Shaping cocaine abstinence by successive approximation.

Cocaine-using methadone-maintenance patients were randomized to standard contingency management (abstinence group, n = 49) or to a contingency designed to increase contact with reinforcers (shaping group, n = 46). For 8 weeks, both groups earned escalating-value vouchers based on thrice-weekly urinalyses: The abstinence group earned vouchers for cocaine-negative urines only; the shaping group earned vouchers for each urine specimen with a 25% or more decrease in cocaine metabolite (first 3 weeks) and then for negative urines only (last 5 weeks). Cocaine use was lower in the shaping group, but only in the last 5 weeks, when the response requirement was identical. Thus, the shaping contingency appeared to better prepare patients for abstinence. A 2nd phase of the study showed that abstinence induced by escalating-value vouchers can be maintained by a nonescalating schedule, suggesting that contingency management can be practical as a maintenance treatment.