Role of ethnicity in predicting antipsychotic medication adherence

BACKGROUND: Clinicians treating schizophrenia face increasingly diverse ethnic populations. Ethnic groups may have different approaches to the management of schizophrenia, which could impact antipsychotic medication adherence. OBJECTIVE: To examine the association between adherence and ethnicity or the specific medication used after controlling for other factors. METHODS: Texas Medicaid claims were retrieved for persons aged 21-65 years, diagnosed with schizophrenia or schizoaffective disorder, after initiating treatment with olanzapine (n = 1875), risperidone (n = 982), or haloperidol (n = 726) between January 1997 and August 1998. The association between ethnicity (African American, Mexican American, white) or medication and days' use of the medication in the year following initiation was assessed using multivariate linear regression. Covariates included other patient demographics, region, comorbid mental health conditions, and prior medication and healthcare resource use. RESULTS: African American and Mexican American patients were significantly less adherent than white patients (19 d less, p < 0.001 for African Americans; 18 d less, p = 0.003 for Mexican Americans). For patients of all ethnicities, olanzapine was associated with 23 more adherent days than risperidone and 55 more adherent days than haloperidol (p < 0.001 for each comparison). CONCLUSIONS: When other factors were controlled for, ethnicity was a significant predictor of medication adherence following initiation on an antipsychotic medication, and patients of all ethnicities were most adherent when taking olanzapine, less adherent when taking risperidone, and least adherent when taking haloperidol.