A remarkable consensus is developing: the way physicians are organized and reimbursed in the United States must change. A leading example is the Medicare Shared Savings Program (MSSP) establishment of accountable care organizations (ACOs), groups of providers who assume collective responsibility for the spending and quality of a defined beneficiary population. The objective of this current project is to synthesize quasi-experimental design, novel data linkages, and primary survey methodology to examine the formation, effects, and mechanisms through which ACOs on spending and quality. We will accomplish this in three aims. First, we will identify key market factors that facilitate or inhibit ACO market penetration as represented by: (1) the presence or absence ACOs in a market, (2) the total number of ACOs, and (3) the share of beneficiaries covered by ACOs. We will examine these relationships for ACOs with both one- and two-sided risk models. Second, will assess the impact of ACOs on spending and quality via instrumental variable analysis. We will instrument for endogenous ACO participation by exploiting exogenous variation in the ?differential supply? of ACO vs. non-ACO physicians surrounding Medicare patients. Third, we will examine frontline physicians? attitudes and perspectives of ACOs through an in-depth survey of the Physician Organization of Michigan (POM) ACO, the largest ACO in the State of Michigan and the tenth largest in the country. Our survey will be one of the first and potentially largest examination of frontline physicians? attitudes and perspectives of ACOs, a perspective largely missing from policy narratives. We posit that ACOs will lead to modest in spending without corresponding decreases in quality but that individual physician uncertainty about ACO tasks and incentives will weaken collective incentives and harm the achievement of shared objectives. We expect our findings to inform Medicare in designing future generations of alternative payment models that are both more effective and more readily extended across the broader universe of Medicare physicians and patients.
Confronted by an increasingly expensive and fragmented health care system, public and private payers have established a series of reforms designed to hold physicians accountable for lower cost, better-coordinated care. Of these, none has drawn more attention than Medicare?s accountable care organizations (ACOs), where groups of providers assume collective responsibility for the spending and quality of a defined beneficiary population. While preliminary studies suggest that ACOs were associated with modest improvements in spending and quality, it remains uncertain whether many markets contain the infrastructure and provider networks necessary to extend such delivery reorganizations to the broader universe of Medicare physicians. Moreover, prior evaluations of ACOs have relied on techniques that may not adequately address self-selection into ACOs and thus may not fully disentangle the effects of ACOs from other unobserved differences between patients and physicians. Finally, current research has not identified how such ACO?s incentives are transmitted to and mediated by individual physicians. The proposed research will combine quasi-experimental techniques, survey methods, and novel data linkages to estimate if ACOs work, why, and whether they hold potential for broader systematic reform. First, we will identify key market factors that facilitate or inhibit formation and market penetration of one- and two-sided MSSP ACOs. Second, we will assess the impact of ACOs on spending and quality via instrumental variable analysis that directly addresses nonrandom selection into ACOs. Finally, we will evaluate the mechanisms through which ACOs? shared accountability and collective incentives are transmitted to and mediated by frontline physicians. We will examine frontline physicians? attitudes and perspectives of ACOs through an in-depth survey of one of the largest ACOs in the country, surveying thousands of individuals physicians and integrating those data with information gleaned from our econometric analyses. We expect our findings to be of immediate interest to decision makers at Medicare, as our findings will help guide the ongoing development and implementation of alternative payment models nationwide. PUBLIC HEALTH RELEVANCE: The current approach of separately paying physicians, hospitals, and other providers for discrete medical services has fragmented care, inflated spending, and harmed patient health. The proposed research is relevant to public health because it will improve our understanding of how collective incentives and shared accountability affect physician behavior, health care spending, and ultimately patient health. This information is particularly relevant given widespread hope that payment reform will incentivize health care systems to one day assume responsibility for population health.
