Quality of care, health care costs, and utilization among Medicare Part D enrollees with and without low-income subsidy

Abstract

The objective of this cross-sectional, retrospective, claims-based analysis was to evaluate disease-specific quality measures, use of acceptable therapies, and health care cost and utilization among Medicare Advantage Part D (MAPD) enrollees overall and by income/subsidy eligibility status. Individuals aged ≥65 years with evidence of ≥1 of 8 common conditions and continuously enrolled in a MAPD plan throughout 2007 were assigned to low-income/dually eligible (LI/DE) or non-LI/DE cohorts. Quality of care metrics were calculated for asthma, chronic obstructive pulmonary disease (COPD), diabetes, and new episode depression. Persistence (proportion with percentage of days covered ≥80%), compliance (proportion with medication possession ratio ≥80%), health care costs, and utilization metrics were assessed by condition. All measures were evaluated for calendar year 2007. Bivariate comparisons were made between all LI/DE and non-LI/DE subgroups. A total of 183,213 patients were included. Metrics showed deficiencies in quality of care overall but generally favored non-LI/DE patients. The proportion of patients filling acceptable medication was suboptimal for most conditions, ranging from 40% to 96% across conditions and cohorts, with COPD the lowest and heart failure (HF) the highest. LI/DE patients were significantly more likely than non-LI/DE patients to fill acceptable therapy in each disease group (P<0.001) except HF. Percentages persistent and compliant with acceptable therapies were lowest for asthma and COPD, and highest for HF; percentages were generally higher among LI/DE patients. Mean disease-specific health care costs ranged from $345 (hyperlipidemia) to $2086 (HF) and were significantly higher for LI/DE than for non-LI/DE enrollees (P<0.001) for all diseases except coronary artery disease and HF. Overall, quality indicators, use of acceptable medications, and persistence/compliance metrics were suboptimal. Quality metrics favored non-LI/DE patients but medication metrics favored LI/DE patients. With an aging population and increasing health care costs, the deficits identified highlight the need for comprehensive strategies to improve clinical and economic outcomes across diseases.