• Narrow time window. In general, most performance measures assess care at only one point in time. The committee recommends measuring the quality, costs, and outcomes of care over a longer time frame. Doing so will necessitate further development of longitudinal measures that can capture the performance of multiple providers caring for a patient; examine how well care is provided across transitions to different settings (e.g., hospital to nursing home); and, most important, evaluate patient outcomes over time. The committee believes that focusing on chronic illness, care across time and locations, and clinical and functional outcomes will move performance measurement much closer to a patient-centered perspective.

  • Provider-centric focus. Current measures tend to focus on specific settings of care, so that measure specifications are applicable to only one setting, such as a physician office or hospital. The committee recommends moving toward individual patient-level measurement, even for the starter set of measures, because of the markedly increased value and flexibility offered by this approach. This approach to data collection would allow for aggregation along three important dimensions: (1) composite measures that can document whether a patient received all recommended services for a given condition within a specified time window; (2) population-based measures, whose aggregation for defined strata of the population on the basis of socioeconomic status, race, and ethnicity would allow for assessment of disparities in treatment at the provider, system, or community level; and (3) systems-level measures that can characterize the overall performance of an organization or entity across conditions and service lines, and can better identify gaps in performance and foster accountability at each level of care, from the individual clinician to the community. The committee strongly recommends that data collection protocols be planned and implemented to support such reporting.

  • Narrow focus of accountability. The committee endorses the principle of shared accountability among all providers involved in a patient’s care. This strategy represents perhaps the most significant explicit departure from a traditional guideline for selecting performance measures—that a responsible entity or person be known at the outset. The committee believes that shared accountability can be achieved both by reporting specific measures that are not uniquely the responsibility of a single provider (e.g., care transitions) and by aggregating patient-level measures. In short, the committee recommends that certain important aspects of care be measured even when no single entity can be held accountable for the results. Qualities such as population mortality rates, efficiency through time, chronic disease complication rates, and measures of oversupply of services may be among the most important of these aspects from the viewpoint of patients and society. Left unmeasured, they are certain to be left unaddressed.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement