Waters H, Goyal R, Strouss L, Legacy S, Nagar S, Wilbanks J. Regional differences in HEDIS measure results for schizophrenia treatment adherence in state Medicaid programs. Poster presented at the 2016 AMCP Managed Care & Specialty Pharmacy Annual Meeting; April 19, 2016. San Francisco, CA. [abstract] J Manag Care Pharm. 2016 Apr; 22(4-a):S62.

BACKGROUND: Following methodology for the Healthcare Effectiveness Data and Information Set (HEDIS) measure of 'Adherence to Antipsychotic Medications for Individuals with Schizophrenia (SAA)', this study estimated the proportion of Medicaid beneficiaries with schizophrenia who are adherent to antipsychotics (APs) across state Medicaid programs.

OBJECTIVE: To examine variation in AP adherence and identify predictors of improvement in adherence and decreased inpatient utilization and cost. METHODS: Analyses utilized claims data from 25 state Medicaid programs from 2006 to 2010. Patients were aged 19-64 years, with ICD-9-CM diagnosis code of 295.xx (excluding 295.4x and 295.7x) on ≥ 1 inpatient or ≥ 2 outpatient claims at least 30 days apart. Adherence was analyzed separately for each measurement year using proportion of days covered (PDC) methodology. Patients with a PDC ≥ 0.80 were considered adherent. Multivariable logistic regression was used to assess predictors of nonadherence using patient-level data for mea - surement year 2010. In addition, we analyzed state-level aggregated data (n = 25) for patients with ≥ 12 months of follow-up available after the schizophrenia diagnosis to assess predictors of inpatient admis - sions and cost.

Between 19,500 and 29,000 patients were included in the analysis across the study period. Results showed an increasing trend in adherence to APs between 2006 and 2010. The average PDC for antipsychotic medications for all 25 states combined was 0.76 in 2006, which increased to 0.80 in 2010. The percentage of adherent patients for all 25 states combined increased from 58.3% in 2006 to 64.4% in 2010. Regression analyses indicated that initiation of long-acting injectable (LAI) AP, and percentage of residents with at least a high school (HS) education, were predictive of improved adherence, while being African American or Hispanic (versus White) and being younger were predictive of poorer adherence. The aggregated state-level analy ses showed that with each one-month delay in starting an LAI, there was a 1.6% increase in inpatient admissions and a 7% increase in hospital costs. Also, with each 1% decrease in the proportion of a state’s population with at least a HS diploma, inpatient admissions and hospital costs increased by 1% and 8%, respectively.

CONCLUSIONS: Overall, substantial variations were observed in terms of adherence to AP among patients with schizophrenia across Medicaid programs. LAI initiation and higher levels of education were predictive of improved adherence and decreased inpatient utilization and cost.

Share on: