Comparison of the PSC-17 and alternative mental health screens in an at-risk primary care sample.

OBJECTIVE To validate the 17-item version of the Pediatric Symptom Checklist (PSC-17) as a screen for common pediatric mental disorders in primary care. METHOD Patients were 269 children and adolescents (8-15 years old) whose parents completed the PSC-17 in primary care waiting rooms. Children were later assessed using the Schedule for Affective Disorders and Schizophrenia for School-Age Children-Present and Lifetime version (K-SADS-PL). The PSC-17's subscales were compared with K-SADS-PL diagnoses and measures of anxiety, depression, general psychopathology, functioning, and impairment. RESULTS In receiver operating characteristics analyses, the PSC-17 subscales performed as well as competing screens (Child Depression Inventory, the parent and child Screens for Child Anxiety-Related Disorders) and Child Behavior Checklist subscales (Aggressive, Anxious-Depressed, Attention, Externalizing, Internalizing, and Total) in predicting diagnoses of attention-deficit/hyperactivity disorder, externalizing disorders, and depression (area under the curve > or =0.80). The instrument was less successful with anxiety (area under the curve = 0.68). None of the screens were highly sensitive, many were insensitive, and all would have low positive predictive value in low-risk primary care populations. CONCLUSIONS The PSC-17 and its subscales are briefer than alternative questionnaires, but performed as well as those instruments in detecting common mental disorders in primary care. Continued research is needed to develop brief yet sensitive assessment instruments appropriate for primary care.