Our meta-analysis estimated the pooled prevalence rates of the individual diagnostic groups that were commonly reported by the studies. The funnel plots were rather asymmetrical for all of the mental disorders, which suggested potential publication bias (Appendix, http://links.lww.com/MD/A660). However, the P value of the Egger regression intercept was >0.05, and the asymmetry was thus statistically nonsignificant.
Any Depressive Disorder
Eight studies including 3104 subjects provided data about any depressive disorder, including major depressive disorder, bipolar depressive disorder, dysthymia, and other and minor depressive disorders not otherwise specified (NOS) (Figure 2).4,5,9–11,13–15 The prevalence estimates ranged from 3% to 38%. The random effects pooled prevalence estimate was 11% (95% CI 7–15, P < 0.001, I2 = 93.7%).
Major Depressive Disorder
The major depressive disorder was analyzed subsequently because this subtype of depressive disorder is one of the most important challenges in global mental health,36 especially in youths.37 Five studies including 1339 subjects provided data about major depressive disorder (Figure 3).4,9–11,14 The prevalence estimates ranged from 1% to 23%. The random effects pooled prevalence estimate was 12% (95% CI 5–18, P < 0.001, I2 = 96.2%).
Any Disruptive Disorder
Eight studies including 3104 subjects provided data about any disruptive disorder including conduct disorder, oppositional defiant disorder (ODD), and other disruptive disorders (Figure 4).4,5,9–11,13–15 The prevalence estimates ranged from 15% to 39%. The random effects pooled prevalence estimate was 27% (95% CI 20–34, P < 0.001, I2 = 94.8%).
Seven studies including 2731 subjects provided data about conduct disorder (Figure 5).5,9–11,13–15 The prevalence estimates ranged from 6% to 28%. The random effects pooled prevalence estimate was 20% (95% CI 13–27, P < 0.001, I2 = 95.3%).
Oppositional Defiant Disorder
Five studies including 2084 subjects provided data about ODD (Figure 6).5,9,11,13,14 The prevalence estimates ranged from 6% to 14%. The random effects pooled prevalence estimate was 12% (95% CI 10–14, P < 0.214, I2 = 31.0%).
Eight studies including 3104 subjects provided data for attention-deficit/hyperactivity disorder (ADHD) (Figure 7).4,5,9–11,13–15 The prevalence estimates ranged from 2% to 21%. The random effects pooled prevalence estimate was 11% (95% CI 6–15, P < 0.001, I2 = 95.4%).
Any Anxiety Disorder
Seven studies including 2731 subjects provided data about any anxiety disorder, including generalized anxiety disorder, overanxious disorder, separation-anxiety disorder, specific phobia, social phobia, panic disorder, obsessive compulsive disorder, posttraumatic stress disorder, and other anxiety disorders NOS (Figure 8).5,9–11,13–15 The prevalence estimates ranged from 4% to 32%. The random effects pooled prevalence estimate was 18% (95% CI 12–24, P < 0.001, I2 = 95.7%).
Posttraumatic Stress Disorder
Five studies including 2379 subjects provided data about posttraumatic stress disorder (Figure 9).4,5,9,13,14 The prevalence estimates ranged from 2% to 8%. The random effects pooled prevalence estimate was 4% (95% CI 2–6, P < 0.001, I2 = 81.3%).
Any Mental Disorder
Eight studies including 3104 patients provided data about any mental disorder (Figure 10).4,5,9–11,13–15 The prevalence estimates ranged from 37% to 67%. The random effects pooled prevalence estimate was 49% (95% CI 43–54, P < 0.001, I2 = 87.3%).
The results of the individual variable meta-regression models for each mental disorder are presented in Table 5. The final multivariate model identified diagnostic criteria and functional impairment (β = −0.12, se[β] = 0.04, P < 0.01; and β = −0.15, se[β] = 0.04, P < 0.001, respectively) as significant moderators of the prevalence estimate of any depressive disorder, informants (β = −0.12, se[β] = 0.04, P < 0.01) as a significant moderator of the prevalence estimates of major depressive disorder, functional impairment (β = 0.06, se[β] = 0.03, P < 0.05) as a significant moderator of the prevalence estimates of ODD, mean age (β = 0.03, se[β] = 0.01, P < 0.001) as a significant moderator of the prevalence estimates of ADHD, sample size (β = −0.17, se[β] = 0.05, P < 0.001) as a significant moderator of the prevalence estimates of any anxiety disorder, and sex ratio and country (β = 0.09, se[β] = 0.03, P < 0.001 and β = −0.09, se[β] = 0.04, P < 0.05, respectively) as significant moderators of the prevalence estimates of any mental disorders. No significant moderator was found for disruptive and conduct disorders.
This systematic review of children and adolescents in the CWS identified 8 surveys that included 3104 subjects.4,5,9–11,13–15
Our findings suggest that mental disorders affect a substantially greater proportion of children and adolescents in the CWS that in the general population. The 49% pooled prevalence for any mental disorder estimated by our meta-analysis is nearly 4-fold greater than the 13.4% pooled prevalence among the general children and adolescent population.2 Bowlby attachment theory underscores the central role of child-to-parent attachment in a child's development and mental health and may explain the high prevalence observed in our work.38–40 Several empirical studies and review reported connections between attachment insecurities and vulnerability to mental disorders.41,42 In children and adolescents in the CWS, adverse experiences, such as maltreatment and serious neglect, contribute to reducing the likelihood of creating a secure attachment that is crucial for developmental health.9,12 In addition, although the CWS should provide safe alternative homes, multiple placements and temporary or disrupted relationships with caregivers can also potentially prevent the children from forming secure attachments.43
Our findings indicate that externalized disorders are the primary main problem in children and adolescents in the CWS. The most common mental disorder was disruptive disorder (27%). The prevalence of conduct disorder was 20%, and the prevalence for ODD was 12%, which are 10 and 3 times more frequent than the prevalences in the general population, respectively. Notably, ADHD was also approximately 3 times more frequent in the CWS children than in the general population.2,44,45 This high prevalence of externalized disorders may be explained by the fact that the symptoms of conduct disorder, such as property loss or damage, aggressive conduct, and serious violations of rules, constitute direct causes of placement in the CWS. In addition, several adverse experiences (eg, multiple placements and maltreatment) during the time of placement may also contribute to the worsening of externalized disorders that are already present or the promotion of the emergence of such disorders. These results elicit some concerns for the children and adolescents in the CWS regarding the poor prognoses for these disorders that including snow-balling negative outcomes, such as the risk of developing antisocial personality disorder and substance use disorders.15,46–48 These disorders are known to be risk factors for delinquency, interactions with the juvenile justice systems, and homelessness.49 For example, in France 25% of homeless people and 20% of adults in jail were formerly youths in the CWS.50,51
The prevalence of internalized disorders was far from uncommon; 18% of the subjects had anxiety disorder, and 11% had depressive disorder, and these percentages are approximately 3- and 4-fold greater than those of the general population, respectively.2 Therefore, it is necessary to acknowledge the gravity of these health problems in this population, particularly considering deleterious effects of these problems on psychosocial functioning and quality of life and their associations with increased suicide rates and drug- and alcohol-use disorders.52,53
The high prevalence of externalized and internalized disorders is in line with studies that have reported that attachment insecurities nonspecifically contribute to many types of psychopathologies.41,54 Our findings highlight the complexity of screening and care in this population in which externalized and internalized disorders are associated and complexly entangled.
Our meta-analysis identified significant heterogeneity across all of the reported random effect models. Comparable levels of heterogeneity have been identified in other recent systematic reviews and meta-analyses in the general population.2,55,56 The significant heterogeneity was attributable to several factors, including country, sex ratio, mean age, sample size, informants, diagnostic criteria, and functional impairment. The rate of any mental disorder was higher in Europe than in the US. Culture may influence the identification and interpretation of symptoms and their attributed meaning.2 However, we cannot exclude the possibility that this heterogeneity resulted from structural and organizational differences between Europe and the United States of America. Indeed, alternatives to institutional care, such as kinship care, have been developed in the United States of America but not in some of the European countries included in our meta-analysis,5,57,58 and it was not possible to adjust for the structure of placement in our multiple meta-regressions. The prevalence of any mental disorder was higher among the males, but the sex ratios were not significantly different for the individual diagnoses. This finding contradicts the literature regarding the general population, which indicates that there do not appear to be sex differences in the overall prevalence of mental disorders, but there are significant differences in the patterns and symptoms of the disorders (ie, higher prevalences of internalized disorders among girls and externalized disorders among boys).59 The absence of differences between girls and boys suggests that externalized disorders in girls and internalized disorders in boys deserve increased attention by professionals and may be underdiagnosed and undertreated.
The other factors related to the heterogeneity (ie, sample size, informants, diagnostic criteria, and functional impairments) may contribute to the methodological issues that future epidemiological studies should consider to produce more accurate estimates. The sampling strategy, including the sample size, is a major factor in the generalization of the estimated prevalence to the entire population. In our work, greater sample sizes were associated with lower estimates of anxiety disorders, which suggests that the estimates may have been overvalued in several studies with small samples. We observed lower estimates of the prevalence of major depressive disorder when the informants were parents, caregivers, teachers, children or adolescents compared with children and adolescents alone. This finding is not surprising; the concordance between informants is known to be low,2 and children and adolescents generally report more internalizing disorders than parents, who tend to report more externalizing disorders.60 The challenge is thus to provide a strategy for reliable diagnoses that integrate information from different sources. Although diagnostic criteria were standardized and this was one of criteria for selecting the articles, we observed that differences in the diagnostic criteria resulted in differences in the prevalence rates as previously reported in several meta-analyses.2,61,62 In a recent study using 2 major nosological systems, the DSM-IV-TR consistently classified more children and adolescents than ICD-10 with an anxiety disorder.63 Lastly, functional impairment measurements result in differences in prevalence rates that include lower prevalence estimates for any depressive disorder. Previous studies have reported that the inclusion of an impairment criterion has a significant influence on reducing the prevalence rates of mental disorders, particularly for internalizing disorders.64 Surprisingly, we observed the opposite effect of higher prevalence estimates for ODD. However, this association was moderate in strength. The presence of both symptomatic and impairment criteria appears to be the most robust approach for case definition.65
Limits. We observed a relatively low number of psychiatric epidemiological surveys that employed standardized diagnostic criteria and psychiatric interviews (n = 8). In contrast, 2 recent reviews identified 41 surveys of the general population that explored the prevalence of mental disorders in children and adolescent over 27 countries2 and 174 surveys of adults over 63 countries.56 In addition, the methodological quality of the included studies was heterogeneous, and few studies used a whole population approach with random selection. Most of these studies were from single towns or regions and focused on out-of-home children and adolescents, which thus limited the generalizability of the findings. The exploration of heterogeneity was limited by the relatively low number of studies and the lack of information about covariates in several of the studies. Only 5 studies presented the mean age, and only 6 presented the sex ratio, despite the importance of these data. Several important covariates had too many categories and could not be included in the meta-regression analyses (eg, the type of placement, the diagnostic instrument, the type of functional measurement). Other data were not available (eg, age of first placement, the number of changes in placement, and adverse childhood experiences such as maltreatment and serious neglect). These characteristics that we were unable to test might be responsible for the heterogeneity and should be accounted for in future studies because some previous research has reported the importance of maltreatment and adverse experiences in the development of mental disorders.12,66
A potential source of bias was the implementation of appropriate search strategies to identify the relevant studies. Specifically, there are important variabilities in the organizational and denomination structures in the child welfare systems between countries that make searches and comparisons difficult. Concordantly, there was evidence of moderate publication bias based on the inspections of the funnel plots, although these results were found to be statistically nonsignificant. Lastly, the majority of the studies did not report prevalence estimates for less-frequent mental disorders, such as eating, elimination, obsessive-compulsive, psychotic, and substance-use disorders. For example, some of the studies reported noticeable rates of psychotic symptoms and highlighted the necessity of reporting them in future research and improving their early detection.10,11 The reports of comorbid disorders were also inconsistent between studies, and the high rates of the associations of multiple mental disorders in the studies that reported such rates highlight the fact that these association should be investigated in more detail in future works.10,15
Perspectives. The availability of accurate epidemiological data about children and adolescents in the CWS appears to be necessary to guide public policy. Interestingly, the “Best Practices for Mental Health in Child Welfare Consensus Conference” published in 2009 developed guidelines in 5 key areas including systematic screening and assessment.67 This systematic screening and assessment could serve as basis for the creation of national registries that could enable more accurate tracking. Altogether, these findings highlight the need for additional studies that specifically target children and adolescents in the CWS to improve the diagnoses and treatments of mental disorders in this population.
Although the high prevalences that were reported for mental disorders in children and adolescents involved in the CWS highlight the need for qualified service provisions, the substantial heterogeneity of our findings is also indicative of the need for accurate epidemiological data to inform and guide effective public policy. Given the importance of mental disorders in this particular population, the poor prognoses of the relevant complex mental states and the high cost to society, it is unfortunate that this population of youths suffering from mental disorders in the CWS does not benefit from greater attention. Thus, this population should be investigated in greater detail in future studies.
1. Achenbach TM, Rescorla LA, Ivanova MY. International epidemiology of child and adolescent psychopathology I: diagnoses, dimensions, and conceptual issues. J Am Acad Child Adolesc Psychiatry
2. Polanczyk GV, Salum GA, Sugaya LS, et al. Annual research review: a meta-analysis of the worldwide prevalence of mental disorders in children and adolescents. J Child Psychol Psychiatry
3. Rescorla L, Ivanova MY, Achenbach TM, et al. International epidemiology of child and adolescent psychopathology ii: integration and applications of dimensional findings from 44 societies. J Am Acad Child Adolesc Psychiatry
4. McMillen JC, Zima BT, Scott LD Jr, et al. Prevalence of psychiatric disorders among older youths in the foster care system. J Am Acad Child Adolesc Psychiatry
5. Ford T, Vostanis P, Meltzer H, et al. Psychiatric disorder among British children looked after by local authorities: comparison with children living in private households. Br J Psychiatry
6. Rosenfeld AA, Pilowsky DJ, Fine P, et al. Foster care: an update. J Am Acad Child Adolesc Psychiatry
7. Bailleau G, Trespeux F. Les bénéficiaires de l’aide sociale des départements en 2006. Document de travail, Série statistiques DREES
8. Munro E, Manful E. Safeguarding children: a comparison of England's data with that of Australia, Norway and the United States. 2012.
9. Lehmann S, Havik OE, Havik T, et al. Mental disorders in foster children: a study of prevalence, comorbidity and risk factors. Child Adolesc Psychiatry Ment Health
10. Bronsard G, Lancon C, Loundou A, et al. Prevalence rate of DSM mental disorders among adolescents living in residential group homes of the French child welfare system. Child Youth Serv Rev
11. McCann JB, James A, Wilson S, et al. Prevalence of psychiatric disorders in young people in the care system. BMJ
12. Oswald SH, Heil K, Goldbeck L. History of maltreatment and mental health problems in foster children: a review of the literature. J Pediatr Psychol
13. Blower A, Addo A, Hodgson J, et al. Mental health of looked after children: a needs assessment. Clin Child Psychol Psychiatry
14. Garland AF, Hough RL, McCabe KM, et al. Prevalence of psychiatric disorders in youths across five sectors of care. J Am Acad Child Adolesc Psychiatry
15. Schmid M, Goldbeck L, Nuetzel J, et al. Prevalence of mental disorders among adolescents in German youth welfare institutions. Child Adolesc Psychiatry Ment Health
16. Bronsard G, Lancon C, Loundou A, et al. Quality of life and mental disorders of adolescents living in French residential group homes. Child Welfare
17. Burns BJ, Phillips SD, Wagner HR, et al. Mental health need and access to mental health services by youths involved with child welfare: a national survey. J Am Acad Child Adolesc Psychiatry
18. Levitt JM. Identification of mental health service need among youth in child welfare. Child Welfare
19. Pecora PJ, Jensen PS, Romanelli LH, et al. Mental health services for children placed in foster care: an overview of current challenges. Child Welfare
20. Yampolskaya S, Chuang E. Effects of mental health disorders on the risk of juvenile justice system involvement and recidivism among children placed in out-of-home care. Am J Orthopsychiatry
21. Fang X, Brown DS, Florence CS, et al. The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse Negl
22. Scott S, Knapp M, Henderson J, et al. Financial cost of social exclusion: follow up study of antisocial children into adulthood. BMJ
23. von Elm E, Altman DG, Egger M, et al. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ
24. DerSimonian R, Laird N. Meta-analysis in clinical trials. Controlled Clin Trials
25. Higgins J, Thompson S, Deeks J, et al. Statistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practice. J Health Serv Res Policy
26. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ
27. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Introduction to Meta-Analysis
. Wiley. 2008.
28. Ambrosini PJ. Historical development and present status of the schedule for affective disorders and schizophrenia for school-age children (K-SADS). J Am Acad Child Adolesc Psychiatry
29. Breton JJ, Bergeron L, Valla JP, et al. Diagnostic Interview Schedule for Children (DISC-2.25) in Quebec: reliability findings in light of the MECA study. J Am Acad Child Adolesc Psychiatry
30. Chambers WJ, Puig-Antich J, Hirsch M, et al. The assessment of affective disorders in children and adolescents by semistructured interview. Test-retest reliability of the schedule for affective disorders and schizophrenia for school-age children, present episode version. Arch Gen Psychiatry
31. Döpfner M, Lehmkuhl G. Diagnostik-System für Psychische Störungen im Kindes- und Ju-gendalter nach ICD-10 und DSM-IV (DISYPS-KJ), (2. korrigierte und ergänzte Auflage). Bern: Huber. 2000
32. Kaufman J, Birmaher B, Brent D, et al. Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime Version (K-SADS-PL): initial reliability and validity data. J Am Acad Child Adolesc Psychiatry
33. Robins LN, Helzer JE, Croughan J, et al. National Institute of Mental Health Diagnostic Interview Schedule. Its history, characteristics, and validity. Arch Gen Psychiatry
34. Shaffer D, Fisher P, Lucas CP, et al. NIMH Diagnostic Interview Schedule for Children Version IV (NIMH DISC-IV): description, differences from previous versions, and reliability of some common diagnoses. J Am Acad Child Adolesc Psychiatry
35. Vandenbroucke JP, von Elm E, Altman DG, et al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Ann Intern Med
36. Collins PY, Patel V, Joestl SS, et al. Grand challenges in global mental health. Nature
37. Birmaher B, Brent D, Bernet W, et al. Practice parameter for the assessment and treatment of children and adolescents with depressive disorders. J Am Acad Child Adolesc Psychiatry
38. Bowlby J. The making and breaking of affectional bonds. I. Aetiology and psychopathology in the light of attachment theory. An expanded version of the Fiftieth Maudsley Lecture, delivered before the Royal College of Psychiatrists, 19 November 1976. Br J Psychiatry
39. Bowlby J. The making and breaking of affectional bonds. II. Some principles of psychotherapy. The fiftieth Maudsley Lecture. Br J Psychiatry
40. Bowlby J. Developmental psychiatry comes of age. Am J Psychiatry
41. Cassidy J, Jones JD, Shaver PR. Contributions of attachment theory and research: a framework for future research, translation, and policy. Dev Psychopathol
42. Mikulincer M, Shaver PR. An attachment perspective on psychopathology. World Psychiatry
43. Bruskas D. Developmental health of infants and children subsequent to foster care. J Child Adolesc Psychiatr Nurs
44. Kessler RC, Avenevoli S, Costello EJ, et al. Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Comorbidity Survey Replication Adolescent Supplement. Arch Gen Psychiatry
45. Merikangas KR, Nakamura EF, Kessler RC. Epidemiology of mental disorders in children and adolescents. Dialogues Clin Neurosci
46. Brinkman WB, Epstein JN, Auinger P, et al. Association of attention-deficit/hyperactivity disorder and conduct disorder with early tobacco and alcohol use. Drug Alcohol Depend
47. Chamberlain P, Leve LD, Smith DK. Preventing Behavior Problems and Health-risking Behaviors in Girls in Foster Care. Int J Behav Consult Ther
48. Mannuzza S, Klein RG, Bessler A, et al. Adult outcome of hyperactive boys. Educational achievement, occupational rank, and psychiatric status. Arch Gen Psychiatry
49. Doria GM, Antoniuk SA, Assumpcao Junior FB, et al. Delinquency and association with behavioral disorders and substance abuse. Rev Assoc Med Bras
50. Firdion JM. Influence des événements de jeunesse et héritage social au sein de la population des utilisateurs des services d’aide aux sans-domicile. Économie et Statistiques
51. Duburcq A, Coulomb S, Bonte J, et al. Enquête de prévalence sur les troubles psychiatriques en milieu carcéral. Rapport final. Etude pour le ministère de la santé (DGS) et le ministère de la Justice
52. Crum RM, Green KM, Storr CL, et al. Depressed mood in childhood and subsequent alcohol use through adolescence and young adulthood. Arch Gen Psychiatry
53. Marmorstein NR, White HR, Loeber R, et al. Anxiety as a predictor of age at first use of substances and progression to substance use problems among boys. J Abnorm Child Psychol
54. Groh AM, Roisman GI, van Ijzendoorn MH, et al. The significance of insecure and disorganized attachment for children's internalizing symptoms: a meta-analytic study. Child Dev
55. Polanczyk GV, Willcutt EG, Salum GA, et al. ADHD prevalence estimates across three decades: an updated systematic review and meta-regression analysis. Int J Epidemiol
56. Steel Z, Marnane C, Iranpour C, et al. The global prevalence of common mental disorders: a systematic review and meta-analysis 1980-2013. Int J Epidemiol
57. Browne K, Hamilton-Giachritsis C, Johnson R, et al. Overuse of institutional care for children in Europe. BMJ
58. Fisher PA, Chamberlain P, Leve LD. Improving the lives of foster children through evidenced-based interventions. Vulnerable Child Youth Stud
60. Romano E, Tremblay RE, Vitaro F, et al. Sex and informant effects on diagnostic comorbidity in an adolescent community sample. Can J Psychiatry
61. Fazel M, Wheeler J, Danesh J. Prevalence of serious mental disorder in 7000 refugees resettled in western countries: a systematic review. Lancet
62. Fazel S, Khosla V, Doll H, et al. The prevalence of mental disorders among the homeless in western countries: systematic review and meta-regression analysis. PLoS Med
63. Adornetto C, Suppiger A, In-Albon T, et al. Concordances and discrepancies between ICD-10 and DSM-IV criteria for anxiety disorders in childhood and adolescence. Child Adolesc Psychiatry Ment Health
64. Romano E, Tremblay RE, Vitaro F, et al. Prevalence of psychiatric diagnoses and the role of perceived impairment: findings from an adolescent community sample. J Child Psychol Psychiatry
65. Costello EJ, Egger H, Angold A. 10-year research update review: the epidemiology of child and adolescent psychiatric disorders: I. Methods and public health burden. J Am Acad Child Adolesc Psychiatry
66. Fisher PA, Mannering AM, Van Scoyoc A, et al. A translational neuroscience perspective on the importance of reducing placement instability among foster children. Child Welfare
67. Romanelli LH, Landsverk J, Levitt JM, et al. Best practices for mental health in child welfare: screening, assessment, and treatment guidelines. Child Welfare
Supplemental Digital Content
Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.