During the first year of the Medicare Physician Value-Based Payment Modifier Program, physician practices that served more socially high-risk patients had lower quality and lower costs, and practices that served more medically high-risk patients had lower quality and higher costs.

Disparities by economic status are observed in the health status and health outcomes of Medicare beneficiaries. For health services and health policy researchers, one barrier to addressing these disparities is the ability to use Medicare data to ascertain information about an individual's income level or poverty, because Medicare administrative data contains limited information about individual economic status. Information gleaned from other sources-such as the Medicaid and Supplemental Security Income programs-can be used in some cases to approximate the income of Medicare beneficiaries. However, such information is limited in its availability and applicability to all beneficiaries. Neighborhood-level measures of income can be used to infer individual-level income, but level of neighborhood aggregation impacts accuracy and usability of the data. Community-level composite measures of economic status have been shown to be associated with health and health outcomes of Medicare beneficiaries and may capture neighborhood effects that are separate from individual effects, but are not readily available in Medicare data and do not serve to replace information about individual economic status. There is no single best method of obtaining income data from Medicare files, but understanding strengths and limitations of different approaches to identifying economic status will help researchers choose the best method for their particular purpose, and help policymakers interpret studies using measures of income.

No abstract

Increasing emphasis on value in health care has spurred the development of value-based and alternative payment models. Inherent in these models are choices around program scope (broad vs. narrow); selecting absolute or relative performance targets; rewarding improvement, achievement, or both; and offering penalties, rewards, or both. We examined and classified current Medicare payment models-the Hospital Readmissions Reduction Program (HRRP), Hospital Value-Based Purchasing Program (HVBP), Hospital-Acquired Conditions Reduction Program (HACRP), Medicare Advantage Quality Star Rating program, Physician Value-Based Payment Modifier (VM) and its successor, the Merit-Based Incentive Payment System (MIPS), and the Medicare Shared Savings Program (MSSP) on these elements of program design and reviewed the literature to place findings in context. We found that current Medicare payment models vary significantly across each parameter of program design examined. For example, in terms of scope, the HRRP focuses exclusively on riskstandardized excess readmissions and the HACRP on patient safety. In contrast, HVBP includes 21 measures in five domains, including both quality and cost measures. Choices regarding penalties versus bonuses are similarly variable: HRRP and HACRP are penalty-only; HVBP, VM, and MIPS are penalty-orbonus; and the MSSP and MA quality star rating programs are largely bonus-only. Each choice has distinct pros and cons that impact program efficacy. Unfortunately, there are scant data to inform which program design choice is best. While no one approach is clearly superior to another, the variability contained within these programs provides an important opportunity for Medicare and others to learn from these undertakings and to use that knowledge to inform future policymaking. Take-Home Points • Increasing emphasis on value in health care has spurred the development of value-based and alternative payment models. Inherent in these models are choices around program scope; selecting absolute or relative performance targets; rewarding improvement, achievement, or both; and offering penalties, rewards, or both.• We examined current Medicare value-based and alternative payment models on these elements of program design and found that while current models vary significantly across each parameter, there are scant prior data to inform which choice is best.• Such variability in program design may represent an important opportunity to learn from existing models and create new ones in the future.