Content uploaded by Ron de Graaf
Author content
All content in this area was uploaded by Ron de Graaf on Sep 04, 2015
Content may be subject to copyright.
168
WWoorrlldd PPssyycchhiiaattrryy 66::33 --
October 2007
Lifetime prevalence and age-of-onset distributions
of mental disorders in the World Health Organization’s
World Mental Health Survey Initiative
RESEARCH REPORT
RONALD C. KESSLER
1
, MATTHIAS ANGERMEYER
2
, JAMES C. ANTHONY
3
, RON DE GRAAF
4
, KOEN
DEMYTTENAERE
5
, ISABELLE GASQUET
6
, GIOVANNI DE GIROLAMO
7
, SEMYON GLUZMAN
8
, OYE GUREJE
9
,
J
OSEP MARIA HARO
10
, NORITO KAWAKAMI
11
, AIMEE KARAM
12
, DAPHNA LEVINSON
13
, MARIA ELENA MEDINA
MORA
14
, MARK A. OAKLEY BROWNE
15
, JOSÉ POSADA-VILLA
16
, DAN J. STEIN
17
, CHEUK HIM ADLEY TSANG
18
,
S
ERGIO AGUILAR-GAXIOLA
19
, JORDI ALONSO
20
, SING LEE
21
, STEVEN HEERINGA
22
, BETH-ELLEN PENNELL
22
,
P
ATRICIA BERGLUND
22
, MICHAEL J. GRUBER
1
, MARIA PETUKHOVA
1
, SOMNATH CHATTERJI
23
,
T. B
EDIRHAN ÜSTÜN
23
, FOR THE WHO WORLD MENTAL HEALTH SURVEY CONSORTIUM
1
Department of Health Care Policy, Harvard Medical School, 180 Longwood Avenue, Boston, MA 02115, USA;
2
Department of Psychiatry, University
of Leipzig, Germany;
3
Department of Epidemiology, Michigan State University, East Lansing, MI, USA;
4
Netherlands Institute of Mental Health and
Addiction, Utrecht, The Netherlands;
5
Department of Neurosciences and Psychiatry, University Hospital Gasthuisberg, Leuven, Belgium;
6
Hôpitaux de Paris,
Paris, France;
7
Department of Mental Health, Local Health Unit, Bologna, Italy;
8
Ukrainian Psychiatric Association, Kyiv, Ukraine;
9
Department
of Psychiatry, University College Hospital, Ibadan, Nigeria;
10
Sant Joan de Deu – Mental Health Services, Barcelona, Spain;
11
Department of Mental Health,
University of Tokyo Graduate School of Medicine, Tokyo, Japan;
12
Institute for Development, Research, Advocacy and Applied Care (IDRAAC), Beirut,
Lebanon;
13
Research and Planning, Mental Health Services, Ministry of Health, Jerusalem, Israel;
14
Department of Epidemiology, National Institute
of Psychiatry, Mexico City, Mexico;
15
Department of Rural and Indigenous Health, School of Rural Health, Faculty of Medicine, Nursing and Health Sciences,
Monash University, Victoria, Australia;
16
Colegio Mayor de Cundinamarca University, Saldarriaga Concha Foundation, Bogota, Colombia;
17
Department of
Psychiatry and Mental Health, University of Cape Town, South Africa;
18
Hong Kong Mood Disorders Centre, Hong Kong, People’s Republic of China;
19
Center
for Reducing Health Disparities, UC Davis School of Medicine, Sacramento, CA, USA;
20
Health Services Research Unit, Institut Municipal d’Investigacio
Medica (IMIM), Barcelona, Spain;
21
Department of Psychiatry, Chinese University of Hong Kong, People’s Republic of China;
22
Institute for Social Research,
University of Michigan, Ann Arbor, MI, USA;
23
Global Programme on Evidence for Health Policy, World Health Organization, Geneva, Switzerland
Although psychiatric epidemiological surveys have been
carried out since after World War II (1), absence of a com-
mon format for diagnosis hampered cross-national synthe-
ses. This situation changed in the early 1980s, with the de-
velopment of fully structured research diagnostic interviews
(2) and the implementation of large-scale psychiatric epi-
demiological surveys in many countries (3-5). The World
Health Organization (WHO) developed a diagnostic in-
strument, the WHO Composite International Diagnostic
Interview (CIDI) (6,7), based on extensive cross-national
field trials, for use in cross-national epidemiological sur-
veys (8-14). In 1998, the WHO created the WHO Interna-
tional Consortium in Psychiatric Epidemiology (ICPE) to
coordinate comparative analyses of these surveys. The
ICPE launched the WHO World Mental Health (WMH)
Survey Initiative shortly thereafter to conduct coordinated
CIDI surveys in all parts of the world. The current report
presents the first cross-national results regarding age of on-
set, lifetime prevalence, and projected lifetime risk of men-
tal disorders from the 17 WMH surveys so far completed.
Data of this sort are sorely needed by policy planners to
assess the societal burden of mental disorders, unmet need
for treatment, and barriers to treatment. These data are es-
pecially important given evidence from the WHO Global
Burden of Disease Study that mental disorders impose
enormous burdens worldwide, due to their combination of
high prevalence and high disability (15), and evidence that,
despite efficacious treatments, substantial unmet need for
Data are presented on the lifetime prevalence, projected lifetime risk, and age-of-onset distributions of mental disorders in the World Health
Organization (WHO)’s World Mental Health (WMH) Surveys. Face-to-face community surveys were conducted in seventeen countries in
Africa, Asia, the Americas, Europe, and the Middle East. The combined numbers of respondents were 85,052. Lifetime prevalence, project-
ed lifetime risk, and age of onset of DSM-IV disorders were assessed with the WHO Composite International Diagnostic Interview (CIDI),
a fully-structured lay administered diagnostic interview. Survival analysis was used to estimate lifetime risk. Median and inter-quartile
range (IQR) of age of onset is very early for some anxiety disorders (7-14, IQR: 8-11) and impulse control disorders (7-15, IQR: 11-12). The
age-of-onset distribution is later for mood disorders (29-43, IQR: 35-40), other anxiety disorders (24-50, IQR: 31-41), and substance use dis-
orders (18-29, IQR: 21-26). Median and IQR lifetime prevalence estimates are: anxiety disorders 4.8-31.0% (IQR: 9.9-16.7%), mood disor-
ders 3.3-21.4% (IQR: 9.8-15.8%), impulse control disorders 0.3-25.0% (IQR: 3.1-5.7%), substance use disorders 1.3-15.0% (IQR: 4.8-9.6%),
and any disorder 12.0-47.4% (IQR: 18.1-36.1%). Projected lifetime risk is proportionally between 17% and 69% higher than estimated life-
time prevalence (IQR: 28-44%), with the highest ratios in countries exposed to sectarian violence (Israel, Nigeria, and South Africa), and
a general tendency for projected risk to be highest in recent cohorts in all countries. These results document clearly that mental disorders
are commonly occurring. As many mental disorders begin in childhood or adolescents, interventions aimed at early detection and treat-
ment might help reduce the persistence or severity of primary disorders and prevent the subsequent onset of secondary disorders.
Key words: Mental disorders, lifetime prevalence, projected lifetime risk, age-of-onset distribution
(World Psychiatry 2007;6:168-176)
IMP. 168-176 24-09-2007 16:11 Pagina 168
169
treatment exists throughout the world (16). While earlier
studies found high lifetime prevalence and generally early
age-of-onset distributions of mental disorders, they did
not make systematic disorder-specific age-of-onset compar-
isons. The latter are important for targeting early interven-
tions, which are coming to be seen as critical for an effec-
tive public health response to mental disorders (17-19).
Previous studies also focused on lifetime prevalence (the
proportion of the population with a lifetime disorder up to
age at interview) rather than projected lifetime risk (the es-
timated proportion of the population who will have the
disorder by the end of their life), even though the latter is
more important for policy planning purposes. We consider
both prevalence and risk in this report.
METHODS
Samples
WMH surveys were administered in Africa (Nigeria, South
Africa); the Americas (Colombia, Mexico, United States),
Asia and the Pacific (Japan, New Zealand, Beijing and
Shanghai in the People’s Republic of China, henceforth re-
ferred to as Metropolitan PRC), Europe (Belgium, France,
Germany, Italy, the Netherlands, Spain, Ukraine) (20); and
the Middle East (Israel, Lebanon). Seven of these countries
are classified by the World Bank as less developed (China,
Colombia, Lebanon, Mexico, Nigeria, South Africa,
Ukraine), while the others are classified as developed (21).
Most WMH surveys were based on stratified multistage
clustered area probability household samples. Samples of
areas equivalent to counties or municipalities in the US
were selected in the first stage, followed by one or more
subsequent stages of geographic sampling (e.g., towns
within counties, blocks within towns, households within
blocks) to arrive at a sample of households. In each of
them, a listing of household members was created and one
or two people were selected to be interviewed. No substi-
tution was allowed when the originally sampled household
resident could not be interviewed. The household samples
were selected from census area data in all countries other
than France (where telephone directories were used) and
the Netherlands (where postal registries were used). Sev-
eral WMH surveys (Belgium, Germany, Italy) used munic-
ipal resident registries to select respondents without listing
households. The Japanese sample is the only totally un-
clustered sample, with households randomly selected in
each of the four sample areas and one random respondent
selected in each sample household. Nine of the 17 surveys
were based on nationally representative household sam-
ples, while two others were based on nationally represen-
tative household samples in urbanized areas (Colombia,
Mexico).
All surveys were conducted face-to-face by trained lay
interviewers in multi-stage household probability samples,
with 85,052 respondents. Country-level samples ranged
from 2372 (Netherlands) to 12,992 (New Zealand). The
weighted average cross-national response rate was 71.1%,
with a 45.9-87.7% range (Table 1).
The Part I interview schedule, completed by all respon-
dents, assessed core diagnoses. All respondents who met
criteria for any diagnosis plus a probability sub-sample of
other Part I respondents were administered Part II, which
assessed disorders of secondary interest and a wide range
of correlates. Part I data were weighted to adjust for differ-
ential probabilities of selection and to match population
distributions on socio-demographic and geographic data.
The Part II sample was additionally weighted for the over-
sampling of Part I respondents with core disorders. The in-
terview schedule and other study materials were translated
using standardized WHO translation and back-translation
protocols. Consistent interviewer training procedures and
quality control monitoring were used in all surveys (22,23).
Informed consent was obtained in all countries using pro-
cedures approved by local Institutional Review Boards.
Measures
Diagnoses were based on CIDI Version 3.0 (24), which
generates both ICD-10 (25) and DSM-IV (26) diagnoses.
DSM-IV criteria are used here to facilitate comparison
with previous epidemiological surveys. Core diagnoses in-
cluded anxiety disorders (panic disorder, agoraphobia
without panic disorder, specific phobia, social phobia, gen-
eralized anxiety disorder, post-traumatic stress disorder,
and separation anxiety disorder), mood disorders (major
depressive disorder, dysthymic disorder, bipolar disorder I
or II or subthreshold bipolar disorder), impulse control
disorders (intermittent explosive disorder, oppositional-
defiant disorder, conduct disorder, attention-deficit/hyper-
activity disorder), and substance use disorders (alcohol
and drug abuse with or without dependence). Not all dis-
orders were assessed in all countries. The Western Euro-
pean countries did not assess bipolar disorders and drug
dependence. Only three countries (Colombia, Mexico,
United States) assessed all impulse control disorders.
The disorders that require childhood onset (oppositional
defiant disorder, conduct disorder, and attention-deficit/hy-
peractivity disorder) were included in Part II and limited to
respondents in the age range 18-39/44, because of con-
cerns about recall bias among older respondents. All other
disorders were assessed for the full sample age range. Or-
ganic exclusion rules and hierarchy rules were used to
make all diagnoses other than substance use disorders,
which were diagnosed without hierarchy, because abuse
often is a stage in the progression to dependence. Clinical
calibration studies (27) found CIDI to assess these disor-
ders with generally good validity in comparison to blinded
clinical reappraisal interviews using the Structured Clini-
cal Interview for DSM-IV (SCID) (28). CIDI prevalence es-
IMP. 168-176 24-09-2007 16:11 Pagina 169
170
WWoorrlldd PPssyycchhiiaattrryy 66::33 --
October 2007
timates were not higher than SCID prevalence estimates.
Retrospective age-of-onset reports were based on a ques-
tion series designed to avoid the implausible response pat-
terns obtained in using the standard CIDI age-of-onset
question (29). Experimental research has shown that this
question sequence yields responses with a much more
plausible age-of-onset distribution than the standard CIDI
age-of-onset question (30). Predictor variables included
cohort (defined by ages at interview 18-34, 35-49, 50-64,
65+), sex, and education (students versus non-students
with low, low-average, average-high, and high education
categories based on country-specific distributions). Educa-
tion was coded as a time-varying predictor by assuming an
orderly educational history.
Analysis procedures
Age of onset and projected lifetime risk as of age 75 were
estimated using the two-part actuarial method implement-
ed in SAS 8.2 (31). Predictors were examined using dis-
crete-time survival analysis with person-year as the unit of
analysis (32). Standard errors were estimated using the
Taylor series linearization method (33) implemented in the
SUDAAN software system (34). Multivariate significance
tests were made with Wald χ
2
tests, using Taylor series de-
sign-based coefficient variance-covariance matrices. Stan-
dard errors of lifetime risk were estimated using the jack-
knife repeated replication method (35) implemented in a
SAS macro (31). Significance tests were all evaluated at the
.05 level with two-sided tests.
RESULTS
Lifetime prevalence
The estimated lifetime prevalence of having one or more
of the disorders considered here varies widely across the
WMH surveys, from 47.4% in the United States to 12.0% in
Nigeria. The inter-quartile range (IQR; 25th-75th percentiles
across countries) is 18.1-36.1%. Symptoms consistent with
the existence of one or more lifetime mental disorders were
reported by more than one-third of respondents in five coun-
tries (Colombia, France, New Zealand, Ukraine, United
States), more than one-fourth in six (Belgium, Germany,
Lebanon, Mexico, The Netherlands, South Africa), and more
than one-sixth in four (Israel, Italy, Japan, Spain). The re-
Table 1 Sample characteristics of the World Mental Health Surveys
Country Survey Field dates Age range Sample size Response rate
Part I Part II Part II and age
≤ 44
a
Belgium ESEMeD 2001-2 18+ 12419 1043 1486 50.6
Colombia NSMH 2003 18-65 14426 2381 1731 87.7
France ESEMeD 2001-2 18+ 12894 1436 1727 45.9
Germany ESEMeD 2002-3 18+ 13555 1323 1621 57.8
Israel NHS 2002-4 21+ 14859 - - 72.6
Italy ESEMeD 2001-2 18+ 14712 1779 1853 71.3
Japan WMHJ 2002-2003 2002-3 20+ 12436 1887 1282 56.4
Lebanon LEBANON 2002-3 18+ 12857 1031 1595 70.0
Mexico M-NCS 2001-2 18-65 15782 2362 1736 76.6
Netherlands ESEMeD 2002-3 18+ 12372 1094 1516 56.4
New Zealand NZMHS 2004-5 16+ 12992 7435 4242 73.3
Nigeria NSMHW 2002-3 18+ 16752 2143 1203 79.3
People’s B-WMH 2002-3 18+ 15201 1628 1570 74.7
Republic of China S-WMH
South Africa SASH 2003-4 18+ 14315 - - 87.1
Spain ESEMeD 2001-2 18+ 15473 2121 1960 78.6
Ukraine CMDPSD 2002 18+ 14725 1720 1541 78.3
United States NCS-R 2002-3 18+ 19282 5692 3197 70.9
ESEMeD - European Study of the Epidemiology of Mental Disorders; NSMH - Colombian National Study of Mental Health; NHS - Israel National Health Sur-
vey; WMHJ 2002-2003 - World Mental Health Japan Survey; LEBANON - Lebanese Evaluation of the Burden of Ailments and Needs of the Nation; M-NCS - Mex-
ico National Comorbidity Survey; NZMHS - New Zealand Mental Health Survey; NSMHW - Nigerian Survey of Mental Health and Wellbeing; B-WMH - Beijing
World Mental Health Survey; S-WMH - Shanghai World Mental Health Survey; SASH - South Africa Health Survey; CMDPSD - Comorbid Mental Disorders dur-
ing Periods of Social Disruption; NCS-R - U.S. National Comorbidity Survey Replication
The response rate is calculated as the ratio of the number of households in which an interview was completed to the number of households originally sampled, ex-
cluding from the denominator households known not to be eligible either because of being vacant at the time of initial contact or because the residents were un-
able to speak the designated languages of the survey
a
All countries were age restricted to
≤44, with the exception of Nigeria, People’s Republic of China, and Ukraine, which were age restricted to ≤39
IMP. 168-176 24-09-2007 16:11 Pagina 170
171
maining two countries, Metropolitan PRC (13.2%) and
Nigeria (12.0%), had considerably lower prevalence esti-
mates, that are likely to be downwardly biased (36, 37).
Prevalence estimates for other developing countries were all
above the lower bound of the inter-quartile range (Table 2).
All four classes of disorder were important components
of overall prevalence. Anxiety disorders were the most pre-
valent in ten countries (4.8-31.0%, IQR 9.9-16.7%) and
mood disorders in all but one other country (3.3-21.4%,
IQR 9.8-15.8%). Impulse control disorders were the least
prevalent in most countries that included a relatively full
assessment of these disorders (0.3-25.0%, IQR 3.1-5.7%).
Substance use disorders were generally the least prevalent
elsewhere (1.3-15.0%, IQR 4.8-9.6). The Western Euro-
pean countries did not assess illicit drug abuse-depend-
ence, though, leading to artificially low prevalence esti-
mates (1.3-8.9%) compared to other countries (2.2-
15.0%). Substance dependence was also assessed only in
the presence of abuse, possibly further reducing estimated
prevalence (38). Lifetime disorder co-occurrence was quite
common, as seen by noting that the sum of prevalence
across the four disorder types was generally between 30%
and 50% higher than the prevalence of any disorder. With-
in-class co-occurrence cannot be seen in the reported re-
sults, but is even stronger than between-class co-occur-
rence (results available on request).
Age-of-onset distributions
Despite the wide cross-national variation in estimated
lifetime prevalence, considerable cross-national consisten-
cy exists in standardized age-of-onset distributions (de-
tailed results are not reported here, but are available on re-
quest).
Impulse control disorders have the earliest age-of-onset
distributions, both in terms of early median ages of onset
(7-9 years of age for attention-deficit/hyperactivity disor-
der, 7-15 for oppositional-defiant disorder, 9-14 for con-
duct disorder, and 13-21 for intermittent explosive disor-
der) and an extremely narrow age range of onset risk, with
80% of all lifetime attention-deficit/hyperactivity disorder
beginning in the age range 4-11 and the vast majority of op-
positional-defiant disorder and conduct disorder begin-
ning between ages 5 and 15. Although the age-of-onset dis-
tribution is less concentrated for intermittent explosive dis-
order, fully half of all lifetime cases have onsets in child-
hood and adolescence.
The situation is more complex with anxiety disorders, as
the age-of-onset distributions fall into two distinct sets.
The phobias and separation anxiety disorder all have very
early ages of onset (medians in the range 7-14, IQR 8-11).
Generalized anxiety disorder, panic disorder, and post-
traumatic stress disorder, in comparison, have much later
age-of-onset distributions (median 24-50, IQR 31-41), with
much wider cross-national variation than for the impulse
control disorders or the phobias or separation anxiety dis-
order.
The age-of-onset distributions for mood disorders are
similar to those for generalized anxiety disorder, panic dis-
order, and post-traumatic stress disorder. Prevalence is
consistently low until the early teens, at which time a
roughly linear increase begins that continues through late
middle age, with a more gradual increase thereafter. The
median age of onset of mood disorders ranges between the
late 20s and the early 40s (29-43, IQR 35-40).
The age-of-onset distribution of substance use disorders
is consistent across countries, in that few onsets occur pri-
or to the mid teens and cumulative increase in onset is rap-
id in adolescence and early adulthood. Considerable cross-
national variation exists, though, in the sharpness of the
change in the slope as well as in the age range of this
change. This cross-national variation leads to wider cross-
national variation in both the median and the inter-quar-
tile range of the age-of-onset distributions than for impulse
control disorders or phobias or separation anxiety disor-
der, but lower variation than for mood disorders or other
anxiety disorders.
Projected lifetime risk
Projected lifetime risk of any disorder as of age 75 is be-
tween 17% (United States) and 69% (Israel) higher than es-
timated lifetime prevalence (IQR 28-44%) (Table 2). The
highest risk-to-prevalence ratios (57-69%) are in countries
exposed to sectarian violence (Israel, Nigeria, and South
Africa). Excluding these three, there is no strong difference
in ratios of less developed (28-41%) versus developed (17-
49%) countries. The highest class-specific proportional in-
crease in projected risk is for mood disorders (45-170%,
IQR 61-98%) and the lowest for impulse control disorders
(0-14%, IQR 0-2%), consistent with the former having the
latest and the latter having the earliest age-of-onset distri-
bution. The projected lifetime risk estimates suggest that ap-
proximately half the population (47-55%) will eventually
have a mental disorder in six countries (Colombia, France,
New Zealand, South Africa, Ukraine, United States), ap-
proximately one-third (30-43%) in six other countries (Bel-
gium, Germany, Israel, Lebanon, Mexico, the Netherlands),
approximately one-fourth (24-29%) in three others (Italy,
Japan, Spain), and approximately one-fifth (18-19%) in the
remaining countries (Metropolitan PRC, Nigeria).
Cohort effects
Previous research has suggested that projected lifetime
risk might be increasing in recent cohorts (39). Prospective
tracking studies are required to monitor cohort effects di-
rectly. However, indirect approximations can be obtained
in cross-sectional data using retrospective age-of-onset re-
IMP. 168-176 24-09-2007 16:11 Pagina 171
172
WWoorrlldd PPssyycchhiiaattrryy 66::33 --
October 2007
ports. This was done in the WMH data using discrete-time
survival analysis to predict onset of disorders across age
groups 18-34, 35-49, 50-64, and 65+. As these surveys were
completed between 2002 and 2005, the most recent cohorts
(aged 18-34 at interview) roughly correspond to those born
in the years from 1968+. Respondents aged 35-49 at inter-
view correspond roughly to cohorts born in 1953-1970,
while those aged 50-64 were born in 1938-1955, and those
aged 65+ were born before 1938. Survival analysis finds that
the odds ratios for anxiety, mood, and substance use disor-
ders are generally higher in recent compared to older co-
horts, while not for impulse control disorders (Tables 3-5).
No meaningful difference exists between less developed
and developed countries, although cross-national variation
exceeds chance expectations.
DISCUSSION
Three possible biases could have led to under-estimating
prevalence. First, people with mental illness have been found
to be less likely than others to participate in surveys, because
of sample frame exclusions (e.g., excluding homeless people),
differential mortality, or greater reluctance to participate (40).
Variation in the magnitude of such under-representation
across countries could help account for the wide between-
country variation in prevalence-risk estimates. Second, pre-
vious research suggests that lifetime prevalence is sometimes
under-reported because of respondent reluctance to admit
mental illness (41). This bias might be especially strong in less
developed countries with no strong tradition of independent
public opinion research, which could help account for the es-
pecially low prevalence-risk estimates in Nigeria and Metro-
politan PRC. Third, interviewer error might have led to un-
der-reporting, especially in countries where there was an in-
direct incentive to rush through interviews, because inter-
viewers were paid by the interview rather than by the hour.
The most plausible bias that could have led to over-estimat-
ing prevalence, in comparison, is that the interview thresh-
olds for defining disorders might have been too liberal. How-
ever, as noted in the section on measures, clinical reappraisal
studies carried out in some of the countries with the highest
prevalence estimates found no evidence of such bias (27).
Two possible biases of other sorts are also noteworthy.
First, the method used to estimate lifetime risk was based
on the assumption of constant conditional risk of first on-
set in a given year of life across cohorts. The existence of an
apparent cohort effect means that this assumption is incor-
rect, probably causing an under-estimation of lifetime risk
in younger cohorts. Second, age of onset might have been
recalled with error related to age at interview, which could
produce the data pattern found here as indirect evidence for
a cohort effect (42). Evidence for age-related bias has been
documented in previous epidemiological research (29), al-
though the novel probing strategy used in the WMH surveys
has been shown to minimize this problem (30).
Based on these considerations, the wide cross-national
variation in WMH prevalence and risk estimates should be
Table 2 Lifetime prevalence and projected lifetime risk as of age 75 of DSM-IV disorders
Country Any anxiety disorder Any mood disorder Any impulse control disorder Any substance use disorder Any disorder
Prevalence Projected Prevalence Projected Prevalence Projected Prevalence Projected Prevalence Projected
lifetime risk lifetime risk lifetime risk lifetime risk lifetime risk
%N
a
SE%SE%N
a
SE%SE%N
a
SE%SE%N
a
SE%SE%N
a
SE % SE
Belgium 13.1 3219 1.9 15.7 2.5 14.1 3367 1.0 22.8 1.7 35.2 3331 1.4 35.2 1.4 18.3 1195 0.9 10.5 1.1 29.1 1519 2.3 37.1
d
3.0
Colombia 25.3 3948 1.4 30.9 2.5 14.6 3666 0.7 27.2 2.0 39.6 273 0.8 10.3 0.9 19.6 1345 0.6 12.8 1.0 39.1 1432 1.3 55.2
d
6.0
France 22.3 3445 1.4 26.0 1.6 21.0 3648 1.1 30.5 1.4 37.6 3371 1.3 37.6 1.3 17.1 1202 0.5 18.8 0.6 37.9 1847 1.7 47.2
d
1.6
Germany 14.6 3314 1.5 16.9 1.7 39.9 3372 0.6 16.2 1.3 33.1 3331 0.8 33.1 0.8 16.5 1228 0.6 18.7 0.9 25.2 1573 1.9 33.0
d
2.5
Israel 15.2 3252 0.3 10.1 0.9 10.7 3524 0.5 21.2 1.6 -
b
-- - -15.3 1261 0.3 1
6.3 0.4 17.6 1860 0.6 29.7
d
1.5
Italy 11.0 3328 0.9 13.7 1.2 39.9 3452 0.5 17.3 1.2 31.7 3327 0.4 -
c
- 11.3 1156 0.2 11.6 0.3 18.1 1612 1.1 26.0
d
1.9
Japan 16.9 3155 0.6 39.2 1.2 37.6 3183 0.5 14.1 1.7 32.8 3311 1.0 -
c
- 14.8 1169 0.5 16.2 0.7 18.0 1343 1.1 24.4
d
1.8
Lebanon 16.7 3282 1.6 20.2 1.8 12.6 3352 0.9 20.1 1.2 34.4 3353 0.9 34.6 1.0 12.2 1127 0.8 - -
c
25.8 1491 1.9 32.9
d
2.1
Mexico 14.3 3684 0.9 17.8 1.6 39.2 3598 0.5 20.4 1.7 35.7 3152 0.6 35.7 0.6 17.8 1378 0.5 11.9 1.0 26.1 1148 1.4 36.4
d
2.1
Netherlands 15.9 3320 1.1 21.4 1.8 17.9 3476 1.0 28.9 1.9 3
4.7 3337 1.1 34.8 1.1 18.9 1210 0.9 11.4 1.2 31.7 1633 2.0 42.9
d
2.5
New Zealand 24.6 3171 0.7 30.3 1.5 20.4 2755 0.5 29.8 0.7 -
b
- - - - 12.4 1767 0.4 14.6 0.5 39.3 4815 0.9 48.6
d
1.5
Nigeria 16.5 3169 0.9 37.1 0.9 33.3 3236 0.3 38.9 1.2 30.3 3339 0.1 -
c
- 13.7 1119 0.4 16.4 1.0 12.0 1440 1.0 19.5
d
1.9
PR China 14.8 3159 0.7 36.0 0.8 33.6 3185 0.4 37.3 0.9 34.3 3337 0.9 34.9 0.9 14.9 1128 0.7 16.1 0.8 13.2 1419 1.3 18.0
d
1.5
South Africa 15.8 3695 0.8 30.1 4.4 39.8 3439 0.7 20.0 2.4 -
b
- - - - 13.3 1505 0.9 17.5 1.2 30.3 1290 1.1 47.5
d
3.7
Spain 19.9 3375 1.1 13.3 1.4 10.6 3672 0.5 20.8 1.2 32.3 3340 0.8 3
2.3 0.8 13.6 1180 0.4 14.6 0.5 19.4 1842 1.4 29.0
d
1.8
Ukraine 10.9 3371 0.8 17.3 2.0 15.8 3814 0.8 25.9 1.5 38.7 3391 1.1 39.7 1.3 15.0 1293 1.3 18.8 1.7 36.1 1074 1.5 48.9
d
2.5
United States 31.0 2692 1.0 36.0 1.4 21.4 2024 0.6 31.4 0.9 25.0 1051 1.1 25.6 1.1 14.6 1144 0.6 17.4 0.6 47.4 3929 1.1 55.3
d
1.2
a
The numbers reported here are the numbers of respondents with the disorders indicated in the column heading. The denominators used to calculate prevalence
estimates based on these numbers of cases are reported in Table 1. In the case of anxiety disorders and substance use disorders, the denominators are the numbers
of respondents in the Part II sample. In the case of mood disorders, the denominators are the numbers of respondents in the Part I sample. In the case of impulse
control disorders and any disorders, the denominators are the numbers of respondents aged
≤44 in the Part II sample
b
Impulse control disorders not assessed
c
Cell size was too small to be included in analysis
d
Projected lifetime risk to age 65 due to the sample including only respondents up to age 65
IMP. 168-176 24-09-2007 16:11 Pagina 172
173
interpreted with caution, because it is likely over-estimat-
ed due to between-country differences in some of the biases
enumerated above. The overall prevalence-risk estimates,
which are consistent with previous cross-national research
(8-14,39), are likely to be conservative, as the most plausi-
ble biases lead to under-estimation. The evidence for co-
hort effects is more difficult to judge, as both substantive
and methodological interpretations are plausible. The op-
tions are either that the prevalence of mental disorders is
on the rise or that prevalence is stable but under-estimated
among older respondents.
Given the high prevalence-risk estimates even with the
possibility of conservative bias, a question can be raised
about the meaningfulness of these estimates. Our clinical
reappraisal studies, consistent with comparable studies
carried out in conjunction with previous community psy-
chiatric epidemiological surveys (43), show that the high
prevalence estimates are genuine (i.e., consistent with ex-
pert clinician judgments) rather than due to CIDI errors. It
is important to recognize, though, that not all mental dis-
orders are severe. WMH measures of disorder severity were
applied only to 12-month cases, so we have no way to es-
timate severity of lifetime cases. Analysis of 12-month cases,
though, finds the majority rated mild on a clinical rating
scale with categories mild, moderate, and severe (22).
These cases are nonetheless meaningful, because even mild
cases can be impairing and often evolve into more serious
disorders over time (44).
Table 3 Inter-cohort differences in lifetime risk of any DSM-IV anxiety disorder
a
Country 18-34 35-49 50-64 65+
b
χχ
2
df N
OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI N
Belgium 2.6* 1.3-5.011254 1.6* 0.8-3.2 1331 1.3* 0.6-2.6 1278 1.0 - 180 114.2* 3 1043
Colombia 1.6* 1.2-2.11 1125 1.3* 0.9-1.8 1818 1.0* - 1438 - - - 110.0* 2 2381
France 3.1* 1.5-6.411388 3.2* 1.5-6.7 1472 1.6* 0.8-3.3 1362 1.0 - 214 121.3* 3 1436
Germany 3.1* 1.9-5.111316 2.3* 1.4-3.9 1436 2.3* 1.3-4.1 1345 1.0 - 226 121.8* 3 1323
Israel 4.7* 2.6-8.31 1627 2.7* 1.6-4.4 1302 2.1* 1.4-3.3 1069 1.0 - 861 127.3* 3 4859
Italy 1.5* 0.7-3.011496 1.6* 0.9-2.8 1516 1.3* 0.8-2.2 1454 1.0 - 313 113.3* 3 1779
Japan 5.6* 2.2-13.8 1155 2.8* 1.3-6.1 1219 2.6* 1.2-5.6 1295 1.0 - 218 114.9* 3 1887
Lebanon 3.2* 1.6-6.211349 2.5* 1.2-5.1 1348 1.0* 0.5-2.1 1199 1.0 - 135 124.1* 3 1031
Mexico 2.4* 1.6-3.41 1183 1.6* 1.1-2.4 1750 1.0* - 1429 - - - 1
25.3* 2 2362
Netherlands 3.6* 2.1-6.111264 4.5* 3.0-6.8 1358 3.0* 2.0-4.6 1302 1.0 - 170 160.6* 3 1094
New Zealand 3.4* 2.7-4.21 2394 2.6* 2.1-3.1 2474 2.1* 1.7-2.7 1517 1.0 - 927 126.3* 3 7312
Nigeria 3.1* 1.4-6.911971 2.3* 1.1-4.9 1549 2.8* 1.5-5.4 1369 1.0 - 254 111.1* 3 2143
PR China 1.7* 0.6-4.411379 1.1* 0.5-2.5 1726 1.6* 0.7-3.9 1357 1.0 - 166 113.3* 3 1628
South Africa 2.3* 1.3-4.01 2172 1.8* 1.1-3.1 1264 1.3* 0.8-2.1 1638 1.0 - 241 116.5* 3 4315
Spain 3.8* 2.2-6.511545 2.8* 1.5-5.2 1556 1.3* 0.8-2.2 1456 1.0 - 564 128.7* 3 2121
Ukraine 1.7* 1.1-2.611420 1.0* 0.6-1.6 1434 1.0* 0.7-1.6 1412 1.0 - 454 116.5* 3 1720
United States 3.5* 2.8-4.41 1939 3.4* 2.7-4.1 1831 2.5* 2.0-3.0 1213 1.0 - 709 159.2* 3 5692
a
Based on discrete-time survival models with person-year as the unit of analysis, controls are time intervals
b
Referent category
*Significant at the .05 level, two-sided test
Table 4 Inter-cohort differences in lifetime risk of any DSM-IV mood disorder
a
Country 18-34 35-49 50-64 65+
b
χχ
2
df N
OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI N
Belgium 11.3* 16.1-20.9 1573 4.9* 3.2-7.511775 3.6* 2.0-6.4 1570 1.0 - 1501 187.3* 3 12419
Colombia 16.3* 14.2-9.31 2000 2.3* 1.6-3.11 1577 1.0* - 1849 - - 1530 192.7* 2 14426
France 19.0* 16.0-13.5 1743 3.0* 2.2-4.211942 1.8* 1.2-2.6 1719 1.0 - 1490 146.4* 3 12894
Germany 12.2* 17.1-21.0 1815 5.2* 3.5-7.71 1180 2.4* 1.6-3.4 1893 1.0 - 1667 194.4* 3 13555
Israel 16.5* 14.5-9.41 1627 2.8* 2.0-4.01 1302 1.8* 1.3-2.5 1069 1.0 - 1861 118.4* 3 14859
Italy 15.7* 13.8-8.41 1326 3.6* 2.6-5.01 1393 2.3* 1.6-3.3 1153 1.0 - 1840 191.3* 3 14712
Japan 23.7* 13.4-42.0 1410 7.7* 4.5-13.2 1571 3.8* 2.4-5.8 1764 1.0 - 1
691 146.2* 3 12436
Lebanon 16.2* 13.0-12.8 1965 3.1* 1.4-6.711931 1.7* 0.8-3.2 1553 1.0 - 1408 160.5* 3 12857
Mexico 14.0* 12.6-6.11 2871 1.6* 1.1-2.31 1888 1.0* - 1023 - - 1646 165.0* 2 15782
Netherlands 11.7* 16.6-20.8 1564 6.4* 4.0-10.2 1729 2.9* 1.7-4.8 1627 1.0 - 1452 115.7* 3 12372
New Zealand 10.0* 18.2-12.2 3747 5.0* 4.1-6.01 4102 2.9* 2.4-3.6 2697 1.0 - 2244 653.9* 3 12790
Nigeria 13.7* 11.8-7.61 3175 1.8* 0.9-3.61 1631 1.2* 0.7-2.1 1104 1.0 - 1842 119.4* 3 16752
PR China 20.8* 19.4-45.8 1209 4.4* 2.3-8.41 2261 2.5* 1.4-4.4 1184 1.0 - 1547 176.5* 3 15201
South Africa 19.6* 15.5-16.7 2172 5.5* 3.1-9.91 1264 2.5* 1.4-4.4 1638 1.0 - 1241 195.6*
3 14315
Spain 19.6* 16.6-13.9 1567 4.2* 3.0-5.91 1431 2.2* 1.6-3.0 1024 1.0 - 1451 176.3* 3 15473
Ukraine 11.9* 11.4-2.41 1194 1.0* 0.8-1.31 1225 0.9* 0.8-1.1 1180 1.0 - 1126 138.2* 3 14725
United States 19.5* 17.3-12.4 3034 5.0* 3.7-6.61 2865 3.0* 2.3-3.9 1922 1.0 - 1461 383.6* 3 19282
a
Based on discrete-time survival models with person-year as the unit of analysis, controls are time intervals
b
Referent category
*Significant at the .05 level, two-sided test
IMP. 168-176 24-09-2007 16:11 Pagina 173
174
WWoorrlldd PPssyycchhiiaattrryy 66::33 --
October 2007
The age-of-onset distributions reported here are consis-
tent with those in previous epidemiological surveys (39,45).
Given the enormous personal and societal burdens of men-
tal disorders, the finding that many cases have early ages of
onset suggests that public health interventions might prof-
itably begin in childhood. Importantly, studies of initial
contact with the treatment system (46-48) show that people
with these early-onset disorders often wait more than a
decade before seeking treatment, and present with serious-
ly impairing disorders that might have been easier to treat if
they had sought treatment earlier in the course of illness. In-
terventions aimed at early detection and treatment might
help reduce the persistence or severity of these largely pri-
mary anxiety and impulse control disorders and prevent the
onset of secondary disorders. More preclinical and clinical
research is needed on treatments of early cases, though, to
determine whether this is true. Epidemiological research is
also needed on the long-term consequences of early inter-
ventions for long-term secondary prevention.
Acknowledgements
The surveys discussed in this article were carried out in
conjunction with the World Health Organization’s World
Mental Health (WMH) Survey Initiative. We thank the
WMH staff for assistance with instrumentation, fieldwork,
and data analysis. These activities were supported by
the United States National Institute of Mental Health
(R01-MH070884), the John D. and Catherine T. MacArthur
Foundation, the Pfizer Foundation, the US Public Health
Service (R13-MH066849, R01-MH069864, and R01-DA016
558), the Fogarty International Center (FIRCA R01-
TW006481), the Pan American Health Organization, Eli
Lilly and Company, Ortho-McNeil Pharmaceutical, Inc.,
GlaxoSmithKline, and Bristol-Myers Squibb. The Chinese
World Mental Health Survey Initiative is supported by the
Pfizer Foundation. The Colombian National Study of
Mental Health (NSMH) is supported by the Ministry of So-
cial Protection, with supplemental support from the Sal-
darriaga Concha Foundation. The ESEMeD project is
funded by the European Commission (Contracts QLG5-
1999-01042; SANCO 2004123), the Piedmont Region
(Italy), Fondo de Investigación Sanitaria, Instituto de Salud
Carlos III, Spain (FIS 00/0028), Ministerio de Ciencia y
Tecnología, Spain (SAF 2000-158-CE), Departament de
Salut, Generalitat de Catalunya, Spain, and other local
agencies, and by an unrestricted educational grant from
GlaxoSmithKline. The Israel National Health Survey is
funded by the Ministry of Health, with support from the Is-
rael National Institute for Health Policy and Health Ser-
vices Research and the National Insurance Institute of Is-
rael. The World Mental Health Japan (WMHJ) Survey is
supported by the Grant for Research on Psychiatric and
Neurological Diseases and Mental Health (H13-SHOGAI-
023, H14-TOKUBETSU-026, H16-KOKORO-013) from
the Japan Ministry of Health, Labour and Welfare. The
Lebanese National Mental Health Survey (LEBANON) is
supported by the Lebanese Ministry of Public Health,
the WHO (Lebanon), anonymous private donations to
IDRAAC, Lebanon, and unrestricted grants from Janssen
Cilag, Eli Lilly, GlaxoSmithKline, Roche, and Novartis.
The Mexican National Comorbidity Survey (MNCS) is
supported by the National Institute of Psychiatry Ramon
de la Fuente (INPRFMDIES 4280) and by the National
Council on Science and Technology (CONACyT-G30544-
H), with supplemental support from the Pan American
Health Organization. Te Rau Hinengaro: The New Zealand
Table 5 Inter-cohort differences in lifetime risk of any DSM-IV substance use disorder
a
Country 18-34 35-49 50-64 65+
b
χχ
2
df N
OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI N
Belgium 15.0* 2.6-9.811254 3.6* 1.7-7.311331 2.6* 1.2-5.4 1278 1.0 - 1180 126.7* 3 11043
Colombia 12.3* 1.6-3.31 2000 1.11 0.7-1.61 1577 1.01 - 1849 - - 1530 139.3* 2 14426
France 15.8* 3.3-10.0 1388 3.3* 2.0-5.711472 2.5* 1.4-4.2 1362 1.0 - 1214 144.1* 3 11436
Germany 15.6* 2.9-10.7 1316 3.7* 2.0-6.811436 3.9* 2.1-7.1 1345 1.0 - 1226 135.0* 3 11323
Israel 11.3* 5.9-21.6 1627 4.6* 2.4-9.01 1302 2.5* 1.2-5.1 1069 1.0 - 1861 119.9* 3 14859
Italy 12.6* 1.0-6.711496 1.81 0.8-4.111516 1.61 0.6-3.9 1454 1.0 - 1313 115.51 3 11779
Japan 11.91 0.6-6.011155 2.3* 1.1-4.911
219 2.5* 1.1-5.7 1295 1.0 - 1218 116.71 3 11887
Lebanon
c
-- -- -- -- ---
Mexico 11.7* 1.3-2.41 2871 1.21 0.9-1.71 1888 1.01 - 1023 - - 1646 112.8* 2 15782
Netherlands 12.4* 7.0-21.8 1264 7.0* 3.8-13.1 1358 6.8* 3.4-13.9 1302 1.0 - 1170 185.3* 3 11094
New Zealand 18.1* 6.1-10.7 3747 3.5* 2.7-4.71 4102 2.5* 1.9-3.3 2697 1.0 - 2244 283.7* 3 12790
Nigeria 13.4* 1.1-10.1 1971 4.9* 1.8-13.3 1549 2.91 1.0-8.7 1369 1.0 - 1254 111.8* 3 12143
PR China 18.2* 1.0-67.2 1379 4.01 0.6-28.2 1726 1.51 0.2-11.2 1357 1.0 - 1166 131.9* 3 11628
South Africa 12.6* 1.3-5.41 2172 1.51 0.8-2.91 1264 1.01 0.6-1.9 1638 1.0 - 1241 1
29.11 3 14315
Spain 19.3* 3.6-24.2 1545 5.0* 1.8-13.7 1556 1.51 0.6-4.2 1456 1.0 - 1564 138.1* 3 12121
Ukraine 10.8* 5.8-20.1 1420 5.0* 2.4-10.4 1434 2.8* 1.3-5.8 1412 1.0 - 1454 116.4* 3 11720
United States 16.7* 4.6-10.0 1939 4.9* 3.5-7.01 1831 3.5* 2.4-5.3 1213 1.0 - 1709 111.0* 3 15692
a
Based on discrete-time survival models with person-year as the unit of analysis, controls are time intervals
b
Referent category
c
Cell size too small to be included in analysis
*Significant at the .05 level, two-sided test
IMP. 168-176 24-09-2007 16:11 Pagina 174
175
Mental Health Survey (NZMHS) is supported by the New
Zealand Ministry of Health, Alcohol Advisory Council,
and the Health Research Council. The Nigerian Survey of
Mental Health and Wellbeing (NSMHW) is supported by
the World Health Organization (Geneva), the World
Health Organization (Nigeria), and the Federal Ministry of
Health, Abuja, Nigeria. The South Africa and Health Study
(SASH) is supported by the US National Institute of Men-
tal Health (R01-MH059575) and National Institute of
Drug Abuse, with supplemental funding from the South
African Department of Health and the University of Michi-
gan. The Ukraine Comorbid Mental Disorders during Pe-
riods of Social Disruption (CMDPSD) study is funded by
the US National Institute of Mental Health (R01-
MH61905). The US National Comorbidity Survey Replica-
tion (NCS-R) is supported by the National Institute of
Mental Health (U01-MH60220), with supplemental sup-
port from the National Institute of Drug Abuse, the Sub-
stance Abuse and Mental Health Services Administration
(SAMHSA), the Robert Wood Johnson Foundation (Grant
044780), and the John W. Alden Trust.
References
1. Cooper B. Psychiatric epidemiology. London: Croom Helm, 1987.
2. Robins LN, Helzer JE, Croughan JL et al. National Institute of
Mental Health Diagnostic Interview Schedule: its history, charac-
teristics and validity. Arch Gen Psychiatry 1981;38:381-9.
3. Cross-National Collaborative Group. The changing rate of major
depression: cross-national comparisons. JAMA 1992;268:3098-
105.
4. Weissman MM, Bland RC, Canino GJ et al. The cross-national
epidemiology of panic disorder. Arch Gen Psychiatry 1997;54:305-
9.
5. Weissman MM, Bland RC, Canino GJ et al. Cross-national epi-
demiology of major depression and bipolar disorder. JAMA 1996;
276:293-9.
6. Wittchen HU. Reliability and validity studies of the WHO Com-
posite International Diagnostic Interview (CIDI): a critical review.
J Psychiatr Res 1994;28:57-84.
7. World Health Organization. Composite International Diagnostic
Interview (CIDI, Version 1.0). Geneva: World Health Organiza-
tion, 1990.
8. Andrade L. Lifetime prevalence of mental disorders in a catch-
ment area in São Paulo, Brazil. Presented at the 7th Congress of
the International Federation of Psychiatric Epidemiology, Santia-
go, August 1996.
9. Bijl RV, van Zessen G, Ravelli A et al. The Netherlands Mental
Health Survey and Incidence Study (NEMESIS): objectives and
design. Soc Psychiatry Psychiatr Epidemiol 1998;33:581-6.
10. Caraveo J, Martinez J, Rivera B. A model for epidemiological stud-
ies on mental health and psychiatric morbidity. Salud Mental
1998;21:48-57.
11. Kessler RC, McGonagle KA, Zhao S et al. Lifetime and 12-month
prevalence of DSM-III-R psychiatric disorders in the United
States. Results from the National Comorbidity Survey. Arch Gen
Psychiatry 1994;51:8-19.
12. Kylyc C. Mental health profile of Turkey: main report. Ankara:
Ministry of Health Publications, 1998.
13. Vega WA, Kolody B, Aguilar-Gaxiola S et al. Lifetime prevalence
of DSM-III-R psychiatric disorders among urban and rural Mexi-
can Americans in California. Arch Gen Psychiatry 1998;55:771-8.
14. Wittchen H-U, Perkonigg A, Lachner G et al. Early Developmen-
tal Stages of Psychopathology study (EDSP): objectives and de-
sign. Eur Addict Res 1998;4:18-27.
15. Murray CJL, Lopez AD. The global burden of disease: a compre-
hensive assessment of mortality and disability from diseases, in-
juries and risk factors in 1990 and projected to 2020. Cambridge:
Harvard University Press, 1996.
16. Bijl RV, de Graaf R, Hiripi E et al. The prevalence of treated and
untreated mental disorders in five countries. Health Aff 2003;22:
122-33.
17. Amminger GP, Leicester S, Yung AR et al. Early-onset of symp-
toms predicts conversion to non-affective psychosis in ultra-high
risk individuals. Schizophr Res 2006;84:67-76.
18. McGue M, Iacono WG. The association of early adolescent prob-
lem behavior with adult psychopathology. Am J Psychiatry 2005;
162:1118-24.
19. Thompson KN, Conus PO, Ward JL et al. The initial prodrome to
bipolar affective disorder: prospective case studies. J Affect Disord
2003;77:79-85.
20. Alonso J, Angermeyer MC, Bernert S et al. Sampling and methods
of the European Study of the Epidemiology of Mental Disorders
(ESEMeD) project. Acta Psychiatr Scand 2004;420(Suppl.):8-20.
21. World Bank. World development indicators 2003. Washington:
World Bank, 2003.
22. Demyttenaere K, Bruffaerts R, Posada-Villa J et al. Prevalence,
severity, and unmet need for treatment of mental disorders in the
World Health Organization World Mental Health Surveys. JAMA
2004;291:2581-90.
23. Kessler RC, Merikangas KR. The National Comorbidity Survey
Replication (NCS-R): background and aims. Int J Methods Psy-
chiatr Res 2004;13:60-8.
24. Kessler RC, Ustun TB. The World Mental Health (WMH) Survey
Initiative Version of the World Health Organization (WHO) Com-
posite International Diagnostic Interview (CIDI). Int J Methods
Psychiatr Res 2004;13:93-121.
25. World Health Organization. International classification of dis-
eases (ICD-10). Geneva: World Health Organization, 1991.
26. American Psychiatric Association. Diagnostic and statistical man-
ual of mental disorders, 4th ed. Washington: American Psychiatric
Association, 1994.
27. Haro JM, Arbabzadeh-Bouchez S, Brugha TS et al. Concordance
of the Composite International Diagnostic Interview Version 3.0
(CIDI 3.0) with standardized clinical assessments in the WHO
World Mental Health Surveys. Int J Methods Psychiatr Res 2006;
15:167-80.
28. First MB, Spitzer RL, Gibbon M et al. Structured Clinical Inter-
view for DSM-IV Axis I Disorders, research version, non-patient
edition (SCID-I/NP). New York: Biometrics Research, New York
State Psychiatric Institute, 2002.
29. Simon GE, VonKorff M. Recall of psychiatric history in cross-sec-
tional surveys: implications for epidemiologic research. Epidemi-
ol Rev 1995;17:221-7.
30. Knauper B, Cannell CF, Schwarz N et al. Improving the accuracy
of major depression age of onset reports in the US National Co-
morbidity Survey. Int J Methods Psychiatr Res 1999;8:39-48.
31. SAS Institute. SAS/STAT software: changes and enhancements,
Release 8.2. Cary: SAS Publishing, 2001.
32. Efron B. Logistic regression, survival analysis, and the Kaplan-
Meier curve. J Am Stat Assoc 1988;83:414-25.
33. Wolter KM. Introduction to variance estimation. New York:
Springer, 1985.
34. Research Triangle Institute. SUDAAN: Professional Software for
Survey Data Analysis. Research Triangle Park: Research Triangle
Institute, 2002.
35. Kish L, Frankel MR. Inferences from complex samples. J Royal
Stat Soc 1974;36:1-37.
IMP. 168-176 24-09-2007 16:11 Pagina 175
176
WWoorrlldd PPssyycchhiiaattrryy 66::33 --
October 2007
36. Gureje O, Lasebikan VO, Kola L et al. Lifetime and 12-month
prevalence of mental disorders in the Nigerian Survey of Mental
Health and Well-Being. Br J Psychiatry 2006;188:465-71.
37. Shen YC, Zhang MY, Huang YQ et al. Twelve-month prevalence,
severity, and unmet need for treatment of mental disorders in met-
ropolitan China. Psychol Med 2006;36:257-67.
38. Hasin DS, Grant BF. The co-occurrence of DSM-IV alcohol abuse
in DSM-IV alcohol dependence: results of the National Epidemi-
ologic Survey on Alcohol and Related Conditions on heterogene-
ity that differ by population subgroup. Arch Gen Psychiatry 2004;
61:891-6.
39. WHO International Consortium in Psychiatric Epidemiology. Cross-
national comparisons of the prevalences and correlates of mental
disorders. Bull World Health Organ 2000;78:413-26.
40. Allgulander C. Psychoactive drug use in a general population sam-
ple, Sweden: correlates with perceived health, psychiatric diag-
noses, and mortality in an automated record-linkage study. Am J
Publ Health 1989;79:1006-10.
41. Cannell CF, Marquis KH, Laurent A. A summary of studies of in-
terviewing methodology: 1959-1970. Vital Health Stat 1977;2:69.
42. Giuffra LA, Risch N. Diminished recall and the cohort effect of ma-
jor depression: a simulation study. Psychol Med 1994;24:375-83.
43. Kessler RC, Wittchen H-U, Abelson JM et al. Methodological
studies of the Composite International Diagnostic Interview (CI-
DI) in the US National Comorbidity Survey. Int J Methods Psy-
chiatr Res 1998;7:33-55.
44. Kessler RC, Merikangas KR, Berglund P et al. Mild disorders
should not be eliminated from the DSM-V. Arch Gen Psychiatry
2003;60:1117-22.
45. Christie KA, Burke JDJ, Regier DA et al. Epidemiologic evidence
for early onset of mental disorders and higher risk of drug-abuse
in young-adults. Am J Psychiatry 1988;145:971-5.
46. Christiana JM, Gilman SE, Guardino M et al. Duration between
onset and time of obtaining initial treatment among people with
anxiety and mood disorders: an international survey of members
of mental health patient advocate groups. Psychol Med 2000;
30:693-703.
47. Olfson M, Kessler RC, Berglund PA et al. Psychiatric disorder on-
set and first treatment contact in the United States and Ontario.
Am J Psychiatry 1998;155:1415-22.
48. Wang PS, Angermeyer M, Borges G et al. Delay and failure in
treatment seeking after first onset of mental disorders in the World
Health Organization’s World Mental Health Survey Initiative.
World Psychiatry 2007;6:177-85.
IMP. 168-176 24-09-2007 16:11 Pagina 176
