Exploring Medicare Provider Networks: Implications for Adoption of CER Findings

点击翻译按钮获取中文摘要

Abstract If comparative effectiveness research (CER) is to reduce health expenditure growth without compromising health outcomes, providers will need to adopt long-term cost-saving strategies that are identified as being no less effective than higher-cost clinical strategies. Understanding how provider "virtual" networks for Medicare influence the cost and content of care may be important when considering health reforms designed to promote the use of lower-cost effective medical care. We propose identifying factors that determine or influence the use of cost-saving yet effective clinical strategies. We further propose investigating the influence of Medicare provider networks (virtual or social networks) and their structure on the cost and content (e.g., evidence-based practices or not) of care provided. Our specific aims are to: 1) determine the extent to which supply and demand factors predict provider adoption of cost-saving evidence-based findings, 2) analyze provider network structure for differences in and influence on the use of evidence-based practices and costs of care, 3) estimate potential Medicare expenditure savings based on policies implementing these findings (perhaps through the use of payment or health reforms such as bundled payments or ACOs). We will base our analysis on the complete Medicare claims files for Parts A and B from years 2005 to 2010 and Part D from 2006 to 2010. Specifically, we will apply these aims to the treatment of patients with chronic stable angina and the use of revascularization with coronary stents before and after evidence-based findings (the COURAGE trial in 2007) were published supporting the initial use of medical management. We will identify the relevant cohort from the Medicare claims files for the time period before and after the time of publication, and identify factors that predicted a change in provider practices from revascularization with stents to medical management in the data. We will control for demographic and patient-level characteristics, investigate ecological and supply determinants, and investigate differences across geographies and patient sub-populations such as women, minorities, and those with diabetes. Next, we will identify provider networks of care for chronic stable angina in the Medicare data, and explore whether provider network structure influences the cost or content of care provided (concordant with evidence-based practices or not). We will explore whether network structures differ across geographies, and whether this partially explains the difference in practice variation and cost between regions. We will identify ways that our findings might be integrated into health reform policies either to optimize the factors we identify as determinants of low-cost yet effective care or to encourage efficient provider network formation. Finally, we will estimate potential Medicare savings based upon these policies.