Comorbid mental disorders: implications for treatment and sample selection.

Disorders from the Diagnostic and Statistical Manual of Mental Disorders (3rd ed., rev.; American Psychiatric Association, 1987) were assessed in a birth cohort of 961 young adults. Comorbid cases exceeded single-disordered cases in chronic history of mental illness, use of treatments, physical health problems, functional interference in daily life, and impaired adaptation across domains such as work, education, health, and social-support networks. Single-disorder cases were also more impaired than nondisordered cases, but comorbid cases were the most severely impaired. Our findings suggest that (a) samples that underrepresent comorbidity (pure single-disorder cases or student samples) will underestimate effect sizes for relations between a disorder and its correlates, whereas samples that overrepresent comorbidity (clinical or adjudicated samples) will overestimate effect sizes, (b) comorbidity is accompanied by complications that challenge treatment planning, compliance, and coordination of service delivery, and (c) comorbidity is associated with physical, educational, and economic problems that make it a broad societal concern.