Antipsychotic adherence over time among patients receiving treatment for schizophrenia: a retrospective review.

OBJECTIVE: Approximately 40% of patients with schizophrenia are poorly adherent to their antipsychotics at any given time. However, little is known about patients' adherence over time, although this has important services implications. We examined antipsychotic adherence over 4 years at the aggregate and the individual level among a large cohort of patients.

METHOD: We identified 34,128 Veterans Affairs patients who received a schizophrenia diagnosis and an antipsychotic fill in fiscal year (FY) 1999, completed schizophrenia visits in each of the next 4 years (FY2000, FY2001, FY2002, FY2003), and had valid medication possession ratios (MPRs) in each of these years. We examined whether patients had consistently good adherence (MPRs >or= 0.8 in all 4 years), consistently poor adherence (MPRs < 0.8 in all years), or inconsistent adherence. We examined predictors of consistently poor or inconsistent adherence.

RESULTS: The cross-sectional prevalence of poor adherence among the patient population remained stable over time; 36%-37% were poorly adherent in each year. However, 61% of patients had adherence difficulties at some point over the 4-year period. Approximately 18% had consistently poor adherence, 43% were inconsistently adherent, and 39% had consistently good adherence. Patients who were younger and nonwhite, with a substance use diagnosis, a psychiatric hospitalization, or predominant treatment with first-generation antipsychotics, were more likely to have consistently poor adherence.

CONCLUSIONS: Antipsychotic adherence is not a stable trait; most patients have difficulties with adherence over time. Health organizations and clinicians must emphasize adherence-enhancing interventions that can be provided on a longer term basis to the majority of patients.