Health Affairs, 22, no. 3 (2003): 122-133 doi: 10.1377/hlthaff.22.3.122 © 2003 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (50) Citing Articles via Google Scholar Google Scholar Articles by Bijl, R. V. Articles by Wittchen, H.-U. Search for Related Content PubMed PubMed Citation Articles by Bijl, R. V. Articles by Wittchen, H.-U. Related Collections International Issues Mental Health/Substance Abuse

International Comparisons

The Prevalence Of Treated And Untreated Mental Disorders In Five Countries

PROLOGUE: The U.S. surgeon general and the World Health Organization (WHO) have both released studies in the past few years with alarming estimates of the prevalence of mental disorders, the burden these conditions create, and high rates of undertreatment. Assessing the policy implications of these findings is difficult, however, in part because the surveys on which prevalence estimates are based cannot capture degrees of severity with great precision. The following paper uses data from five countries to focus on the problem of undertreatment of severe mental illnesses. Serious cases are a relatively small proportion of all mental disorders, and "the probability of receiving treatment is strongly related to illness severity in each country," according to this multinational team of researchers.

Rob Bijl is head of the Department of Crime Prevention and Sanctions in the Research and Documentation Center of the Ministry of Justice in the Netherlands. Ron de Graaf is a physician and a senior researcher at the Netherlands Institute of Mental Health and Addiction. Eva Hiripi is a statistician and a senior analyst in the Department of Health Care Policy, Harvard Medical School; Ron Kessler is a professor in that department; and Ellen Walters is a senior biostatistician there. Robert Kohn is a psychiatrist and an assistant professor in the Department of Psychiatry and Human Behavior at Brown University (U.S.). David Offord is a psychiatrist and professor emeritus of psychiatry at McMaster University in Toronto. He also directs the Canadian Centre for Studies of Children at Risk. Bedirhan Ustun is a psychiatrist and head of the WHO Classification, Assessment, Surveys, and Terminology Unit in Geneva. Benjamin Vicente is a psychiatrist and a professor in the Department of Psychiatry and Mental Health at the Universidad de Concepción in Chile. Wilma Vollebergh is a psychiatrist and head of the Research Program on Developmental Psychiatry at the National Institute on Mental Health and Addiction in the Netherlands. Hans-Ulrich Wittchen is a psychologist and a professor of psychology at the Technical University of Dresden, Germany, and professor of clinical psychology at the Max Planck Institute of Psychiatry in Munich.

Abstract

 
We analyzed survey data from Canada, Chile, Germany, the Netherlands, and the United States to study the prevalence and treatment of mental and substance abuse disorders. Total past-year prevalence estimates range between 17.0 percent (Chile) and 29.1 percent (U.S.). Many cases are mild. Although disorder severity is strongly related to treatment, one- to two-thirds of serious cases receive no treatment each year. Most treatment goes to minor and mild cases. Undertreatment of serious cases is most pronounced among young, poorly educated males. Outreach is needed to reduce barriers to care among serious cases and young people at risk of serious disorders.

Three questions arise in evaluating these results. Are the high prevalence estimates in these surveys accurate? If so, is there a severity gradient among these cases such that only a small number are serious cases in need of treatment? Third, what is the relationship between disorder severity and treatment?

Only the first of these three questions can be answered with available data. A number of methodological studies that reinterviewed survey respondents with state-of-the-art, clinician-administered diagnostic assessments have confirmed the prevalence estimates in epidemiological surveys.⁶ The second question cannot be answered because no data on severity have as yet been reported from cross-national mental health surveys. In this paper we present the first data of this sort. Preliminary results based on U.S. data suggest that only a minority of current mental disorders are severe and that a substantial proportion are mild, but it is not clear whether this pattern holds in the other countries considered here.⁷ The third question, whether treatment resources are allocated in relation to severity, has been studied in the United States and Canada.⁸ A significant relationship was found in that study between severity and treatment. It is not known, though, whether this pattern also holds in other countries.

The data presented here come from five countries participating in the World Health Organization (WHO) International Consortium in Psychiatric Epidemiology (ICPE), a group established by WHO to coordinate the analysis of community surveys of mental disorders around the world.⁹ The five first-generation ICPE surveys that distinguish between recently treated and untreated mental disorders are included in the analysis.¹⁰ These surveys were administered between 1990 and 1999 and have a combined sample size of more than 23,000 respondents.

Participating countries. The surveys were fielded in Canada, Chile, Germany, the Netherlands, and the United States. Chile is a higher-middle-income country, based on World Bank criteria, while the others are high-income countries.

The five countries differ in their mental health care systems. All but the United States have universal health insurance coverage that includes coverage for mental health and substance abuse treatment services (with supplemental insurance required to cover the costs of medications in Canada).¹¹ The Canadian system is the only single-payer system; Chile, Germany, and the Netherlands have dual systems in which consumers either have a choice between public and private coverage (Chile and Germany) or have a first tier of basic public coverage supplemented with a second tier of either public or private coverage depending on income (the Netherlands).

The public-private distribution differs across the three countries with dual systems (60 percent public and 25 percent private in Chile, 90 percent public and 9 percent private in Germany, and 65 percent public and 35 percent private in the second tier of coverage in the Netherlands) and consistently overrepresents people with higher incomes in the private sector. All four countries with universal coverage use supply-side controls in the public sector, while the three with dual systems use demand-side controls in the public sector. The two most notable variations on these general cost control strategies are in Chile, where strict limits are placed on the number of mental health specialists (leading to the vast majority of mental health treatment’s being provided in the primary care sector), and in Germany, where supply-side controls are imposed on public budgets but not on health personnel or on restriction of direct access to mental health specialists.

In the United States, by comparison, roughly 60 percent of the population had private health insurance at the time the U.S. survey was carried out, with provisions for mental health treatment varying widely in these plans.¹² Low-income people were eligible for Medicaid or could receive free mental health care through state-funded community mental health clinics. Approximately 15 percent of the population had no health insurance, and another 10 percent had insurance that did not cover mental health treatment.

Study Methods

Top Study Methods Study Findings Discussion NOTES

 
All five surveys were based on general population probability samples and administered in face-to-face interviews. The surveys in the United States, the Netherlands, and Germany are nationally representative, while the Canadian survey is representative of the province of Ontario and the Chile survey is representative of four provinces in geographically distinct regions of the country (Santiago, Concepción, Iquique, and Caulin).

The response rates vary widely (53.8–90.3 percent) across surveys, raising a concern that the accuracy of prevalence estimates might differ across surveys (Exhibit 1). Somewhat tempering this concern is evidence in the United States that prevalence estimates are not strongly related to survey response rates.¹³ The surveys also differ in age range. We focus on respondents ages eighteen to fifty-four because this is the widest age range covered by all of the surveys. This means that no information is reported here on geriatric mental disorders.

The surveys all used versions of the WHO Composite International Diagnostic Interview (CIDI) to make diagnoses.¹⁴ The criteria of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III-R) were used to make diagnoses in all countries except Germany, where DSM-IV criteria were used.¹⁵ The disorders considered are mood disorders (major depression, dysthymia, and mania), anxiety disorders (panic disorder, generalized anxiety disorder, phobias, and post-traumatic stress disorder), and substance abuse disorders (alcohol and drug abuse and dependence). CIDI organic exclusion rules were imposed in making all diagnoses. Because of the complex sample designs and weighting of the surveys, standard errors of the various descriptive statistics reported here were estimated using the Taylor Series method implemented in the SUDAAN software package.¹⁶ These estimates adjust for the clustering and weighting of cases. Results described below as being "significant" were evaluated using two-sided design-based statistical tests evaluated at the .05 level of significance.

Study Findings

Top Study Methods Study Findings Discussion NOTES

 
Past-year prevalence. The overall prevalence estimates (Exhibit 2) range from 17.0 percent in Chile to 29.1 percent in the United States. Prevalence was in the range of 4.9–11.9 percent for mood disorders, 5.0–17.0 percent for anxiety disorders, and 5.2–11.5 percent for substance abuse disorders. Prevalence estimates for anxiety disorders, substance abuse disorders, and any disorders were highest in the United States, while the prevalence estimate for mood disorders was highest in Germany.

In addition to the possible biases just mentioned, the higher prevalence estimates in the United States might be attributable to special procedures in the U.S. survey, which featured the use of commitment probes and a review of lifetime diagnostic stem questions at the beginning of the CIDI interview, in an effort to motivate complete and accurate reporting. A methodological experiment showed that these methods generated significantly higher prevalence estimates and improved validity compared with the standard version of the CIDI by decreasing the problem of false negatives.¹⁸ This cannot be the entire reason for the high U.S. prevalence estimates, though, as the same procedures were used in the Canadian survey, where the prevalence estimates are considerably lower than in the U.S. survey.

To assess severity, we classified respondents with disorders into mild, moderate, and serious cases based on their multivariate disorder profiles.¹⁹ This is only a rough classification because no direct data on severity were collected consistently across the surveys. Cases classified as mild made up a majority of all cases in Canada and the Netherlands and close to a majority in the other countries (Exhibit 2). Both serious and moderate cases were estimated to be more prevalent in the United States than in the other countries, while the prevalence of mild cases was quite similar in Canada, the Netherlands, and the United States but lower in Chile and Germany.

Past-year treatment Respondents were asked about receiving professional treatment in the past twelve months (six months in Chile) for problems with their emotions or substance use.²⁰ Treatment rates varied significantly across countries (Exhibit 3), from a low of 7.0 percent in Canada to a high of 20.3 percent in Germany, with a U.S. rate of 10.9 percent. There is no relationship between the prevalence rates in Exhibit 2 and the overall treatment rates in Exhibit 3.

Even though the lowest treatment rate was consistently among respondents classified as noncases, the fact that some people in this category received treatment is noteworthy in light of the fact that noncases make up 70.9–83.0 percent of the respondents across surveys. Noncases in treatment presumably consist of people with mild disorders not assessed in the CIDI or with subthreshold psychological distress. Even though their treatment rates were low, noncases made up a large proportion of all people in treatment (from 24.2 percent in the Netherlands to 69.1 percent in Chile).

Noncases play a critical role in explaining between-country differences in overall treatment rates. The low overall treatment rate in Canada would increase to a rate higher than in the United States and close to the rate in the Netherlands, for example, if the proportion of Canadian noncases in treatment increased to the average of the other countries. It is not clear whether the especially low treatment rate among Canadian noncases is attributable to greater reluctance to seek help for minor emotional problems, higher barriers to care of such problems in the Canadian health care system, or some combination of these factors. An earlier comparative analysis of treatment in the United States and Canada shed some light on this issue by showing that Canadians in the noncase segment of the population had a much lower perceived need for treatment than did their U.S. counterparts, presumably reflecting a lower predisposition to define minor emotional problems as sufficiently serious to seek medical attention.²¹ Importantly, the treatment rate among noncases who perceive themselves as needing health treatment was found to be the same in Canada as in the United States, implying that system barriers to seeking care for minor emotional problems do not differ in the two countries.

The high overall treatment rates in Chile and Germany are also importantly influenced by the treatment of noncases. In Chile the conditional treatment rate among noncases was significantly higher than in Canada, the Netherlands, and the United States, while the conditional treatment rates for mild, moderate, and serious cases were generally not higher than in these other countries. The same general patterns holds in Germany, except that the elevated German treatment rate is found among both mild cases and noncases.

The high treatment rates of minor and mild cases in Chile and Germany might be related to their differences from other countries in sectors of mental health care. The majority of patients were seen in the general medical sector in Canada (65.4 percent), the Netherlands (74.0 percent), and Chile (80.3 percent), but not in the United States (33.9 percent) or Germany (39.0 percent). Specialty treatment, in comparison, was most common among patients in Germany (70.0 percent), least common among patients in Chile (36.5 percent), and intermediate in the other countries (48.5–50.6 percent). Chile and Germany, importantly, are the only countries in which severity was unrelated to the proportion of treatment delivered in the specialty sector (Exhibit 3).

These differences in sectors of care might help explain the high treatment rates among noncases in Chile, because, unlike in the other counties, the primary care sector provides the vast majority of this treatment, and national health insurance allows easy access to the primary care sector. There is also low restraint on access to mental health care among noncases in Germany because the specialty sector provides the vast majority of this treatment, and the German health care system allows patients to go directly to these specialists without a primary care referral.

Consistent with this line of reasoning, Exhibit 3 shows that the proportion of patients in the specialty sector increased significantly with severity in Canada, the Netherlands, and the United States. This variation is also noteworthy in light of the fact that more severe cases have a greater need for specialty care and the fact that specialty care is more likely than nonspecialty care is to conform with published guidelines for adequate treatment of serious cases.²² This might mean that receipt of adequate treatment was lower in Chile than the other countries even through the percentage of cases in treatment was higher in Chile than in all other countries but Germany. There is no direct way to evaluate this possibility in the ICPE surveys, though, as no consistent data were collected on quality of care.

Demographic correlates of treatment. In light of the fact that one- to two-thirds of people with serious disorders in the five countries received no treatment, the correlates of treatment among serious cases are of interest. As it turns out, we found that the demographic correlates of treatment did not vary depending on illness severity. Across all levels of severity, treatment was consistently and positively related to age (other than in Chile), education, and being female.²³ Specialty treatment among people in treatment, in addition, was positively related to education in all countries.

Discussion

Top Study Methods Study Findings Discussion NOTES

 
The ICPE results are consistent with those of earlier surveys in showing that mental disorders are highly prevalent in a number of countries. However, we also presented evidence suggesting that many of these disorders are mild. Caution is needed in interpreting this second result, though, as the disorder severity gradient used here is crude. A much more refined severity classification scheme is being used in a new second generation of ICPE surveys that are now in progress.

The tentative conclusion that many mental disorders are mild also should not be taken to suggest that these disorders have low need for treatment. On the contrary, longitudinal research shows that people with mild mental disorders, if left untreated, have a higher risk of future serious outcomes, such as attempted suicide, hospitalization, and work disability.²⁴

Even though the severity measure is coarse, it is strongly related to probability of treatment in all countries. This is most reasonably interpreted to mean that demand for treatment was related to severity, presumably mediated by distress and impairment. There is also indirect evidence that the treatment system was responsive to severity in at least three of the countries, as indicated by proportional treatment in the specialty sector increasing with severity. The fact that these associations are not significant in Chile or Germany is part of a larger pattern: A much higher proportion of all mental health care was provided in a single sector in those countries than in the other countries. It would be interesting to see if more direct measures of treatment intensity, such as number of visits, are as strongly related to illness severity in Chile and Germany as in the other countries. The only available information of this sort is in the Canadian and U.S. surveys, in which severity was strongly related to number of visits.²⁵

Despite the evidence of rationality in allocation of treatment resources, we found that most patients in each of the countries had either a minor or mild disorder, while one-third to two-thirds of people with serious disorders received no treatment. The low treatment rate among serious cases is most striking in the United States, where only about one-third of serious cases received treatment. Although there are disagreements among specialists about the relationship between severity and need for treatment, it is agreed that serious cases should be treated.²⁶ The low treatment rate among serious cases is consequently a matter of concern.

The serious cases least likely to receive treatment were young, poorly educated males. Outreach programs to increase their treatment are unlikely to succeed because the psychological barriers to treatment in this population are difficult to overcome. A more feasible approach might be to reach these young men in adolescence through school-based interventions. Consistent with this suggestion, retrospective ICPE data show that the vast majority of young, poorly educated males (and, for that matter, females) with serious mental illness reported that their disorders started in childhood or adolescence. Many of them also reported, though, that their disorders became serious only after they left school and became adults. As a result, adolescent treatment might best be conceptualized as early intervention aimed at preventing progression from mild to more serious disorders.

Early intervention among adolescents with mental disorders could be of great value in reducing the documented adverse effects of adolescent mental disorders on such critical life-course transitions as educational attainment, teenage child-bearing, and timing of marriage.²⁷ It is likely that the progression from mild adolescent disorders to serious young adult disorders is mediated by such effects. While focused treatment of adolescent disorders might be seen as leading to an increase in the already high proportion of treatment resources devoted to mild cases, this need not be so. The mild cases in treatment in all of the ICPE surveys include high proportions of older, well-educated females who are at low risk of progressing to serious mental illness. Reallocation of treatment resources from these cases to adolescents could have a substantial public health benefit by preventing future serious disorders in all of these countries without changing the current proportion of treatment resources allocated to mild cases. At the same time, efforts are needed to develop programs that reach a higher proportion of current severe cases.

Editor's Notes

 
Preparation of this paper was supported by grants from the U.S. Public Health Service (R01 DA11121) and the Pfizer Foundation. This report presents data from five countries. The MHS-OHS survey (Canada) was funded by the Ontario Ministry of Community and Social Services. The CPPS survey (Chile) was funded by FONDECYT (90-229, 92-233, 1971315, 1990325), Dirección de Investigación de la Universidad de Concepción (201.087.027-1.0), and the Pan American Health Organization/World Health Organization. The GHS-MHS survey (Germany) was funded by the German Federal Ministry of Research, Education, and Science; BMBF (01EH970/8). The NEMESIS survey (the Netherlands) was funded by the Netherlands Institute of Mental Health and Addiction (Trimbosinstituut); the Netherlands Ministry of Health, Welfare, and Sport (VWS); the Medical Sciences Department of the Netherlands Organization of Scientific Research (NWO); and the National Institute for Public Health and Environment (RIVM). The NCS survey (U.S.) was funded by the U.S. National Institute of Mental Health (R01 MH46376 and R01 MH 49098) and the W.T. Grant Foundation (90135190).

NOTES

Top Study Methods Study Findings Discussion NOTES

 

1. A.H. Leighton, My Name Is Legion, Stirling County Study, vol. 1 (New York: Basic Books, 1959); T.S. Langner and S.T. Michael, Life Stress and Mental Health: The Midtown Manhattan Study (London: Collier-MacMillan, 1963); and L.N. Robins et al., "National Institute of Mental Health Diagnostic Interview Schedule: Its History, Characteristics, and Validity," Archives of General Psychiatry 38, no. 4 (1981): 381–389.[Abstract]
2. L.N. Robins and D.A. Regier, Psychiatric Disordersin America: The Epidemiologic Catchment Area Study (New York: Free Press, 1991). Examples of later surveys based on the ECA study are R.C. Bland, H. Orn, and S.C. Newman, "Lifetime Prevalence of Psychiatric Disorders in Edmonton," Acta Psychiatrica Scandinavica, Supplement 338 (1988): 24–32; H.G. Hwu, E.K. Yeh, and L.Y. Cheng, "Prevalence of Psychiatric Disorders in Taiwan Defined by the Chinese Diagnostic Interview Schedule," Acta Psychiatrica Scandinavica 79, no. 2 (1989): 136–147[Medline]; and J.P. Lépine et al., "Anxiety and Depressive Disorders in a French Population: Methodology and Preliminary Results," Psychiatric and Psychobiology 4, no. 5 (1989): 267–274.
3. Cross-National Collaborative Group, "The Changing Rate of Major Depression: Cross-National Comparisons," Journal of the American Medical Association 268, no. 21 (1992): 3098–3105[Abstract]; M.M. Weissman et al., "Cross-National Epidemiology of Major Depression and Bipolar Disorder," Journal of the American Medical Association 276, no. 4 (1996): 293–299[Abstract]; and M.M. Weissman et al., "The Cross-National Epidemiology of Panic Disorder," Archives of General Psychiatry 54, no. 4 (1997): 305–309.[Abstract]
4. C.J.L. Murray and A.D. Lopez, The Global Burden of Disease: A Comprehensive Assessment of Mortality and Disability from Diseases, Injuries, and Risk Factors in 1990 and Projected to 2020 (Cambridge, Mass.: Harvard University Press, 1996).
5. K.B. Wells et al., "The Functioning and Well-Being of Depressed Patients: Results from the Medical Outcomes Study," Journal of the American Medical Association 262, no. 7 (1989): 914–919[Abstract]; and R.C. Kessler et al., "The Effects of Chronic Medical Conditions on Work Loss and Work Cutback," Journal of Occupational and Environmental Medicine 43, Suppl. 3 (2001): 218–225.
6. H.U. Wittchen, "Reliability and Validity Studies of the WHO Composite International Diagnostic Interview (CIDI): A Critical Review," Journal of Psychiatric Research 28, no. 1 (1994): 57–84[Medline]; and R.C. Kessler et al., "Methodological Studies of the Composite International Diagnostic Interview (CIDI) in the U.S. National Comorbidity Survey," International Journal of Methods in Psychiatric Research 7, no. 1 (1998): 33–55.
7. R.C. Kessler et al., "The Twelve-Month Prevalence and Correlates of Serious Mental Illness (SMI)," in Mental Health, United States, 1996, ed. R.W. Manderscheid and M.A. Sonnenschein (Washington: U.S. Department of Health and Human Services, 1996), 59–70; R.C. Kessler et al., "Population-Based Analyses: A Methodology for Estimating the Twelve-Month Prevalence of Serious Mental Illness," in Mental Health, United States, 1998, ed. R.W. Manderscheid and M.J. Henderson (Washington: U.S. Government Printing Office, 1998), 99–109; and W.E. Narrow et al., "Revised Prevalence Estimates of Mental Disorders in the United States: Using a Clinical Significance Criterion to Reconcile Two Surveys’ Estimates," Archives of General Psychiatry 59, no. 2 (2002): 115–123.[Abstract/Free Full Text]
8. R.C. Kessler et al., "Differences in the Use of Psychiatric Outpatient Services between the United States and Ontario," New England Journal of Medicine 336, no. 8 (1997): 551–557.[Abstract/Free Full Text]
9. R. Kessler, "The World Health Organization International Consortium in Psychiatric Epidemiology (ICPE): Initial Work and Future Directions—the Nape Lecture, 1998," Acta Psychiatrica Scandinavica 99, no. 1 (1999): 2–9.[Medline]
10. R.C. Kessler and T.B. Ustun, "The World Health Organization World Mental Health 2000 Initiative," Hospital Management International (2000): 195–196.
11. S.V. Parikh, E. Lin, and A.D. Leasage, "Mental Health Treatment in Ontario: Selected Comparisons between the Primary Care and Specialty Sectors," Canadian Journal of Psychiatry 42, no. 9 (1997): 929–934[Medline]; A. Manuel, "The Chilean Health System: Twenty Years of Reforms," Salud Publica de Mexico 44, no. 1 (2002): 60–68[Medline]; C. Altenstetter, "Insights from Health Care in Germany," International Perspectives Forum 93, no. 1 (2003): 38–44; and A.J.P. Schrijvers and L. Droyan Kodner, Health and Health Care in the Netherlands (Utrecht: de Tijdstroom, 1997).
12. R.G. Frank and T.G. McGuire, "Health Reform and Financing of Mental Health Services: Distribution Issues," in Mental Health, United States, 1994, ed. R. Manderschied and M. Sonnenschein (Washington: DHHS, 1994), 8–20; and J.C. Cantor, S.H. Long, and M.S. Marquis, "Challenge of State Health Reform: Variations in Ten States," Health Affairs (Jan/Feb 1998): 191–200.
13. R.C. Kessler, R.J.A. Little, and R.M. Groves, "Advances in Strategies for Minimizing and Adjusting for Survey Nonresponse," Epidemiologic Reviews 17, no. 1 (1995): 192–204.[Free Full Text]
14. The original WHO CIDI was used in Chile and the Netherlands. L.N. Robins et al., "The Composite International Diagnostic Interview: An Epidemiologic Instrument Suitable for Use in Conjunction with Different Diagnostic Systems and in Different Cultures," Archives of General Psychiatry 45, no. 12 (1988): 1069–1077. [Abstract]The University of Michigan version of the CIDI was used in Canada and the United States. Kessler et al., "Methodological Studies." The Munich version of the CIDI was used in Germany. H.U. Wittchen et al., "Test-Retest Reliability of the DSM-IV Version of the M-CIDI," Social Psychiatry and Psychiatric Epidemiology 33, no. 11 (1998): 568–578.[Medline]
15. American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (Washington: APA, 1987); and APA, Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (Washington: APA, 1994).
16. K. Rust, "Variance Estimation for Complex Estimators in Sample Surveys," Journal of Official Statistics 1, no. 4 (1985): 382–397; and Research Triangle Institute, SUDAAN, Version 8.0.1.
17. Kessler et al., "Methodological Studies"; and Wittchen et al., "Test-Retest Reliability."
18. Kessler et al., "Methodological Studies."
19. A variable ranging between 1 and 20 was constructed for all respondents who met criteria for at least one of the disorders. Some disorders were given one point (dysthymia and simple phobia), others two points (agoraphobia, social phobia, and substance abuse disorders), and others four points (generalized anxiety disorder, major depression, mania, and panic disorder), based on preliminary analyses of the effects of the disorders in predicting summary measures of role impairment. Severity categories were defined based on summary scores as follows: 1–2, mild; 3–4, moderate; and 5–20, serious.
20. The versions of the CIDI used in the first-generation ICPE surveys did not include a treatment section. As a result, treatment questions differed across surveys but were sufficiently similar to allow parallel coding.
21. Kessler et al., "Differences in the Use of Psychiatric Outpatient Services."
22. P.S. Wang, O. Demler, and R.C. Kessler, "Adequacy of Treatment for Serious Mental Illness in the United States," American Journal of Public Health 92, no. 1 (2002): 92–98.[Abstract/Free Full Text]
23. Results not presented in the text are archived at www.hcp.med.harvard.edu/icpe/relatedmaterials.htm.
24. R.C. Kessler et al., "Should Mild Disorders Be Eliminated from the DSM-V?" Archives of General Psychiatry (forthcoming).
25. Kessler et al., "Differences in the Use of Psychiatric Outpatient Services."
26. Narrow et al., "Revised Prevalence Estimates;"; and J.C. Wakefield and R.L. Spitzer, "Lowered Estimates—But of What?" Archives of General Psychiatry 59, no. 2 (2002): 129–130.[Free Full Text]
27. R.C. Kessler et al., "The Social Consequences of Psychiatric Disorders, I. Educational Attainment," American Journal of Psychiatry 152, no. 7 (1995): 1026–1032[Abstract/Free Full Text]; R.C. Kessler et al., "The Social Consequences of Psychiatric Disorders, II. Teenage Parenthood," American Journal of Psychiatry 154, no. 10 (1997): 1405–1411[Abstract]; and R.C. Kessler, E.E. Walters, and M.S. Forthofer, "The Social Consequences of Psychiatric Disorders, III. Probability of Marital Stability," American Journal of Psychiatry 155, no. 8 (1998): 1092–1096.[Abstract/Free Full Text]

                      What's this?

This article has been cited by other articles:

				B. G. Druss, P. S. Wang, N. A. Sampson, M. Olfson, H. A. Pincus, K. B. Wells, and R. C. Kessler Understanding Mental Health Treatment in Persons Without Mental Diagnoses: Results From the National Comorbidity Survey Replication Arch Gen Psychiatry, October 1, 2007; 64(10): 1196 - 1203. [Abstract] [Full Text] [PDF]

				J. Sareen, B. J. Cox, T. O. Afifi, M. B. Stein, S.-L. Belik, G. Meadows, and G. J. G. Asmundson Combat and Peacekeeping Operations in Relation to Prevalence of Mental Disorders and Perceived Need for Mental Health Care: Findings From a Large Representative Sample of Military Personnel Arch Gen Psychiatry, July 1, 2007; 64(7): 843 - 852. [Abstract] [Full Text] [PDF]

				G. Borges, M. E. Medina-Mora, P. S. Wang, C. Lara, P. Berglund, and E. Walters Treatment and Adequacy of Treatment of Mental Disorders Among Respondents to the Mexico National Comorbidity Survey Am J Psychiatry, August 1, 2006; 163(8): 1371 - 1378. [Abstract] [Full Text] [PDF]

				O. GUREJE, V. O. LASEBIKAN, L. KOLA, and V. A. MAKANJUOLA Lifetime and 12-month prevalence of mental disorders in the Nigerian Survey of Mental Health and Well-Being The British Journal of Psychiatry, May 1, 2006; 188(5): 465 - 471. [Abstract] [Full Text] [PDF]

				R. C. Kessler, W. T. Chiu, O. Demler, and E. E. Walters Prevalence, Severity, and Comorbidity of 12-Month DSM-IV Disorders in the National Comorbidity Survey Replication Arch Gen Psychiatry, June 1, 2005; 62(6): 617 - 627. [Abstract] [Full Text] [PDF]

				The WHO World Mental Health Survey Consortium Prevalence, Severity, and Unmet Need for Treatment of Mental Disorders in the World Health Organization World Mental Health Surveys JAMA, June 2, 2004; 291(21): 2581 - 2590. [Abstract] [Full Text] [PDF]