Age of onset of mental disorders: a review of recent literature

Ronald C. Kesslera, G. Paul Ammingerb, Sergio Aguilar-Gaxiolac, Jordi Alonsod,

Sing Leee and T. Bedirhan Üstünf

Purpose of review Abbreviations

The aim of this article is to review recent epidemiological AOO age of onset
research on age-of-onset of mental disorders, focusing on CIDI Composite Diagnostic Interview
IQR inter-quartile range
the WHO World Mental Health surveys. NAP nonaffective psychosis
Recent findings SAD separation anxiety disorder
SSD schizophrenia spectrum diagnoses
Median and inter-quartile range (IQR; 25th–75th WMH World Mental Health
percentiles) of age-of-onset is much earlier for phobias
ß 2007 Lippincott Williams & Wilkins
(7–14, IQR 4–20) and impulse–control disorders (7–15; 0951-7367
IQR 4–35) than other anxiety disorders (25–53,
IQR 15–75), mood disorders (25–45, IQR 17–65), and
substance disorders (18–29, IQR 16–43). Although less Introduction
data exist for nonaffective psychosis, available evidence The purpose of this paper is to review recent evidence
suggests that median age-of-onset is in the range late teens from epidemiological surveys on the age of onset (AOO)
through early 20s. Roughly half of all lifetime mental distributions of commonly occurring mental disorders.
disorders in most studies start by the mid-teens and three Although AOO is one of the least commonly studied
quarters by the mid-20s. Later onsets are mostly secondary aspects of descriptive epidemiology, it is important
conditions. Severe disorders are typically preceded by less for reasons described below. The recent publication
severe disorders that are seldom brought to clinical of comprehensive AOO results from the WHO’s World
attention. Mental Health (WMH) Surveys provides unprecedented
Summary data on the AOO distributions of many commonly occur-
First onset of mental disorders usually occur in childhood or ring mental disorders. These data have several important
adolescence, although treatment typically does not occur implications for clinical practice and research that are
until a number of years later. Although interventions discussed in the second half of the review.
with early incipient disorders might help reduce
severity-persistence of primary disorders and prevent Practical difficulties in studying age of onset
secondary disorders, additional research is needed on The dearth of information on AOO of mental disorders is
appropriate treatments for early incipient cases and on presumably due to reluctance on the part of epidemiol-
long-term evaluation of the effects of early intervention ogists to rely on the retrospective reports obtained in
on secondary prevention. general population surveys that, as a practical matter,
must be used to generate the survival distributions
Keywords needed to study AOO. Two theoretical alternatives exist
age of onset, early intervention, mental disorders, to relying on these retrospective reports, although neither
prevention, WHO world mental health survey initiative of the two is broadly feasible. The first applies largely
to psychosis: to use information obtained about total-
Curr Opin Psychiatry 20:359–364. ß 2007 Lippincott Williams & Wilkins. population incidence of treated disorder from studies of
a
psychotic disorders in catchment areas that monitor all
Department of Healthcare Policy, Harvard Medical School, Boston, USA, bThe
ORYGEN Research Center, Department of Psychiatry, University of Melbourne,
points of contact with the treatment system [1,2]. The
Australia, cCenter for Reducing Health Disparities, Department of Internal implicit assumption in this approach, that the vast
Medicine, School of Medicine, University of California-Davis, Sacramento, USA,
d
Health Services Research Unit, Institut Municipal d’Investigacio Medica-Institut
majority of psychotic patients eventually come to clinical
Municipal d’Assistencia Sanitaria (IMIM-IMAS), Barcelona, Spain, eDepartment of attention, is likely to be true, making this approach useful
Psychiatry, The Chinese University of Hong Kong, Hong Kong SAR, China and
f
Global Programme on Evidence for Health Policy, World Health Organization,
for studying AOO of nonaffective psychosis. The same
Geneva, Switzerland approach would not be nearly as useful, though, for less
Correspondence to R.C. Kessler, PhD, Department of Healthcare Policy, Harvard severely impairing disorders, as many people with such
Medical School, 180 Longwood Avenue, Boston, MA 02115, USA disorders never come to clinical attention. Even for
Tel: +1 617 432 3587; fax: +1 617 432 3588;
e-mail: [email protected] nonaffective psychosis, epidemiological surveys show
that the time between onset of the first episode and first
Current Opinion in Psychiatry 2007, 20:359–364
contact with the treatment system is sometimes quite
long [3], although analysis of incident treated cases is still
the best available approach due to the substantial
359

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
360 Services research and outcomes

underrepresentation of psychosis in community epidemio- population surveys in 28 countries [17]. The main
logical surveys [4]. aim is to provide estimates of the prevalence, distri-
bution, societal burden, and patterns of treatment of
The second theoretical alternative to relying on retro- mental disorders to health policy makers for planning
spective AOO reports is to carry out large long-term purposes. The WMH interview schedule and all other
prospective surveys to estimate incidence directly. study materials were translated using standardized
Numerous studies of this sort exist in such well funded WHO protocols. Consistent interviewer training and
fields of cancer epidemiology [5] and cardiovascular quality control procedures were used in all surveys.
epidemiology [6]. Comparable studies of mental dis- Informed consent was obtained in each country using
orders do not exist. Although some large long-term procedures approved by the Institutional Review
community epidemiological studies have included some Boards of the organizations coordinating the survey in
information on mental disorders (e.g., website http:// the country.
www.nshd.mrc.ac.uk), the data collection waves are
not sufficiently frequent and the assessment mental The WHO regions and 16 countries that have completed
disorder interval incidence not sufficiently central to and published WMH results up to now are Africa
generate accurate prospective estimates of incidence. (Nigeria, South Africa), the Americas (Colombia, Mexico,
Prospective birth cohort studies in schizophrenia that United States), Asia and the Pacific (Japan, New Zealand,
cover the whole age range of risk for the onset of Beijing and Shanghai in the People’s Republic of
the disorder are sparse [7,8]. As a result, we have to China), Europe (Belgium, France, Germany, Italy, The
rely largely on retrospective reports in cross-sectional Netherlands, Spain, Ukraine); and the Middle East
community surveys to estimate AOO distributions of (Israel, Lebanon). The surveys in most of these countries
commonly occurring mental disorders. were nationally representative (the exceptions being
China, Japan, and Nigeria, which were regionally repre-
Why study age of onset? sentative). Seven of the countries are classified by
An examination of AOO distributions is important for at the World Bank as less developed (China, Colombia,
least two reasons. One is that information on AOO Lebanon, Mexico, Nigeria, South Africa, Ukraine), while
allows us to distinguish between lifetime prevalence the others are classified as developed [18]. All surveys
(the proportion of the population who had a disorder at were conducted face-to-face by trained lay interviewers
some time in their life up to their age at interview) and in multistage household probability samples. A total of
projected lifetime risk (the estimated proportion of the 85 052 interviews were completed in these 16 countries,
population who will have the disorder by the end of their with samples ranging from 2372 in the Netherlands to
life). Lifetime risk cannot be estimated directly from 12 992 in New Zealand. The weighted average response
community surveys because respondents differ in age rate was 71.1%. More details on WMH samples, designs,
and, therefore, number of years at risk. Projections of and field procedures are presented at the website http://
estimated future risk can be made from AOO distri- www.hcp.med.harvard.edu/wmh.
butions, though, using either the Kaplan–Meier [9]
method or the slightly more precise actuarial method The WMH diagnostic interview was the WHO Compo-
[10] to estimate survival distributions. Second, an site Diagnostic Interview (CIDI) Version 3.0 [19], a fully-
understanding of AOO is important for targeting structured lay-administered interview that generates
research on prevention of mental disorders [11], early both ICD-10 and DSM-IV diagnoses. DSM-IV diagnoses
intervention with prodromal or incipient mental were used in the analyses reviewed here. Included were
disorders [12], and primary prevention of secondary anxiety disorders (panic disorder, agoraphobia without
disorders [13]. In the absence of AOO information, panic disorder, specific phobia, social phobia, generalized
we would have no way of knowing the appropriate anxiety disorder, posttraumatic stress disorder, separation
age range to target preventive interventions. A related anxiety disorder), mood disorders (major depressive
issue is that early AOO is often found to be associated disorder, dysthymic disorder, bipolar disorder I and II,
with greater disorder severity [14], persistence [15], sub-threshold bipolar disorder), impulse-control dis-
and lack of treatment response [16]. Based on these orders (intermittent explosive disorder, oppositional-
associations, AOO information can be useful in making defiant disorder, attention-deficit/hyperactivity disorder,
projections of aggregate illness course associated with conduct disorder), and substance use disorders (alcohol
primary and secondary disorders. and drug abuse with or without dependence). Not all
disorders were assessed in all countries. Nonaffective
World Mental Health survey methods psychosis (NAP) was not assessed because the CIDI
As the main focus of the following review is on the WMH assessment of NAP was not sufficiently reliable and
surveys, we present a brief overview of WMH methods. valid [20]. Clinical calibration studies [21] found that
The WMH Survey Initiative consists of coordinated the CIDI assessed other disorders with good concordance

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Age of onset of mental disorders Kessler et al. 361

to blinded clinical reappraisal interviews using the Struc- consistently low prevalence until the early teens followed
tured Clinical Interview for DSM-IV (SCID) [22]. by a roughly linear increase through late middle age and a
declining increase thereafter. The median AOO of mood
Research carried out in conjunction with the landmark disorders has a very wide range across countries (25–45)
Epidemiological Catchment Area (ECA) [23] study and an even wider IQR (17–65).
found that retrospective AOO reports often generate
implausible response patterns, indicating the existence The AOO distributions of substance use disorders,
of retrospective recall bias [24]. Special procedures devel- finally, are quite consistent across the WMH countries
oped by survey methodologists to minimize this bias were in that few onsets occur prior to the mid teens and
used in the WMH surveys both to decrease recall failure cumulative increase in onset is rapid in adolescence
involving the lifetime occurrence of disorders [19] and to and early adulthood. Considerable cross-national vari-
decrease imprecision in dating the AOO of reported ation exists, though, in the sharpness of the change in
disorders [25]. Experimental research has shown that this the slope as well as in the age range of this change. This
question sequence yields responses with a much more cross-national variation leads to wider cross-national vari-
plausible AOO distribution than standard AOO questions ation in both the median (18–29) and the IQR (16–43) of
[25]. the AOO distributions than for impulse-control disorders
or phobias or SAD, but lower variation than for mood
Age of onset distributions of commonly disorders or other anxiety disorders.
occurring disorders
Disorder-specific estimates of AOO distributions have No strong consistency in between-country differences
been published for WMH surveys in Italy [26], metro- in AOO distributions was detected across disorders.
politan China [27], Mexico [28], New Zealand [29], Furthermore, between-country differences in these
Nigeria [30], Spain [31], Ukraine [32], and the USA distributions were found to be unrelated to economic
[33]. In addition, a comparative analysis of these distri- development, to region of the world, or to other structural
butions across 16 WMH countries has been completed correlates.
[34]. Systematic comparison of these results documents a
number of clear cross-national consistencies both within The age of onset distribution of treated
and between disorders. psychosis
Psychotic disorders rarely occur before age 14 years, but
The impulse-control disorders have the earliest AOO show a marked increase in prevalence between ages 15
distributions, with median AOO across countries of and 17 [35]. Schizophrenia spectrum diagnoses (SSDs)
7–9 years for attention-deficit/hyperactivity disorder account for approximately two-thirds of all psychotic
(ADHD), 7–15 for oppositional-defiant disorder (ODD), disorders. Schizophrenia usually begins in the age range
9–14 for conduct disorder, and 13–21 for intermittent 15–35. Disorder-specific estimates of AOO distributions
explosive disorder (IED). Impulse-control disorders for nonaffective and affective psychotic disorders have
also have an extremely narrow age range of onset risk. not been separately reported in any of the WMH surveys
For example, 80% of all lifetime ADHD begins in the age due to the underrepresentation of these cases in surveys.
range 4–11, while the vast majority of ODD and conduct
disorder begins between ages 5 and 15 years. Fully half As noted above, studies that either establish the treated
of all lifetime IED begins in childhood or adolescence. incidence of psychosis in a well defined catchment area or
that observe onsets in long-term prospective general
Some anxiety disorders – the phobias and separation population cohorts are preferred. An integrated com-
anxiety disorder (SAD) – also have very early AOO distri- munity-based treatment system is needed to obtain
butions, with median AOO in the range 7–14 and inter- accurate data on treated incidence. A good example of
quartile range (IQR; 25th–75th percentiles of the AOO such a system can be found at the Early Psychosis
distributions) of 4–20. The other anxiety disorders (panic Prevention and Intervention Centre (EPPIC) in Mel-
disorder, generalized anxiety disorder, and posttraumatic bourne, Australia. This is a community-based specialized
stress disorder), in comparison, have considerably later service mandated to treat all people between 15 and
AOO distributions, although the cross-national variation 29 years who present with a first psychotic episode to
in both median AOO (25–53) and in inter-quartile range public mental health services in a geographically defined
(IQR) AOO (15–75) is considerably wider than for the catchment area. Between 1997 and 2000, 1019 individuals
impulse-control disorders or the phobias or SAD. were registered at EPPIC with a first episode of psycho-
sis. The median age of initial presentation in this cohort
The mood disorder AOO distributions in the WMH was 22 with an IQR of 19–25. The same median and
surveys are quite similar to those for the later-onset IQR existed for patients with SSD (69%) and those with
anxiety disorders. Mood disorder AOO curves show non-SSD (31%).

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
362 Services research and outcomes

Similar results were found in the 1966 North Finland might have been recalled incorrectly as well [24] despite
Birth Cohort, where median AOO and IQR for schizo- the WMH surveys using a novel probing strategy to
phrenia were 23 and 19–27 [8], and in the British 1946 reduce bias in AOO reports. Estimates of psychosis
birth cohort (median AOO for schizophrenia 22) [7]. AOO were limited by their being based on incident
The majority of psychoses manifest in the third life treatment. The AOO results are further limited by focus-
decade with a median in the early 20s and a narrow ing on syndrome onset, ignoring any prodrome at an
IQR more similar to impulse-control disorders and some earlier age. A number of at-risk mental states are known
anxiety disorders than mood disorders. to occur prior to the onset of many cases of NAP [39] and
bipolar disorder [40]. A number of problem behaviors
Projected lifetime risk in comparison to indicative of impulsivity are known to occur prior to
lifetime-prevalence-to-date the onset of many cases of impulse-control disorders
As noted in the introduction, one important reason for [41]. Childhood behavioral inhibition is known to occur
estimating AOO distributions is to obtain data on pro- prior to the onset of many cases of panic disorder and
jected lifetime risk. It is noteworthy that the estimates of depression [42]. Epidemiological analysis of these early
projected lifetime risk of any DSM-IV disorder in indicators of incipient disorder would almost certainly
the WMH surveys was roughly one-third higher (IQR lead to much earlier estimates of AOO than those
28–44%) than estimated lifetime prevalence-to-date. reported here.
This means that three to four people in the populations
of these countries are likely to develop a first mental Implications
disorder at some time in the future for every 10 Within the context of these limitations, the AOO distri-
people who already had a disorder. The highest risk- butions reported here are consistent with those in
to-prevalence ratios (57–69%) were found in countries previous epidemiological surveys [43,44]. We know of
exposed to sectarian violence (Israel, Nigeria, South no research prior to the WMH surveys that examined
Africa). Excluding these three, no strong difference in AOO distributions of posttraumatic stress disorder, but
risk-to-prevalence ratios were found for less developed one would expect these to be quite variable due to trauma
countries (28–41%) versus developed countries (17–49%). exposure occurring throughout the life course. Nor are we
aware of previous research on the AOO distributions of
Not surprisingly, the highest class-specific proportional impulse-control disorders, although the lifetime preva-
increase in projected lifetime risk versus prevalence was lence estimates of these disorders in the WMH surveys
associated with mood disorders (IQR 61–98%) and the [29,33] are in the range reported in previous surveys of
lowest with impulse-control disorders (IQR 0–2%). adolescents [45,46]. No previous research has examined
These differences reflect the fact that many mood dis- the temporal concentration of AOO or highlighted the
orders begin in middle age or old age, while the vast concentration of onset ages for most disorders in a very
majority of impulse-control disorders begin in childhood narrow time span. It is striking in this regard that the
or adolescence. upper bounds of the AOO IQRs for disorders with narrow
ranges are all quite early: the mid-teens for impulse-
The high comorbidity known to exist among mental control disorders and anxiety disorders with narrow IQRs,
disorders [36–38] would be expected to result in many and the late 20s for substance use disorders. These
WMH respondents who developed impulse-control dis- are opposite to the patterns found for chronic physical
orders or early-onset anxiety disorders subsequently disorders, where conditional risk increases with age and
developing substance, mood, or later-onset anxiety dis- the upper bound of the IQR is in late middle age or old
orders. This possibility was investigated by comparing age [47].
the risk-to-prevalence ratios of any disorder versus indi-
vidual disorders. This showed that the vast majority of Investigations of initial contact with the treatment system
projected new onsets of individual disorders would be based on data collected in a number of community
secondary disorders, as indicated by the fact that the risk- epidemiological surveys [48] have consistently found that
to-prevalence ratio for any disorder is close to 1.0 in many people wait more than a decade after first onset of a
most countries. mental disorder before seeking treatment. These people
often present with highly comorbid conditions that might
Limitations have been easier to treat if they had sought treatment
The WMH results are limited by the possibility that earlier in the course of illness. Early treatment of first-
people with a history of mental illness might be less likely onset disorders would often mean childhood–adolescent
than others to participate in community surveys or might treatment. Unfortunately, we know little about treatment
underreport their disorders. Although innovative strat- of mild child–adolescent cases and even less about treat-
egies were used to minimize the latter problem [19], it is ment of incipient child–adolescent cases. A complicating
unlikely that they were completely successful. AOO factor is that the effects of psychotropic medications

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Age of onset of mental disorders Kessler et al. 363

9 Kaplan EL, Meier P. Nonparametric estimation from incomplete observations.

can be quite different on children–adolescents than J Am Stat Assoc 1958; 53:457–481.
adults [49]. The most active research of this sort deals 10 Halli SS, Rao KV, Halli SS. Advanced Techniques of Population Analysis. NY:
with treating incipient psychosis [11] and early stages Plenum; 1992.
of adolescent substance use disorder [50], although much 11 Amminger GP, Leicester S, Yung AR, et al. Early-onset of symptoms predicts
conversion to nonaffective psychosis in ultra-high risk individuals. Schizophr
remains to be learned about the best ways to treat these Res 2006; 84:67–76.
early cases. The paper examined AOO of psychiatric symptoms as predictors of conversion to
nonaffective and affective psychosis in help-seeking individuals at ultra-high risk
for schizophrenia. Early-onset of psychiatric symptoms were found to be spe-
Conclusion cifically associated with conversion to nonaffective psychosis but not affective
psychosis. This finding suggests that early AOO of certain psychiatric symptoms
The results reported here show clearly that first onset of is an independent factor relevant for defining samples specifically at risk for
mental disorders usually occurs in childhood or adoles- schizophrenia.
12 Klosterkotter J, Ruhrmann S, Schultze-Lutter F, et al. The European Prediction
cence. Although it is thought that timely interventions of Psychosis Study (EPOS): integrating early recognition and intervention in
with early-onset cases might help reduce severity- Europe. World Psychiatry 2005; 4:161–167.
persistence of primary disorders, prevent or delay onset 13 Kendall PC, Kessler RC. The impact of childhood psychopathology interven-
tions on subsequent substance abuse: policy implications, comments, and
of secondary disorders, or reduce persistence-severity recommendations. J Consult Clin Psychol 2002; 70:1303–1306.
of secondary disorders, much preclinical and clinical 14 Kessler RC, Keller MB, Wittchen HU. The epidemiology of generalized anxiety
research is needed on treatments of early cases to deter- disorder. Psychiatr Clin North Am 2001; 24:19–39.
mine whether this is true. Epidemiological research is 15 Clark DB, Jones BL, Wood DS, Cornelius JR. Substance use disorder
trajectory classes: diachronic integration of onset age, severity, and course.
also needed on the long-term consequences of early Addict Behav 2006; 31:995–1009.
interventions for long-term secondary prevention. 16 Nierenberg AA, Quitkin FM, Kremer C, et al. Placebo-controlled continuation
treatment with mirtazapine: acute pattern of response predicts relapse.
Neuropsychopharmacology 2004; 29:1012–1018.
Acknowledgements 17 Kessler RC, Haro JM, Heeringa SG, et al. The World Health Organization
This report was prepared in collaboration with the WHO World Mental World Mental Health Survey Initiative. Epidemiol Psichiatr Soc 2006;
Health (WMH) Survey Initiative. The core activities of the WMH are 15:161–166.
supported by the National Institute of Mental Health (R01MH070884), 18 Kessler RC, Andrews G, Mroczek D, et al. The World Health Organization
the John D. and Catherine T. MacArthur Foundation, the Pfizer Composite International Diagnostic Interview Short Form (CIDI-SF). Int J
Foundation, the US Public Health Service (R13-MH066849, R01- Methods Psyciatr Res 1998; 7:171–185.
MH069864, and R01 DA016558), the Pan American Health Organiza- 19 Kessler RC, Ustun TB. The World Mental Health (WMH) Survey Initiative
tion, Eli Lilly and Company, and GlaxoSmithKline. A complete list of WMH Version of the World Health Organization (WHO) Composite International
publications can be found at http://www.hcp.med.harvard.edu/wmh/. Diagnostic Interview (CIDI). Int J Methods Psychiatr Res 2004; 13:93–121.
20 Kessler RC, Birnbaum H, Demler O, et al. The prevalence and correlates of
nonaffective psychosis in the National Comorbidity Survey Replication
References and recommended reading (NCS-R). Biol Psychiatry 2005; 58:668–676.
Papers of particular interest, published within the annual period of review, have
been highlighted as: 21 Haro JM, Arbabzadeh-Bouchez S, Brugha TS, et al. Concordance of the
of special interest Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with
of outstanding interest standardized clinical assessments in the WHO World Mental Health surveys.
Additional references related to this topic can also be found in the Current Int J Methods Psychiatr Res 2006; 15:167–180.
World Literature section in this issue (pp. 431–432). 22 First MB, Spitzer RL, Gibbon M, Williams JBW. Structured Clinical Interview
for DSM-IV axis I disorders, research version, nonpatient edition (SCID-I/NP).
1 Amminger GP, Harris MG, Conus P, et al. Treated incidence of first-episode New York: Biometrics Research, New York State Psychiatric Institute; 2002. .
psychosis in the catchment area of EPPIC between 1997 and 2000. Acta
23 Robins LN, Regier DA, editors. Psychiatric Disorders disorders in America:
Psychiatr Scand 2006; 114:337–345.
The Epidemiologic Catchment Area Study. New York: The Free Press; 1991.
2 Bromet EJ, Fennig S. Epidemiology and natural history of schizophrenia. Biol
24 Simon GE, VonKorff M. Recall of psychiatric history in cross-sectional
Psychiatry 1999; 46:871–881.
surveys: implications for epidemiologic research. Epidemiol Rev 1995;
3 Norman RM, Malla AK. Duration of untreated psychosis: a critical examination 17:221–227.
of the concept and its importance. Psychol Med 2001; 31:381–400.
25 Knauper B, Cannell CF, Schwarz N, et al. Improving the accuracy of major
4 Perala J, Suvisaari J, Saarni SI, et al. Lifetime prevalence of psychotic and depression age of onset reports in the US National Comorbidity Survey. Int J
bipolar I disorders in a general population. Arch Gen Psychiatry 2007; Methods Psyciatr Res 1999; 8:39–48.
64:19–28.
26 de Girolamo G, Polidori G, Morosini PL, et al. Prevalence of common mental
The limitations of screening for psychosis in population surveys were highlighted
disorders in Italy. Results from the European Study of the Epidemiology of
by comparing clinician-based assessments of psychosis based on CIDI screens
Mental Disorders (ESEMeD). Soc Psychiatry Psychiatr Epidemiol 2006;
with data from a population registry and documenting substantial underrepresen-
41:853–861.
tation in the survey.
27 Lee S, Tsang A, Zhang MY, et al. Lifetime prevalence and inter-cohort variation
5 Jenab M, Riboli E, Ferrari P, et al. Plasma and dietary carotenoid, retinol and
in DSM-IV disorders in metropolitan China. Psychol Med 2007; 37:61–71.
tocopherol levels and the risk of gastric adenocarcinomas in the European
prospective investigation into cancer and nutrition. Br J Cancer 2006; 28 Medina-Mora M-E, Borges G, Lara C, et al. Prevalence, age of onset, and
95:406–415. demographic correlates of DSM-IV psychiatric disorders in Mexico: Results
from the Mexican Psychiatric Survey. Psychol Med (in press).
6 Yan LL, Liu K, Daviglus ML, et al. Education, 15-year risk factor progression,
and coronary artery calcium in young adulthood and early middle age: the 29 Oakley Browne MA, Wells JE, Scott KM, McGee MA. Lifetime prevalence and
Coronary Artery Risk Development in Young Adults study. JAMA 2006; projected lifetime risk of DSM-IV disorders in Te Rau Hinengaro: the New
295:1793–1800. Zealand Mental Health Survey. Aust N Z J Psychiatry 2006; 40:865–874.
7 Jones P, Rodgers B, Murray R, Marmot M. Child development risk factors 30 Gureje O, Lasebikan VO, Kola L, Makanjuola VA. Lifetime and 12-month
for adult schizophrenia in the British 1946 birth cohort. Lancet 1994; prevalence of mental disorders in the Nigerian Survey of Mental Health and
344:1398–1402. Well Being. Br J Psychiatry 2006; 188:465–471.
8 Lauronen E, Miettunen J, Veijola J, et al. Outcome and its predictors in 31 Haro JM, Palacin C, Vilagut G, et al. Prevalence of mental disorders and
schizophrenia within the Northern Finland 1966 Birth Cohort. Eur Psychiatry associated factors: results from the ESEMeD-Spain study. Med Clin (Barc)
2007; 22:129–136. 2006; 126:445–451.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
364 Services research and outcomes

32 Bromet EJ, Gluzman SF, Paniotto VI, et al. Epidemiology of psychiatric and 41 McGue M, Iacono WG. The association of early adolescent problem behavior
alcohol disorders in Ukraine: findings from the Ukraine World Mental Health with adult psychopathology. Am J Psychiatry 2005; 162:1118–1124.
survey. Soc Psychiatry Psychiatr Epidemiol 2005; 40:681–690.
42 Rosenbaum JF, Biederman J, Hirshfeld-Becker DR, et al. A controlled study of
33 Kessler RC, Berglund P, Demler O, et al. Lifetime prevalence and age-of- behavioral inhibition in children of parents with panic disorder and depression.
onset distributions of DSM-IV disorders in the National Comorbidity Survey Am J Psychiatry 2000; 157:2002–2010.
Replication. Arch Gen Psychiatry 2005; 62:593–602.
43 Christie KA, Burke JDJ, Regier DA, et al. Epidemiologic evidence for early
34 The WHO World Mental Health Survey Consortium. Lifetime prevalence and onset of mental disorders and higher risk of drug-abuse in young-adults. Am J
age-of-onset distributions of mental disorders in the WHO World Mental Psychiatry 1988; 145:971–975.
Health (WMH) Surveys. Bull World Health Organ (in press).
44 WHO International Consortium in Psychiatric Epidemiology. Cross-national
35 Thomsen PH. Schizophrenia with childhood and adolescent onset–a comparisons of the prevalences and correlates of mental disorders.
nationwide register-based study. Acta Psychiatr Scand 1996; 94:187– Bull World Health Organ 2000; 78:413 –426.
193.
45 Loeber R, Burke JD, Lahey BB, et al. Oppositional defiant and conduct
36 Jacobi F, Wittchen HU, Holting C, et al. Prevalence, co-morbidity and disorder: a review of the past 10 years, part I. J Am Acad Child Adolesc
correlates of mental disorders in the general population: results from the Psychiatry 2000; 39:1468–1484.
German Health Interview and Examination Survey (GHS). Psychol Med 2004;
46 Scahill L, Schwab-Stone M. Epidemiology of ADHD in school-age children.
34:597–611.
Child Adolesc Psychiatr Clin N Am 2000; 9:541–555; vii.
37 Kessler RC. The prevalence of psychiatric comorbidity. In: Wetzler S, 47 Murray CJL, Lopez AD. Global Health Statistics. Cambridge: Harvard
Sanderson WC, editors. Treatment strategies for patients with psychiatric University Press; 1996.
comorbidity. New York, NY: John Wiley & Sons; 1997. pp. 23–48.
48 Christiana JM, Gilman SE, Guardino M, et al. Duration between onset and time
38 Krueger RF, Chentsova-Dutton YE, Markon KE, et al. A cross-cultural study of of obtaining initial treatment among people with anxiety and mood disorders:
the structure of comorbidity among common psychopathological syndromes an international survey of members of mental health patient advocate groups.
in the general healthcare setting. J Abnorm Psychol 2003; 112:437– Psychol Med 2000; 30:693–703.
447.
49 Murrin LC, Sanders JD, Bylund DB. Comparison of the maturation of the
39 Yung AR, Yuen HP, McGorry PD, et al. Mapping the onset of psychosis: the adrenergic and serotonergic neurotransmitter systems in the brain: Implica-
Comprehensive Assessment of At-Risk Mental States. Aust N Z J Psychiatry tions for differential drug effects on juveniles and adults. Biochem Pharmacol
2005; 39:964–971. 2007; 73:1225–1236.
40 Thompson KN, Conus PO, Ward JL, et al. The initial prodrome to bipolar 50 White HR, Morgan TJ, Pugh LA, et al. Evaluating two brief substance-use
affective disorder: prospective case studies. J Affect Disord 2003; 77:79– interventions for mandated college students. J Stud Alcohol 2006; 67:309–
85. 317.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.