Accounting for Person- vs Neighborhood-Level Social Risk in Quality Measurement | Health Policy | JAMA Health Forum | JAMA Network

O ver the past 2 decades, the US government has steadily shifted care for Medicare beneficiaries into value-based payment models. Central to their success is the ability to measure and pay fairly for performance on quality measures, including clinical outcome measures. While there is broad consensus that outcome measures, such as age and comorbidities, should be adjusted for clinical risk, there is considerable controversy regarding whether they should be adjusted for social risk. Proponents of social risk adjustment argue that failure to account for social factors that are independently associated with health outcomes leads to fundamentally inaccurate characterization of clinicians’ performance and inappropriately penalizes those caring for historically marginalized patients.1 Opponents of social risk adjustment argue that it obscures disparities, reduces incentives to invest in efforts that improve equity, and allows for substandard care to persist.2 Within this context, the National

O ver the past 2 decades, the US government has steadily shifted care for Medicare beneficiaries into value-based payment models. Central to their success is the ability to measure and pay fairly for performance on quality measures, including clinical outcome measures. While there is broad consensus that outcome measures, such as age and comorbidities, should be adjusted for clinical risk, there is considerable controversy regarding whether they should be adjusted for social risk. Proponents of social risk adjustment argue that failure to account for social factors that are independently assoc