Accountable Care Organizations (ACOs) are rapidly developing and have the goals of improving quality of care and containing health care costs. Previous research on ACOs has been conducted piecemeal, relying mainly on narrow studies of individual ACO pilots, demonstrations, and experiments, or on pre-ACO data because ACO were not yet formed. Health care providers and policy makers lack understanding of what drives ACO development, and what kind of ACO structures and competencies (alone or in combination) are important for achieving ACO goals. This study will provide new insights that can have a near-term effect on refining ongoing ACO development. Given its broad national scope and use of a pre-/post-ACO design, the proposed study fills many gaps in existing knowledge.
Specific aims for the proposed study include quantitative analyses of drivers conducive to ACO development across the country. Development of a taxonomy of hospitals participating in the Centers for Medicaid and Medicare Services (CMS) Medicare Shared Savings Program (MSSP) and the Pioneer ACO Program. Finally, the study will evaluate the effects of changes in hospital ACO structures and competencies on changes in hospital quality of care and costs for Medicare patients over the period 2008-2013. First, a novel database of all existing hospitals participating in the federally-designated CMS ACO programs will be developed by identifying and comprehensively searching for information on these ACOs. The study focuses on hospitals given their key role in the continuum of care and ACO implementation, as well as persistent concerns about the quality and costs of hospital care. Second, external (e.g., social, market, and geographic) and internal (e.g., hospital organizational) drivers conducive to ACO development will be linked in and evaluated. This latter analysis not only will provide insights on why particular hospitals are participating in ACO but also may provide information critical to addressing potential endogeneity issues in the longitudinal quality and cost analyses that will be conducted. Third, hospitals participating in CMS ACOs will be systematically analyzed through use of cluster analytic techniques to develop the first hospital-based ACO taxonomy. Fourth, a series of analyses of hospital involvement in ACOs and their related competencies will follow comparing hospitals in the pre-ACO and the post-ACO periods with ACO measures relative to non-ACO hospitals. We will link in Healthcare Cost and Utilization Project State Inpatient Databases (HCUP SID) to develop quality of care measures (based on inpatient mortality, adverse events, preventable hospitalizations, 30-day hospital readmissions, and 30-day mortality) and measures related to hospital costs for Medicare patients. Finally, we will estimate multivariate difference-in-difference and fixed effects models to achieve the specific aims of the project for estimating changes in hospital quality and cost outcomes.
The proposed project presents a unique opportunity to provide empirical evidence based on the experiences of hospitals participating in the federally-designated Centers for Medicaid and Medicare Services (CMS) Medicare Shared Savings Program (MSSP) and the Pioneer Accountable Care Organization (ACO) Program. The project will develop a novel database of all hospital participants of CMS ACOs;evaluate the external and internal drivers conducive to ACO development;and analyze the effects of changes in a broad range of hospital-related ACO features and competencies on changes in hospital quality of care and costs for Medicare patients.