OBJECTIVE: Psychiatric advance directives allow patients with severe mental illness to document their preferences for particular medications. This study investigated the role of psychiatric advance directives in treatment choice and medication adherence. METHODS: A total of 123 persons with severe mental illness recorded medication preferences in psychiatric advance directives. The authors compared medication preferences to prescribed medications over 12 months, determined concordance between preferred and prescribed medications, and examined the effect of concordance on medication adherence at 12 months. RESULTS: Participants requested a median of two medications in their psychiatric advance directives (range from zero to six) and refused a median of one medication (range from zero to ten). Between baseline and follow-up there was a 27% increase in the number of medications prescribed that had been requested on the psychiatric advance directive (Wilcoxon matched pairs, p<.001). After correction for the number of medications listed in the psychiatric advance directive, a 10% increase in concordance remained significant (p<.001). Being prescribed at least one medication requested in the psychiatric advance directive predicted higher medication adherence at 12 months, after the analysis controlled for relevant covariates (odds ratio=7.8, 95% confidence interval=1.8-34.0). CONCLUSIONS: Providing information about medication preferences in psychiatric advance directives may increase prescribing of patient-preferred medications even in noncrisis settings. Patients who were prescribed medications that they requested in advance were significantly more likely to adhere to medications, supporting the benefit of patient participation in medication choice. Psychiatric advance directives appear to be a clinically useful conduit for communicating patient medication preferences.