OBJECTIVE Laboratory studies in which drugs of abuse are self- or experimenter-administered to non-treatment-seeking research volunteers provide valuable data about new pharmacotherapies for substance use disorders, as well as behavioral and performance data for understanding the neurobiology of drug abuse. This paper analyzed follow-up data from six smoked cocaine self-administration laboratory studies, in order to determine whether changes in substance use occurred 1 and 3 months after study participation compared to pre-study baseline. METHODS Ninety-eight healthy, non-treatment-seeking cocaine users were admitted to inpatient and combined inpatient/outpatient studies lasting from 12 to 105 days. The studies allowed participants to self-administer repeated doses of smoked cocaine (0, 6, 12, 25, and/or 50mg per dose) on multiple occasions. Participants returned for follow-up at 1 and 3 months, at which time self-reported consumption of cocaine, alcohol, marijuana, and nicotine was assessed. RESULTS Compared to baseline ($374.04/week, S.D. $350.09), cocaine use significantly decreased at 1 month ($165.13/week, S.D. $165.56) and 3 months ($118.59/week, S.D. $110.48) after study participation (p<0.001; results based on the 39 participants who completed all 3 time points). This decrease was not accompanied by a change in other drug use, e.g., a compensatory increase in alcohol, marijuana or nicotine use. CONCLUSIONS Study participation was not associated with increased post-study cocaine, alcohol, marijuana, or nicotine use. Thus, human laboratory models of cocaine self-administration, conducted in non-treatment-seeking research volunteers, are relatively safe, and study participation does not exacerbate ongoing drug use.