This article was originally posted on Health Affairs. Read the full article here.
While value-based purchasing (VBP) programs have proliferated in the past two decades, quality measures have become increasingly important as they are employed to calculate the “value” of care. The health care industry, however, is struggling with a quality measurement dilemma. On the one hand, as the number of measures is increasing rapidly, health care organizations—both payers and providers—are investing significant resources and time in data collection and reporting of measure results. One recent study reported that a single hospital spent more than $5 million, along with an additional expenditure of more than half a million dollars in vendor fees, for the preparation and report of 162 unique quality metrics in a single calendar year. On the other hand, despite these efforts and investment, recent evidence of quality improvement is mixed. As seen in the case of chronic conditions such as diabetes, despite the adoption of numerous discrete diabetes care measures, diabetes-related outcomes such as lower extremity amputation are still far from being optimal. This raises questions about whether we are measuring and incentivizing the right drivers of health. According to the Centers for Medicare and Medicaid Services’ (CMS’s) Innovation Center’s synthesis review of their 30 payment models launched in the past decade, most of the models did not demonstrate consistent and significant improvements in quality.
In response, CMS recently announced a new strategy, the Universal Foundation, to scale back to a core set of just about twenty quality measures and align them across programs. While this approach is laudable, there are already questions about an excessive focus on disease-focused measures and plan-level measures, applicability to different specialties, testing of new models, and their alignment between programs.
The central question is: What is the most crucial aspect of “value” of care that we can measure for value-based arrangements without needing an excessive number of quality measures or risking too few? Michael Chernew and Mary Beth Landrum suggested a targeted supplemental data collection approach, in which only low performers of a core set of metrics would be required to provide additional data on other measures, reducing costs while avoiding gaps in a core measure set. Michael McWilliams proposed other strategies that focus on physicians’ intrinsic motivation and professionalism, without necessarily relying on quality measures. In this commentary, we propose a third approach: measuring what matters most to patients from a whole health perspective.