OBJECTIVES: Unlike many other health care organizations, central Texas veterans do not have day supply limits on medications. Thus, the potential for medication over-adherence exists. The objectives of this study were to: 1) examine the adherence rate (AR) and medication possession ratio (MPR) among veterans taking antipsychotics; 2) determine whether over-adherence (AR > 1 or MPR > 1) exists among veterans taking antipsychotics; and 3) evaluate factors related to over-adherence. METHODS: Data were extracted from the Central Texas Veterans Health Care System (CTVHCS) from September 1995 - October 2002 for continuously enrolled adult outpatients who were on monotherapy, and had at least two prescriptions within a year. Demographic and relevant clinical information, including age, gender, ethnicity, psychiatric diagnoses, and antipsychotic prescription medication information were collected. Subjects were followed up to 12 months. RESULTS: Among the eligible patients (N = 3252), the mean AR was 1.56 (sd = 2.58) and MPR was 1.23 (sd = 0.75). Over one-half (54.4%) of the subjects were over-adherent with ARs and MPRs >1. For those over-adherent patients, the mean AR was 2.38 (sd = 3.27) and MPR was 1.70 (sd = 0.71). Multiple logistic analyses revealed that schizophrenic patients (OR = 1.241, p = 0.0085) were significantly more likely to be over-adherent (MPR > 1), while nonwhite (OR = 0.617, p <0.0001), depressed patients (OR = 0.801, p = 0.0085) and atypical users (OR = 0.620, p <0.0001) were significantly less likely to be over-adherent. A second logistic analysis to assess individual antipsychotic agents showed that, compared to haloperidol, quetiapine users (OR = 2.309, p = 0.0073) were significantly more likely to be over-adherent, and olanzapine (OR = 0.557, p<0.0001) users were significantly less likely to be over-adherent. CONCLUSION: Over one-half of central Texas veterans were over-adherent with their antipsychotic therapy. Further studies are warranted to determine the impact of excessive medication possession on patient outcomes as well as the economic impact on the VA system.