The Epidemiology of Child and Adolescent Mental Disorders

STEPHEN L. BUKA, MICHAEL MONUTEAUX, and FELTON EARLSI Harvard School of Public Health, Boston, MA 02115


Considerable scientific ground has been covered regarding the epidemiologic study of child and adolescent psychopathology in the past twenty years. Most notably, much more is known about the measurement, community study, prevalence and risk factors for psychiatric disorders, particularly those of older children and adolescents. This was not the case in 1980 when our earlier review noted ''there are few published accounts of investigations which attempt to replicate and refine existing techniques in populations different from [the ones in which they] were originally developed'' (Earls, 1980a). It was then noted that among the largest and most influential epidemiologic studies in child psychiatry there was little overlap in terms of definition of psychopathology, sources of information (parents, teachers or children), measurement instruments, or procedures for case definition. This is clearly not the case in 2002, where common nosology, instruments, study designs, and statistical procedures have been developed, implemented, and replicated. This chapter reviews some of the major accomplishments and milestones in our field over the past twenty years, while identifying notable challenges that remain.

Evidence that the mental health of children and adolescents is an area warranting continued scientific and public health attention is well documented, most recently by the Office of the U.S. Surgeon General (USDHHS, 1999, 2001), the World Health Organization and the National Institute of Mental Health (NIMH, 2001). Epidemiologic data suggest that roughly 20% of children ages 1 -18 are in need of mental health services in the United States, half of which have mental illnesses severe enough to cause some level of impairment (Burns et al., 1995; Shaffer et al., 1996). Despite the high prevalence among children and adolescents, only about 5-7% have received services for their disability (NIMH, 2001). This number varies greatly by the child's age with 2% of preschoolers, 6-8% of children

Textbook in Psychiatric Epidemiology, Second Edition, Edited by Ming T. Tsuang and Mauricio Tohen. ISBN 0-471-40974-X © 2002 John Wiley & Sons, Inc.

ages 6-11 and 8-9% of adolescents ages 12-17 receiving services (NIMH, 2001) This unmet need for services has raised concerns as children with emotional and behavior problems are at increased risk for dropping out of school, being in trouble with the law and having an overall lower quality of life. Furthermore, mental disorders, particularly if left untreated, are likely to persist into adulthood. According to the 2001 Surgeon Generals Conference on Children's Mental Health, about 74% of 21-year olds with mental disorders had prior mental health problems.

In addition to these social costs, there is a high fiscal burden associated with child and adolescent mental disorders. A study in California found that 8.1% of hospital discharges for children ages 6-12 were for mental illness and that these children accounted for close to 90,000 days of hospitalization and 85 million dollars in hospital charges in 1992 (Chabra, Chavez, and Harris, 1999). A national estimate of child mental health expenditures was recently produced by Sturm and colleagues (2000). In 1998, the total treatment expenditures for child and adolescent mental health were 11.75 billion dollars including inpatient, outpatient, and medication costs. This, too, varied by age group with adolescents accounting for roughly 60% of the total expenditures. Unfortunately, the negative impact of poor child and adolescent mental health is felt worldwide and is likely to get worse. Based on recent data, the WHO predicts that by the year 2020, childhood neuropsychiatric disorders will become one of the leading causes of morbidity, mortality, and disability among children worldwide (USDHHS, 1999).

This summary of the global burden associated with child and adolescent mental disorder underscores the need for and utility of epidemiologic methods and data. Epidemiologic investigations in child psychiatry have been described as having three major purposes (Kellam and Ensminger, 1980). For public health planning purposes, epidemiology provides critical information regarding the prevalence of disorders, service utilization, treatment outcome, and costs. As epidemiology is the study of the distribution, determinants and causes of disease in human populations, a second purpose of epidemiologic investigations is to advance understand -ing of the origins and course of child psychiatric illness and disorder. This includes understanding the significance of early circumstances, both biological and social, that contribute to the etiology and progression of psychiatric diseases. Such data ideally lead to the third major application of epidemiology, the design and assessment of preventive and treatment interventions.

This chapter has been designed to provide to provide an introduction to and overview of some of the major developments in the epidemiologic study of child and adolescent mental disorders of the past twenty years. We summarize several of the current measures available for use with community samples, major community studies of the past decade, resulting information on the prevalence of and risk factors for child and adolescent disorders, and some concluding observations on future directions.


As noted in our previous review (Earls, 1980a), epidemiologic research in child psychiatry has generally assumed either a clinical or statistical approach to diagnosis and classification. Clinically oriented studies assume that cases of psychiatric disorder existing in the general population are broadly similar to declared cases in a clinic population (Ingham and Miller, 1976) and that the same tools that guide clinical diagnosis should be applied in epidemiologic studies. Such an orientation was applied in the classic Isle of Wight study, in which a series of reliability and validity studies based on the capacity of parent and child interview methods to select cases (children attending a psychiatric facility) from controls preceded application of this technique to a general population (Rutter and Graham, 1968). This clinical orientation has been accelerated by the refinement of diagnostic classification systems in adult and child psychiatry, most notably the publication of the modern Diagnostic and Statistical Manuals for the Classification of Mental Disorders (DSM-III, IIIR, and IV), beginning in 1980.

An alternative approach to measurement and classification derives from studies originating from a basic social science orientation, in which previous clinical practice and conditions are not taken as the starting point for definitional purposes. "The definition of disorder in these studies is arrived at as a result of carrying out a survey in a nonclinical population and not on a priori assumptions of what constitutes a psychiatric disorder'' (Earls, 1980a, p. 8). This approach has yielded several self-report scales and symptom checklist inventories administered to parents, teachers, and youth, that query about a host of problem behaviors that may be indicative of underlying psychopathology. Such measures have been used both to generate categorical diagnoses and provide continuous, dimensional values for constructs such as depression, conduct problems, and the like. Below, we review some of the most commonly used measures of child and adolescent psychopathology in contemporary epidemiologic studies. These can be divided into two categories according to the quantitative approach utilized: categorical assessments and dimensional assessments.

0 0

Post a comment