Although several studies have examined the relationship between symptoms of depression or psychological distress and medication adherence, this is the first published study of HIV antiretroviral adherence and its correlates among persons diagnosed with serious mental illness. Forty-five of 47 (96%) participants completed a two-week study in which their adherence to antiretroviral medication was measured using electronic monitoring caps. Mean adherence (proportion of prescribed doses taken) was 66% (SD=34). There were several correlates (p<0.05) of adherence among background and medical characteristics, physical symptoms and side effects, cognitive and psychosocial functioning, and treatment-related attitudes and beliefs. However, in a forward stepwise regression, attendance at recent clinical appointments was the sole predictor that entered the model--accounting for 49% of the variance in adherence. Using attendance at recent clinic appointments as the criterion, adherence readiness (90+% adherence to antiretrovirals) was correctly determined for 72% of the sample. Although not sufficient to serve as the basis for treatment decision making, review of appointment-keeping records may provide clinicians with a simple, cost-effective method for predicting adherence to ongoing treatment, as well as for evaluating adherence readiness to inform the decision of whether to prescribe or defer treatment.