Do accountable care organizations affect race mediated differences in cancer screening?

Abstract

131 Background: Accountable care organizations (ACO) reflect a payment innovation aimed to orient incentives to improve quality and reduce waste. The objective of this study was to determine whether ACO enrollment affects racial disparities in cancer screening, and to characterize the impact on the appropriateness of prostate cancer screening in minority populations. Methods: We built a cohort of Medicare beneficiaries from 2007 to 2013 were comprising a cohort of 11,087,056 person-years among ACO beneficiaries and 37,187,979 person-years among non-ACO beneficiaries. A difference-in-difference-in-differences (DDD) approach was utilized to identify the effect of ACO enrollment on cancer screening in racial/ethnic minorities relative to non-Hispanic whites. We then characterized differences in screening appropriateness after ACO enrollment using age (65-74 vs. 75+) and predicted survival (top vs. bottom quartile). Results: ACO enrollment was associated with approximately a 5% reduction in prostate cancer overscreening for white beneficiaries, namely among the elderly and those with unfavorable predicted survival. Compared to white men in the lowest quartile of predicted survival, Asian and Hispanic men with similarly low survival had a 4.8% and 13.0% relative increase in prostate cancer screening associated with ACO enrollment (DDD p = 0.015, p = 0.011, respectively). Prostate cancer overscreening was common among Asian men, with 46% of elderly Asian men attributed to an ACO undergoing cancer screening compared to 28% of elderly white men. Furthermore, ACO enrollment was associated with a 2.7% increase in screening relative to whites (DDD p = 0.0005). Compared to white beneficiaries, black men had consistently lower rates of prostate cancer screening. ACO enrollment did not narrow the disparity of prostate cancer screening between healthier (DDD p = 0.75) or younger (DDD p = 0.27) black and white beneficiaries. Conclusions: This study provides evidence of ACO-mediated increases in low-value screening for prostate cancer among sick and elderly Hispanics and Asians. Furthermore, ACO enrollment did not narrow known disparities in high-value prostate cancer screening among healthier and younger black men.