Anda di halaman 1dari 9

The PHQ-15: Validity of a New Measure for Evaluating the Severity of Somatic

Symptoms
KURT KROENKE, MD, ROBERT L. SPITZER, MD,

AND

JANET B. W. WILLIAMS, DSW

Objective: Somatization is prevalent in primary care and is associated with substantial functional impairment and
healthcare utilization. However, instruments for identifying and monitoring somatic symptoms are few in number
and not widely used. Therefore, we examined the validity of a brief measure of the severity of somatic symptoms.
Methods: The Patient Health Questionnaire (PHQ) is a self-administered version of the PRIME-MD diagnostic
instrument for common mental disorders. The PHQ-15 comprises 15 somatic symptoms from the PHQ, each
symptom scored from 0 (not bothered at all) to 2 (bothered a lot). The PHQ-15 was administered to 6000 patients
in eight general internal medicine and family practice clinics and seven obstetrics-gynecology clinics. Outcomes
included functional status as assessed by the 20-item Short-Form General Health Survey (SF-20), self-reported sick
days and clinic visits, and symptom-related difficulty. Results: As PHQ-15 somatic symptom severity increased,
there was a substantial stepwise decrement in functional status on all six SF-20 subscales. Also, symptom-related
difficulty, sick days, and healthcare utilization increased. PHQ-15 scores of 5, 10, 15, represented cutoff points for
low, medium, and high somatic symptom severity, respectively. Somatic and depressive symptom severity had
differential effects on outcomes. Results were similar in the primary care and obstetrics-gynecology samples.
Conclusions: The PHQ-15 is a brief, self-administered questionnaire that may be useful in screening for somatization and in monitoring somatic symptom severity in clinical practice and research. Key words: somatization,
somatization disorder, depression, screening, quality of life, utilization.

DSM-IV ! Diagnostic and Statistical Manual of Mental Disorders, fourth edition; PHQ-9 ! Patient Health
Questionnaire depressive symptom severity scale;
PHQ-15 ! Patient Health Questionnaire somatic
symptom severity scale; SF-20 ! 20-item Short-Form
General Health Survey.

INTRODUCTION
Somatic symptoms account for more than half of all
outpatient encounters (1), and at least one third of
these somatic symptoms are medically unexplained
(2 6). Somatization is the association of medically
unexplained somatic symptoms with psychological
distress and health-seeking behavior and is present in
at least 10% to 15% of primary care patients (711).
Indeed, somatization together with depression and
anxiety constitute the three most common psychiatric
problems seen in primary care (8, 12, 13). The detrimental impact of somatization on multiple domains of
health-related quality of life remains considerable
even after controlling for comorbid depression and
From the Regenstrief Institute for Health Care and Department of
Medicine (K.K.), Indiana University, Indianapolis, IN; and the New
York State Psychiatric Institute and Department of Psychiatry
(R.L.S., J.B.W.), Columbia University, New York, NY.
Address reprint requests to: Kurt Kroenke, MD, Regenstrief Institute for Health Care, RG-6 1050 Wishard Blvd., Indianapolis, IN
46202. Email: Kkroenke@regenstrief.org. For a complimentary copy
of reproducible PHQ materials, contact: Robert L. Spitzer, MD.
Email: rls8@columbia.edu
Received for publication February 27, 2001; revision received
June 11, 2001.

258
0033-3174/02/6402-0258
Copyright 2002 by the American Psychosomatic Society

anxiety (79, 11). Moreover, compared with depression and anxiety, somatization results in more healthcare utilization and greater clinician frustration (6,
14 17). Although somatization is often comorbid with
depression and/or anxiety (just as depression and anxiety frequently coexist), a third or more of patients
have somatization alone (8, 10 12, 18). The importance of recognizing and evaluating somatization has
been heightened by recent evidence of the effectiveness of specific treatment strategies (19 21).
Measures to identify and monitor somatic symptoms are important if researchers are to study somatization and clinicians are to evaluate and manage it.
Unlike depressive symptom measures, measures to assess somatic symptoms are less well established. Limitations of existing measures (11, 2225) include one
or more of the following: their length, the need to
inquire about lifetime as well as current symptoms, a
predominant focus on identifying DSM-IV somatization disorder (which accounts for only a small proportion of clinically significant somatization in primary
care), validation in psychiatric rather than general
medical patient populations, and an assessment of
symptom counts alone rather than both the severity
and number of somatic symptoms. The few studies
comparing multiple somatic symptom measures in the
same sample have not demonstrated the superiority of
any one particular measure (26, 27). Consensus is further complicated by the ongoing debate about the optimal classification of somatoform disorders (17,
28 32).
PRIME-MD (Pfizer Inc, New York, NY) is a brief
instrument for making criteria-based diagnoses of
mental disorders commonly encountered in primary
care (8, 33). The Patient Health Questionnaire (PHQ) is

Psychosomatic Medicine 64:258 266 (2002)

PHQ-15 AS A MEASURE OF SOMATIZATION


an entirely self-administered version of the PRIME-MD
that was recently validated in two studies involving
6000 patients in eight primary care clinics and seven
obstetrics-gynecology clinics, respectively (34, 35).
The PHQ assesses eight diagnoses, divided into threshold disorders (disorders that correspond to specific
DSM-IV diagnoses: major depressive disorder, panic
disorder, and bulimia nervosa) and subthreshold disorders (disorders whose criteria encompass fewer
symptoms than are required for any specific DSM-IV
diagnoses: other depressive disorder, other anxiety
disorder, probable alcohol abuse/dependence, bingeeating disorder, and probable somatoform disorder).
In this article we analyze data for the 15-item somatic symptom scale, which we call the PHQ-15, to
address two major questions. First, is the PHQ-15 a
valid measure of somatic symptom severity as determined by its association with multiple domains of
functional status as well as disability days and utilization? Second, how do somatic and depressive symptoms differ in their effects on these outcomes?
METHODS
Description of the PHQ-15
The PHQ-15 is a somatic symptom subscale (Appendix) derived
from the full PHQ. It inquires about 15 somatic symptoms or symptom clusters that account for more than 90% of the physical complaints (excluding upper respiratory tract symptoms) reported in the
outpatient setting (1, 36). Also, the symptoms inquired about in the
PHQ-15 include 14 of the 15 most prevalent DSM-IV somatization
disorder somatic symptoms (ie, those with a prevalence of 3% or
greater in the general population) (32). Thirteen of the PHQ-15
somatic symptoms are included in the PHQ somatic symptom module, in which subjects are asked to rate the severity of each symptom
as 0 (not bothered at all), 1 (bothered a little), or 2 (bothered a
lot). Two additional physical symptomsfeeling tired or having
little energy, and trouble sleepingare contained in the PHQ depression module, in which subjects are asked: Over the last 2
weeks, how often have you been bothered by any of the following
problems? For scoring, response options for these two symptoms
are coded as 0 (not at all), 1 (several days), or 2 (more than half
the days or nearly every day). Thus, in determining the PHQ-15
score, each individual symptom is coded as 0, 1, or 2, and the total
score ranges from 0 to 30. Appendix 1 displays the recommended
format for the PHQ-15 if used as a somatic symptom severity scale
separate from the full PHQ.

PHQ Study Samples and Procedures


From May 1997 to November 1998, 3890 patients, aged 18 years
or older, were invited to participate in the PHQ Primary Care Study
(34). One hundred ninety declined to participate, and 266 started
but did not complete the questionnaire (often because there was
inadequate time before seeing their physician). Data for 434 patients
were not entered into the data set because the equivalent of approximately one page (20 items) of the PHQ was incomplete. This resulted in 3000 primary care patients (1422 from five general internal
medicine clinics and 1578 from three family practice clinics).

Psychosomatic Medicine 64:258 266 (2002)

From May 1997 to March 1999, 3636 patients, aged 18 years or


older, were approached to participate in the PHQ Obstetrics-Gynecology Study (35). Two hundred forty-five patients declined to participate, and 127 started but did not complete the questionnaire.
Data (equivalent of approximately one page) were missing for 264 of
these patients, resulting in 3000 subjects from seven obstetricsgynecology sites. All sites used one of two subject selection methods
to minimize sampling bias: either consecutive patients for a given
clinic session or every nth patient until the intended quota for that
session was achieved. Patient characteristics are summarized in
Table 1. Besides being entirely women, the obstetrics-gynecology
sample had a younger average age, more Hispanic subjects, lower
average education, and less medical comorbidity. Medical comorbidity was assessed by asking physicians to indicate the presence or
absence of nine types of physical disorders: hypertension, heart
disease, diabetes, liver disease, renal disease, arthritis, pulmonary
disease, cancer, or other disorders.
Sixty-two physicians participated in the PHQ Primary Care
Study (21 general internal medicine and 41 family practice, 19 of
whom were family practice residents). Their mean age was 37 years
(SD ! 6.5), and 63% were male. A total of 40 physicians and 21
nurse practitioners participated in the PHQ Obstetrics-Gynecology
Study. Their mean age was 39 years (SD ! 8.9), and 48% were male.
Before seeing the physician, all patients completed the PHQ as
well as the Medical Outcomes Study Short-Form General Health
Survey (SF-20) (37). The SF-20 measures functional status in six
domains (all scores from 0 to 100, where 100 ! best health). Patients
also reported the number of physician visits and disability days
during the past 3 months and provided a global rating of symptomrelated difficulty by responding to the following question: How
difficult have these problems made it for you to do your work, take
care of things at home, or get along with other people? Response
categories for this global rating are not difficult at all, somewhat
difficult, very difficult, and extremely difficult.
TABLE 1.

Characteristics of Patients in the PHQ Primary Care


and Obstetrics-Gynecology Studies

Patient Characteristic
No. of subjects
Established clinic patient, %
Age (mean " SD), y
Women, %
Race, %
White
African American
Hispanic
Marital status, %
Married
Never married
Divorced, separated, widowed
Education, %
College graduate
Partial college
High school graduate only
Less than high school
Physical disorders, %
Hypertension
Arthritis
Diabetes
Pulmonary

Study 1
PHQ Primary Care

Study 2
ObstetricsGynecology

3000
80
46 " 17
66

3000
71
31 " 11
100

79
13
4

39
15
39

48
23
29

52
33
15

27
27
33
13

16
25
32
27

25
11
8
7

2
1
1
2

259

K. KROENKE et al.
Analysis
The PHQ-15 is intended to function as a continuous measure of
somatic symptom severity. For this article the PHQ-15 score was
divided into several categories to illustrate more clearly the relationship between graded increases in somatic symptom severity and
various health outcomes. The categories were minimal (PHQ-15
score ! 0 4), low (score ! 59), medium (score ! 10 14), and high
(score ! 1530) levels of somatic symptom severity. These categories were chosen for several reasons. The first was pragmatic: the
cutoff points of 5, 10, and 15 are simple for clinicians to remember
and apply. The second reason was empiric: using different cutoff
points did not noticeably change the associations between increasing PHQ-15 severity and measures of construct validity. The third
reason is that patients with the most severe symptoms (score of 15 or
higher) constituted approximately 10% of the sample, a prevalence
comparable with the lower boundary of prevalence estimates for
clinically significant somatization in primary care (7, 8, 11).
The internal reliability of the PHQ-15 was assessed using Cronbachs !. Construct validity of the PHQ-15 as a measure of somatization severity was assessed by examining functional status (the six
SF-20 scales), disability days, symptom-related difficulty, and
healthcare utilization (clinic visits) over the four PHQ-15 intervals.
The independent effects of somatic symptoms, depressive symptoms, and medical comorbidity on functional status and other outcomes were assessed using stepwise linear regression models. The
PHQ has a nine-item depressive symptom severity scale (the PHQ-9)
that ranges from 0 to 27. The PHQ-15 score, PHQ-9 score, and
number of physical disorders were entered as independent variables
in each model, adjusting for age, gender, minority status, education,
and study site.

RESULTS
Distribution and Reliability of PHQ-15 Scores
Table 2 shows the distribution of PHQ-15 scores.
Each sample was roughly divided into thirds, with
TABLE 2.

Association Between PHQ-15 Severity and


Functional Status
Table 3 summarizes the strong associations between
increasing PHQ-15 severity and worsening function on
all six SF-20 scales. Several findings should be noted.
First, the effects of increasing PHQ-15 severity are in
the same direction for all scales in both samples, although the magnitude was somewhat less in obstetrics-gynecology patients. Second, the stepwise decrements in SF-20 scores with increasing PHQ-15 scores
show a consistent pattern across all six domains.
Third, most pairwise comparisons within each SF-20
scale between successive PHQ-15 levels of severity
were highly significant (p # .001).
Figure 1 illustrates graphically the relationship between increasing PHQ-15 scores and worsening func-

Distribution of PHQ-15 Somatic Symptom Severity Scores and Patient Characteristics in Primary Care and ObstetricsGynecology Samples

Patient Characteristic
Primary care sample
No. of patientsa
Percentage of patients
Mean age, y
Women, %
White, %
Some college education, %
Mean no. of physical disorders
Obstetrics-gynecology sample
No. of patientsa
Percentage of patients
Mean age, y
Women, %
White, %
Some college education, %
Mean no. of physical disorders
a

scores of 0 to 4 representing the lower tertile of somatic


symptom severity; 5 to 9, the middle tertile; and 10 or
greater, the upper tertile. Within the upper tertile, onethird (or approximately 10% of both samples) had
scores of 15 or greater. The internal reliability of the
PHQ-15 was excellent, with a Cronbachs ! of 0.80 in
both the primary care and obstetrics-gynecology samples. The 15 individual symptoms showed moderate
associations with one other: the majority of item-item
correlations in both samples were in the 0.20-to-0.29
(45%) or the 0.10-to-0.19 range (33%). Only 6% of the
item-item correlations were less than 0.10, and 9%
exceeded 0.40, with the highest being the correlation
between trouble sleeping and fatigue (0.55).

PHQ-15 Somatic Symptom Severity Score


04

59

1014

1530

1012
35
47.6
57
81
58
0.9

1012
35
45.1
69
79
56
1.0

594
20
44.5
74
79
51
1.1

291
10
44.0
78
75
43
1.2

1021
36
31.7
100
39
44
0.1

1036
36
30.6
100
43
43
0.2

562
20
30.0
100
44
42
0.2

250
9
31.7
100
44
35
0.3

Total is less than 3000 per sample because of missing PHQ-15 items for some patients.

260

Psychosomatic Medicine 64:258 266 (2002)

PHQ-15 AS A MEASURE OF SOMATIZATION


TABLE 3.

Relationship Between PHQ-15 Somatic Symptom Severity and SF-20 Functional Status

Mean SF-20 Scorea (SD)


Level of Somatic
Symptom
Mental
Social
Role
General
Pain
Physical
Severity
(PHQ-15
Primary Obstetrics- Primary Obstetrics- Primary Obstetrics- Primary Obstetrics- Primary Obstetrics- Primary ObstetricsScore)
Care Gynecology Care Gynecology Care Gynecology Care Gynecology Care Gynecology Care Gynecology
Minimal (14)
Low (59)
Medium (1014)
High (1530)

82 (14)
72 (18)
61 (21)
51 (23)

82 (15)
74 (17)
63 (20)
56 (21)

94 (17)
86 (23)
71 (31)
53 (34)

92 (21)
88 (23)
78 (29)
67 (30)

91 (25)
78 (38)
55 (45)
36 (36)

91 (27)
84 (33)
74 (40)
50 (46)

76 (20)
61 (24)
45 (24)
29 (22)

80 (18)
67 (23)
53 (25)
39 (25)

76 (24)
60 (25)
44 (23)
35 (21)

81 (21)
66 (23)
54 (24)
40 (22)

88 (20)
77 (25)
64 (28)
50 (29)

88 (22)
82 (21)
75 (23)
63 (25)

a
SF-20 scores are adjusted for age, sex, education, and number of physical disorders. Means are presented with standard deviations in
parentheses. Most pairwise comparisons of mean SF-20 scores between adjacent PHQ-9 levels within each scale were highly significant (p #
.001).

the primary care sample; results for the obstetricsgynecology sample (not shown) were similar.
Table 4 shows the association between PHQ-15 severity levels and three other measures of construct
validity: self-reported disability days, clinic visits, and
the amount of difficulty patients globally attribute to
their symptoms. Greater levels of somatization severity
were associated with a stepwise increase in disability
days, healthcare utilization, and symptom-related difficulty in activities and relationships.
Independent Effects of Somatic and Depressive
Symptoms

Fig. 1. Relationship between somatic symptom severity as measured by the PHQ-15 and decline in functional status as
measured by the six subscales of the SF-20. The decrement
in SF-20 scores are shown as the difference between each
PHQ-15 severity group and the reference group (ie, those
with PHQ-15 scores of 0 4). SF-20 scores are adjusted for
age, sex, education, and number of physical disorders. Effect size is the difference in adjusted means divided by the
pooled standard deviation for that scale.

tional status. Decrements in SF-20 scores are shown in


terms of effect size, which is the difference in mean
SF-20 scores, expressed as the number of standard
deviations, between each PHQ-15 interval subgroup
and the reference group. The reference group is the
group with the lowest PHQ-15 scores (ie, 0 4), and for
each SF-20 scale we used the pooled standard deviation for that scale. Effect sizes of 0.5 and 0.8 are typically considered moderate and large between-group
differences, respectively (38). Moving from a lower
level of somatic symptom severity to the next level
typically approximated a moderate effect size for all
SF-20 domains. The figure shows the effect sizes for

Psychosomatic Medicine 64:258 266 (2002)

Table 5 summarizes the results of the multivariate


linear regression models examining the relative effects
of somatic symptoms, depressive symptoms, and medical comorbidity on health outcomes. The partial R2
values shown in the table reflect the proportion of
variance explained by these three independent variables, controlling for one another plus age, gender,
minority status, education, and study site. For example, somatic symptom severity accounted for 35.2% of
the variance in patients general health perceptions in
the primary care sample, whereas depressive symptom
severity and the number of physical disorders each
accounted for 4.5% of the variance. Not only did the
number of physical disorders explain only a small
proportion of the variance for most health outcomes, it
was only weakly correlated with somatic symptom
severity in both the primary care (r ! .10) and obstetrics-gynecology (r ! .14) samples.
Of note, somatic and depressive symptoms have
differential effects. Somatic symptom severity has the
strongest association with general health perceptions,
bodily pain, and physical and role functioning,
whereas depressive symptom severity has its predominant effects on mental health and social functioning.
With respect to clinic visits, somatic symptoms have a

261

K. KROENKE et al.
TABLE 4.

Relationship Between PHQ-15 Somatic Symptom Severity and Disability Days, Symptom-Related Difficulty, and Clinic
Visits

Level of Somatic
Symptom
Severity (PHQ15 Score)
Minimal (14)
Low (59)
Medium (1014)
High (1530)

Mean Disability Daysa (95% CI)

Symptom-Related Difficultyb (%)

Mean Physician Visitsa (95% CI)

Primary Care

ObstetricsGynecology

Primary Care

ObstetricsGynecology

Primary Care

ObstetricsGynecology

1.4 (1.11.7)
4.7 (3.95.5)
8.7 (7.310.1)
18.2 (14.721.7)

1.6 (1.32.0)
4.0 (3.44.6)
5.2 (4.46.1)
12.0 (9.314.7)

1.4
7.8
17.3
28.0

0.7
3.6
10.9
21.7

0.8 (0.70.9)
1.4 (1.21.6)
1.9 (1.72.1)
2.9 (2.43.4)

0.7 (0.70.9)
1.2 (1.01.4)
1.2 (1.01.3)
1.9 (1.62.3)

a
Disability days refers to number of days in past 3 months that symptoms interfered with usual activities. Physician visits refers to past 3
months also. Both are based on self-report. Means are adjusted for age, sex, education, and number of physical disorders.
b
Response to single question: How difficult have these problems made it for you to do your work, take care of things at home, or get along
with other people? The four response categories are not difficult at all, somewhat difficult, very difficult, and extremely difficult.
Symptom-related difficulty in this table refers to those patients reporting very or extremely difficult.

TABLE 5.

Proportion of Variance in Health Outcomes Attributable to Somatic Symptoms, Depressive Symptoms, and Number of
Physical Disorders
Percentage of Variance in Health Outcome Explained bya

Health Outcome

General health perceptions


Bodily pain
Physical functioning
Role functioning
Social functioning
Mental health
Clinic visits
Disability days

PHQ-15 Somatic Symptom


Severity

PHQ-9 Depressive Symptom


Severity

No. of Physical Disorders

Primary
Care

ObstetricsGynecology

Primary
Care

ObstetricsGynecology

Primary
Care

ObstetricsGynecology

35.2
28.2
21.2
20.5
4.1
0.2
7.8
1.4

26.9
29.2
11.4
9.7
0.8
0.2

4.5

5.3

1.1
3.3
27.2
54.4
0.9
16.2

0.5
1.6
12.5
43.5
2.4
9.5

4.5
1.3
9.7
3.1
1.5
0.3
0.8

0.5
0.1
1.3
1.1
1.2

0.6

1.3
0.4

Partial R2 from stepwise linear regression model, which included as independent variables PHQ-15 score, PHQ-9 score, number of physical
disorders, age, gender, minority status, education, and study site.

stronger association, whereas self-reported disability


days are more strongly influenced by depressive
symptoms.
DISCUSSION
Our data establish several desirable psychometric
characteristics of the PHQ-15 as a measure of somatic
symptom severity. First, the internal reliability of the
PHQ-15 was high in both samples. Second, convergent
validity was established by the strong association between PHQ-15 scores and functional status, disability
days, and symptom-related difficulty. Third, discriminant validity was shown by the differential effects of
somatic and depressive symptoms on various outcomes. Fourth, findings from the 3000 primary care
patients were replicated in a second sample of 3000
obstetrics-gynecology patients, although the amount of
change in various health outcomes with increasing

262

PHQ-15 scores was somewhat less in the obstetricsgynecology sample. The differences in magnitude of
effect are probably due to demographic and comorbidity differences between the two samples, although the
possibility of some other unmeasured construct cannot
be excluded.
One limitation of the PHQ-15 as a self-administered
measure is that it cannot distinguish between medically explained and unexplained symptoms, a distinction that typically requires a directed interview and
clinical judgment. The PHQ-15 is therefore best characterized as a measure of somatic symptom severity
rather than a diagnostic instrument for somatoform
disorders. Patients who have high screening scores on
the PHQ-15 should be further questioned to determine
which symptoms might be medically unexplained.
Unexplained symptom counts are more specific for
somatoform disorders and predict adverse health consequences at lower thresholds. Still, total symptom

Psychosomatic Medicine 64:258 266 (2002)

PHQ-15 AS A MEASURE OF SOMATIZATION


counts (including unexplained and explained) are predictive of somatoform disorders (39, 40) and correlate
strongly with psychological distress, functional impairment, and healthcare utilization (5, 14, 17, 41).
Two other study limitations should be noted. Because our sample was disproportionately female, studies involving more male patients are needed to determine the degree to which our findings can be
extrapolated to men (42). Also, validity coefficients are
based almost exclusively on self-report measures. Although patients are typically the criterion standard for
evaluating somatic and depressive symptoms as well
as functional status and health-related quality of life,
independent measures of healthcare utilization would
be desirable in subsequent studies.
The PHQ-15 score functions as a continuous measure. At the same time, scores of 5, 10, and 15 do
represent valid and easy-to-remember thresholds demarcating low, medium, and high levels of somatic
symptom severity. Moving from a lower to the next
TABLE 6.

higher level of severity typically represented a moderate effect size for all functional status domains. In
particular, scores of 15 or higher were associated with
considerable impairment and high utilization. The fact
that 8% to 10% of patients in the two samples have
scores of 15 or greater is notable in that this is also the
lower boundary of prevalence estimates for clinically
significant somatization in primary care (79, 11).
Table 6 shows how the symptom coverage provided
by the PHQ-15 compares favorably with other screeners for somatization: the World Health Organization
Screener for Somatoform Disorders (22), the somatization scale from the Hopkins Symptom Checklist (23),
and two screeners for somatization disorder developed
by Swartz et al. (24) and Othmer and DeSouza (25).
The PHQ-15 assesses 9 of the 12 WHO items, 7 of the
12 Hopkins items, 8 of the 11 Swartz et al. screen
items, and 4 of the 7 Othmer and DeSouza screen
items. This concordance rate of the PHQ-15 with the
other instruments is superior to that of any two other

Comparison of PHQ-15 With Other Somatization Screening Measuresa

Characteristic
No. of symptoms
Time frame of symptoms
Symptoms includedb
Joint or limb pain
Dizziness
Fatigue
Headaches
Back pain
Abdominal pain
Chest pain
Breathing trouble
Fainting
Gas or indigestion
Sleeping trouble
Palpitations
Menstrual problems
Diarrhea (constipation)
Sexual pain/problems
Vomiting
Numbness or tingling
Weakness
Lump in throat
Nausea
Heavy- or light-headedness
Dry mouth
Heaviness of arms/legs
Hot or cold spells
Feeling sickly
Amnesia

PHQ-15

WHO SSD

SCL-12

Swartz et al.

Othmer and
DeSouza

15
1 mo

12
6 mo

12
1 wk

11
Lifetime

7
Lifetime

X
X
X
X
X
X
X
X
X
12
X
X
X
X
X

X
X
X
X
X
X

X
12
X
X
X

X
X

X
X
12
X
X
X

X
X
X
X
X

X
X
X
X

X
X
X

X
X

X
X
X

12

X
X
X
X
X
X

a
WHO-SSD ! World Health Organization Schedule for Somatoform Disorders screener; SCL-12 Hopkins Symptom Checklist somatization
scale; Swartz et al. (24) and Othmer and DeSouza (25) represent 11-item and 7-item screeners, respectively, for somatization disorder.
b
X indicates that this symptom (or equivalent) is included on the scale; 12 indicates that the two symptoms so designated constitute a single
item for the scale; X indicates that a symptom close in type to that in the table is represented on the scale.

Psychosomatic Medicine 64:258 266 (2002)

263

K. KROENKE et al.
instruments with one another. Of note, other measures
designed to assess particular domains of somatization
(eg, hypochondriasis and somatosensory amplification) or to screen for the somatic manifestations of
depressive and anxiety disorders also correlate highly
with one another (43). Although core diagnostic symptoms for depression, fatigue and sleep complaints are
included in the PHQ-15 for several reasons. First, they
are also included in one or more other somatization
screening measures (Table 6). Second, 40% to 50% of
primary care patients with fatigue or sleep complaints
do not have a depressive or anxiety disorder diagnosis
(44 46).
Somatic and depressive symptoms have differential
effects on various measures of health. The results in
our two PHQ studies are similar to findings from the
original PRIME-MD study, where somatoform disorders were most strongly associated with general health
perceptions, bodily pain, role functioning, and clinic
visits (17, 47). By controlling for depression severity
and number of physical disorders, we have shown that
the adverse consequences of somatic symptoms as
measured by the PHQ-15 are not entirely mediated
through coexisting depressive symptoms or medical
comorbidity. There has been a long-standing focus in
general medicine on treatment of physical disorders
and recent attention to improved detection and management of depression. Our findings suggest that in
certain subgroups of patients, that is, high utilizers or
those with poor self-rated health, persistent pain, or
impaired role functioning, the identification and management of residual or unresolved somatic symptoms
may also be important.
The association between number of physical disorders
and various health outcomes is surprisingly weak. In part
this may be because we used a simple count of physical
disorders rather than a more sophisticated medical comorbidity measure that can capture the severity as well
as number of disorders. Also, it may be only those disorders that are symptomatic (and the severity of those
symptoms) that determines impairment. Many physical
disorders (eg, hypertension, well-controlled diabetes,
and stable coronary artery disease) are minimally symptomatic and thus produce less impairment than mental
disorders or symptomatic physical disorders (47, 48). In
both our samples, there was only a weak correlation
between the number of physical disorders and somatic
symptoms. Also, the conventional wisdom that somatic
symptoms in medical patients commonly are secondary
to the side effects of prescribed medications is challenged by studies showing that symptom prevalence may
be equally high in patients receiving placebo or no medication (49 51).
Valid measures for assessing somatization severity

264

are important given the emerging evidence for effective treatments. Two recent critical reviews of the literature have shown that somatizing disorders can respond to antidepressants (96 controlled trials) as well
as cognitive-behavioral therapy (31 controlled trials)
(20, 21). Although depression also responds to these
types of treatment, there may still be reasons for differentiating somatization and depression. First, the
benefits of these two treatment modalities in reducing
somatic symptoms do not appear to be mediated entirely through amelioration of depressive or other psychological symptoms (20, 21). Second, the majority of
antidepressant trials conducted in patients with somatic symptom disorders have focused on specific
symptom syndromes rather than patients with multiple unexplained complaints. Third, the cognitions and
behaviors targeted in depressed patients receiving cognitive-behavioral therapy may be differ from those emphasized in somatizing patients.
Nonpharmacologic treatments other than cognitivebehavioral therapy, including operant behavioral therapy, relaxation therapy, biofeedback, and problemsolving therapy, may also be beneficial for patients
with chronic symptoms, especially pain (5256).
There may be more to offer the somatizing patient than
the rather noninterventionist approach shown by
Smith et al. (19, 57) to reduce costs with modest to no
impact on the patients quality of life. In patients with
milder or less chronic versions of somatization, even
simple measures, such as attention to symptom-specific concerns and expectations, reassurance, and follow-up, may be useful (58 60).
Treatment trials of somatizing patients using the
PHQ-15 as an outcome measure are necessary to establish its sensitivity to change. Also, test-retest reliability
should be evaluated because it is possible that somatization, like depression, may peak the day of the primary care visit and diminish shortly afterward before
treatment can potentially have an effect. Additional
somatization measures may be warranted in some trials because patients with multiple somatic symptoms
frequently have one or several symptoms that cause
greater distress or impairment than the others. Thus,
instruments that focus on the predominant symptom(s), such as 1-to-10 severity scales (61) or other
symptom-specific measures (62), may complement generic somatization scales such as the PHQ-15 in monitoring treatment outcomes. Meanwhile, our validation
data from two studies involving a total of 6000 patients
establish the PHQ-15 as a promising measure for identifying patients with potential somatization in clinical
practice as well as assessing somatic symptom counts
and severity in clinical research.

Psychosomatic Medicine 64:258 266 (2002)

PHQ-15 AS A MEASURE OF SOMATIZATION


The development of the PHQ-15 was underwritten
by an educational grant from Pfizer US Pharmaceuticals Inc., New York, New York.
REFERENCES
1. Schappert SM. National Ambulatory Medical Care Survey: 1989
summary. Vital Health Stat 13 1992;(110):1 80.
2. Kroenke K, Mangelsdorff AD. Common symptoms in ambulatory
care: incidence, evaluation, therapy, and outcome. Am J Med
1989;86:262 6.
3. Marple RL, Kroenke K, Lucey CR, Wilder J, Lucas CA. Concerns
and expectations in patients presenting with physical
complaints: frequency, physician perceptions and actions, and
2-week outcome. Arch Intern Med 1997;157:1482 8.
4. Kroenke K, Price RK. Symptoms in the community: prevalence,
classification, and psychiatric comorbidity. Arch Intern Med
1993;153:2474 80.
5. Kroenke K, Spitzer RL, Williams JBW, Linzer M, Hahn SR,
deGruy FV, Brody D. Physical symptoms in primary care: predictors of psychiatric disorders and functional impairment.
Arch Fam Med 1994;3:774 9.
6. Hartz AJ, Noyes R, Bentler SE, Damiano PC, Willard JC, Momany
ET. Unexplained symptoms in primary care: perspectives of
doctors and patients. Gen Hosp Psychiatry 2000;22:144 52.
7. Kirmayer LJ, Robbins JM. Three forms of somatization in primary care: prevalence, co-occurrence, and sociodemographic
characteristics. J Nerv Ment Dis 1991;179:64755.
8. Spitzer RL, Williams JBW, Kroenke K, Linzer M, deGruy FV,
Hahn SR, Brody D, Johnson JG. Utility of a new procedure for
diagnosing mental disorders in primary care: the PRIME-MD
1000 study. JAMA 1994;272:1749 56.
9. Gureje O, Simon GE, Ustun TB, Goldberg D. Somatization in
cross-cultural perspective: a World Health Organization study in
primary care. Am J Psychiatry 1997;154:989 95.
10. Peveler R, Kilkenny L, Kinmonth AL. Medically unexplained
physical symptoms in primary care: a comparison of self-report
screening questionnaires and clinical opinion. J Psychosom Res
1997;42:24552.
11. Escobar JI, Waitzkin H, Silver RC, Gara M, Holman A. Abridged
somatization: a study in primary care. Psychosom Med 1998;60:
46672.
12. Ormel J, Von Korff M, Ustun TB, Pini S, Korten A, Oldehinkel T.
Common mental disorders and disability across cultures: results
from the WHO Collaborative Study on Psychological Problems
in General Health Care. JAMA 1994;272:1741 8.
13. Kroenke K. Somatization in primary care: its time for parity.
Gen Hosp Psychiatry 2000;22:1413.
14. Katon W, Lin E, Von Korff M, Russo J, Lipscomb P, Bush T.
Somatization: a spectrum of severity. Am J Psychiatry 1991;148:
34 40.
15. Smith GR. The course of somatization and its effects on utilization of health care resources. Psychosomatics 1994;35:2637.
16. Hahn SR, Kroenke K, Spitzer RL, Williams JBW, Brody D, Linzer
M, deGruy FV. The difficult patient in primary care: prevalence,
psychopathology and impairment. J Gen Intern Med 1996;11:1 8.
17. Kroenke K, Spitzer RL, deGruy FV, Hahn SR, Linzer M, Williams JBW, Brody D, Davies M. Multisomatoform disorder: an
alternative to undifferentiated somatoform disorder for the somatizing patient in primary care. Arch Gen Psychiatry 1997;54:
352 8.
18. Simon GE, Von Korff M. Somatization and psychiatric disorder
in the NIMH Epidemiologic Catchment Area Study. Am J Psychiatry 1991;148:1494 500.

Psychosomatic Medicine 64:258 266 (2002)

19. Smith GR, Rost K, Kashner TM. A trial of the effect of a standardized psychiatric consultation on health outcomes and costs
in somatizing patients. Arch Gen Psychiatry 1995;52:238 43.
20. OMalley PG, Jackson JL, Tomkins G, Santoro J, Balden E, Kroenke
K. Antidepressant therapy for unexplained symptoms and symptom syndromes: a critical review. J Fam Pract 1999;48:98093.
21. Kroenke K, Swindle R. Cognitive-behavioral therapy for somatization and symptom syndromes: a critical review of controlled
clinical trials. Psychother Psychosom 2000;69:20515.
22. Tacchini G, Janca A, Isaac M. Somatoform Disorders Schedule,
version 2.0. Geneva: World Health Organization; 1994.
23. Derogatis LR, Lipman RS, Rickels K, Uhlenhuth EH, Covi L. The
Hopkins Symptom Checklist (HSCL): a self-report symptom inventory. Behav Sci 1974;19:115.
24. Swartz M, Hughes D, George L, Blazer D, Landerman R, Bucholz
K. Developing a screening index for community studies of somatization disorder. J Psychiatr Res 1986;20:335 43.
25. Othmer E, DeSouza C. A screening test for somatization disorder
(hysteria). Am J Psychiatry 1985;142:1146 9.
26. Bucholz KK, Dinwiddie SH, Reich T, Shayka JJ, Cloninger CR.
Comparison of screening proposals for somatization disorder:
empirical analyses. Compr Psychiatry 1993;34:59 64.
27. Smith GR, Brown FW. Screening indexes in DSM-III-R somatization disorder. Gen Hosp Psychiatry 1990;12:148 52.
28. Lipowski ZJ. Somatization: the concept and its clinical application. Am J Psychiatry 1988;145:1358 68.
29. Kellner R. Psychosomatic syndromes, somatization and somatoform disorders. Psychother Psychosom 1994;61:4 24.
30. Fava GA, Freyberger HJ, Bech P, Christodoulou G, Sensky T,
Theorell T, Wise TN. Diagnostic criteria for use in psychosomatic research. Psychother Psychosom 1995;63:1 8.
31. Rief W, Hiller W. Toward empirically based criteria for somatoform disorders. J Psychosom Res 1999;46:50718.
32. Liu G, Clark MR, Eaton WW. Structural factor analyses for
medically unexplained somatic symptoms of somatization disorder in the Epidemiologic Catchment Area Study. J Psychosom
Res 1997;42:24552.
33. Hahn SR, Kroenke K, Williams JBW, Spitzer RL. Evaluation of
mental disorders with the PRIME-MD. In: Maruish M, editor.
Handbook of Psychological Assessment in Primary Care Settings. Mahwah (NJ): Lawrence Erlbaum; 2000:191253.
34. Spitzer RL, Kroenke K, Williams JBW, Patient Health Questionnaire
Study Group. Validity and utility of a self-report version of PRIMEMD: the PHQ Primary Care Study. JAMA 1999;282:173744.
35. Spitzer RL, Williams JBW, Kroenke K, Hornyak R, McMurray J,
Heartwell SF, for the Patient Health Questionnaire Obstetrics Gynecology Study Group. Validity and utility of the Patient Health
Questionnaire in assessment of 3000 obstetric-gynecologic
patients: the PRIME-MD Patient Health Questionnaire ObstetricsGynecology Study. Am J Obstet Gynecol 2000;183:759 69.
36. Kroenke K, Arrington ME, Mangelsdorff AD. The prevalence of
symptoms in medical outpatients and the adequacy of therapy.
Arch Intern Med 1990;150:16859.
37. Stewart AL, Hays RD, Ware JE. The MOS Short-Form General
Health Survey: reliability and validity in a patient population.
Med Care 1988;26:724 32.
38. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting
changes in health status. Med Care 1989;27:S178 S189.
39. Kisely S, Goldberg D, Simon G. A comparison between somatic
symptoms with and without clear organic cause: results of an
international study. Psychol Med 1997;27:10119.
40. Kroenke K, Spitzer RL, deGruy FV, Swindle R. A symptom
checklist to screen for somatoform disorders in primary care.
Psychosomatics 1998;39:26372.

265

K. KROENKE et al.
41. Simon G, Gater R, Kisely S, Piccinelli M. Somatic symptoms of
distress: an international primary care study. Psychosom Med
1996;58:481 8.
42. Barsky AJ, Peekna HM, Borus JF. Somatic symptom reporting in
women and men. J Gen Intern Med 2001;16:266 75.
43. Wyshak G, Barsky AJ, Klerman GL. Comparison of psychiatric
screening tests in a general medical setting using ROC analysis.
Med Care 1991;29:775 85.
44. Buchwald D, Pearlman T, Umali J, Schmaling K, Katon W.
Functional status in patients with chronic fatigue syndrome,
other fatiguing illnesses, and healthy individuals. Am J Med
1996;171:364 70.
45. Cathebras PJ, Robbins JM, Kirmayer LJ, Hayton BC. Fatigue in
primary care: prevalence, psychiatric comorbidity, illness behavior, and outcome. J Gen Intern Med 1992;7:276 86.
46. Simon GE, Von Korff M. Prevalence, burden, and treatment of
insomnia in primary care. Am J Psychiatry 1997;154:141723.
47. Spitzer RL, Kroenke K, Linzer M, Hahn SR, Williams JBW,
deGruy FV, Brody D, Davies M. Health-related quality of life in
primary care patients with mental disorders: results from the
PRIME-MD 1000 study. JAMA 1995;274:15117.
48. Wells KB, Stewart A, Hays RD, Burnam A, Rogers W, Daniels M,
Berry S, Greenfield S, Ware J. The functioning and well-being of
depressed patients: results from the Medical Outcomes Study.
JAMA 1989;262:914 9.
49. Di Tullio M, Alli C, Avanzini F, Bettelli G, Colombo F, Devoto MA,
Marchioli R, Mariotti G, Radice M, Taioli E, Tognoni G, Villella M,
Zussino A. Prevalence of symptoms generally attributed to hypertension or its treatment: study on blood pressure in elderly outpatients (SPAA). J Hypertens 1988;6(Suppl 1):S8790.
50. TOMHS Research Group. The Treatment of Mild Hypertension
Study. Arch Intern Med 1991;151:141323.
51. Reidenberg MM, Lowenthal DT. Adverse nondrug reactions.
N Engl J Med 1968;279:678 9.

52. Mayou R, Bass C, Sharpe M, eds. Treatment of functional somatic symptoms. Oxford: Oxford University Press; 1995.
53. Nicholas MK, Wilson PH, Goyen J. Operant-behavioural and
cognitive-behavioural treatment for chronic low back pain. Behav Res Ther 1991;29:22538.
54. Compas BE, Haaga DAF, Keefe FJ, Leitenberg H, Williams DA.
Sampling of empirically supported psychological treatments
from health psychology: smoking, chronic pain, cancer, and
bulimia nervosa. J Consult Clin Psychol 1998;66:89 112.
55. Keefe FJ, Dunsmore J, Burnett R. Behavioral and cognitivebehavioral approaches to chronic pain: recent advances and
future directions. J Consult Clin Psychol 1992;60:528 36.
56. Wilkinson P, Mynors-Wallis L. Problem-solving therapy in the
treatment of unexplained physical symptoms in primary care: a
preliminary study. J Psychosom Res 1994;38:591 8.
57. Smith GR, Monson RA, Ray DC. Psychiatric consultation in
somatization disorder: a randomized, controlled study. N Engl
J Med 1986;314:140713.
58. Thomas KB. General practice consultations: is there any point in
being positive? BMJ 1995.
59. Rao JK, Weinberger M, Kroenke K. Visit-specific expectations and
satisfaction with care: a literature review. Arch Fam Med 2000;9:
114855.
60. Jackson JL, Kroenke K. The effect of unmet expectations among
adults presenting with physical symptoms. Ann Intern Med
2001;134:889 97.
61. Cohen SR, Mount BM, Bruera E, Provost M, Rowe J, Tong K.
Validity of the McGill Quality of Life Questionnaire in the
palliative care setting: a multi-centre Canadian study demonstrating the importance of the existential domain. Palliat Med
1997;11:320.
62. Kroenke K. Studying symptoms: sampling and measurement
issues. Ann Intern Med 2001;134:844 55.

APPENDIX
Patient Health Questionnaire 15-Item Somatic
Symptom Severity Scale
During the past 4 weeks, how much
Not
Bothered Bothered
have you been bothered by any of the bothered
a little
a lot
following problems?
at all
a. Stomach pain
b. Back pain
c. Pain in your arms, legs, or joints
(knees, hips, etc.)
d. Menstrual cramps or other problems
with your periods [Women only]
e. Headaches
f. Chest pain
g. Dizziness
h. Fainting spells
i. Feeling your heart pound or race
j. Shortness of breath
k. Pain or problems during sexual
intercourse
l. Constipation, loose bowels, or
diarrhea
m. Nausea, gas, or indigestion
n. Feeling tired or having low energy
o. Trouble sleeping

266

!
!
!

!
!
!

!
!
!

!
!
!
!
!
!
!

!
!
!
!
!
!
!

!
!
!
!
!
!
!

!
!
!

!
!
!

!
!
!

Psychosomatic Medicine 64:258 266 (2002)