Purpose: To provide preliminary prevalence estimates of common DSM-IV (Diagnostic and Statistical Manual of Mental Disorders-4th Edition) disorders in a sample of Hong Kong Chinese adolescents.

Methods: 541 Chinese adolescents were recruited from Grades 7, 8 and 9 of 28 mainstream high schools in Hong Kong (mean age=13.8 years; SD=1.2). The adolescents and their parents were separately administered the Youth and Parent versions of DISC-IV (Diagnostic Interview Schedule for Children-Version 4), respectively.

Results: Based upon both symptom and impairment criteria, as required by DSM-IV, the overall prevalence estimate of DSM-IV disorders in our sample of Chinese adolescents was 16.4%. Estimates for such individual disorders/diagnostic groupings as anxiety disorders, depressive disorders, attention deficit/hyperactivity disorder (ADHD), oppositional defiant disorder (ODD), conduct disorder (CD), and substance use disorders were 6.9, 1.3, 3.9, 6.8, 1.7, and 1.1%, respectively. These rates were largely compatible with those reported in previous studies with perhaps lower rates of generalized anxiety disorder (GAD), depressive disorders, CD, and substance use disorders, but a higher rate of ODD. The rate of ADHD was somewhat higher, but this might reflect the current DSM-IV diagnostic practice. The rate of anxiety disorders was not as high as predicted from some previous questionnaire surveys. The application of an impairment criterion had discernible impacts on prevalence estimates, greater on anxiety and substance use disorders, but smaller on depressive and disruptive behavior disorders. There was a lack of gender difference in rates of ODD and CD.

Discussion and conclusion: While the findings reported here are broadly compatible with those of other studies, there may be cross-cultural differences in rates of some individual disorders, e.g., GAD, depressive disorders, ODD, CD, and substance use disorders, as well as in gender difference regarding rates of ODD and CD. However, exact comparison between studies is confounded by methodological differences in sample characteristics, measures, and case definition. Standardization of methodology in epidemiological surveys should allow more precise identification of any within- or between-culture variations in prevalence estimation.