter 5
Conclusion and Recommendation

5.1 Conclusion
The study of stress, since its receiving more attention after World War II,
has achieved an understanding of a rather complex scientific concept
encompassed various faculties of knowledge including psychology (mind,
emotion and behaviour), medicine, human physiology, prevention and treatment
of disease, and philosophy. The diversity and variability of factors affecting stress
process need no more elaboration.
Since stress is associated with a wide range of health problems, it is
reasonable to speculate that stress may significantly affect ones oral health,
including periodontal condition. Sutton (1990) postulated that stress would
adversely affect the efficacy of the immune system which, in turn, would reduce
the defense against bacterial attack, permitting the development of acute dental
caries. This theory can also be extended to periodontal disease.
The present study found that stressors, revealed as job strain or financial
strain, and the negative affective state of depression were associated with greater
levels of periodontal disease manifested as greater clinical attachment level which
gives an estimate of the historical amount of periodontal destruction (Loe et al.,
1978; Locker & Leake, 1993). Individuals with more adaptive coping behaviours
(problem-focused coping) were at lower odds for periodontal disease while those
with poor coping behaviours (emotion-focused coping) were at higher odds. Trait
dispositions of anxiety and depression as well as dental anxiety were also
212
suggested as significant risk indicators.
Given the chronic nature of periodontal disease as a long-term health
outcome and, therefore, it is logical to expect that a chronic pattern of adverse
psychosocial effects is required to give a measurable disease impact. Of the 6
dimensions of Measure of Chronic Stress, strain in job and finance showed a
positive relationship with clinical attachment level. The negative results from
analysis of the contribution of acute life events or changes are not surprising.
Coping is the response of a person in an attempt to manage, control,
reduce, or avoid the negative and unpleasant consequences of stress. Personality
dispositions of negative affectivities determine the propensity of an individual to
develop the respective affective features in response to stress. Both coping and
personality dispositions are of the chronic and durable nature in predisposing an
individual to behave in a particular way in a variety of situations over a long
period of time in his or her life. Of considerable interest is the finding that traits of
anxiety and depression, coping behaviours were found to have significant
relationship with periodontal disease in such a way as moderating factor or
mediating process, respectively, of the impact of stressor.
It was also found that those individuals who possessed adaptive coping
behaviours, even when under job strain or financial strain, exhibited no more
periodontal disease than those individuals not under job strain or financial strain.
Those individuals with lower trait disposition of anxiety or depression were found
to express no more odds of periodontal disease even if they were under job strain
or financial strain. Adequate/adaptive coping behaviours as evidence by high level
of problem-focused coping or low level of emotion-focused coping, and
favourable trait dispositions as evidenced by low anxiety or depression traits were
213
found to reduce the stress-associated risk for periodontal disease.
The influence of dental anxiety on clinical attachment level was
statistically significant. A significant association between oral health-related
quality of life and periodontal status was detected. Dental anxiety is also found to
be associated with oral health-related quality of life.
The main conclusions of the present study are summarized as follow:
1. Dental Anxiety Inventory (DAxI) was successfully translated to
Chinese and validated for use in the Hong Kong population;
2. Chinese Dental Anxiety Inventory short form (SDAxI) was
successfully constructed and validated for use in the Hong Kong
population;
3. A community survey on the periodontal health of subjects sampled
from general dental practices was successfully conducted. The
demographic parameters, periodontal status and dental health care
profile of the studied sample were comparable with the local
population data indicating that this study sample could represent
the broader adult population of Hong Kong;
4. Stressors revealed as job stain or financial strain, and the negative
affective state of depression were significant risk indicators of
periodontal disease manifested as greater clinical attachment level
or periodontal attachment loss;
5. Individuals with adaptive coping behaviours (problem-focused
coping) were at lower risk for periodontal attachment loss while
those with less desirable coping behaviours (emotion-focused
coping) were at higher risk;
214
6. Trait dispositions of anxiety and depression were also significant
risk indicators for periodontal attachment loss;
7. Individauls with adequate/adaptive coping behaviours as evidence
by high level of problem-focused coping or low level of
emotion-focused coping, and favourable trait dispositions as
evidenced by low anxiety or depression traits were at reduced job
or financial stress-associated risk for periodontal disease.
8. Dental anxiety was significantly associated with periodontal
attachment loss;
9. The impact of periodontal health on oral health-related quality of
life was significant;
10. Dental anxiety was significantly associated with inferior oral
health-related quality of life.

5.2 Recommendation
The present study attempted to evaluate rigorously the stress-associated
risk for periodontal disease using periodontal clinical attachment level as outcome
variable, with stress, traits and coping being assessed by validated instruments,
and had at-risk behaviours (such as smoking and oral health regimes) and other
possible confounding factors (such as age, gender, education level, systemic
disease) measured and adjusted. Although not all of the psychosocial factors
investigated were related to periodontal disease, the results as a whole were
compatible with previous reports in the literatures.
A general concern in this area of research has been the clinical
significance of stress induced alternations of immune functions. Longitudinal
215
studies are needed to address the specific temporal relationships between stress,
diminished immunocompetence and the onset and progression of periodontal
disease; the magnitude of the associations and the dose-response relationships
should be explored as well. Studies directing towards the biochemical and
physiological mechanisms by which psychosocial stress contributes to periodontal
disease are needed to establish the biological rationale for this relationship.
Animal studies are needed to investigate the mechanisms and to exclude the
impact of various behavioural changes, such as smoking and poor oral hygiene,
which are associated with stress. Psychoneuroimmunological studies of the
mechanism by which psychosocial stress exerts effects on periodontal disease are
needed to establish the biological rationale for this relationship and confirm their
roles as risk factors. Such studies may include investigation of biochemical
mediators of stress, immune functions, neurological and endocrine changes, as
well as behavioural concomitants. Evaluation of these mechanisms with animal
models is deemed necessary and instructive.
The present cross-sectional study on predominantly Chinese Hong Kong
adults confirmed common risk factors like smoking, increasing age, gender and
diabetes mellitus were associated with periodontal attachment loss while high
education attainment was associated with better periodontal condition.
Psychological factors in terms of chronic stress of job and financial strain,
inadequate or ineffective stress moderation and mediation abilities and depressive
stress response could increase the risk for periodontal attachment loss in those
affected. Proper coping, low anxiety or depression trait appeared to be able to
compensate the negative effects towards periodontal health by job and financial
strains. Stress management training in general or the contemporary Cognitive
216
Behavioural Therapy in particular, which have been advocated in managing daily
living stress, enhancing coping strategies and allowing adaptive adjustment of
trait disposition (Corey, 2001; Powell, 2001) seem to be the possible adjunctive
regimes in treatment of periodontal disease. Intervention studies employing stress
management to reduce stress and/or cognitive behavioural training for
enhancement of coping with randomized or matched control and trial
methodology are necessary to establish efficacy of modification of stress as a
treatment modality in stress-related disease.
A longitudinal study on a subgroup of the present study sample has been
planned to further explore if intervention measures such stress management
enhancement training may provide adjunctive approaches for prevention and
treatment of periodontal disease. Literature available up to the present moment
including the present project just illustrated we are at the early experimental phase
towards the full understanding the role of stress, within the boarder
biopsychosocial perspectives, as a contributor to periodontal disease. Further
longitudinal studies about periodontal changes in a cohort of periodontally healthy
subjects, including those with adequate or inadequate coping strategies, with or
without significant job or financial strains are recommended to allow a more
in-depth analysis of the effects and interaction of these psychosocial factors.
Integrated clinical, sociological and molecular based studies are needed for full
understanding the role of stress as a contributor to periodontal disease.
The influence of dental anxiety appears pervasive. It contributes to
impaired dental health as well as compromised oral-health related quality of life
(McGrath & Bedi, 2004). The clinical significance regarding the odds of suffering
severe periodontal attachment loss exacerbated due to dental anxiety was
217
appreciable (OR=1.2, CI: 1.10-1.50; mean 15.815.51). That means one SD above
mean equivalent to the odds of 2.73 in having high/severe CAL as compared with
subject with health/low CAL. The need to explore in further details the pathway
and extent of its influence and implication in treatment planning of periodontal
and other dental diseases remains meaningful. Study would be recommended in
general population to include objective clinical parameters, subjective indicators,
as well as details in treatment decision and planning process.
A significant association between oral health-related quality of life and
periodontal status was evident in the present study. These findings have
significant implications for the employment of patient-centered outcome measures,
in addition to objective clinical parameters of periodontal disease, in assessment,
planning and provision of treatment, and subsequent evaluation of care. Dentists
perhaps need to utilize this tool to evaluate if successful therapist-centered
outcome co-relates with patient-centered outcome. Greater understanding of the
difference in oral health that exists between periodontally healthy versus
periodontally compromised patients beyond clinical parameters is important
because it will provide an insight into the consequence of periodontal problems
for patients daily life and quality of life, as well as illustrating the need for
addressing these disparities. Further research is also recommended to assess
whether the measure of oral health-related quality of life as a patient-centered
outcome is sensitive to changes in clinical periodontal status over time and also at
the level of the individual.
The finding that dental anxiety is correlated with oral health-related
quality of life also implies that future study would be necessary to evaluate if
further attention and resources should be allocated for dentally anxious people to
218
enable them to maintain better oral health condition and hence enjoy better quality
of life.


219
Appendix 1
Brockprobe
TM
Periodontal Probe


220
Appendix 2
List of questionnaires used in this study:
1. Personal data
2. Dental habit
3. Life Event Questionnaire (LEQ)
4. Social Readjustment Rating Scale (SRRS)
5. Measure of Chronic Stress
6. COPE Inventory (COPE)
7. Depression Anxiety Stress Scales Trait (DASS-T)
8. Symptom Checklist 90 (SCL-90)
9. Depression Anxiety Stress Scales State (DASS-S)
10. Dental Anxiety Inventory Short form (SDAxI)
11. Oral Health Impact Profile (OHIP-14S)

221
In accordance with the copyright ordinance, original full set questionnaires are not
displayed. Please refer to the respective sources for original copy or further
information:

Life Event Questionnaire (LEQ) (Brugha & Cragg, 1990)
English & Chinese
versions:
Brugha, TS. University of Leicester, Department of Health Sciences, Brandon
Mental Health Unit, Leicester General Hospital, Leicester, UK.
Social Readjustment Rating Scale (SRRS) (Holmes & Rahe, 1967)
English & Chinese
versions:
Holmes, TH. Department of Psychiatry and. Behavioral Science. University
of Washington. Pearson Assessments. http://www.pearsonassessments.com.
Measure of Chronic Stress (Pearlin & Schooler, 1978)
English & Chinese
versions:
Pearlin, LI. Department of Sociology, Art-Sociology Building, University of
Maryland, College Park, MD 20742-1315, USA.
Symptom Checklist 90 (SCL-90) (Derogatis, 1994)
English & Chinese
versions:
Pearson Assessments. http://www.pearsonassessments.com.
Depression Anxiety Stress Scales State (DASS-S) (Lovibond & Lovibond, 1995a & 1995b)
English version: Lovibond PF. School of Psychology, University of New South Wales,
Sydney, Australia.
Chinese version: Chan, CKY. calaischan@cuhk.edu.hk.
COPE Inventory (COPE) (Carver, 1989)
Carver, C. Department of Psychology, University of Miami, USA.
Depression Anxiety Stress Scales Trait (DASS-T) (Lovibond & Lovibond, 1995a & 1995b)
English version: Lovibond PF. School of Psychology, University of New South Wales,
Sydney, Australia.
Chinese version: Chan, CKY. calaischan@cuhk.edu.hk.
Dental Anxiety Inventory Short form (SDAxI) (Southhard, 1989)
English version: Stouthard, ME. University of Amsterdam, The Netherlands.
marlies@joopbuinink.demon.nl.
Chinese version: Ng, SKS. Tuen Mun Hospital, HKSAR. samng@hkstar.com.
Oral Health Impact Profile (OHIP-14S) (Slade & Spencer, 1994)
English & Chinese
versions:
Australian Research Centre for Population Oral Health, Dental School, The
University of Adelaide, SA, Australia.

222
Copy of recruitment notice




223

Copy of information sheet



224





225

Copy of consent form








226



227
Personal Data



228


229





230
Dental habit


231


232
Life Event Questionnaire (LEQ)
(Full version of questionnaire not shown)


233
Social Readjustment Rating Scale (SRRS)
(Full version of questionnaire not shown)


234
Measure of Chronic Stress
(Full version of questionnaire not shown)

235
COPE Inventory (COPE)
(Full version of questionnaire not shown)

236
Depression Anxiety Stress Scales Trait (DASS-T)
(Full version of questionnaire not shown)


237
Symptom Checklist 90 (SCL-90)
(Full version of questionnaire not shown)



238
Depression Anxiety Stress Scales State (DASS-S)
(Full version of questionnaire not shown)


239
Dental Anxiety Inventory Short form (SDAxI)
(Full version of questionnaire not shown)

240
Oral Health Impact Profile (OHIP 14S)
(Full version of questionnaire not shown)

241
Appendix 3
Table extracted from:
Department of Health, HKSAR (2002).
Oral health survey 2001. Common dental diseases and oral health related
behaviour. HKSAR: The Printing Department.


242
243

244

245
Appendix 4
Table extracted from:
Holmgren CJ, Corbet EF, Lim LP. 1994. Periodontal conditions among the
middle-aged and the elderly in Hong Kong. Community Dent Oral Epidemiol
22:396-402.





246
Appendix 5

Publications:
Parts of the research work reported in this thesis have been published or are in
press:

1. Ng SKS, Stouthard MEA, Leung WK. 2005. Validation of a Chinese version
of Dental Anxiety Inventory. Community Dent Oral Epidemiol 33:107-14.

2. Ng SKS, Leung WK. 2006a. Oral health-related quality of life and periodontal
status. Community Dent Oral Epiedmiol 34:114-22.

3. Ng SKS, Leung WK. 2006b. A community study on the relationship between
stress, coping, affective dispositions and periodontal attachment loss.
Community Dent Oral Epidemiol. In press.

(COPY ATTACHED TO THE BACK OF THIS THESIS)




247
References
Aartman IH, de J ongh A, Makkes PC, Hoogstraten J . 2000. Dental anxiety reduction
and dental attendance after treatment in a dental fear clinic: a follow-up study.
Community Dent Oral Epidemiol 28: 435-42.
Abdellatif HM, Burt BA. 1987. An epidemiological investigation into the relative
importance of age and oral hygiene status as determinants of periodontitis. J
Dent Res 66:13-8.
Abrahamsson KH, Berggren U, Carlsson SG. 2000. Psychosocial aspects of dental and
general fears in dental phobic patients. Acta Odontol Scand 58: 37-43.
Abrahamsson KH, Berggren U, Hallberg L, Carlsson SG. 2002. Dental phobic
patients' view of dental anxiety and experiences in dental care: a qualitative study.
Scand J Caring Sci 16: 188-96.
Ader R, Cohen N, Felten D. 1990. Interactions between the brain and the immune
system. Ann Rev Pharmacol Toxicol 30:561-602.
Ader R, Felton DL, Cohen N. 1991. Psychoneuroimmunology. 2nd Ed. New York:
Academic Press.
Albandar J M. 2002. Global risk factors and risk indicators for periodontal diseases.
Periodontol 2000 29:177-206.
Albandar J M, Kingman A. 1999. Gingival recession gingival bleeding and dental
calculus in adults 30 years of age and older in the United States 1988-1994. J
Periodontol 70:30-43.
Albandar J M, Brunelle J A, Kingman A. 1999. Destructive periodontal disease in
adults 30 years of age and older in the United States, 1988-1994. J Periodontol
70:13-29.
Aleksejuniene J , Holst D, Eriksen HM, Gjermo P. 2002. Psychosocial stress lifestyle
and periodontal health. J Clin Periodontol 29:326-35.
Allen J , Walsh J A. 2000. A construct-based approach to equivalence: Methodologies
for cross-cultural/multicultural personality assessment research. In: Dana RH
(ed.), Handbook of cross-cultural and multicultural personality assessment.
Personality and clinical psychology series, pp. 63-85. Mahwah: Lawrence

248
Erlbaum Associates.
Allen PF. 2003. Assessment of oral health related quality of life. Health Qual Life
Outcomes 1:40. Available at: http://www.hqlo.com/content/1/1/40.
Allen PF, Locker D. 1997. Do item weights matter? An assessment using the oral
health impact profile. Community Dent Health 14:133-8.
Allison PJ , Locker D, Feine J S. 1997. Quality of life: a dynamic construct. Soc Sci
Med 45:221-30.
Allison PJ , Locker D, J okovic A, Slade G. 1999. A cross-cultural study of oral health
values. J Dent Res 78:643-9.
Allred KD, Smith TW. 1989. The hardy personality: Cognitive and physiological
responses to evaluative threat. J Pers Soc Psychol 56:257-256.
Altman, DG. 1991. Practical statistics for medical research. London: Chapman and
Hall.
Alvir J M, Schooler NR, Borenstein MT, Woerner MG, Kane J M. 1988. The reliability
of a shortened version of the SCL-90. Psychopharmacol Bull 24:242-6.
American Academy of Periodontology. 1992. The etiology and pathogenesis of
periodontal diseases (position paper). Chicago: The American Academy of
Periodontoloty; November 1992.
American Academy of Periodontology. 2003. Periodontal Disease.
http://www.perio.org.
American Psychiatric Association. 1994. Diagnostic and statistical manual of mental
disorders. 4th Ed. (DSM-IV). Washington DC: American Psychiatric
Association.
Andersen BL, Kiecolt-Blaser J K, Glaser R. 1994. A biobehavioral model of cancer
stress and disease course. Am Psychol 49:3-12.
Aneshensel CS, Pearlin LI, Schuler RH. 1993. Stress role captivity and the cessation
of caregiving. J Health Soc Behav 34:54-70.
Angst J , Dobler-Mikola A, Binder J . 1984. The Zurich study--a prospective

249
epidemiological study of depressive, neurotic and psychosomatic syndromes. I.
Problem, methodology. Eur Arch Psychiatry Neurol Sci 234:13-20.
Anisman H, Baines MG, Berczi I, Bernstein CN, Blennerhassett MG, Gorczynski RM,
Greenberg AH, Kisil FT, Mathison RD, Nagy E, Nance DM, Perdue MH,
Pomerantz DK, Sabbadini ER, Stanisz A, Warrington RJ . 1996. Neuroimmune
mechanisms in health and disease: 1. Health. CMAJ 155:867-74. Review.
Antony MM, Bieling PJ , Cox BJ , Enns MW, Swinson RP. 1998. Psychometric
properties of the 42-item and 21-item versions of the Depression Anxiety Stress
Scales (DASS) in clinical groups and a community sample. Psychol Assess
10:176-181.
Arnetz BB, Wasserman J , Petrini B, Brenner SO, Levi L, Eneroth P, Salvoaara H,
Hjelm R, Salvoaara L, Theorell T, Petterson IL. 1987. Immune function in
unemployed women. Psychosom Med 49:3-12.
Atanackovic D, Kroger H, Serke S, Deter HC. 2004. Immune parameters in patients
with anxiety or depression during psychotherapy. J Affect Disord 81:201-9.
Axtelius B, Soderfeldt B, Edwardsson S, Attstrm R. 1997a. Therapy-resistant
periodontitis (I). Compliance and general and dental health experience. J Clin
Periodontol 24:640-645.
Axtelius B, Soderfeldt B, Edwardsson S, Attstrm R. 1997b. Therapy-resistant
periodontitis (II). Compliance and general and dental health experience. J Clin
Periodontol 24:646-653.
Axtelius B, Soderfeldt B, Nilsson A, Edwardsson S, Attstrom R. 1998.
Therapy-resistant periodontitis. Psychosocial characteristics. J Clin Periodontol
25:482-91.
Bacic M, Plancak D, Granic M. 1988. CPITN assessment of periodontal disease in
diabetic patients. J Periodontol 59:816-822.
Baker EG, Crook GH, Schwacher ED. 1961. Personality correlates of periodontal
diseases. J Dent Res 40:396-403.
Ballieux RE. 1991. Impact of stress on the immune response. J Clin Periodontol
18:427-430.

250
Barmes DE. 1977. Epidemiology of dental disease. J Clin Periodontol 58:639-42.
Bartold PM, Kylstra A, Lawson R. 1994. Substance P: an immunohistochemical and
biochemical study in human gingival tissues. A role for neurogenic inflammation?
J Periodontol 65:1113-21.
Bartrop RW, Luckhurst E, Lazarus L, Kiloh LG, Penny R. Depressed lymphocyte
function after bereavement. Lancet. I. 1997;834-836.
Beck AT. 1967. Depression: Clinical experimental and theoretical aspects. 1st Ed.
p333. New York: Harper & Row.
Beck AT, Steer RA. 1990. Manual for the Beck Anxiety Inventory. San Antonio: The
Psychological Corporation.
Beck AT, Epstein N, Brown G, Steer RA. 1988. An inventory for measuring clinical
anxiety: psychometric properties. J Consult Clin Psychol 56: 893-7.
Beck J D. 1996. Periodontal implications: Older adults. Ann Periodontol 1:322-357.
Beck J D, Koch GG, Rozier RG, Tudor GE. 1990. Prevalence and risk indicators for
periodontal attachment loss in a population of older community-dwelling blacks
and whites. J Periodontol 61:521-528.
Beck J D, Koch GG, Zambon J J , Genco RJ , Tudor GE. 1992. Evaluation of oral
bacteria as risk indicators for periodontitis in older adults. J Periodontol
63:93-99.
Beck J S. 1995. Cognitive Therapy: Basics and Beyond. New York: The Guilford
Press.
Becker BE, Karp CL, Becker W, Berg L. 1988. Personality differences and stressful
life events. Differences between treated periodontal patients with and without
maintenance. J Clin Periodontol 15:49-52.
Bedi R, Sutcliffe P, Donnan P, Barrett N, McConnachie J . 1992. Dental caries
experience and prevalence of children afraid of dental treatment. Community
Dent Oral Epidemiol 20:368-71.
Belting CM, Gupta OP. 1961. The influence of psychiatric disturbances on the
severity of periodontal disease. J Periodontol 32:219-226.

251
Berggren U, Meynert G. 1984. Dental fear and avoidance: causes, symptoms, and
consequences. J Am Dent Assoc 109:247-51.
Berggren U, Pierce CJ , Eli I. 2000. Characteristics of adult dentally fearful individuals.
A cross-cultural study. Eur J Oral Sci 108: 268-74.
Birch S, Ismail AI. 2002. Patient preferences and the measurement of utilities in the
evaluation of dental technologies. J Dent Res 81:446-50.
Black S. 1963a. Inhibition of immediate-type hypersensitivity response by direct
suggestion under hypnosis. BMJ 1:925-929.
Black S. 1963b. Shift in dose-response curve of Prausnitz-Kustner reaction by direct
suggestion under hypnosis. BMJ 1:990-992.
Black S, Friedman M. 1965. Adrenal function and the inhibition of allergic re-sponses
under hypnosis. BMJ 1:562-567.
Blalock J E. 1994. The syntax of immune-endocrine communication. Immunol Today
15:504-511.
Bland, M. 2000. An introduction to medical statistics. 3rd Ed. Oxford, UK: Oxford
University Press.
Bolger N, Schilling EA. 1991. Personality and problems in everyday life: The role of
neuroticism in exposure and reactivity to daily stressors. J Pers 59:355-386.
Bovbjerg DH, Redd WH, Maier LA, Holland J C, Lesko IM, Niedzwiecki D, Rubin
SC, Hakes TB. 1990. Anticipatory immune suppression and nausea in women
receiving cyclic chemotherapy for ovarian cancer. J Consult Clin Psychol
58:153-157.
Brandtzaeg P, Fjellanger I, Gjeruldsen ST. 1970. Human secretory immunolglobulins.
I. Salivary secretion of individuals with normal or low levels of serum
immunoglobulins. Scand J Haematol 12:3-83.
Brett J F, Brief AP, Burke MJ , George J M, Webster J . 1990. Negative affectivity and
the reporting of stressful life event. Health Psychol 9:57-68.
Brevik T, Thrane PS, Gjermo P, Opstad PK, Padst R, von Horsten S. 2001.
Hypothalamic-pituitary-adrenal axis activation by experimental periodontal

252
disease in rats. J Periodontal Res 36:295-300.
Brevik T, Thrane PS, Murison R, Gjermo P. 1996. Review: Emotional stress effects on
immunity gingivitis and periodontitis. Eur J Oral Sci 104:327-334.
Bricken H. 1953. The effect of mental stress on the pH of the saliva: a preliminary
study. J Dent Med 8:96-104.
Briggs SR. 1992. Assessing the Five-Factor Model of personality description. J Pers
60:253-93.
Broadhead WE, Blazer DG, George LK, Tse CK. 1990. Depression disability days and
days lost from work in a prospective epidemiologic survey. J AMA
264:2524-2528.
Brown GW, Harris TO. 1978. The social origins of depression. London: Tavistock.
Brown LJ , Loe H. 1993. Prevalence, extent, severity and progression of periodontal
disease. Periodontol 2000 2:57-71.
Brown LJ , Oliver RC, Loe H. 1989. Periodontal diseases in the U.S. in 1981:
Prevalence severity extent and role in tooth mortality. J Periodontol 60:363-370.
Brown TA, Chorpita BF, Korotitsch W, Barlow DH. 1997. Psychometric properties of
the Depression Anxiety Stress Scales (DASS) in clinical samples. Behav Res
Ther 35:79-89.
Brugha. TS, Cragg D. 1990. The list of threatening experiences: The reliability and
validity of a brief Life Events Questionnaire. Acta Psychiatr Scand 82:77-81.
Buck D, Newton J T. 2001. Non-clinical outcome measures in dentistry: publishing
trends 1988-98. Community Dent Oral Epidemiol 29:2-8.
Buxton MN, ARkey Y, Lagos J , Deposito F, Lowenthal F, Simring S. 1981. Stress and
platelet aggregation in hemophiliac children and their family members. Research
Communication in Psychology, Psychiatry and Behavior, 6(1), 21-48.
Byrt T, Bishop J , Carlin J B. 1993. Bias, prevalence and kappa. J Clin Epidemiol
46:423-429.
Cannon WB. 1932. The wisdom of the body. New York: Norton.

253
Canter A, Cluff LF, Imboden J B. 1972. Hypersensitive reactions to immunization
innoculations and antecedent psychological vulnerability. J Psychosom Res
16:99-101.
Canter D. 1985. Facet theory. New York: Springer Verlag.
Cappel R, Gregoire F, Thiry L, Sprecher S. 1978. Antibody and cell-mediated
immunity to herpes simplex virus in psychotic depression. J Clin Psychiatry
39:266-8.
Carver CS. 1997. You want to measure coping but your protocols too long: Consider
the brief COPE. Int J Behav Med 4:92-100.
Carver CS, Scheier MF, Weintraub J K. 1989. Assessing coping strategies: A
theoretically based approach. J Pers Soc Psychol 56:267-283.
Cattell RB, Scheier IH. 1958. The nature of anxiety: A review of thirteen multi-variate
analyses comprising 814 variables. Psychol Rep 4:351-388.
Cattell RB, Scheier IH. 1961. The meaning and measurement of neuroticism and
anxiety. New York: Ronald Press.
Census and Statistics Department of the HKSAR. 2002. Population census 2001.
HKSAR: The Printing Department.
Chan CK, Lovibond PF. 1996. Expectancy bias in trait anxiety. J Abnorm Psychol
105:637-47.
Chan DW. 2000. Chinese way of coping questionnaire (English & Chinese) [CWCQ]
(1998). In: Corcoran K, Fischer J (eds.), Measures for clinical practice: A
sourcebook. 3rd Ed. Vol. 1, pp. 521-523. New York: Free Press.
Chandler DC, Silverman NS, Lundblad RL, McFall WT. 1974. Human parotid IgA
and periodontal disease. Arch Oral Biol 19:733-735.
Chatterton RT J r, Vogelsong KM, Lu YC, Hudgens GA. 1997. Hormonal responses to
psychological stress in men preparing for skydiving. J Clin Endocrinol Metab
82:2503-9.
Cheng SKW, Wong CW, Wong KC, Chong GSC, Wong MTP, Chang SSY. 2002. A
study of psychometric properties, normative scores, and factor structure of the

254
Beck Anxiety Inventory - the Chinese version. Chin J Clin Psychol 10: 4-6. [In
Chinese]
China Association of Mental Health. 1993. Rating scales for mental health. Beijing:
Chinese Mental Health J ournal. [In Chinese]
Clark A, Friedman M. 1983. Factor structure and discriminant validity of the SCL-90
in a veteran psychiatric population. J Pers Assess 47:396-404.
Clark KK, Bormann CA, Cropanzano RS, J ames K. 1995. Validation evidence for
three coping measures. J Pers Assess 65:434-55.
Clarke NG, Hirsch RS. 1995. Personal risk factors for generalized periodontitis. J Clin
Periodontol 22:136-145.
Clarke NG, Shephard BC, Hirsch RS. 1981. The effects of intra-arterial epinephrine
and nicotine on gingival circulation. Oral Surg Oral Med Oral Pathol
52:577-582.
Coates TJ , McKusick L, Kuno R, Stites DP. 1989. Stress reduction training changed
number of sexual partners but not immune function in men with HIV. Am J
Public Health 79:885-886.
Cogen RB, Stevens AW J r, Cohen-Cole S, Kirk K, Freeman A. 1983. Leukocyte
function in the etiology of acute necrotizing ulcerative gingivitis. J Periodontol
54:402-7.
Cohen BH, Lea RB. 2004. Essentials of statistics for the social and behavioral
sciences. Hoboken NJ : Wiley.
Cohen J . 1968. Weighted kappa: nominal scale agreement with provision for scaled
disagreement or partial credit. Psychol Bull 70, 213-20.
Cohen J . 1977. Statistical power analysis for the social sciences. New York: Academic
press.
Cohen LH, Simons RF, Rose SC, McGowan J , Zelson MF. 1986. Relationships
among negative life events physiological reactivity and health symptomatology. J
Human Stress 12:142-8.
Cohen ME. 1985. Dental anxiety and DMFS status: association within a US naval

255
population versus differences between groups. Community Dent Oral Epidemiol
13:75-8.
Cohen MM, Shusterman S, Shklar G. 1969. The effect of stressor agents on the grey
lethal mouse strain periodontium. J Periodontol 40:462-6.
Cohen S, Edwards J R. 1989. Personality characteristics as moderators of the
relationship between stress and disorder. In: Neufeld RWJ (ed.), Advances in the
investigation of psychological stress, pp. 235-383. New York NY: Wiley.
Cohen S, Hajioff J . 1972. Life events and the onset of acute closed-angle glaucoma. J
Psychosom Res 16:335-41.
Cohen S, Williamson GM. 1991. Stress and infectious disease in humans. Psychol
Bull 109:5-24
Cohen S, Tyrrell DAJ , Smith AP. 1993. Negative life events perceived stress negative
affect and susceptibility to the common cold. J Pers Soc Psychol 64:131-140.
Cohen-Cole SA, Cogen RB, Stevens AW J r, Kirk K, Gaitan E, Bird J , Cooksey R,
Freeman A. 1983. Psychiatric, psychosocial, and endocrine correlates of acute
necrotizing ulcerative gingivitis (trench mouth): a preliminary report. Psychiatr
Med 1:215-25.
Comstock GW, Helsing KJ . 1976. Symptoms of depression in two communities.
Psychol Med 6:551-563.
Condon KM. 1993. Assessing clinical significance: Application to the State-Trait
Anxiety Inventory. Thesis, Provo: Brigham Young University.
Cook RJ . 1998. Kappa and its dependence on marginal rates. In: Armitage P, Colton T
(eds.), The Encyclopedia of Biostatistics, pp. 2160-2168. New York: Wiley.
Cooper CL. 1984. The social-psychological precursors to cancer. J Human Stress
10:4-11.
Cooper CL, Faragher EB. 1993. Psychosocial stress and breast cancer: the
inter-relationship between stress events coping strategies and personality.
Psychol Med 23:653-62.
Corah NL. 1969. Development of a dental anxiety scale. J Dent Res 48:596.

256
Corby AE. 1947. Dental Caries and periodontoelasia: etiology and control. Ann Dent
6:220-227.
Corey G. 2001. Theory and practice of counseling and psychotherapy. 6th Ed.
Australia: Brooks/Cole.
Corless IB, Nicholas PK, Nokes KM. 2001. Issues in cross-cultural quality-of-life
research. J Nurs Scholarsh 33:15-20.
Costa PT J r, McCrae RR. 1985. Hypochondriasis, neuroticism, and aging: When are
somatic complaints unfounded? Am Psychol 40:1928.
Craufurd DI, Creed F, J ayson MI. 1990. Life events and psychological disturbance in
patients with low-back pain. Spine 15:490-4.
Creed F. 1989. Appendectomy. In Brown GW, Harris TO (eds.), Life event and illness,
pp. 46-68. New York: Guilford Press.
Creed F. 1993. Stress and psychosomatic disorders. In: Goldberger L, Breznitz S
(eds.), Handbook of stress: Theoretical and clinical aspects. 2nd Ed. pp. 496-510.
New York: Free Press.
Croucher R, Marcenes WS, Torres MC, Hughes F, Sheiham A. 1997. The relationship
between life-events and periodontitis. A case-control study. J Clin Periodontol
24:39-43.
Cunningham SJ , Hunt NP. 2001. Quality of life and its importance in orthodontics. J
Orthod 28:152-8.
Cupps TR, Fauci AS. 1982. Corticosteroid-mediated immunoregulation in man.
Immunol Rev 65:133-155.
Cushing A, Sheiham A, Maisels J . 1986. Developing socio-dental indicators-the social
impact of dental disease. Community Dent Health 3:3-17.
Cyr J J , McKenna-Foley J M, Peacock E. 1985. Factor structure of the SCL-90-R: is
there one? J Pers Assess 49:571-8.
Dahlstrom WG, Welsh GS. 1960. An MMPI handbook: A guide to use in clinical
practice and research. 1st Ed. Minneapolis: University of Minnesota Press.

257
Davies RM, Davies GM. 2005. Periodontal disease and general health. Dent Update
32:438-40, 442.
Davies RM, Smith RG, Porter SR. 1985. Destructive forms of periodontal disease in
adolescents and young adults. Br Dent J 158:429-436.
Davis CH, J enkins D. 1962. Mental stress and oral disease. J Dent Res 41:1045-9.
Davis J M, Albert J D, Tracy KJ , Calvano SE, Lowry SF, Shires GT, Yurt RW. 1991.
Increased neutrophil mobilization and decreased chemotaxis during cortisol and
epinephrine infusions. J Trauma 31:725-31; discussion 731-2.
De Marco TJ . 1976. Periodontal emotional stress syndrome. J Periodontol 47:67-68.
DeLongis A, Coyne J C, Dakof G, Folkman S, Lazarus RS. 1982. Relationship of daily
hassles uplifts and major life events to health status. Health Psychol 1:119-136.
DeLongis A, Folkman S, Lazarus RS. 1988. The impact of daily stress on health and
mood: Psychological and soical resources as mediators. J Pers Soc Psychol
54:486-95.
Department of Health of the HKSAR. 2002. Oral health survey 2001. Common dental
diseases and oral health related behaviour. HKSAR: The Printing Department.
Department of Health of the HKSAR. 2004. Report on Population Health Survey
2003/2004. HKSAR: Department of Health. http://www.info.gov.hk/dh/diseases/
index.htm
Derogatis LR. 1994. SCL-90-R: administration scoring and procedures manual, 3rd
Ed. Minneapolis MN: National Computer Systems.
Derogatis LR. 1996. Brief Symptom Inventory-18 manual. Minnetonka MN: NCS
Assessments.
Derogatis LR, Cons HL. 1993. Self-report measures of stress. In: Goldberger L,
Breznitz S (eds.), Handbook of stress: Theoretical and clinical aspects. 2nd Ed.
pp. 200-33. New Your: Free Press.
Derogatis LR, Melisaratos N. 1983. The Brief Symptom Inventory: an introductory
report. Psychol Med 13:595-605.

258
Dietrich T, Garcia RI. 2005. Associations between periodontal disease and systemic
disease: evaluating the strength of the evidence. J Periodontol 76:2175-84.
Dimsdale J E, Moss J . 1980. Plasma catecholamines in stress and exercise. J AMA
243:340-2.
Dohrenwend BS, Dohrenwend BP. 1974. Stressful life events: Their nature and effects.
New York: Wiley.
Dorfer CE, Becher H, Ziegler CM, Kaiser C, Lutz R, J orss D, Lichy C, Buggle F,
Bultmann S, Preusch M, Grau AJ . 2004. The association of gingivitis and
periodontitis with ischemic stroke. J Clin Periodontol 31:396-401.
Dorian B, Garfinkel PE. 1987. Stress immunity and illness--a review. Psychol Med
17:393-407.
Dorian B, Garfinkel P, Brown G, Shore A, Gladman D, Keystone E. 1982. Aberration
in lymphocyte sub-populations and function during psychological stress. Clin
Experimental Immunol 50:132-8.
Dorian BJ , Garfinkel PE, Keystone FC, Gorezynski R, Gardner DM, Darby P, Shore
A. 1985. Occupational stress and immunity. Psychosom Med 47:77.
Downer MC. 1998. The changing pattern of dental disease over 50 years. Br Dent J
185:36-41.
Edelmann RJ . 1992. Anxiety: theory, research, and intervention in clinical and health
psychology. Chichester: J ohn Wiley and Sons Ltd.
Egan KJ , Kogan HN, Garber A, J arrett M. 1983. The impact of psychological distress
on the control of hypertension. J Human Stress 9:4-10.
Eli I. 1992. Oral psychophysiology: stress, pain, and behaviour in dental care. Florida:
CRC Press.
Ellard K, Beaurepaire J , J ones M, Piper D, Tennant C. 1990. Acute and chronic stress
in duodenal ulcer disease. Gastroenterology 99:1628-32.
Ellis A. 1997. Albert Ellis on rational emotive behavior therapy. Interview by Lata K.
McGinn. Am J Psychother 51:309-16.

259
Elter J R, White BA, Gaynes BN, Bader J D. 2002. Relationship of clinical depression
to periodontal treatment outcome. J Periodontol 73:441-449.
Emrich LJ , Shlossman M, Genco RJ . 1991. Periodontal disease in
non-insulin-dependent diabetes mellitus. J Periodontol 62:123-130.
Endler NS, Edward J . 1982. Stress and Personality. In: Goldberg L, Breznitz S (eds.),
Handbook of Stress: Theoretical and clinical aspects. pp. 21-39. New York: Free
Press.
Erickson SJ , Feldman SS, Steiner H. 1996. Defense mechanisms and adjustments in
normal adolescents. Am J Psychiatry 153:826-828
Evans P, Der G, Ford G, Hucklebridge F, Hunt K, Lambert S. 2000. Social class, sex,
and age differences in mucosal immunity in a large community sample. Brain
Behav Immun 14:41-8.
Eysenck HJ . 1967. The biological basis of personality. Springfield I.L.: Charles C.
Thomas.
Eysenck HJ . 1980. Psychological theories of anxiety. In: Burrows GD, Davies B
(eds.), Handbook of studies in anxiety, pp. 101-23. Amsterdam: Elsevier.
Eysenck HJ , Eysenck MW. 1985. Personality and individual differences: A natural
science approach. New York: Plenum.
Eysenck HJ , Eysenck SBG. 1965. The Eysenck Personality Inventory. London:
University of London.
Eysenck HJ , Eysenck SBG. 1975. The manual of the Eysenck Personality
Questionnaire. London: Hodder and Stroughton.
Eysenck MW. 1989. Anxiety and Cognition: Theory and research. In: Archer T,
Nilsson LG (eds.), Aversion, avoidance and anxiety: Perspectives on aversively
motivated behaviour, pp. 437-65. Hillsdale, NJ : Erlbaum.
Eysenck MW, Mathews A. 1987. Trait anxiety and cognition. In: Eysenck HJ , Martin
I (eds.), Theoretical foundations of behaviour therapy, pp. 197-216. New York:
Plenum Press.
Farber EM, Lanigan SW, Rein G. 1990. The role of psychoneuroimmunology in the

260
pathogenesis of psoriasis. Cutis 46:314-316.
Fardal O, J ohannessen AC, Linden G. 2002. Patient perceptions of periodontal therapy
completed in a periodontal practice. J Periodontol 72:1060-6.
Fava GA, Perini GI, Santonastaso P, Fornasa CV. 1989. Life events and psychological
distress in dermatologic disorders: Psoriasis, chronic urticaria, and fungal
infections. In Miller TW (ed.), Stressful life events, pp. 701-39. Madison, CT:
international Universities Press.
Featherstone HJ , Beitman BD. 1984. Marital migraine: a refractory daily headache.
Psychosomatics 25:30-1, 35, 38.
Fedi PF J r. 1958. The effects of stress on the periodontium of the Syrian hamster. J
Periodontol 29:292-300.
Fibiger W, Singer G, Miller A. 1984. A relationship between catecholamines in urine
and physical and mental effort. Int J Psycholphysiol 1:325-333.
Fleiss, J . L. 1981. Statistical Methods for Rates and Proportions. New York: Wiley.
Florian V, Mikulincer M, Taubman O. 1995. Does hardiness contribute to mental
health during a stressful real-life situation? The roles of appraisal and coping. J
Pers Soc Psychol 68:687-95.
Foder EM. 1984. The power motive and reactivity to power stresses. J Pers Soc
Psychol 47:853-859.
Folkman S. 1984. Personal control and stress and coping process: a theoretical
analysis. J Pers Soc Psychol 46:839-52.
Folkman S, Lazarus RS. 1988. The relationship between coping and emotion:
implications for theory and research. Soc Sci med 26:309-317.
Formicola AJ , Witte ET, Curran PM. 1970. A study of personality traits and acute
necrotizing ulcerative gingivitis. J Periodontol 41:36-38.
Frandsen A, Pindborg J . 1949. Tobacco and gingivitis. III. Difference in action of
cigarette and pipe smoking. J Dent Res 28:264-265
Fredrikson M, Furst CJ , Lekander M, Rotstein S, Blomgren H. 1993. Trail anxiety

261
and anticipatory immune reactions in women receiving adjuvant chemo-therapy
for breast cancer. Brain Behav Immun 7:79-90.
Freeman R, Goss S. 1993. Stress measures as predictors of periodontal disease-a
preliminary communication. Community Dent Oral Epidemiol 21:176-177.
Freud S. 1940. An outline of psychoanalysis. Int J Psychoanal 21:27-84.
Friedman M, Rosenman RF. 1974. Type A behaviour and your heart. New York:
Knopf.
Fries H, Nillius SJ , Pettersson F. 1974. Epidemiology of secondary amenorrhea. II. A
retrospective evaluation of etiology with special regard to psychogenic factors
and weight loss. Am J Obstet Gynecol 118:473-9.
Gale E, Ayer W. 1969. Treatment of dental phobias. J Am Dent Assoc 78:1304-7.
Garcia ML, Cutress TW. 1986. A national survey of periodontal treatment needs of
adults in the Philippines. Community Dent Oral Epidemiol 14:313-6.
Gardell B. 1971. Alienation and mental health in the modern industrial environment.
In: Live L (ed.), Society Stress and Disease, pp. 148-180. London: Oxford
University Press.
Gardell B. 1982a. Scandinavian research on stress in working life. Int J Health Serv
12:31-41.
Gardell B. 1982b. Worker participation and autonomy: a multilevel approach to
democracy at the workplace. Int J Health Serv 12:527-58.
Garrett VD, Brantley PJ , J ones GN, McKnight GT. 1991. The relation between daily
stress and Crohn's disease. J Behav Med 14:87-96.
Gaspersic R, Stiblar-Martincic D, Skaleric U. 2002. Influence of restraint stress on
ligature-induced periodontitis in rats. Eur J Oral Sci 110:125-9.
Genco RJ . 1996. Current view of risk factors for periodontal diseases. J Periodontol
67:1041-1049
Genco RJ , Loe H. 1993. The role of systemic conditions and disorders in periodontal
disease. Periodontol 2000 2:98-116. Review.

262
Genco RJ , Grossi SG, Ho A, Nishimura F, Murayama Y. 2005. A proposed model
linking inflammation to obesity, diabetes, and periodontal infections. J
Periodontol 76:2075-84.
Genco RJ , Ho AW, Grossi SG, Dunford RG, Tedesco LA. 1999. Relationship of stress
distress and inadequate coping behaviors to periodontal disease. J Periodontol
70:711-23.
Genco RJ , Ho AW, Kopman J , Grossi SG, Dunford RG, Tedesco LA. 1998. Models to
evaluate the role of stress in periodontal disease. Ann Periodontol 3:288-302.
Gerra G, Monti D, Panerai AE, Sacerdote P, Anderlini R, Avanzini P, Zaimovic A,
Brambilla F, Franceschi C. 2003. Long-term immune-endocrine effects of
bereavement: relationships with anxiety levels and mood. Psychiatry Res
121:145-58.
Giddon DB, Goldhaber P, Dunning J M. 1963. Prevalence of reported cases of acute
necrotizing ulcerative gingivitis in a university population. J periodontol
34:366-370.
Giddon DB, Zackin SJ , Goldhaber P. 1964. Acute necrotizing ulcerative gingivitis in
college students. J Am Dent Assoc 68:381-386.
Gift H, Reisine S, Larach D. 1989. The social impact of dental problems and visits.
Am J Public Health 82:1163-8.
Gilbert RS. 1947. Psychogenic factor in oral disease. Oral Health 37:199-201.
Giri NC. 2004. Multivariate statistical analysis. 2nd Ed. New York: Basel Marcel
Dekker.
Goldberg D. 1972. The detection of psychiatric illness by questionnaire. London:
Oxford University Press.
Goldberg D. 1978. Manual of the general health questionnaire 1st edition Windsor:
NFER Publishing Company.
Goldberg D, Williams P. 1988. A user's guide to the General Health Questionnaire.
Windsor: NFER-NELSON.
Goldhaber P, Giddon DB. 1964. Present concepts concerning the etiology and

263
treatment of acute necrotizing ulcerative gingivitis. Int Dent J 14:468-96.
Gonder-Frederick LA, Carter WR, Cox DJ , Clarke WL. 1990. Environmental stress
and blood glucose change in insulin-dependent diabetes mellitus. Health Psychol
9:503-15.
Gooch B, Dolan TA, Bourque L. 1989. Correlates of self-reported dental health status
upon enrollment in the Rand Health Insurance Experiment. J Dental Educ
53:629-37.
Grant I, McDonald WI, Patterson T, Trimber MR. 1989. Multiple sclerosis. In Brown
GW, Harris TO (eds.), Life event and illness, pp. 321-31. New York: Guilford
Press.
Green LW, Tryon WW, Marks B, J uryn J . 1986. Periodontal disease as a function of
life-events stress. J Human Stress 12:32-36.
Green RM, Green A. 1985. Adult attitudes to dentistry among dental attenders in
South Wales. Br Dent J 159:157-60.
Greene WA, Swisher SN. 1969. Psychological and somatic variables associated with
the development and course of monozygotic twins discordant for leukemia. Ann
N Y Acad Sci 164:394-408.
Greenstein G. 1997. Contemporary interpretation of probing depth assessments:
Diagnostic and therapeutic implications. A literature review. J Periodontol
68:1194-205.
Grossi S, Dundford R, Reynolds H, Zambon J , Christersson L. 1990. Clinical and
microbiological parameters in the diagnosis of severe periodontal disease in
older adults. J Dent Res 69:361.
Grossi SG, Genco RJ , Machtei EE, Ho AW, Koch G, Dunford R, Zambon J J ,
Hausmann E. 1995. Assessment of risk for periodontal disease. II. Risk
indicators for alveolar bone loss. J Periodontol 66:23-9.
Grossi SG, Skrepcinski FB, DeCaro T, Zambon J J , Cummins D, & Genco RJ . 1996.
Response to periodontal therapy in diabetics and smokers. J Periodontol.
67:1094-102. Review.
Grossi SG, Zambon J J , Ho AW, Koch G, Dunfort RG, Machtei EE, Norderyd OM,

264
Genco R. 1994. Assessment of risk for periodontal disease. I. Risk indicators for
attachment loss. J Periodontol 65:260-267.
Groth-Marnat G. 1990. The handbook of psychological assessment. 2nd Ed. New
York: J ohn Wiley & Sons .
Grupe HE, Wilder LS. 1956. Observations of necrotizing gingivitis in 870 military
trainees. J Periodontol 27:225-61.
Guggenmoos-Holzmann I. 1996. The meaning of Kappa: Probabilistic concepts of
reliability and validity revisited. J Clin Epidemiol 49:775-82.
Gupta OP. 1966. Psychosomatic factors in periodontal disease. Dent Clin North Am
March:11-9.
Gupta OP. 1968. Psychosomatic factors in periodontal disease. J Indian Dent Assoc
40:101-7.
Gupta OP, Blechman H, Stahl SS. 1960. Effects of stress on the periodontal tissues of
young adult male rats and hamsters. J Periodontol 31:413-7.
Gwartney J , Lawson R. 2004. Economic freedom of the world: 2004 annual report.
Canada: The Fraser Institute National Book Network.
Haaga DA, Davison GC. 1993. An appraisal of rational-emotive therapy. J Consult
Clin Psychol 61:215-20.
Haber J , Kent RL. 1992. Cigarette smoking in a periodontal practice. J Periodontol
63:100-6.
Hafkenscheid A. 1993. Reliability of a standardized and expanded Brief Psychiatric
Rating Scale: a replication study. Acta Psychiatr Scand 88:305-10.
Hakeberg M, Berggren U, Carlsson SG. 1992. Prevalence of dental anxiety in an adult
population in a major urban area in Sweden. Community Dent Oral Epidemiol
20: 97-101.
Hakeberg M, Berggren U, Grondahl HG. 1993. A radiographic study of dental health
in adult patients with dental anxiety. Community Dent Oral Epidemiol 21:27-30.
Hardt J , Gerbershagen HU. 2001. Cross-validation of the SCL-27: a short

265
psychometric screening instrument for chronic pain patients. Eur J Pain
5:187-97.
Harmsen P, Rosengren A, Tsipogianni A, Wilhelmsen L. 1990. Risk factors for stroke
in middle-aged men in Goteborg, Sweden. Stroke 21:223-9.
Helke CJ . 1991. Tachykinins in autonomic control systems: The Company they keep.
Ann NY Acad Sci 632:154-69.
Hendrix WH. Factors Predictive of Stress Organizational Effectiveness and Coronary
Heart Disease Potential. 1985. Aviat Space Environ Med 56:654-659.
Hennessy CH, Moriarty DG, Zack MM, Scherr PA, Brackbill R. 1994. Measuring
health-related quality of life for public health surveillance. Public Health Rep
109:665-72.
Herbert TB, Cohen S. 1993. Depression and immunity: A meta-analytic review.
Psychol Bull 113:472-86.
Herulf G. 1950. [Roentgenographic measurement of the height of the alveolar ridge in
adolescents.] Sven Tandlak Tidskr 43:42-82. [In German]
Hildebrand HC, Epstein J , Larjava H. 2000. The influence of psychological stress on
periodontal disease. J West Soc Periodontol Periodontal Abstr 48:69-77.
Hishinuma ES, Miyamoto RH, Nishimura ST, Nahulu LB, Andrade NN, Makini GK
J r, Yuen NY, J ohnson RC, Kim SP, Goebert DA, Guerrero AP. 2000.
Psychometric properties of the state-trait anxiety inventory for
Asian/Pacific-islander adolescents. Assessment 7:17-36.
Hjelle LA, Ziegler DJ . 1992. Personality theories: Basic assumptions research and
applications. New York: McGraw-Hill.
Hoffmann N, Overall P. 1978. Factor structure of the SCL-90 in a psychiatric
population. J Consult Clin Psychol 46:1187-91.
Holcomb WR, Adams NA, Ponder HM. 1983. Factor structure of the Symptom
Checklist-90 with acute psychiatric inpatients. J Consult Clin Psychol 51:535-8.
Holmes TH, Rahe RH. 1967. The Social Readjustment Rating Scale. J Psychosom
Res 11:213-18.

266
Holmgren CJ , Corbet EF, Lim LP. 1994. Periodontal conditions among the
middle-aged and the elderly in Hong Kong. Community Dent Oral Epidemiol
22:396-402.
Horning GM, Cohen ME. 1995. Necrotizing ulcerative gingivitis periodontitis and
stomatitis: clinical staging and predisposing factors. J Periodontol 66:990-8.
Hugoson A, Koch G, Bergendal T, Hallonsten AL, Laurell L, Lundgren D, Nyman J E.
1986. Oral health of individuals aged 2-80 years in J onkoping Sweden in 1973
and 1983. Swed Dent J 10:175-94.
Hugoson A, Ljungquist B, Breivik T. 2002. The relationship of some negative events
and psychological factors to periodontal disease in an adult Swedish population
50 to 80 years of age. J Clin Periodontol 29:247-53.
Hujoel PP, Drangsholt M, Spiekerman C., DeRouen, TA. 2000. Periodontal disease
and coronary heart disease risk. J Am Med Assoc 284:1406-10.
Hujoel PP, Leroux BG, DeRouen TA, Powell LV, Kiyak HA. 1997. Evaluating the
validity of probing attachment loss as a surrogate for tooth mortality in a clinical
trial on the elderly. J Dent Res 76:858-66.
Hynes GJ , Callan VJ , Terry DJ , Gallois C. 1992. The psychological well-being of
infertile women after a failed IVF attempt: the effects of coping. Br J Med
Psychol 65:269-78.
IMD (International Institute for Management Development). 2004. The World
competitiveness yearbook 2004. Lausanne Switzerland: International Institute for
Management Development (IMD).
Irwin M, Daniels M, Bloom ET, Smith TI, Weiner H. 1987. Life events depressive
symptoms and immune function. Am J Psychiatry 144:437-441.
Ismail AI, Burt BA, Eklund SA. 1983. Epidemiologic patterns of smoking and
periodontal disease in the United States. J Am Dent Assoc 106:617-21.
J anis IL. 1958. Psychological stress. New York: Wiley.
J ohnson DB, Engel D. 1986. Acute necrotizing ulcerative gingivitis: A review of
diagnosis etiology and treatment. J Periodontol 57:141-150.

267
J ohnson GK, Hill M. 2004. Cigarette smoking and the periodontal patient. J
Periodontol 75:196-209.
Kalkwarf KL, Kaldahl WB, Patil KD. 1992. Patient preference regarding 4 types of
periodontal therapy following 3 years of maintenance follow-up. J Clin
Periodontol 19:788-93.
Kammer CH. 1994. Stress and coping of family members responsible for nursing
home placement. Res Nurs Health 17:89-98.
Kaplan HB. 1991. Social psychology of the immune system: a conceptual framework
and review of the literature. Soc Sci Med 33:909-23.
Kaplan HI. 1989. History of psychosomatic medicine. In: Kaplan HI, Sadock BJ
(eds.), Comprehensive textbook of psychiatry. 5
th
Ed. pp. 877-91. Baltimore:
Williams & Wilkins.
Karren KO, Ingle J I. 1964. Statistical analysis of the effect of natural and artificial
stress on the periodontium of rats. J Dent Res 43:793.
Kemeny ME, Cohen F, Zegans LS, Conant MA. 1989. Psychological and
immunological predictors of genital herpes recurrence. Psychosom Med
51:195-208.
Kendall PC, Finch AJ J r, Auerbach SM, Hooke J F, Mikulka PJ . 1976. The State-Trait
Anxiety Inventory: a systematic evaluation. J Consult Clin Psychol 44:406-12.
Kessler TA. 1998. The cognitive appraisal of health scale: development and
psychometric evaluation. Res Nurs Health 21:7382.
Kiecolt-Glaser J K, Glaser R. 1992. Psychoneuro-immunology: can psychological
interventions modulate immunity? J Consult Clin Psychol 60:569-75.
Kiecolt-Glaser J K, Dura J R, Speicher CE, Trask OJ , Glaser R. 1991. Spousal
caregivers of dementia victims: Longitudinal changes in immunity and health.
Psychosom Med 53:345-62.
Kiecolt-Glaser J K, Fisher L, Ogrocki P, Stout J C, Speicher CE, Glaser R. 1987.
Marital quality marital disruption and immune function. Psychosom Med
49:13-43.

268
Kiecolt-Glaser J K, Garner W, Speicher C, Penn GM, Holliday J , Glaser R. 1984a.
Psychosoical modifiers of immun-ocompetence in medical students. Psychosom
Med 46:7-14.
Kiecolt-Glaser J K, Glaser R, Williger D, Stout J , Messick G, Sheppard S, Ricker D,
Romisher SC, Briner W, Bonnell G, Donnerberg R. 1985. Psychosocial
enhancement of immunocompetence in a geriatric population. Health Psychol
4:25-41.
Kiecolt-Glaser J K, Kennedy S, Malkoff S, Fisher L, Speicher CE, Glaser R. 1988.
Marital discord and immunity in males. Psychosom Med 50:213-29.
Kiecolt-Glaser J K, Marucha PT, Malarkey WB, Mercado AM, Glaser R. 1995.
Slowing of wound healing by psychological stress. Lancet 346:1194-6.
Kiecolt-Glaser J K, Ricker D, George J , Messick G, Speicher CE, Garner W, Glaser R.
1984b. Urinary cortisol levels cellular immunocompetency and loneliness in
psychiatric inpatients. Psychosom Med 46:15-23.
Kinane DF. 2000. Aetiology and pathogenesis of periodontal disease. Ann R Australas
Coll Dent Surg 15:42-50.
Kleinknecht RA, Klepac RK, Alexander LD. 1973. Origins and characteristics of fear
of dentistry. J Am Dent Assoc 86: 842-8.
Kobasa SC. 1979. Stressful life events personality and health: an inquiry into
hardiness. J Pers Soc Psychol 37:1-11.
Kobasa SC. 1984. How much stress can you survive? Am Health 3:64-77.
Kobasa SC, Puccetti MC. 1983. Personality and social resources in stress resistance. J
Pers Soc Psychol 45:839-50.
Kocher T, Schwahn C, Gesch D, Bernhardt O, J ohn U, Meisel P, Baelum V. 2005.
Risk determinants of periodontal disease--an analysis of the Study of Health in
Pomerania (SHIP 0). J Clin Periodontol 32:59-67.
Kohlmann CW. 1993. Rigid and flexible modes of coping: Related to coping style?
Anxiety Stress Coping 6:107-23.
Kohn PM, Lafreniere K, Gurevich M. 1991. Hassles health and personality. J Pers Soc

269
Psychol 61:478-82.
Kojima M, Furukawa TA, Takahashi H, Kawai M, Nagaya T, Tokudome S. 2002.
Cross-cultural validation of the Beck Depression Inventory-II in J apan.
Psychiatry Res 110: 291-9.
Kovar M, Borzova E, Erdelski J , J any Z, Pechan J . 1989. Changes in capillary blood
flow in the gingiva during emotional excitation. Stomatologiia (Mosk) 68:23-5.
[in Russian]
Krasner D. 1978. Stress as a distal etiological factor in dental disease. J Prev Dent
5:24-7.
Kronfol Z, Silva J J r, Greden J , Dembinski S, Gardner R, Carroll B. 1983. Impaired
lymphocyte function in depressive illness. Life Sci 33:241-7.
Landis RJ , Koch GG. 1977. The measurement of observer agreement for categorical
data. Biometrics 33:159-74.
LaPerriere AR, Antoni MH, Schneiderman N, Ironson G, Klimas N, Caralis P,
Fletcher MA. 1990. Exercise intervention attenuates emotional distress and
natural killer cell decrements following notification of positive scrologic status
for HIV-1. Biofeedback Self Regul 15:229-42.
Laughlin H. 1979. The ego and its defenses. New York: Aronson.
Laurenzi MA, Persson MAA, Dalsgaard CJ , Ringden O. 1989. Stimulation of human
B lymphocyte differentiation by the neruopeptide substance P and neurokinin A.
Scand J Immunol 30:695-701.
Lautch H. 1971. Dental phobia. Br J Psychiatry 119:151-8.
Lazarus RS. 1981. The stress and coping paradigm. In: Eisdorfer C, Cohen D,
Kleinmen A, Maxim P, (eds.), Models for clinical psychopathology, pp. 177-214.
New York: Spectrum.
Lazarus RS. 1993a. Coping theory and research: Past present and future. Psychosom
Med 55:234-47
Lazarus RS. 1993b. Why we should think of stress as a subset of emotion. In:
Goldberger L, Breznitz S (eds.), Handbook of stress: Theoretical and clinical

270
aspects, 2nd Ed. pp. 21-39. New York: Free Press.
Lazarus RS. 1999. Stress and emotion: a new synthesis. New York: Springer.
Lazarus RS. 2000. Toward better research on stress and coping. Am Psychol
55:665-73.
Lazarus RS, Folkman S. 1984. Stress appraisal and coping. New York: Springer.
Lazarus RS, Lazarus BN. 1994. Passion and reason: Making sense of our emotions.
New York: Oxford University Press.
Lee J M, Song KB, Sohn HY, J eong SH, Kwon HK. 2002. Comparison between
patient expectation before and satisfaction after periodontal surgical treatment. J
Periodontol 73:1037-42.
Lefcourt HM. 1976. Locus of control: current trends in theory and research. 1st Ed.
Hillsdale: Erlbaum.
Levy S, Herberman R, Maluish A, Schlien. B, Lippman M. 1985. Prognostic risk
assessment in primary breast cancer by behavioral and immunological
parameters. Health Psychol 4:99-113.
Liedemann RR, Prilipiko LL. 1978. The behavior of T lymphocytes in schizophrenia.
In: Bergsma D, Goldstein AL, Cohen GH, Paul NW, Mudge J . (eds.),
Neurochemical and immunologic components in schizophrenia, pp. 365-377.
New York: Alan R. Liss.
Linden GJ , Mullally BH, Freeman R. 1996. Stress and progression of periodontal
disease. J Clin Periodontol 23:675-80.
Link BG, Mesagno FP, Lubner ME, Dohrenwend BP. 1990. Problems in measuring
role strains and social functioning in relation to psychological symptoms. J
Health Soc Behav 31:354-69.
Linn MW, Linn BS, J ensen J . 1984. Stressful events dysphoric mood and immune
responsiveness. Psychological reports 54:219-22.
Lipp MEN, Guevara AJ H. 1994. Empirical validation of Stress Symptoms Inventory
(ISS). Estudos de Psicologia 43-9. [in Italian]

271
Lo ECM, Schwarz E. 1994. Tooth and root conditions in the middle-aged and the
elderly in Hong Kong. Community Dent Oral Epidemiol 22:381-5.
Locker D. 1988. Measuring oral health: a conceptual framework. Community Dent
Health 5:3-13.
Locker D. 1992. The burden of oral disorders in a population of older adults.
Community Dent Health 9:109-24.
Locker D. 1995. Health outcomes of oral disorders. Int J Epidemiol 24 Suppl 1:S85-9.
Locker D. 1997. Concepts of oral health, disease and the quality of life. In: Slade GD
(ed.), Measuring oral health and quality of life, pp. 11-24. Chapel Hill, NC:
Department of Dental Ecology, University of North Carolina.
Locker D. 2003. Psychosocial consequences of dental fear and anxiety. Community
Dent Oral Epidemiol 31:144-51.
Locker D. 2004. Oral health quality of life. Oral Health Prev Dent 2:247-53. Review.
Locker D, Allen PF. 2002. Developing short-form measures of oral health-relation
quality of life. J Public Health Dent 62:13-20.
Locker D, Leake J L. 1993. Risk indicators and risk markers for periodontal disease
experience in older adults living independently in Ontario Canada. J Dent Res
72:9-17.
Locker D, Liddell A. 1991. Correlates of dental anxiety among older adults. J Dent
Res 70:198-203.
Locker D, Liddell A. 1992. Clinical correlates of dental anxiety among older adults.
Community Dent Oral Epidemiol 20:372-5.
Locker D, Miller Y. 1994. Evaluation of subjective oral health status indicators. J
Public Health Dent. 54:167-76.
Locker D, Slade GD. 1994. Association between clinical and subjective indicators of
oral health status in an older adult population. Gerodontology 11:108-114.
Locker D, Liddell A, Burman D. 1991. Dental fear and anxiety in an older adult
population. Community Dent Oral Epidemiol 19:120-4.

272
Locker D, Liddell A, Shapiro D. 1999. Diagnostic categories of dental anxiety: a
population-based study. Behav Res Ther 37: 25-37.
Locker D, Poulton R, Thomson WM. 2001. Psychological disorders and dental
anxiety in a young adult population. Community Dent Oral Epidemiol
29:456-63.
Loe H. 1967. The gingival index the plaque index and the retention index systems. J
Periodontol 38:610-6.
Loe H, Anerud A, Boysen H, Smith M. 1978. The natural history of periodontal
disease in man. The rate of periodontal destruction before 40 years of age. J
Periodontol 49:607-20.
Lovdal A, Arno A, Waerhaug J . 1958. Incidence of clinical manifestations of
periodontal disease in light of oral hygiene and calculus formation. J Am Dent
Assoc 56:21-33
Lovibond PF, Lovibond SH. 1995a. The structure of negative emotional states:
Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck
Depression and Anxiety Inventories. Behav Res Ther 33:335-43.
Lovibond SH, Lovibond PF. 1995b. Manual for the Depression Anxiety Stress Scales.
2nd Ed. Sydney: Psychology Foundation.
Lowental U. 1981. Stress anxiety and the dental patient: the missing specification. Int
Dent J 31:193-7.
Ludenia K, Donham GW. 1983. Dental outpatients; health locus of control correlates.
J Clin Psychol 39:854-8.
Lycke E, Norrby R, Roos BE. 1974. A serological study on mentally ill patients with
particular reference to the prevalence of herpes virus infections. Br J Psychiatry
124:273-9
Lynch DJ , McGrady A, Alvarez E, Forman J . 2005. Recent life changes and medical
utilization in an academic family practice. J Nerv Ment Dis 193: 633-5.
MacCallum RC, Zhang S, Preacher KJ , Rucker DD. 2002. On the practice of
dichotomization of quantitative variables. Psychol Methods 7:19-40.

273
Machtei EE, Christersson LA, Grossi SG, Dunford R, Zambon J J , Genco RJ . 1992.
Clinical criteria for the definition of established periodontitis. J Periodontol
63:206-14.
Maggirias J , Locker D. 2002. Five-year incidence of dental anxiety in an adult
population. Community Dent Health 19: 173-9.
Manhold J H, Doyle J L, Weisinger EH. 1971. Effects of social stress on oral and other
bodily tissues: results offering substance to a hypotheisis for the mechanism of
formation of periodontal pathology. J Periodontol 42:109-11.
Manold J H J r. 1970. Relationship of social stress and oral disease. Psychosomatics
11:41-4.
Marcenes WS, Croucher R, Sheiham A. 1992. The relationship between work stress
and oral health status. Soc Sci Med 35:1511-1520.
Marcenes WS, Croucher R, Sheiham A, Marmot MG. 1993. The relationship between
self-reported oral symptoms and life-events. Psychol Health 8:123-134.
Marshall-Day CD. 1951. The epidemiology of periodontal disease. J Periodontal Res
22:13-22.
Marucha PT, Kiecolt-Glaser J K, Favagehi M. 1998. Mucosal wound healing is
impaired by examination stress. Psychosom Med 60:362-5.
Master S, Gershman L. 1983. Physiological responses to rational-emotive
self-verbalizations. J Behav Ther Exp Psychiatry 14:289-96.
Mathews DC, McCulloch CAG. 1993. Evaluating patient perceptions as short-term
outcomes of periodontal treatment. A comparison of surgical and non-surgical
therapy. J Periodontol 64:990-7.
Maupin CC, Bell WB. 1975. The relationship of 17-hydroxy-corticosteriod to acute
necrotizing ulcerative gingivitis. J Periodontol 46:721-2.
McClelland DC, Burnham D. 1976. Power is the great motivator. Harv Bus Rev
March/April:100-11.
McDonald K. 1995. Evloution, the five-factor model and levels of personality. J Pers
63:526-67.

274
McGlynn FD, Gale EN, Glaros AG, LcResche L, Massoth DL, Weiffen-back J M.
1990. Biobehavioral research in dentistry: some directions for the 1990s. Ann
Behav Med 12:133-40.
McGrath CM, Bedi R. 1999. The value and use of "quality of life" measures in the
primary dental care setting. Prim Dent Care 6:53-7.
McGrath CM, Bedi R. 2004. The association between dental anxiety and oral
health-related quality of life in Britain. Community Dent Oral Epidemiol
32:67-72.
McGrath CM, Bedi R, Gilthorpe MS. 2000. Oral health-related quality of life - views
of the public in the United Kingdom. Community Dent Health 17: 3-7.
McGrath CM, Comfort MB, Lo EC, Luo Y. 2003. Can third molar surgery improve
quality of life? A 6-month cohort study. J Oral Maxillofac Surg 61:759-63;
discussion 764-5.
McGrath RE, Burkhart BR. 1983. Measuring life stress: a comparison of the
predictive validity of different scoring systems for the social readjustment rating
scale. J Clin Psychol 39:573-81.
McKeon J , Roa B, Mann A. 1984. Life events and personality trait in
obsessive-compulsive neurosis. Br J Psychiatry 144:185-9.
McMullin RE. 2000. The New Handbook of Cognitive Therapy Techniques. New
York:W W Norton.
Mehta F. 1956. Prevalence of periodontal disease. 5. Epidemiology in an Indian child
population in relation to their socio-economic status. Int Dent J 6:31-40.
Melnick SL, Roseman J M, Engel D, Cogen RB. 1988. Epidemiology of acute
necrotizing ulcerative gingivitis. Epidemiol Rev 10:191-211.
Mercer Human Resource Consulting. 2004. Cost of living.
http://money.cnn.com/pf/features/popups/costofliving/popup.html
Merchant AT, Pitiphat W, Ahmed B, Kawachi I, J oshipura K. 2003. A prospective
study of social support anger expression and risk of periodontitis in men. J Am
Dent Assoc 134:1591-6.

275
Meyer M. 1989. Stress and periodontal disease: a review of the literature. J N Z Soc
Periodonotol 68:23-26.
Michalowicz BS. 1994. Genetic and heritable risk factors in periodontal disease. J
Periodontol 65:479-88.
Michalowicz BS, Aeppli D, Virag J G, Klump DG, Hinrichs J E, Segal NL, Bouchard
TJ J r, Pihlstrom BL. 1991. Periodontal findings in adult twins. J Periodontol
62:293-9.
Michalowicz BS, Diehl SR, Gunsolley J C, Sparks BS, Brooks CN, Koertge TE,
Califano J V, Burmeister J A, Schenkein HA. 2000. Evidence of a substantial
genetic basis for risk of adult periodontitis. J Periodontol 71:1699-707.
Michelberger D, Matthews D. 1996. Periodontal manifestations of systemic diseases
and their management. J Can Dent Assoc 62:313-4,317-21.
Milgrom P, Fiset L, Melnick S, Weinstein P. 1988. The prevalence and practice
management consequences of dental fear in a major US city. J Am Dent Assoc
116:641-7.
Milgrom P, Weinstein P, Kleinknecht R, Getz T. 1985. Treating fearful dental patients:
a patient management handbook. Reston, Virginia: Reston Pub. Co..
Miller A, Brunelle J , Carlos J , Brown L, Le H. 1987. Oral health of United States
adults. The National Survey of Oral Health in U.S. Employed Adults and Seniors:
1985-1986. National Findings (NIH Publication No.87-2868). Bethesda:
National Institutes of Health.
Miller SC, Firestone J M. 1947. Psychosomatic factors in the etiology of periodontal
disease: a critical review of the literature. Am J Orthod Oral Surg 33:675-86.
Miller TW. 1989. Stressful life events. Madison: International University Press.
Minneman MA, Cobb C, Soriano F, Burns S, Schuchman L. 1995. Relationhsips of
personality traits and stress to gingival status or soft-tissue oral pathology: An
exploratory study. J Public Health Dent 55:22-7.
Mishel M, Sorenson D. 1991. Uncertainty in gynecological cancer: a test of the
mediating functions of mastery and coping. Nurs Res 36:98103.

276
Molin C, Seeman K. 1970. Disproportionate dental anxiety. Clinical and nosological
considerations. Acta Odontol Scand 28:197-212.
Monteiro da Silva AM, Newman HN, Oakley DA. 1995. Psychosocial factors in
inflammatory periodontal diseases. A review. J Clin Periodontol 22:516-26.
Monteiro da Silva AM, Oakley DA, Newman HN, Nohl FS, Nohl HM. 1996.
Psychosocial factors and adult rapidly progressive periodontitiis. J Clinc
Periodontol 23:789-94.
Moore R, Birn H, Kirkegaard E, Brodsgaard I, Scheutz F. 1993. Prevalence and
characteristics of dental anxiety in Danish adults. Community Dent Oral
Epidemiol 21: 292-6.
Morrison DF. 2004. Multivariate statistical methods. 4th Ed. Southbank Vic.:
Brooks/Cole Thomson Learning.
Moss ME, Beck J D, Kaplan BH, Offenbacher S, Weintraub J A, Koch GG, Genco RJ ,
Matchei EE, Tedesco LA. 1996. Exploratory case-control analysis of
psychosocial factors and adult periodontitis. J Periodontol 67:1060-9.
Moulton R. 1955. Oral and dental manifestations of anxiety. Psychiatry 18:261-73.
Moulton R, Ewen S, Thieman W. 1952. Emotional factors in periodontal disease. Oral
Surg 5:833-60.
Muldoon MF, Barger SD, Flory J D, Manuck SB. 1998. What are quality of life
measurements measuring? Br Med J 316:542-5.
Murayama Y, Kurihara H, Nagai A, Dompkowski D, Van Dyke TE. 1994. Acute
necrotizing ulcerative gingivitis: Risk factors involving host defense mechanisms.
Periodontol 2000 6:116-24.
Myers A, Hansen CH. 2002. Experimental psychology. 5th Ed. Pacific Grove Calif.:
Wadsworth/Thomson Learning.
Needleman I, McGrath C, Floyd P, Biddle A. 2004. Impact of oral health on the life
quality of periodontal patients. J Clin Periodontol 31:454-7.
Newcomb PA, Carbone PP. 1992. The health consequences of smoking: Cancer. Med
Clin North Am 76:305-31.

277
Newman HN. 1974. Diet, attrition, plaque and dental disease. Br Dent J 136:491-7.
Ng SKS, Leung WK. 2006a. Oral health-related quality of life and periodontal status.
Community Dent Oral Epiedmiol 34:114-22.
Ng SKS, Leung WK. 2006b. A community study on the relationship between stress,
coping, affective dispositions and periodontal attachment loss. Community Dent
Oral Epidemiol in press.
Ng SKS, Chau AW, Leung WK. 2004. The effect of pre-operative information in
relieving anxiety in oral surgery patients. Community Dent Oral Epidemiol
32:227-35.
Ng SKS, Stouthard MEA, Leung WK. 2005. Validation of a Chinese version of
Dental Anxiety Inventory. Community Dent Oral Epidemiol 33:107-14.
Nieuwenhuijsen K, de Boer AG, Verbeek J H, Blonk RW, van Dijk FJ . 2003. The
Depression Anxiety Stress Scales (DASS): detecting anxiety disorder and
depression in employees absent from work because of mental health problems.
Occup Environ Med 60:i77-82.
Oberst MT, Hughes S, Chang A, McCubin M. 1991. Self-care burden, stress appraisal,
and mood among persons receiving radiotherapy. Cancer nurs 14:7178.
Offenbacher S, Katz V, Fertik G, Collins J , Boyd D, Maynor G, McKaig R, Beck J .
1996. Periodontal infection as a possible risk factor for preterm low birth weight.
J Periodontol 67:1103-13.
O'Leary T, Shannon IL, Prigmore J R. 1962. Clinical and systemic findings in
periodontal disease. J Periodontol 33:243-51.
Oliver RC, Brown LJ , Loe H. 1998. Periodontal diseases in the United States
population. J Periodontol 69:269-78.
Owens J D, Dowsett SA, Eckert GJ , Zero DT, Kowolik MJ . 2003. Partial-mouth
assessment of periodontal disease in an adult population of the United States. J
Periodontol 74:1206-13.
Page RC, Beck J D. 1997. Risk assessment for periodontal diseases. Int Dent J
47:61-87.

278
Page RC, Kornman KS. 1997. The pathogenesis of human periodontitis: an
introduction. Periodontol 2000 14:9-11.
Pagel MD, Smilkstein G, Regen H, Montano D. 1990. Psychosocial influences on new
born outcomes: a controlled prospective study. Soc Sci Med 30:597-604.
Palmer RM. 1988. Tobacco smoking and oral health. Br Dent J 164:258-60. Review.
Paquette DW. 2002. The periodontal infection-systemic disease link: a review of the
truth or myth. J Int Acad Periodontol 4:101-9.
Pare WP, Glavin GB. 1986. Restraint stress in biomedical research: A review.
Neurosci Biobehav Rev 10:339-70.
Pearlin LI. 1975. Status inequality and stress in marriage. Am Sociol Rev 40:344-57.
Pearlin LI, Radabaugh CW. 1976. Economic strains and the coping functions of
alcohol. Am J Sociol 82:652-63.
Pearlin LI, Schooler C. 1978. The structure of coping. J Health Soc Beh 19:2-21.
Pearson Assessments. 2004. SCL-90-R. USA: NCS Pearson Inc.
Pedersen BK, Bruunsgaard H, Klokker M, Kappel M, MacLean DA, Nielsen HB,
Rohde T, Ullum H, Zacho M. 1997. Exercise-induced
immunomodulation--possible roles of neuroendocrine and metabolic factors. Int
J Sports 18:S2-7.
Pennebaker J W. 1989. Confession inhibition and disease. In: Berkowitz L (ed.),
Advances in experimental social psychology, vol.22, pp. 211-244. Orlando FL:
Academic Press.
Pennebaker J W, Colder M, Sharp LK. 1990. Accelerating the coping process. J P Soc
Psychol 58:528-37.
Pennebaker J W, Kiecolt-Glaser J K, Glaser R. 1988. Disclosure of traumas and
immune function: Health implications for psychotherapy. J Consult Clin Psychol
56:239-45.
Petersen PE, Ogawa H. 2005. Strengthening the Prevention of Periodontal Disease:
The WHO Approach. J Periodontol 76:2187-93.

279
Pihlstrom BL, Michalowicz BS, J ohnson NW. 2005. Periodontal diseases. Lancet
366:1809-20.
Pike J L, Smith TL, Hauger RL, Nicassio PM, Patterson TL, McClintick J , Costlow C,
Irwin MR. 1997. Chronic life stress alters sympathetic neuroendocrine and
immune responsivity to an acute psychological stressor in humans. Psychosom
Med 59:447-57.
Pilgram TK, Hildebolt CF, Yokojama-Crothers N, Dotson M, Cohen SC, Hauser J F,
Kardaris E. 2000. Relationships between radiographic alveolar bone height and
probing attachment level: data from healthy post-menopausal women. J Clin
Periodontol 27: 341-6.
Pillow DR, Zautra AJ , Sandler I. 1996. Major life events and minor stressors:
Identifying mediation links in the stress process. J Pers Soc Psychol 70:381-94.
Pindborg J J . 1951. Influence of service in Armed Forces on incidence of gingivitis. J
Am Dent Assoc 42:517-22.
Pistorius A, Krahwinkel T, Willershausen B, Boekstegen C. 2002. Relationship
between stress factors and periodontal disease. Eur J Med Res 7:393-8.
Plutchik R, Williams MH J r, J errett I, Karasu TB, Kane C. 1978. Emotions,
personality and life stresses in asthma. J Psychosom Res 22:425-31.
Powell TJ . 1997. Free yourself from harmful stress. New York NY: D. K. Publishing.
Powell TJ . 2001. The mental health handbook [electronic resource]. Great Britain:
Winslow.
Rabkin J G. 1993. Stress and psychiatric disorders. In: Goldberger L, Breznitz S (eds.),
Handbook of stress: Theorectical and clinical aspects, 2nd Ed. pp. 477-95. New
York: Free Press.
Rahe RH, Arthur RH. 1978. Life change and illness studies. J Human Stress 4:3-15.
Rahe RH, Holmes TH. 1965. Social, Psychological and psychophysiologic aspects of
inguinal hernia. J Psychosom Res 148:487-91.
Raphael KG, Cloitre M, Dohrenwend BP. 1991. Problems of recall and
misclassification with checklist methods of measuring stressful life events.

280
Health Psychol 10:62-74.
Ratcliff PA. 1956. The relationship of the general adaptation syndrome to the
periodontal tissues in the rat. J Periodontol 27:40-3.
Regenbaum G. 1970. Psychogenic aspects of periodontal disease. Case history. N Y
State Dent J 36:609-12.
Reisine ST. 1984. Dental disease and work loss. J Dent Res 63:1158-61.
Reisine ST, Fertig J , Weber J , Leder S. 1989. Impact of dental conditions on patients'
quality of life. Community Dent Oral Epidemiol 17:7-10.
Rijsdijk FV, Sham PC, Sterne A, Purcell S, Mcguffin P, Farmer A, Goldberg D, Mann
A, Cherny SS, Webster M, Ball D, Eley TC, Plomin R. 2001. Life events and
depression in a community sample of siblings. Psychol Med 31:401-10.
Rimon R, Halonen P, Anttinen E, Evola K. 1971. Complement fixing antibody to
herpes simplex virus in patients with psychotic depression. Dis Nerv Syst
32:822-4.
Ringsdorf WM, Cheraskin E. 1969. Emotional status and the periodontium. J Tenn
State Dent Assoc 49:5-18.
Rogers MP, Dubey D, Reich P. 1979. The influence of the psyche and brain on
immunity and disease susceptibility. A critical review. Psychosom Med
41:147-64.
Rosengren A, Tibblin G, Wilhelmsen L. 1991. Self-perceived psychological stress and
incidence of coronary artery disease in middle-aged men. Am J Cardiol
68:1171-5.
Rosenman RH. 1993. Relationships of the Type A behavior pattern with coronary
heart disease. In: Goldberger L, Breznitz S (eds.), Handbook of stress:
Theoretical and clinical aspects, 2nd Ed. pp. 449-76. New York: Free Press.
Roskam EECI, Broers NJ . 1996. Constructing questionnaires: An application of facet
design and item response theory to the study of lonesomeness. In: Engelhard G,
Wilson M (eds.), Objective measurement: Theory into practice, pp. 349-85.
Norwood: Ablex.

281
Rost K, Smith GR, Burnam MA, Burns BJ . 1992. Measuring the outcomes of care for
mental health problems. The case of depressive disorders. Med Care 30:266-73.
Rotter J B. 1966. Generalized expectancies for internal versus external control of
reinforcement. Psychol Monogr 80:1-28.
Rotter J B. 1975. Some problems and misconceptions related to the construct of
internal versus external control of reinforcement. J Consult Clin Psychol
43:56-67
Rotter J B. 1990. Internal versus external control of reinforcement: A case history of a
variable. Am Psychol 45:489-93.
Rubin L. 1963. Emotional stress as an etiological factor in periodontal disease. Alpha
Omegan 56:146-9.
Russell A. 1960. Geographical distribution and epidemiology of periodontal disease.
(WHO/DH/33/34). Geneva: World Health Organization.
Saito T, Tazawa K, Yokoyama Y, Saito M. 1997. Surgical stress inhibits the growth of
fibroblasts through the elevation of plasma catecholamine and cortisol
concentrations. Surg Today 27:627-31.
Salvi GE, Lawrence HP, Offenbacher S, Beck J D. 1997. Influence of risk factors on
the pathogenesis of periodontitis. Periodontol 2000 14:173-201.
Sandler HC. 1954. The influence of generalized diseases on clinical manifestations of
periodontal disease. J Am Dent Assoc 49:656-67.
Sandler HC. 1960. Prevalence of periodontal disease in a hospitalized population. J
Dent Res 39:439-49.
Sandler IN, Guenther RT. 1985. Assessment of life stress events. In: Karoly P (ed.),
Measurement strategies in health psychology, pp. 555-600. Singapore: Wiley.
Scannapieco FA, Ho AW. 2001. Potential associations between chronic respiratory
disease and periodontal disease: analysis of National Health and Nutrition
Examination Survey III. J Periodontol 72:50-6.
Scannapieco FA, Bush RB, Paju S. 2003. Associations between periodontal disease
and risk for atherosclerosis, cardiovascular disease, and stroke. A systematic

282
review. Ann Periodontol 8:38-53.
Schei O, Arno A, Lvdal A, Waerhaug J . 1959a. Alveolar bone loss as a function of
tobacco consumption. Acta Odontol Scand 17:3-10.
Schei O, Waerhaug J , Lvdal A, Arno A. 1959b. Alveolar bone loss as related to oral
hygiene and age. J Periodontol 30:7-16
Scheier MF, Carver CS. 1985. Optimism coping and health: Assessment and
implications of generalized outcome expectancies. Health Psychol 4:219-47.
Scheier MF, Carver CS. 1987. Dispositional optimism and physical well-being: The
influence of generalized outcome expectancies on health. J Pers 55:169-210.
Scheier MF, Carver CS, Bridges MW. 1994. Distinguishing optimism from
neuroticism (and trait anxiety self-mastery and self-esteem): A re-evaluation of
the life orientation test. J Pers Soc Psychol 67:1063-78.
Scherwitz L, Berton K, Leventhal H. 1978. Type A behavior self-involvement and
cardiovascular response. Psychosom Med 40:593-609.
Schleifer SJ , Keller SE, Camerino M, Thornton J C, Stein M. 1983. Suppression of
lymphocyte stimulation following bereavement. J Am Med Assoc 250:374-77.
Schleifer SJ , Keller SE, Meyerson AT, Raskin MJ , Davis KI, Stein M. 1984.
Lymphocyte function in major depressive disorder. Arch Gen Psychiatry
41:484-6.
Schleimer RP. 1990. Effects of glucocorticosteriods on inflammatory cells relevant to
their therapeutic applications in asthma. Am Rev Respir Dis 141:59-69.
Schleimer RP, Freeland HS, Peters SP, Brown KE, Derse CP. 1989. An assessment of
the effects of glucocorticorids on degranulation chemotaxis binding to vascular
endothelium and formation of leukotriene-B4 by purified human neutrophils. J
Pharmacol Exp Ther 250:598-605.
Schmitz N, Hartkamp N, Franz M, Buse S, Karig R, Tress W. 2002. Properties of the
Symptom Check List (SCL-90-R) in a psychosomatic consultation-liaison setting.
Psychol Rep 90:1201-7.
Schmitz N, Hartkamp N, Kiuse J , Franke GH, Reister G, Tress W. 2000. The

283
Symptom Check-List-90-R (SCL-90-R): a German validation study. Qual Life
Res 9:185-93.
Schoor RS, Havrilla J . 1986. Acute necrotizing ulcerative gingivitis: etiology and
stress relationships. Int J Psychosom 33:35-40.
Schultz DP. 1990. Psychology and industry today, 5th Ed. New York: MacMillan.
Schuurs AHB, Hoogstraten J . 1993. Appraisal of dental anxiety and fear
questionnaires: a review. Community Dent Oral Epidemiol 21:329-39.
Schuurs AHB, Duivenvoorden HJ , Thoden van Velzen SK, Verhage F. 1980. Three
factors predicting irregular versus regular dental attendance: a model fitting to
empirical data. Community Dent Oral Epidemiol 8:413-9.
Schuurs AH, Duivenvoorden HJ , Thoden van Velzen SK, Verhage F, Eijkman MA,
Makkes PC. 1985. Sociodemographic correlates of dental anxiety. Community
Dent Oral Epidemiol 13:212-5.
Schwarz E, Birn H. 1995. Dental anxiety in Danish and Chinese adults--a
cross-cultural perspective. Soc Sci Med 41:123-30.
Schwarz E, Lo EC. 1994. Dental health knowledge and attitudes among the
middle-aged and the elderly in Hong Kong. Community Dent Oral Epidemiol
22:358-63.
Seffert K, Hosoi J , Torii H, Ozawa H, Ding W, Campton K, Wagner J A, Granstein RD.
2002. Catecholamines inhibit the antigen-presenting capability of epidermal
Langerhans cells. J Immunol 168:6128-35.
Segerstrom SC, Miller GE. 2004. Psychological stress and the human immune system:
a meta-analytic study of 30 years of inquiry. Psychol Bull 130:601-30.
Selye H. 1936. A syndrome produced by diverse nocuous agents. Nature 138:32.
Selye H. 1946. The general adaptation syndrome and the diseases of adaptation. J Clin
Endocrinol 6:117-230.
Selye H. 1948. The alarm reaction and the diseases of adaptation. Ann Intern Med
29:403-15.

284
Selye H. 1956. The stress of life. New York: McGraw-Hill.
Selye H. 1975. Stress without distress, 2nd Ed., p41. London: Hodder and Stoughton
Limited.
Selye H. 1982. History and present status of the stress concept. In: Goldberger L,
Breznitz S (eds.), Handbook of stress: Theoretical and clinical aspects, 7-20.
New York: Free Press.
Senna A, Campus G, Gagliani M, Strohmenger L. 2005. Socio-economic influence on
caries experience and CPITN values among a group of Italian call-up soldiers
and cadets. Oral Health Prev Dent. 3:39-46.
Seta J J , Seta CE, Wang MA. 1991. Feelings of negativity and stress: An averaging
summation analysis of impressions of negative life experiences. Pers Soc
Psychol Bull 17:376-84.
Seymour GJ . 1991. Importance of the host response in the periodontium. J Clin
Periodontol 18:421-6.
Shannon IL, Kilgore WG, O'Leary T. 1969. Stress as a predisposing factor in
necrotizing ulcerative gingivitis. J Periodontol 40:240-242.
Shields WD. 1977. Acute necrotizing ulcerative gingivitis: a study of some of the
contributing factors and their validity in an army population. J Periodontol
48:346-9.
Shimura N, Nakamura C, Hirayama Y, Yonemitsu M. 1983. Anxiety and dental caries.
Community Dent Oral Epidemiol 11:224-7.
Shklar G, Glickman I. 1953. The periodontiam and salivary glands in the alarm
reaction. J Dent Res 32:773-8.
Shlossman M, Knowler W, Pettitt D, Genco R. 1990. Type 2 diabetes mellitus and
periodontal disease. J Am Dent Assoc 121:532-6.
Sieber WJ , Rodin J , Larson L, Ortega S, Cummings N, Levy S, Whiteside T,
Herberman R. 1992. Modulation of human natural killer cell activity by exposure
to uncontrollable stress. Brain Behav Immun 6:141-56.
Siegel J M, J ohnson J H, Sarason IG. 1979. Life changes and menstrual discomfort. J

285
Human Stress 5:41-6.
Simos G. 2002. Cognitive Behaviour Therapy: A Guide for the Practising Clinician.
New York: Brunner-Routledge.
Skaret E, Raadal M, Berg E, Kvale G. 1998. Dental anxiety among 18-yr-olds in
Norway. prevalence and related factors. Eur J Oral Sci 106:835-43.
Slade GD. 1997a. Derivation and validation of a short-form oral health impact profile.
Community Dent Oral Epidemiol 25:284-90.
Slade GD. 1997b. Measuring Oral Health and Quality of Life. Chapel Hill, NC:
University of North Carolina, Dental Ecology.
Slade GD, Spencer AJ . 1994. Development and evaluation of the Oral Health Impact
Profile. Community Dent Health 11:3-11.
Smari J , Arason E, Hafsteinsson H, Ingimarsson S. 1997. Unemployment, coping and
psychological distress. Scand J Psychol 38:151-6.
Smith TA, Heaton LJ . 2003. Fear of dental care: are we making any progress? J Am
Dent Assoc134:1101-8.
Snyder DS, Unanue ER. 1982. Corticosteroids inhibit murine macrophage Ia
expression and interleukin 1 production. J Immunol 129:1803-5.
Solomon GF. 1981. Immunologic abnormalities in mental illness. In: Ader R (ed.),
Psychoneuro-immunology, pp. 259-278. Orlando: Academic Press.
Somerfield MR, commentators. 1997. The utility of systems models of stress and
coping for applied research: The case of cancer adaptation. J Health Psychol
2:133-72.
Sommer R, Sommer B. 2002. A practical guide to behavioral research: tools and
techniques. 5th Ed. New York: Oxford University Press.
Soskolne WA, Klinger A. 2001. The relationship between periodontal diseases and
diabetes: an overview. Ann Periodontol 6:91-8.
Spielberger CD. 1966. Theory and research on anxiety. In: Spielberger CD (ed.),
Anxiety and behaviour, pp. 3-20. New York: Academic Press.

286
Spielberger CD. 1972. Anxiety as an emotional state. In: Spielberger CD (ed.),
Anxiety: Current trends in theory and research (Vol. 1), pp. 23-49. New York:
Academic.
Spielberger CD. 1985. Anxiety Cognition and affect: A State-Trait perspective. In:
Tuma AH, Maser J (eds.), Anxiety and anxiety disorders, pp. 171-82. Hillsdale,
NJ : Erlbaum.
Spielberger CD, Barker L, Russell S, Silva de, Crane R, Westberry L, Knight J , Marks
E. 1979. Preliminary manual for the state-trait personality inventory (STPI).
Unpublished manual University of South Florida.
Spielberger CD, Gorsuch RL, Lushene RE. 1970. Manual for the state-trait anxiety
inventory. Palo Alto CA: Consulting Psychological Press.
SPSS Inc. 2004. SPSS Base 12.0 User's Guide Package. Chicago: SPSS Inc.
Stabholz A, Peretz B. 1999. Dental anxiety among patients prior to different dental
treatments. Int Dent J 49:90-4.
Sternberg EM. 1988. Monokines lymphokines and the brain. In: Cruse J M, Lewis RE
J r (eds.), The Year in Immunology vol. 5, pp. 205-207. Basel Switzerland:
Karger.
Sternberg EM, Licinio J . 1995. Overview of neuroimmune stress interactions.
Implications for susceptibility to inflammatory disease. Ann NY Acad Sci
771:364-71.
Stone AA, Bovbjerg DH, Neale J M, Napoli A, Valdimarsdottir H, Cox D, Hayden FG,
Gwaltney J M J r. 1992. Development of common cold symptoms following
experimental rhinovirus infection is related to prior stressful life events. Behav
Med 18:115-20.
Stouthard ME. 1989. Angst voor de tandheelkundige behandeling. [Dental Anxiety.]
Thesis Amsterdam: University of Amsterdam. [In Dutch]
Stouthard ME, Hoogstraten J . 1990a. Prevalence of dental anxiety in The Netherlands.
Community Dent Oral Epidemiol 18:139-42.
Stouthard ME, Hoogstraten J . 1990b. Dental anxiety: a comparison of dentate and
edentulous subjects. Community Dent Oral Epidemiol 18:267-8.

287
Stouthard ME, Hoogstraten J , Mellenbergh GJ . 1995. A study on the convergent and
discriminant validity of the Dental Anxiety Inventory. Behav Res Ther
33:589-95.
Stouthard ME, Mellenbergh GJ , Hoogstraten J . 1993. Assessment of dental anxiety: a
facet approach. Anxiety Stress Coping 6:89-105.
Sutton PR. 1990. Acute dental caries, mental stress, immunity and the active passage
of ions through the teeth. Med Hypotheses 31:17.
Takada T, Yoshinari N, Sugiishi S, Kawase H, Yamane T, Noguchi T. 2004. Effect of
restraint stress on the progression of experimental periodontitis in rats. J
Periodontol 75:306-15.
Tang QP, Cheng ZH, Yan AH, Den YL. 1999. The use and reanalysis of the SCL-90 in
China. Chin J of Clin Psychol 7:16-20. [in Chinese]
Taylor SE. 1995. Health psychology. 3rd Ed. New York NY: McGraw-Hill.
Taylor SE, Aspinwall LG. 1996. Mediating and moderating processes in psychological
stress: Appraisal coping resistance and vulnerability. In: Kaplan HG (ed.),
Psychological stress: Perspectives on structure theory life-course and methods,
pp. 71-110. San Diego CA: Academic Press.
Teng HC, Lee CH, Hung HC, Tsai CC, Chang YY, Yang YH, Lu CT, Yen YY, Wu YM.
2003. Lifestyle and psychosocial factors associated with chronic periodontitis in
Taiwanese adults. J Periodontol 74:1169-75.
The Mental Health Association of China. 1993. Rating scales for Mental Health.
Beijing: Chinese Mental Health J ournal. [in Chinese]
Thomson WM, Locker D, Poulton R. 2000. Incidence of dental anxiety in young
adults in relation to dental treatment experience. Community Dent Oral
Epidemiol 28:289-94.
Thomason BT, Brantley PJ , J ones GN, Dyer HR, Morris J L. 1992. The relation
between stress and disease activity in rheumatoid arthritis. J Behav Med
15:215-20.
Tomar SL, Asma S. 2000. Smoking-Attributable Periodontitis in the United States:
Findings from NHANES III. J Periodontol 71:743-51.

288
Torabi-Gaarden R, Breivik T, Hansen BF, Malt UF, Gjermo PE. 2004. Negative life
events anxiety depression and coping ability (stress) as related to chronic
periodontitis. Perio 1:35-42.
Torrungruang K, Tamsailom S, Rojanasomsith K, Sutdhibhisal S, Nisapakultorn K,
Vanichjakvong O, Prapakamol S, Premsirinirund T, Pusiri T, J aratkulangkoon O,
Unkurapinun N, Sritara P. 2005. Risk indicators of periodontal disease in older
Thai adults. J Periodontol 76:558-65.
Tsuchiya T, Horii I. 1996. Epidermal cell proliferative activity assessed by
proliferating cell nuclear antigen (PCNA) decreases following
immobilization-induced stress in male Syrian hamsters.
Psychoneuroendocrinology 21:111-7.
Turner J R, Wheaton B. 1995. Checklist measurement of stressful life events. In:
Cohen S, Kessler RC, Gordon LU (eds.), Measuring stress: A guide for health
and social scientists. pp. 29-58. New York: Oxford University Press.
Turner SM, Michelson L. 1984. Conceptual methodological and clinical issues in the
assessment of anxiety. J Behav Assess 6:265-79.
US Department of Health and Human Services. 1992. Healthy people 2000. National
health promotion and disease prevention objectives. Boston: J ones and Bartlett.
Vaillant GE. 1994. Ego mechanisms of defense and personality psychopathology. J
Abnorm Psychol 103:44-50.
VanderPlate C, Aral SO, Magder L. 1988. The relationship among genital herpes
simplex virus, stress, and social support. Health Psychol 7:159-68.
Vassend O. 1993. Anxiety, pain and discomfort associated with dental treatment.
Behav Res Ther 31:659-66.
Vettore MV, Leao AT, Monteiro da, Silva AM, Quintanilha RS, Lamarca GA. 2003.
The relationship of stress and anxiety with chronic periodontitis. J Clin
Periodontol 30:394-402.
Vines RF, Richards J C, Thomson DM, Brechman-Toussaint M, Kluin M, Vesely L.
2004. Clinical psychology in general practice: a cohort study. Med J Aust
181:74-7.

289
Vogel RI, Morante EA, lves C, Diamond R. 1977. Relationships of personality traits
and periodontal disease. Psychosomatics 18:21-4.
Wactawski-Wende J , Grossi SG, Trevisan M, Genco RJ , Tezal M, Dunford RG, Ho
AW, Hausmann E, Hreshchyshyn MM. 1996. The role of osteopenia in oral bone
loss and periodontal disease. J Periodontol 67:1076-84. Review.
Walker LG, J ohnson VC, Eremin O. 1993. Modulation of the immune response to
stress by hypnosis and relaxation training in healthy volunteers. A critical review.
Contemp Hypn 10:19-27.
Wallis BJ , Bogduk N. 1996. Faking a profile: can naive subjects simulate whiplash
responses? Pain 66:223-7.
Wallis BJ , Lord SM, Barnsley L, Bogduk N. 1996. Pain and psychologic symptoms of
Australian patients with whiplash. Spine 21:804-10.
Wallis BJ , Lord SM, Barnsley L, Bogduk N. 1998. The psychological profiles of the
patients with whiplash-associated headache. Cephalalgia 18:101-5.
Wallis BJ , Lord SM, Bogduk N. 1997. Resolution of psychological distress of
whiplash patients following treatment by radiofrequency neurotomy -- a
randomized double-blind placebo-controlled trial. Pain 73:15-22.
Wallston KA, Wallston BS, DeVellis R. 1978. Development of the Multidimensional
Health Locus of Control (MHLC) Scales. Health Educ Monogr 6:160-70.
Wang J H, Zhang FG, Song LL, Wang ZL, Tian J , Zou HG. 2000. The development of
the norm of the SCL-90 in Chinese recruits. Chin Mental Health J 14:228-30.
[In Chinese]
Watson D, Clark LA. 1984. Negative affectivity: The disposition to experience
aversive emotional states. Psychol Bull 96:465-90.
Watson D, Pennebaker J W. 1989. Health complaints stress and distress: exploring the
central role of negative affectivity. Psychol Review 96:234-54.
Watson D, Tellegen A. 1985. Toward a consensual structure of mood. Psychol Bull
98:219-35.
Weinberger M, Hiner SL, Tierney WM. 1987. In support of hassles as a measure of

290
stress in predicting health outcomes. J Behav Med 10:19-31.
Weiner H. 1978. Emotional factors. In Werner SC, Ingbar SH (eds.), The thyroid, pp.
251-70. New York: Harper & Row.
Weintraub J A. 1998. Uses of oral health related quality of life measures in public
health. Community Dent Health 15:8-12.
Weiten W. 2004. Psychology : themes and variations. 6th Ed. Belmont California:
Wadsworth Thomson Learning.
Whitehead SP, Watts TL.1987. Short-term effect of Keyes' approach to periodontal
therapy compared with modified Widman flap surgery. J Clin Periodontol
14:599-604.
Whitlock FA, Sisking M. 1979. Depression and cancer: a follow-up study. Psychol
Med 9:747-52.
Wiebe DJ . 1991. Hardiness and stress moderation: A test of proposed mechanisms. J
Pers Soc Psychol 60:89-99.
Williams NA, Deffenbacher J L. 1983. Life stress and chronic yeast infections. J
Human Stress 9:26-31.
Williams PG, Wiebe DJ , Smith TW. 1992. Coping processes as mediators of the
relationship between hardiness and health. J Behav Med 15:237-55.
Wimmer G, J anda M, Wieselmann-Penkner K, J akse N, Polansky R, Pertl C. 2002.
Coping with stress: its influence on periodontal disease. J Periodontol
73:1343-51.
Wolf S, Goodell H. 1968. Stress and disease. Springfield, IL: Charles D Thomas.
Wolman BB. 1989. Dictionary of behavioral science, 2nd Ed., p351. New York:
Academic Press Inc.
Wong KC. 1996. Depression and its related mood conditions in Hong Kong. Master
Dissertation Hong Kong: Department of Psychology Faculty of Social Science
The University of Hong Kong.
Wong MC, Lo EC, McMillan AS. 2002. Validation of a Chinese version of the Oral

291
Health Impact Profile (OHIP). Community Dent Oral Epidemiol 30:423-30.
Woolfolk RL, Richardson FC. 1978. Stress sanity and survival. New York:
Sovereign/Monarch.
Woolgrove J , Cumberbatch G, Gelbier S. 1987. Understanding dental attendance
behaviour. Community Dent Health 4:215-21.
Zachariae R, Bjerring P. 1990. The effect of hypnotically induced analgesia on flare
reaction of the cutancous histamine prick test. Arch Dermatol Res 282:539-43.
Zaidens SH. 1954. Psychosomatic disorders of the mouth. N Y J Dent 24:218-22.
Zambon J J , Grossi SG, Machtei EE, Ho AW, Dunford R, & Genco RJ . 1996. Cigarette
smoking increases the risk for subgingival infection with periodontal pathogens.
J Periodontol 67:1050-4.
Zheng XH, Shu AL, Zhang L, Huang GL. 1993. Evaluation and application of
State-Trait Anxiety Inventory in Changchu. Chin Mental Health J 7: 60-2. [In
Chinese]
Zigmond AS, Snaith RP. 1983. The hospital anxiety and depression scale. Acta
Psychiatr Scand 67:361-70.
Zuniga J R. 2000. Guidelines for anxiety control and pain management in oral and
maxillofacial surgery. J Oral Maxillofac Surg 58:4-7.