Decreased drug reporting in a cross-sectional student drug use survey.

PURPOSE: Prior longitudinal cohort studies have suggested that reinterviews about drug use often lead to decreased reports of lifetime substance use (recanting). Respondents may edit their answers on reinterview because of perceptions regarding question threat. Since reinterviews usually occur after long periods of time, the influence of inadequate recall (memory), cannot be ruled out. In order to evaluate the relative importance of editing and memory on recanting, we examined a cross-sectional survey administered in 1993 to a probability sample of Illinois students who were in the 7th through 12th grade.
METHODS: Two sets of self-administered survey questions assessed drug use: the I-SAY drug-use questionnaire, and a supplemental questionnaire asked at the end of the main survey. We compared the rates of "new use" (no use for a drug reported in the I-SAY, use reported for a drug on the supplement) with rates of recanting (use for a drug reported on the I-SAY, no use for a drug reported on the supplement).
RESULTS: Recanting was generally more pronounced than new use, especially for cocaine, heroin, marijuana, and inhalants. Those classified as light or inconsistent users on the I-SAY were significantly more likely to recant their drug use reports.