Risky sexual behavior is common among individuals with bipolar and substance use disorders. This 15-month prospective study examined the effects of between-subject differences and within-subject changes in mood symptoms and drug use on sexual risk behavior among 61 patients with both disorders. Participants completed five post-treatment follow-up assessments at 3-month intervals. Using a multivariate mixed-effects model analysis, more average weeks of mania (between-subject difference) was associated with greater sexual risk, but change in weeks of mania (within-subject change) was not; depression was unrelated to sexual risk. In addition, within-subject increases in days of cocaine use predicted increases in sexual risk. Results underscore the importance of substance abuse treatment and suggest that bipolar patients with active and/or recurrent mania are in need of targeted HIV prevention services. Further research is needed to test whether individual differences in impulsivity may explain the association between mania and sexual risk.