Adverse events among patients in a behavioral treatment trial for heroin and cocaine dependence: effects of age, race, and gender.

Safety monitoring is a critical element of clinical trials evaluating treatment for substance dependence, but is complicated by participants' high levels of medical and psychiatric comorbidity. This paper describes AEs reported in a large (N = 286), 29-week outpatient study of behavioral interventions for heroin and cocaine dependence in methadone-maintained outpatients. A total of 884 AEs were reported (3.1 per patient, 0.12 per patient-week), the most common being infections (26.8%), gastrointestinal (20.5%), musculoskeletal (12.3%), and general (10%) disorders. Serious AEs were uncommon (1.6% of total). Female participants reported significantly higher rates of AEs (incidence density ratio, IDR = 1.38, p < 0.0001); lower rates of AEs were reported by African Americans (IDR = 0.73, p<0.0001) and participants over age 40 reported lower rates of AEs (IDR = 0.84, p = 0.0095). AE incidence was not associated with the study intervention or with psychiatric comorbidity. Further work is needed to adapt AE coding systems for behavioral trials for substance dependence; the standard Medical Dictionary for Regulatory Activities, International Federation of Pharmaceutical Manufacturers Associations (MedDRA) coding system used in this report did not contain a separate category for one of the most common types of AE, dental problems. Nonetheless, the data reported here should help provide a context in which investigators and IRBs can interpret the patterns of AEs they encounter.