To evaluate quality and the effect of pay for performance among minority patient groups, during a pay-for-performance program in 22 primary care practice sites.Methods:
Data were collected on 26 standardized measures of care for 2 measurement cycles. Proportions of recommended care received across 5 composite quality domains were analyzed by demographic group. Regression models including significant covariates were constructed. Adjusted odds ratios (ORs) were derived to assess the effect of pay of performance within demographic groups.Results:
Improvements were observed from 2007 to 2009 for all patients in each of 5 composite quality domains of diabetes, coronary artery disease, heart failure, screening and prevention, and all care. With the exception of heart failure care for Hispanic/Latino and Spanish language–preferring patients, improvement was observed in all domains for African American/black race, Hispanic/Latino ethnicity, and Spanish language–preferred groups. Following adjustment for covariates, pay for performance was associated with significant improvement in all-patient diabetes care (adjusted OR = 1.15; [95% confidence interval [CI], 1.09–1.22), screening and prevention (adjusted OR = 1.55; 95% CI, 1.41–1.69), and all care (adjusted OR = 1.27; 95% CI, 1.20–1.35). Significant improvements were also observed within the minority demographic groups noted earlier.Conclusions:
Pay-for-performance programs structured as additional incentive monies for providers improved care for all patients and among minority groups, in whom disparities have historically been observed.