Pharmacy characteristics correlating to performance in a community pharmacy network

Abstract

Objective: To assess the correlation between pharmacy characteristics and a broad set of performance measures used to support a community pharmacy network.

Methods: Baseline characteristics regarding demographics, services provided, technology, and staffing were collected via a 68-item survey for 123 pharmacies participating in the North Carolina community pharmacy enhanced services network. Performance metric data were collected, and scores were calculated for each pharmacy. Outcome measures for this study comprised of 4 adherence measures, a risk-adjusted hospitalization measure, a risk-adjusted emergency department visit measure, a total cost of medical care measure, and a composite pharmacy performance measure. Generalized estimating equations (GEE) were used to create multivariable statistical models measuring the correlation between pharmacy characteristics and performance measures.

Results: After inclusion criteria were applied, 115 pharmacies remained in the analysis. These pharmacies were primarily single and multiple independent pharmacies, at 36.5% and 59.1%, respectively. Five characteristics were significantly associated with 3 measures, and none were associated with 4 or more. Having pharmacists in non dispensing roles was positively associated with total score, diabetes adherence, and chronic medication adherence. Home visits were positively associated with hypertension adherence, ED visit rate, and total score. Offering a smoking cessation program was positively associated with chronic medication, hypertension, and statin adherence. Offering free home delivery was positively associated with diabetes adherence but negatively associated with total medical spending and ED visit rate. Using dispensing automation was negatively associated with adherence to chronic medications, renin-angiotensin system antagonists, and statins.

Conclusion: No pharmacy characteristics were associated with a majority of performance measures chosen. Additional research is needed to identify structural variables that can be used as minimum participation criteria for high-performing pharmacy networks.