Frequency of 5+/4+ drinks as a screener for drug use and drug-use disorders.

OBJECTIVE: The objective of this study was to test the ability of a question on frequency of drinking 5+ (for men) or 4+ (for women) drinks to screen for drug use and drug-use disorders (DUDs) in a general population sample.
METHOD: Using data collected in 2001-2002 from a representative U.S. adult population sample (N= 43,093), including a subsample of those with past-year emergency-department use (n = 8,525), past-year frequency of drinking 5+/4+ drinks was evaluated as a screener for drug use and DUDs for four categories of illicit drugs.
RESULTS: Sensitivities and specificities of the 5+/4+ drinks screener were 72.4% and 76.6% for any drug dependence, 71.9% and 77.3% for any DUD, and 63.3% and 78.9% for any drug use in the general population. Sensitivities and specificities were higher for marijuana and cocaine/crack and lowest for illicit prescription drugs. Optimal screening cut-points were once a month or more for cocaine/crack dependence, either once or more a month or seven or more times a year for cocaine/crack DUDs, seven or more times a year for cocaine/crack use, and once or more a year for the other drug use and DUD measures. Sensitivity and specificity were similar among adults who had visited an emergency department in the past year, and the optimal screening cutpoints were identical.
CONCLUSIONS: Past-year frequency of drinking 5+/4+ drinks was quite accurate as a screener for past-year marijuana and cocaine/crack use and DUDs, but it was less accurate for illicit prescription drug use and DUDs. Its drug-screening potential can be thought of as "added value" from an item already likely to be asked in the interest of detecting problem drinking. Future work may consider using the alcohol consumption screener as a starting point, with follow-up questions to assess illicit drug use among those who screen positive.