The Concurrent Validity of the Problem Oriented Screening Instrument for Teenagers (POSIT) Substance Use/Abuse Subscale in Adolescent Patients in an Urban Federally Qualified Health Center

Abstract

Background

The Problem Oriented Screening Instrument for Teenagers (POSIT) substance use/abuse subscale has been validated with high school students, adolescents with criminal justice involvement, and adolescent substance use treatment samples using the Diagnostic and Statistical Manual of Mental Disorders (DSM)-III-R and DSM-IV. This study examines the concurrent validity of the POSIT's standard 17-item substance use/abuse subscale and a revised, shorter 11-item version using DSM-5 substance use disorder diagnoses.

Methods

Adolescents (N = 525; 93% African American, 55% female) 12–17 years of age awaiting primary care appointments at a Federally Qualified Health Center in Baltimore, Maryland completed the 17-item POSIT substance use/abuse subscale and items from a modified World Mental Health Composite International Diagnostic Interview corresponding to DSM-5 alcohol use disorder (AUD) and cannabis use disorder (CUD). Receiver operating characteristic curves, sensitivities, and specificities were examined with DSM-5 AUD, CUD, and a diagnosis of either or both disorders for the standard and revised subscales using risk cutoffs of either 1 or 2 POSIT “yes” responses.

Results

For the 17-item subscale, sensitivities were generally high using either cutoff (range: 0.79–1.00), although a cutoff of 1 was superior (sensitivities were 1.00 for AUD, CUD, and for either disorder). Specificities were also high using either cutoff (range: 0.81–0.95) but were higher using a cutoff of 2. For the 11-item subscale, a cutoff of 1 yielded higher sensitivities than a cutoff of 2 (ranges for 1 and 2: 0.96–1.00 and 0.79–0.86, respectively). Specificities for this subscale were higher using a cutoff of 2 (ranges for 1 and 2: 0.82–0.89 and 0.89–0.96, respectively).

Conclusions

Findings suggest that the POSIT's substance use/abuse subscale is a potentially useful tool for screening adolescents in primary care for AUD or CUD using a cutoff of 1 or 2. The briefer, revised subscale may be preferable to the standard subscale in busy pediatric practices.