A two-item conjoint screen for alcohol and other drug problems.

TLDR

Nonmedical use of illicit and prescription drugs is common among American adults, yet existing substance‑use disorder screens focus only on alcohol. The study evaluates the criterion validity of a two‑item conjoint screen (TICS) for alcohol and other drug abuse or dependence in primary‑care patients. The TICS, comprising two questions about excess use and desire to cut down, was compared against DSM‑III‑R criteria via the Composite International Diagnostic Interview‑Substance Abuse Module in two random primary‑care samples. The TICS achieved roughly 80% sensitivity and specificity for current substance‑use disorders, was especially sensitive to polysubstance use, and showed consistent performance across two samples, with 0, 1, and 2 positive responses corresponding to 7.3%, 36.5%, and 72.4% disorder probabilities.

Abstract

Although nonmedical use of illicit and prescription drugs is not uncommon among American adults, the currently recommended screens for substance use disorders focus only on alcohol. This study reports on the criterion validity of a two-item conjoint screen (TICS) for alcohol and other drug abuse or dependence for a split sample of primary care patients.Two random samples of primary care patients aged 18 to 59 years responded to several screening items that emanated from a focus group process. The DSM-III-R criteria for substance use disorders, as codified by the Composite International Diagnostic Interview-Substance Abuse Module, served as the criterion standard.At least one positive response to the TICS (In the last year, have you ever drunk or used drugs more than you meant to? and Have you felt you wanted or needed to cut down on your drinking or drug use in the last year?) detected current substance use disorders with nearly 80% sensitivity and specificity. The TICS was particularly sensitive to polysubstance use disorders. Respondents who gave 0, 1, and 2 positive responses had a 7.3%, 36.5%, and 72.4% chance of a current substance use disorder, respectively; likelihood ratios were 0.27, 1.93, and 8.77. The results were consistent across split samples of 434 and 702 participants.Current alcohol or other drug problems can be detected in nearly 80% of young and middle-aged patients by asking two questions that are easily integrated into a clinical interview.