The Underuse of Overuse Research

Abstract

Tremendous progress has been made in the development of practical quality of care measures that are now applied in a variety of health care settings. In this issue of HSR, Swaminathan et al. use Healthcare Effectiveness Data and Information Set (HEDIS) measures to examine the stability of health maintenance organization quality ratings over time, and Scanlon et al. use these measures to assess if market competition improves quality. These two important research questions could not have been addressed without established and tested quality measures, and a strong foundation of quality of care research. While this foundation is laudable, we would like to draw attention to a conspicuous problem for most nationally available measures of quality, i.e. they focus primarily on the underuse and misuse of health services, not their overuse. Underuse refers to the lack of provision of necessary or needed care (e.g., withholding aspirin in a patient with coronary artery disease). Misuse of care refers to the provision of wrong care (e.g., prescribing a subtherapeutic dose of a medication). Overuse, however, refers to providing unnecessary care (e.g., echocardiogram in a young healthy patient) or care where the risk exceeds the benefits (e.g., carotid endarterectomy in an otherwise healthy patient with asymptomatic carotid disease). Unnecessary use of health care services is not just a cost issue but also a health care safety problem as patients may suffer complications from procedures that they did not need in the first place. Nevertheless, among the 27 HEDIS Effectiveness of Care measures proposed for Medicare in 2009, almost all address underuse or misuse of health services. Recent years have seen the funding and advancement of a research agenda designed to develop the tools necessary to assess the quality of care. While there is still much more work that needs to be done, it is notable that most recent advances in characterizing and measuring quality focus on underuse of health care services (McGlynn et al. 2003). While overuse had garnered significant research attention in 1980s and 1990s (Brook et al. 1986; Brook, Park, and Chassin 1990), it has been supplanted with population-based concerns about high versus low utilization in different geographic areas or high versus low utilization physicians (Hayward et al. 1994; Wennberg and Fisher 2002; Fisher et al. 2003; Baker, Fisher, and Wennberg 2008). However, there is little evidence that conclusively finds that higher utilization rates in an area indicate overuse. To the contrary, past research that identified inappropriate use of procedures found no evidence that it occurred more often in high versus low utilization/cost areas (Chassin et al. 1987; Leape et al. 1990). Despite the important cost, quality, and safety implications of overuse, this branch of quality research has moved from the center stage over a decade ago to the periphery today. We believe that this is a result of research, cultural, and political challenges.