To assess the relative cost-effectiveness of lower versus higher cost prize-based contingency management (CM) treatments for cocaine abuse. Cost-effectiveness analyses based on resource utilization, unit costs and outcomes from a previous CM efficacy trial. Two community-based treatment centers. Patients (n = 120) enrolled in out-patient treatment for cocaine abuse. Random assignment to one of three 12-week treatment conditions: standard treatment (STD) alone or two variants of STD combined with prize based CM. In CM, drawing for prizes was available to those submitting drug-free urine samples and completing goal-related activities. There were two levels of pay-out (referred to as $80 versus $240) based on the potential value of prizes won. Costs per participant associated with counseling utilization, urine and breathalyzer testing, and operation of the prize-drawing procedure were derived from a survey conducted at 16 clinics that had participated in CM studies. The three measures of effectiveness were: (1) longest duration of consecutive abstinence; (2) percentage completing treatment; and (3) percentage of samples drug-free. The higher magnitude CM produced outcomes at a lower per unit cost than did the lower magnitude prize CM treatment. This was the case for all three outcome measures examined and held across various assumptions in the sensitivity analysis. Cost-effectiveness analyses can inform policy decisions regarding selection of one treatment model over another. Decisions on adoption of new evidence-based treatments would be aided by more information on society's willingness to pay for incremental gains in effectiveness.