Anda di halaman 1dari 8


Socio-environmental factors associated with

dental occlusion in adolescents
Paulo Frazãoa and Paulo Capel Narvai
Santos and São Paulo, Brazil

Introduction: Information about the distribution of malocclusion in the population and identification of
factors and conditions associated with it could help researchers build models to understand its occurrence
and help public-health policy makers improve interventions. The aim of this study was to assess the severity
of occlusal disorders in Brazilian adolescents, 12 and 18 years old, and to investigate associations between
occlusal disorders and demographic, socio-environmental, and clinical variables. Methods: Secondary data
from a cross-sectional study, including 13,801 dental occlusion status records from a probabilistic sample
randomly selected from public and private schools in 131 cities in the state of São Paulo, Brazil, were
analyzed according to the dental aesthetic index (DAI). The proportion of DAI scores greater than 30 were
compared between ages, sexes, white and nonwhite students, urban and rural dwellers, and private and
public school students; and they were compared with variables such as access to fluoridated tap water and
city population, and with clinical aspects such as the care index (CI) and the decayed, missing, filled teeth
(DMFT) index. Data analysis included frequency distribution calculation and multiple logistic regression
modeling. Results: The mean DAI score for the sample was 24.33 (SD 7.54), and 16.5% of the subjects had
DAI scores of 30 or more (severe or very severe malocclusion). The rate of DAI ⬎31 was significantly higher
among 12-year-olds, nonwhites, public-school students, those from smaller municipalities, those without
fluoridated tap water, and those with a CI ⬍51%, a DMFT score ⬎4 at age 12 years, or a DMFT score ⬎6
at age 18 years. At age 18, fewer subjects had DAI scores ⬎30; the components responsible for this
reduction were spacing in at least 1 incisal segment, midline diastema ⱖ1, and anterior maxillary overjet ⱖ4.
Conclusion: Some socio-environmental factors are associated with severity of malocclusion in adolescents.
(Am J Orthod Dentofacial Orthop 2006;129:809-16)

alocclusion is a variation of normal growth lence and severity of occlusal disorders have increased
and development that affects muscles and in the last 200 years, especially with respect to dental
facial bones during childhood and adoles- crowding.9-11
cence. Maloccluded teeth can cause psychosocial prob- Knowledge concerning the distribution of maloc-
lems related to impaired dentofacial esthetics1,2; distur- clusion in the population and the identification of
bances of oral function, such as mastication, swallowing, factors and conditions associated with it might help
and speech3; and greater susceptibility to trauma4-6 and researchers build models for understanding its occur-
periodontal disease.7 Malocclusion is not a single entity rence and help public-health policy makers improve
but, rather, a collection of situations, each in itself a interventions.
problem, any of which can be complicated by various The aim of this study was to investigate factors
genetic and environmental causes.8 associated with severity of malocclusion by using the
Although much of the available epidemiologic dental aesthetic index (DAI) in Brazilian adolescents,
information comes from cross-sectional studies and 12 and 18 years old.
thus includes the potential biases inherent to this type
of study, it is generally acknowledged that the preva- MATERIAL AND METHODS

Between August and December 1998, official agen-
Professor, Public Health Posgraduate Program, Catholic University of Santos,
Santos, Brazil. cies of the Brazilian Health Authority completed an
Professor, School of Public Health, University of São Paulo, São Paulo, epidemiological survey12 of oral health in the state of
Brazil. São Paulo, Brazil, according to international stan-
Reprint requests to: Paulo Frazão, Universidade Católica de Santos, Programa
de Pós-Graduação (Mestrado) em Saúde Pública, Rua Carvalho de Mendonça, dards established by the World Health Organization
144-4° andar, 11070-906 Santos, SP, Brazil; e-mail, (WHO).13 Twenty-four centers participated, coordi-
Submitted, July 2004; revised and accepted, October 2004. nated by the School of Public Health at São Paulo
Copyright © 2006 by the American Association of Orthodontists. University. Guidelines to assure standardized and uni-
doi:10.1016/j.ajodo.2004.10.016 form procedures were prepared. The participating cen-
810 Frazão and Narvai American Journal of Orthodontics and Dentofacial Orthopedics
June 2006

ters conducted the study and were responsible and the environmental variables of population of the city
accountable for its integrity. of residence (ⱕ50,000 or ⬎50,000 inhabitants) and
The data for this multicenter study were collected availability of fluoridated water in the urban nucleus
from 131 cities. In each city, at least 20 schools (private for at least 5 years.21 A high caries incidence in areas
and public, urban and rural) were randomly selected. In where the disease is declining18 and a high rate of
cities with 20 or fewer units, all schools were selected. untreated caries might indicate limited access to dental
Participating centers identified and calibrated examin- care and low socioeconomic level. Thus, intraoral
ers according to WHO recommendations. In all, 87,918 conditions such as dental caries at age 12 (DMFT ⬍5
students, adults, and senior citizens were evaluated for and DMFT ⬎4) and age 18 (DMFT ⬍7 and DMFT
dental caries, treatment need, prosthesis, periodontal ⬎6) and dental care index,22 expressed as the ratio
disease, fluorosis, and occlusal disorders. The multi- between the number of filled teeth and the DMFT
stage sampling was designed to represent the whole score, were also evaluated.
state and each participating city, and to estimate the Frequency distributions, chi-square association tests,
decayed, missing, and filled teeth (DMFT) index of and odds ratios were calculated. To evaluate the influ-
schoolchildren 5 to 12 years of age. ence of each independent variable on the severity of
As the sponsoring institutions made the data avail- malocclusion, nonconditional multiple logistic analysis
able for public consultation, we evaluated 13,801 oral- was carried out with software (version 11.0, SPSS,
examination records of 12- and 18-year-olds. Results of Chicago, Ill). The dichotomous dependent variable
analyses of dental caries and dental injuries in pre- (DAI ⬍31 and DAI ⬎30) and the independent vari-
school children and schoolchildren, and early tooth loss ables were inserted into the logistic model in ascending
in adults, were published in previous articles.6,14-18 order, based on their statistical significance by using the
To evaluate occlusion in 12- and 18-year-olds, we
stepwise forward procedure. The significance level for
used registration criteria from the DAI, described in the
the inclusion of each variable in the model was mea-
WHO manual.13 According to the authors, the DAI was
sured by the likelihood ratio test. The criterion for
especially designed to measure dental esthetics not
variable inclusion was a P value less than .20, and the
related to other dimensions of poor occlusion and not
final model included all variables with P values up to
based on the subjective perceptions of orthodontists,
.05 after adjustment for all other variables. Values are
patients, or parents. It might therefore aid epidemiolo-
shown in the tables in the second “OR” column. The
gists in evaluating the severity of malocclusion and the
Hosmer-Lemeshow test was used to assess the model’s
need for orthodontic treatment, based on the degree of
deviation from the social norms of dental appearance.19 goodness-of-fit.23 Interactions among variables were
This tool was developed based on 200 images represent- tested to obtain a valid measure of exposure-event
ing dental esthetics, evaluated according to a scale of relationship that accounts for confounding and effect
social acceptability by groups of adolescents and adults. It modification.24 The values of the variances of the
is acknowledged that the need for orthodontic treatment parameter estimates were checked for multicollinearity.
includes at least 3 fundamental elements: objective signs, Significant interactions are given in the third “OR”
subjective symptoms, and social norms.20 column and indicated by dental caries and care index,
The original database comprised 14,522 records. Eth*Sch (ethnic group and type of school) and
From this database, a specific file was generated con- Inh*FTW (inhabitants and fluoridated tap water). The
taining the variables of interest. Data consistency was models for ages 12 and 18 years were compared,
verified, and 721 (4.9%) records with errors were considering the potential effects of distal variables on
discarded. According to the DAI, occlusal status is event regulation, especially the dental caries index and
measured by a whole-number scale from 13 to 230, by the auto-correction phenomenon, described in certain
which a subject’s occlusion can be classified into 1 of stages in the development of dental occlusion.3
4 categories and score ranges: no abnormality or minor
(13-25), definite (26-30), severe (31-35), and very severe
or handicapping malocclusion (36-230). In this study,
subjects with scores higher than 30 were compared with Data were checked, and 13,801 records were ana-
the remaining subjects. This dependent variable was lyzed. Severe or very severe malocclusions were observed
analyzed in relation to the demographic variables of age in 2273 subjects, a 16.5% prevalence. The mean score for
(12 or 18 years) and sex (male or female), the socio- the sample was 24.3 (SD 7.54). Table I shows the
economic variables of type of school (public or private; distribution of malocclusion categories and the DAI vari-
urban or rural) and ethnicity (white or nonwhite), and ables at ages 12 and 18 years. Normal or mild malocclu-
American Journal of Orthodontics and Dentofacial Orthopedics Frazão and Narvai 811
Volume 129, Number 6

Table I. Distribution of malocclusion in 12- and 18-year-old adolescents in state of São Paulo, Brazil, 1998
Age 12 Age 18

n % n % P* value

Malocclusion (DAI score)

No abnormality or minor (ⱕ25) 5507 62.3 3462 69.7
Definite (26-30) 1736 19.6 823 16.6
Severe (31-35) 866 9.8 357 7.2
Very severe or handicappping 728 8.2 322 6.5 .000
95% CI means 24.63 ⌯ 24.95 23.32 ⌯ 23.72
DAI variables (criteria)
Missing visible teeth (ⱖ1) 196 2.2 286 5.8 .000
Crowding (1-2) 3203 36.2 1971 39.7 .000
Spacing (1-2) 2179 24.7 893 18.0 .000
Diastema (ⱖ1) 1530 17.3 591 11.9 .000
Maxillary irregularity (⬎1) 1701 19.2 924 18.6 .362
Mandibular irregularity (⬎1) 1001 11.3 715 14.4 .000
Anterior maxillary overjet (ⱖ4) 2555 28.9 1049 21.1 .000
Anterior mandibular overjet (ⱖ0) 178 2.0 110 2.2 .426
Vertical anterior openbite (ⱖ0) 811 9.2 427 8.6 .256
Molar relation (ⱖ1/2 cuspid) 4430 50.1 2518 50.7 .502

*Chi-square significance.

sion was more frequent among the older subjects. The cantly associated with less favorable dental appearance
components responsible for this reduction were spacing in (Table III).
at least 1 incisal segment, interincisal diastema, and In the multivariate model, the factors age, type of
maxillary overjet. This reduction in severity occurred school, and care index remained significantly associ-
in spite of certain components whose frequency ated with the outcome (Table II). Among the interac-
increased (visible tooth loss, crowding, and mandib- tions explored in the multivariate model, DAI ⬎30 was
ular irregularity). significantly associated with the environmental vari-
As seen in Table II, significant differences were ables number of inhabitants and fluoridated water for 5
observed in severity of malocclusion, with greater rates or more years. The odds for unfavorable dental appear-
among 12-year-olds (P ⫽ .000), nonwhites (P ⫽ .001), ance were 1.51 times higher in smaller cities without
public-school students (P ⫽ .000), those living in cities fluoridated water (P ⫽ .001).
with up to 50,000 inhabitants (P ⫽ .003), those without At age 12, in addition to type of school, the
fluoridated water (P ⫽ .036), and those with less than interaction of clinical variables DC*CI remained sig-
51% of decayed teeth restored (P ⫽ .000). nificantly associated with the outcome after adjustment
Data separation by age showed similar univariate for the remaining variables in the model (Table III).
analysis results for all clinical variables in both groups At age 18, the effects of the variables sex and type of
(Tables III and IV). school remained statistically significant (Table IV). Prox-
As to the remaining factors, more associations were ies of socioeconomic condition (white/nonwhite; private/
observed in the 18-year-old group than in the 12-year- public school), environmental variables (⬎50,000 or
old group. As seen in Table IV, in the older group, ⱕ50,000 inhabitants; fluoridated tap water), and clinical
higher rates of severe and very severe occlusal condi- aspects (DMFT ⬎6 or ⬍7; care index), coupled in
tions were observed among girls (P ⫽ .021). A com- interaction terms Eth*Sch, Inh*FTW, and DC*CI, were
parison of the rates of DAI ⬎30 among nonwhite significantly associated with DAI ⬎30.
public-school students and the remaining subjects in the
sample (white subjects from public and private schools DISCUSSION
and nonwhite subjects from private schools) showed a Epidemiologic measures are affected by both ran-
higher risk for the former group (OR ⫽ 1.32, 95% CI dom and systematic sources of error. In general, re-
[1.11-1.56], P ⫽ .001). search based on large samples has relatively high
At age 12 years, factors related to ethnicity, number precision; this can protect the study from random error.
of inhabitants, and fluoridated water were not signifi- Concerns in these cases can include sample selection
812 Frazão and Narvai American Journal of Orthodontics and Dentofacial Orthopedics
June 2006

Table II. Severe malocclusion (DAI ⬎30) in adolescents from state of São Paulo, Brazil, 1998 — odds ratio (OR) and
confidence intervals (CI) (95%) according to demographic, environmental, and clinical variables

n ⬍31 ⬎30 OR [CI 95%] P* ORa [CI 95%] P* ORb [CI 95%] P*

18 4964 86.3 13.7
12 8837 82.0 18.0 1.39 [1.26-1.53] .000 1.31 [1.19-1.45] .000 1.28 [1.16-1.42] .000
Male 6549 83.6 16.4
† †
Female 7252 83.5 16.5 1.01 [0.92-1.10] .905
White 6818 84.6 15.4

Nonwhite 6983 82.5 17.5 1.17 [1.07-1.28] .001 1.10 [1.01-1.21] .035
Private 845 90.5 9.5
Public 12956 83.1 16.9 1.95 [1.54-2.46] .000 1.85 [1.46-2.34] .000 1.73 [1.36-2.20] .000
Urban 13117 83.6 16.4
† †
Rural 684 81.9 18.1 1.13 [0.93-1.38] .230
⬎50,000 4113 85.0 15.0
ⱕ50,000 9688 82.9 17.1 1.17 [1.05-1.29] .003 † †

Fluoridated water
Yes 8599 84.0 16.0
† †
No 5202 82.7 17.3 1.10 [1.01-1.21] .036
Care index
⬎50% 7954 85.0 15.0
⬍51% 5847 81.5 18.5 1.28 [1.17-1.40] .000 1.20 [1.10-1.32] .000 1.20 [1.10-1.32] .000
else 7083 84.8 15.2
Nonwhite ⫹ public 6718 82.2 17.8 1.21 [1.10-1.32] .000 1.11 [1.01-1.21] .037
All others 9362 84.4 15.6
Inh ⱕ50,000 ⫹ no FTW 4439 81.7 18.3 1.21 [1.10-1.33] .000 1.51 [1.19-1.92] .001

Etn, Ethnicity; Sch, school; Inh, inhabitants; FTW, fluoridated tap water.
*Pearson chi-square test of independence; aadjusted for all other variables; badjusted for all other variables with interactions terms.

Not included in multiple logistic regression analysis after losing statistical significance.

process and measuring-instrument usage. This analysis areas in the Klang District25 and Australian secondary
was restricted to severe malocclusion and was based on schoolchildren,26 with mean DAI scores about 24 and
a large and randomly selected sample from a multi- DAI ⬎30 in 16% to 18% of the population. Mean
center study that included 24 participating centers in scores were slightly lower than those observed among
131 cities with hundreds of sample units (schools). All American high-school students,27 Chinese subjects in
records were checked to ensure data consistency. Taiwan aged 18 to 24,8 and Malaysian schoolchildren
Therefore, the estimates produced were considered to from the entire country.28 Native Americans living on
have acceptable levels of reliability. reservations,27 Japanese youths,29 and 13-year-old New
Schoolchildren with orthodontics bands or appli- Zealand schoolchildren30 showed significantly higher
ances were not examined, but history of orthodontic values in studies with the DAI. The lowest mean DAI
treatment was not registered during data collection. The score (95% CI 21.9-22.8) was found among Nigerian
supply of orthodontic treatment in the study area is adolescents, with only 9.2% with DAI ⬎30.31
limited, and it is mainly directed toward the few In our study, occlusion records were evaluated at ages
adolescents who attend private schools. Thus, the 12 and 18 years, corresponding to the development stage
influence of this aspect on the estimates for private- characterized by the initial maturation of the permanent
school subjects was restricted. dentition, which spans the period between the eruption of
The results of this survey were similar to those the second and third molars. A significant reduction in
among Malaysian schoolchildren from urban and rural unfavorable dental appearance was observed. Estioko et
American Journal of Orthodontics and Dentofacial Orthopedics Frazão and Narvai 813
Volume 129, Number 6

Table III. Severe malocclusion (DAI ⬎30) in 12-year-old adolescents from state of São Paulo, Brazil, 1998 — odds
ratio (OR) and confidence intervals (CI) (95%) according to demographic, environmental, and clinical variables

n ⬍31 ⬎30 OR [CI 95%] P* ORa [CI 95%] P* ORb [CI 95%] P*

Male 4203 81.4 18.6
† †
Female 4634 82.5 17.5 0.93 [0.83-1.03] .168
White 4200 82.5 17.5
† †
Nonwhite 4637 81.5 18.5 1.07 [0.96-1.19] .211
Private 447 87.9 12.1
Public 8390 81.6 18.4 1.64 [1.23-2.19] .001 1.57 [1.18-2.10] .002 1.56 [1.17-2.09] .003
Urban 8295 82.0 18.0
† †
Rural 542 81.5 18.5 1.03 [0.82-1.29] .797
⬎50,000 2142 81.7 18.3
ⱕ50,000 6695 82.1 17.9 0.97 [0.86-1.10] .669 † †

Fluoridated tap water

Yes 5301 81.8 18.2
† †
No 3536 82.2 17.8 0.98 [0.88-1.09] .700
Care index
⬎50% 4419 83.1 16.9
⬍51% 4418 80.8 19.2 1.17 [1.05-1.30] .006 1.21 [1.08-1.35] .001 †

Dental caries
DMFT ⬍5 6028 83.1 16.9
DMFT ⬎4 2809 79.6 20.4 1.26 [1.12-1.41] .000 1.27 [1.13-1.42] .000 †

Dental carries*Care index

All others 7706 82.8 17.2
DMFT ⬎4 ⫹ CI ⬍51% 1131 76.5 23.5 1.43 [1.22-1.69] .000 1.45 [1.25-1.69] .000
a b
*Pearson chi-square test of independence; adjusted for all other variables; adjusted for all other variables with interactions terms.

Not included in multiple logistic regression analysis after losing statistical significance.

al26 also observed lower mean DAI scores among 14- to Assuming that the timing of data collection could be
16-year-olds than among 11- to 13-year-olds. historical, one can consider both event and exposure as
The morphologic component directly related to this having been produced during earlier developmental
reduction, in both studies, was maxillary overjet, a stages.
characteristic that was also found among adolescents by Among demographical variables, age—not sex—
other authors.32,33 Spatial conditions in at least 1 incisal was strongly associated with the outcome in both
segment play an important role. In our study, a reduc- bivariate and multivariate analysis (OR ⫽ 1.28, 95% CI
tion in spacing was observed, whereas Estioko et al26 [1.16-1.42]; P ⫽ .000). This strengthens the notion that
observed a reduction in crowding. There is no agree- adolescence is a period of maturation of dental occlu-
ment concerning other components, and further studies sion, regardless of other factors.
must be carried out to evaluate the longitudinal evolu- At age 18 years, the proportion of more severe
tion of these characteristics during this stage of occlu- malocclusion was 3.5% higher among nonwhite sub-
sion development. jects. This effect remained significant in bivariate
The influence of certain demographic, environmen- and multivariate analyses. This proportion was also
tal, clinical, and socioeconomic variables in the propor- 2% higher among girls and remained statistically
tion of more severe cases was analyzed by multiple significant in multivariate analysis with the interac-
logistic regression. Three models were developed to tion terms. In spite of very slight differences, both
examine these factors. Although the data were origi- findings are provocative. The relationship between
nally from a cross-sectional study, we assumed that the certain physical traits and facial patterns, especially
independent variables analyzed represented character- types associated with the anteroposterior maxillo-
istics that correspond to the subjects’ past and present. mandibular relationship, is widely acknowledged in
814 Frazão and Narvai American Journal of Orthodontics and Dentofacial Orthopedics
June 2006

Table IV. Severe malocclusion (DAI ⬎30) in 18-year-old adolescents from state of São Paulo, Brazil, 1998 — odds
ratio (OR) and confidence intervals (CI) (95%) according to demographic, environmental, and clinical variables

n ⬍31 ⬎30 OR [CI 95%] P* ORa [CI 95%] P* ORb [CI 95%] P*

Male 2346 87.5 12.5
Female 2618 85.3 14.7 1.21 [1.03-1.43] .021 1.22 [1.04-1.44] .020 1.23 [1.04-1.45] .013
White 2618 88.0 12.0

Nonwhite 2346 84.5 15.5 1.34 [1.14-1.58] .000 1.35 [1.14-1.60] .000
Private 398 93.5 6.5
Public 4566 85.7 14.3 2.39 [1.59-3.58] .000 1.95 [1.29-2.94] .001 1.75 [1.15-2.66] .009
Urban 4822 86.4 13.6
† †
Rural 142 83.1 16.9 1.29 [0.83-2.02] .257
⬎50,000 1971 88.6 11.4
ⱕ50,000 2993 84.8 15.2 1.39 [1.17-1.65] .000 † †

Fluoridated tap water

Yes 3298 87.6 12.4

No 1666 83.8 16.2 1.37 [1.16-1.61] .000 1.34 [1.13-1.59] .001
Care index
⬎50% 3535 87.4 12.6
⬍51% 1429 83.8 16.2 1.34 [1.13-1.59] .001 † †

Dental caries
DMFT ⬍7 1928 88.3 11.7
DMFT ⬎6 3036 85.1 14.9 1.32 [1.11-1.57] .001 1.38 [1.15-1.66] .001 †

else 2755 88.2 11.8
Nonwhite ⫹ public 2209 83.9 16.1 1.32 [1.11-1.56] .001 1.33 [1.13-1.58] .001
else 3686 87.9 12.1
Inh ⱕ50,000 ⫹ no FTW 1278 81.7 18.3 1.52 [1.25-1.86] .000 1.55 [1.30-1.85] .000
Dental carries*Care index
All others 4404 87.4 12.6
DMFT ⬎6 ⫹ CI ⬍51% 560 77.9 22.1 2.00 [1.51-2.66] .000 1.77 [1.42-2.21] .000

Etn, Ethnicity; Sch, school; Inh, inhabitants; FTW, fluoridated tap water.
*Pearson chi-square test of independence; aadjusted for all other variables; badjusted for all other variables with interactions terms.

Not included in multiple logistic regression analysis after losing statistical significance.

the literature.34-37 Prior studies of the distribution of normative orthodontic treatment was greater in de-
malocclusion as a whole in populations located in prived socioeconomic groups. At age 18 years,
similar Brazilian areas did not show consistent dif- nonwhite public-school students were 1.33 times
ferences between ethnic groups.38 As to sex, con- more likely to have unfavorable dental appearance (P
cerning the need for orthodontic treatment, most ⫽ .001), when compared with nonwhite private-
studies from various parts of the world did not show school students and white students from both types
significant differences. of schools (synergic effect adjusted for the remaining
Of the variables studied, type of school was variables).
strongly associated with the outcome, independent of The relationship between outcome and dental care
age, and after adjustment for the remaining aspects was tested in the 3 models, showing a strong association
(OR ⫽ 1.73, 95% CI [1.36-2.20]; P ⫽ .000). In the between severity of malocclusion and more than 50%
studied area, private schools are attended by adoles- untreated decayed teeth. This supports the hypothesis that
cents from the higher levels of the social pyramid, greater access to, and regular use of, dental care might
and this factor indicates family socioeconomic stand- hinder the aggravation of occlusal conditions. Greater
ing. Tickle et al39 also noted that the need for caries incidence combined with more untreated teeth
American Journal of Orthodontics and Dentofacial Orthopedics Frazão and Narvai 815
Volume 129, Number 6

increased the risk of unfavorable dental appearance in event, the investigation of the influent aspects of severe
both age groups (OR12y ⫽ 1.45 95% CI [1.25-1.69], P ⫽ malocclusion generates knowledge of the role of distal
.000; OR18y ⫽ 1.77, 95% CI [1.42-2.21], P ⫽ .000). factors on event regulation.
Several studies have identified lower prevalences of
malocclusion in areas with fluoridated tap water than CONCLUSIONS
those without treated water.40-42 This shows the indirect The most important finding in this study was the
effect of fluoridation on the occurrence of malocclu- identification of socio-environmental factors associated
sion, the reduction of which is attributed to the prophy- with severe malocclusion. Adolescents living in large
lactic effects of fluoride on dental caries. It is widely cities, with fluoridated tap water, and of higher socio-
described in the literature, and fluoridated water pro- economic status (private school students, with low
motes dramatic reductions in caries prevalence and caries incidence, and with access to dental care— care
severity as well as in the premature loss of deciduous index ⬎50%) were at lower risk. These socio-environ-
and permanent teeth. mental aspects might be considered distal factors asso-
In this study, we observed the influence of tap water ciated with unfavorable dental appearance that can
fluoridation and caries attack on the distribution of the impair an adolescent’s self-esteem and interpersonal
outcome in the population. The more significant results relationships, reducing his or her range of social aspi-
were observed among 18-year-olds, in which DAI ⬎30 rations and opportunities, with implications that are
showed a negative association with fluoridation (OR ⫽ relevant to dental policy making.
1.34, 95% CI [1.13-1.59], P ⫽ .001) and a positive
association with high caries rate (OR ⫽ 1.38, 95% CI
[1.15-1.66], P ⫽ .001). REFERENCES
Where dental caries is declining, correlations between 1. Shaw WC, Addy M, Ray C. Dental and social effects of
greater severity and worse socioeconomic conditions have malocclusion and effectiveness of orthodontic treatment: a re-
been described.43-48 Both factors were significantly asso- view. Community Dent Oral Epidemiol 1980;8:36-45.
ciated with malocclusion severity in our study. 2. Kenealy P, Frude N, Shaw W. An evaluation of the psycholog-
ical and social effects of malocclusion: some implications for
In the 2 age groups together and in the 18-year-old dental policy making. Soc Sci Med 1989;28:583-91.
group alone, the influence of the interaction of environ- 3. Proffit WR, Fields HW. Contemporary orthodontics. St Louis:
mental variables on malocclusion severity was ob- C. V. Mosby; 1995.
served. This synergic effect could be related to the fact 4. Petti S, Tarsitani G. Traumatic injuries to anterior teeth in Italian
that community size might be an important barrier to schoolchildren: prevalence and risk factors. Endod Dent Trau-
matol 1996;12:94-7.
water fluoridation.49 Peres et al50 observed that, in 5. Marcenes W, Alessi ON, Traebert J. Causes and prevalence of
southern Brazil, cities without water fluoridation were traumatic injuries to the permanent incisors of school children aged
smaller than those with this service. 12 years in Jaragua do Sul, Brazil. Int Dent J 2000;50:87-92.
Many genetic and environmental causes determine the 6. Grimm S, Frazão P, Antunes JLF, Castellanos RA, Narvai PC.
distribution of malocclusion in populations. This distribu- Dental injury among Brazilian schoolchildren in the state of São
Paulo. Dent Traumatol 2004;20:134-8.
tion can assume different configurations, depending on the 7. Greiger AM. Malocclusion as an etiologic factor in periodontal
occlusal characteristics captured by the different measur- disease: a retrospective essay. Am J Orthod Dentofacial Orthop
ing instruments. The study of factors associated with 2001;120:112-5.
dental occlusion must consider, separately, the prevalence 8. Katoh Y, Ansai T, Takehara T, Yamashita Y, Miyazaki H, Jenny
of the effect as a whole, and that of specific occlusal J, et al. A comparison of DAI scores and characteristics of
occlusal traits in three ethnic groups of Asian origin. Int Dent J
deviations. Studies in the literature describe the strong 1998;48:405-11.
influence of environment on the occurrence of anterior 9. Corruccini RS. An epidemiologic transition in dental occlusion
open bite51 and of genetic factors on the characteristics of in world populations. Am J Orthod 1984;86:419-26.
certain craniofacial dimensions such as mandibular prog- 10. Weiland FJ, Jonke E, Bantleon HP. Secular trends in malocclu-
nathism.52 In constructing explanatory models, it is nec- sion in Austrian men. Eur J Orthod 1997;19:355-9.
11. Brin I, Zwilling-Sellam O, Harari D, Koyoumdjisky-Kaye E,
essary to distinguish between malocclusion as a whole Ben-Bassat Y. Does a secular trend exist in the distribution of
and specific occlusal deviations. To better understand the occlusal patterns? Angle Orthod 1998;68:81-4.
causes of occlusal disorders in a populational context, 12. Faculdade de Saúde Pública, Universidade de São Paulo. Oral
researchers must extend focal length and look not only at health survey: state of São Paulo, Brazil, 1998. São Paulo:
specific proximal risk factors, but also at upstream condi- Núcleo de Estudos e Pesquisas de Sistemas de Saúde; 1999.
13. World Health Organization. Oral health surveys: basic methods.
tions associated with the phenomenon.53 Whereas the 4th ed. Geneva: WHO; 1997.
study of specific occlusal deviations allows for the iden- 14. Antunes JLF, Junqueira SR, Frazão P, Bispo CM, Pegoretti T,
tification of proximal and distal factors associated with the Narvai PC. City-level gender differentials in the prevalence of
816 Frazão and Narvai American Journal of Orthodontics and Dentofacial Orthopedics
June 2006

dental caries and restorative dental treatment. Health Place nization standart methods (1962 and 1971) in South Pacific
2003;9:231-9. populations. Am J Orthod 1974;66:251-72.
15. Antunes JLF, Pegoretti T, Andrade FP, Junqueira SR, Frazão P, 35. Brunelle JA, Bhat M, Lipton JA. Prevalence and distribution of
Narvai PC. Ethnic disparities in the prevalence of dental caries selected occlusal characteristics in the US population, 1988-
and restorative dental treatment in Brazilian children. Int Dent J 1991. J Dent Res 1996;75:706-13.
2003;53:7-12. 36. Wheeler TT, McGorray SP, Yurkiewicz L, Keeling SD, King GJ.
16. Frazão P, Antunes JLF, Narvai PC. Perda dentária precoce em Orthodontic treatment demand and need in third and fourth grade
adultos de 35 a 44 anos de idade. Estado de São Paulo, Brasil, schoolchildren. Am J Orthod Dentofacial Orthop 1994;106:22-33.
1998. Rev Bras Epidemiol 2003;6:49-57. 37. Proffit WR, Fields HW, Moray LJ. Prevalence of malocclusion
17. Peres MA, Peres KG, Antunes JLF, Junqueira SR, Frazão P, and orthodontic treatment need in the United States: estimates
Narvai PC. The association between socioeconomic develop- from the NHANES III survey. Int J Adult Orthod Orthognath
ment at the town level and the distribution of dental caries in Surg 1998;13:97-106.
Brazilian children. Panam J Public Health 2003;14:149-57. 38. Frazão P, Narvai PC, Latorre MRDO, Castellanos RA. Prevalên-
18. Antunes JLF, Narvai PC, Nugent ZL. Measuring inequalities in cia de má oclusão na dentição decídua e permanente de crianças
the distribution of dental caries. Community Dent Oral Epide- na cidade de São Paulo, Brasil, 1996. Cad Saúde Pública
miol 2004;32:41-8. 2002;18:1197-205.
19. Cons NC, Jenny J, Kohout FJ, Songpaisan Y, Jotikastira D. 39. Tickle M, Kay EJ, Bearn D. Socio-economic status and orth-
Utility of the dental aesthetic index in industrialized and devel- odontic treatment need. Community Dent Oral Epidemiol 1999;
oping countries. J Public Health Dent 1989;49:163-6. 27:413-8.
20. Prahl-Anderson B. The need for orthodontic treatment. Angle 40. Ast DB, Allaway N, Draker HL. Prevalence of malocclusion,
Orthod 1978;48:1-8. related to dental caries and lost first permanent molars in a
21. Riley JC, Lennon MA, Ellwood RP. The effect of water fluoridated city and a fluoride-deficient city. Am J Orthod
fluoridation and social inequalities on dental caries in 5-year-old 1962;48:106-13.
children. Int J Epidemiol 1999;28:300-5. 41. Erickson DM, Graziano FW. Prevalence of malocclusion in
seventh grade children in two North Carolina cities. J Am Dent
22. Walsh J. International patterns of oral health care—the example
Assoc 1966;73:124-7.
of New Zealand. N Z Dent J 1970;66:143-52.
42. Murray JJ. Prevalence of malocclusion in fifteen-year-old chil-
23. Hosmer D, Lemeshow S. Goodness-of-fit tests for the multiple
dren from fluoride and non-fluoride communities. Dent Prac
logistic regression model. Communications Stat 1980;A9:1043-69.
Dent Rec 1969;19:395-400.
24. Kleinbaum DG. Logistic regression: a self-learning text. New
43. Treasure ET, Dever VG. Relationship of caries with socioeco-
York: Springer-Verlag; 1994.
nomic status in 14-year-old children from communities with
25. Esa R, Razak IA, Allister JH. Epidemiology of malocclusion and
different fluorides stories. Community Dent Oral Epidemiol
orthodontic treatment need of 12-13-year-old Malaysian school-
children. Community Dent Health 2001;18:31-6.
44. Schou L, Uitenbroe KD. Social and behavioral indicators of
26. Estioko LJ, Wright FA, Morgan MV. Orthodontic treatment need
caries experience in 5-year-old children. Community Dent Oral
of secondary schoolchildren in Heidelberg, Victoria: an epide-
Epidemiol 1995;23:276-81.
miologic study using the dental aesthetic index. Community
45. Irigoyen ME, Maupome G, Mejia AM. Caries experience and
Dent Health 1994;11:147-51. treatment needs in a 6- to 12-year-old urban population in
27. Jenny J, Cons NC, Kohout FJ, Jakobsen J. Differences in need relation to socio-economic status. Community Dent Health
for orthodontic treatment between Native Americans and the 1999;16:245-9.
general population based on DAI scores. J Public Health Dent 46. Gillcrist JA, Brumley DE, Blackford JU. Community socioeco-
1991;51:234-8. nomic status and children’s dental health. J Am Dent Assoc
28. Abdullah MS, Rock WP. Assessment of orthodontic treatment 2001;132:216-22.
need in 5,112 Malaysian children using the IOTN and DAI 47. Patussi MP, Marcenes W, Croucher R, Sheiham A. Social
indices. Community Dent Health 2001;18:242-8. deprivation, income inequality, social cohesion and dental caries
29. Ansai T, Miyazaki H, Katoh Y, Yamashita Y, Takehara T, Jenny in Brazilian school children. Soc Sci Med 2001;53:915-25.
J, et al. Prevalence of malocclusion in high school students in 48. Antunes JLF, Frazão P, Narvai PC, Bispo CM, Pegoretti T. Spatial
Japan according to the dental aesthetic index. Community Dent analysis to identify differentials in dental needs by area-based
Oral Epidemiol 1993;21:303-5. measures. Community Dent Oral Epidemiol 2002;30:133-42.
30. Chi J, Johnson M, Harkness M. Age changes in orthodontic 49. Centers for Disease Control and Prevention. Recommendation
treatment need: a longitudinal study of 10 and 13-year-old children, focusing fluoride to prevent and control dental caries in the United
using the dental aesthetic index. Aust Orthod J 2000;16:150-6. States. Morbidity and Mortality Weekly Report 2001;50:1-42.
31. Otuyemi OD, Ogunyinka A, Dosumu O, Cons NC, Jenny J. 50. Peres MA, Fernandes LS, Peres KG. Inequality of water fluori-
Malocclusion and orthodontic treatment need of secondary dation in southern Brazil: the inverse equity hypothesis revisited.
school students in Nigeria according to the dental aesthetic index Soc Sci Med 2004;58:1181-9.
(DAI). Int Dent J 1999;49:203-10. 51. Proffit WR. On the aetiology of malocclusion. Br J Orthod
32. Bjork A. Variability and age changes in overjet and overbite. 1986;13:1-11.
Am J Orthod 1953;39:779-801. 52. Litton SF, Ackerman LV, Isaacson RJ, Shapiro B. A genetic
33. Tarvit DJ, Freer TJ. Assessing malocclusion: the time factor. Br J study of Class III malocclusion. Am J Orthod 1970;58:565-77.
Orthod 1998;25:31-4. 53. McMichael AJ. Prisoners of the proximate: loosening the con-
34. Baume LJ. Uniform methods for the epidemiologic assessment straints on epidemiology in an age of change. Am J Epidemiol
of malocclusion. Results obtained with the World Health Orga- 1999;149:887-97.