| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
|
|
|
|||||||
|
|||||||||
TRACK to:
|
|
Electronic letters published:
![[Read Comment]](/icons/misc/sectionDOWN.gif){kind=icon}
Rather than Deemphasize Measures, Measure What's Important
Framing the quality discussion
Measuring the hard way
Quality measures may require more time.
|
|
|||
|
Arnold S. Milstein MD, MPH, San Francisco, California Medical Director, Pacific Business Group on Health; and Mercer Health & Benefits
Send response to journal:
|
As a spokesperson for health care's consumer and purchaser "customers," I agree with many of the well-argued points about current performance measures made by Drs. Werner and Asch. Teaching to an overly narrow test is perhaps the biggest danger. However I believe that patients will be much better served by rapidly extending performance measures to "what's important" than by reducing the importance of measures. What's important to health care customers is (1) clinical outcomes (i.e., maintenance or improvement of their physical and mental functioning), (2) experiential outcomes (i.e., their subjective experience of care processes), and (3) financial outcomes (i.e., the individual and purchaser spending for clinical care). Version 1.0 physician performance measures for the latter two domains are alive and well, but their adoption and vigorous use by most providers and insurers is the exception rather than the rule. The vision of a highly performance sensitive environment around American physicians is far from realized. Clinical outcome measures remain severely underdeveloped. With the admirable exception of a handful of medical specialty societies (e.g., cardiologists, CV surgeons, intensivists, transplant surgeons, and cystic fibrosis specialists) and a few provider organizations (e.g., Mayo, Intermountain Healthcare), most physicians have been content to not know what impact their services are having on the physical and mental functioning of their patients. Many years after their peer reviewed publication, how many health care providers use "How's Your Health," the SF-12, or other such instruments to track their patients' health–related functional status? Nor have national physician organizations invested their considerable national political clout to reprioritize federal research spending from the biomedical sciences to the evaluative sciences on which better clinical outcome measurement depends. If both forms of neglect by physician leaders cease, we can begin to measure and reward what IS important, rather than limit ourselves to articulating widely acknowledged limitations in current measures. We should not be too harsh on our temporary reliance on process measures. We can minimize the risk of teaching to the test by expanding to over 100 the number of guideline-based process measures that could be produced immediately with insurance claims and enrollment data (cf E. Steinberg, "Physician Clinical Performance Measurement," presented at the NQF Implementing Measures for Ambulatory Care Conference, 12/18/06, Washington, DC). Would better performance on process measures make a difference? The Rand Health Insurance Study (cf R. H. Brook et al, Medical Care, Vol 28, No 5, May 1990) revealed that in the judgment of independent physicians who examined a large sample of chronically ill patients, more than 30% of such patients would be functioning at substantially higher levels of physical and mental health status, if their MDs had scored better on guideline-based process measures. That's important. Arnold Milstein MD, MPH, Medical Director, Consumer-Purchaser Disclosure Project, Pacific Business Group on Health; and Mercer Health & Benefits Competing interests: I work for a consulting firm that advises many health care purchasers on their purchasing approach. |
|||
|
|
|||
|
Randall Longenecker, Bellefontaine, Ohio, United States Program Director, The Ohio State University Rural Program
Send response to journal:
|
This essay is an excellent example of how framing an issue like quality in only one way kidnaps the discussion, carrying it to an end that was never intended. As generalists, we need tools to help explain the Starfield paradox and the conventional wisdom that bigger is better and uniformity is best. This article and the one by James et al about rural vs urban hospital quality are a major contribution to the discussion around quality and I applaud the authors for their work! Competing interests: None declared |
|||
|
|
|||
|
Francois S de Brantes, Washington, DC, USA National Coordinator, Bridges To Excellence
Send response to journal:
|
The authors bring out some important and salient points about performance measurement that should be carefully considered by private and public sector purchasers engaged in developing and implementing quality- based performance incentive programs. The rush to measure should be a rush to measure right, not simply a rush to measure anything. And measuring right means focusing on measures that matter to the patient, the provider and the payer. There is a parsimonious set of measures that accomplishes this goal, for example, measuring blood pressure control in patients with diabetes, or hypertension, or coronary artery disease, or heart failure, or a number of other conditions that affect the cardiac and vascular systems. This simple act of adequately controlling blood pressure in all patients that need it would save countless lives, countless dollars and focus physicians on the management of patients as opposed to focusing on managing to the test. There are a few other important ambulatory care measures that would yield significant clinical and economic benefits, creating very clear and unambiguous accountability while maintaining an appropriate amount of autonomy to the practice of medicine. These measures, closely related to real patient outcomes, are difficult to collect in a systematic way without widespread adoption and use of health information technology tools such as electronic health records. And herein lies the dilemma. The 21st century health care system that Former Speaker Newt Gingrich so eloquently describes as delivering true patient-centered care cannot become a reality under the constant barrage of negative incentives caused by the current perverse and toxic payment models. To break away from “fee-for-volume” or “fee-for-nothing” we need quality-based performance incentive programs until a better payment model comes about. Those programs have to be focused on the right measures, and the right measures are hard to collect. So the choice is simple: focus on the right measures and force change to happen, even if it’s harder for everyone, or focus on the easy measures and watch nothing change. We’ve chosen the former. Competing interests: None declared |
|||
|
|
|||
|
Ian M Bennett, Philadelphia, USA Assistant Professor, University of Pennsylvania
Send response to journal:
|
Thanks to Werner and Asch for this timely article which brings up many important issues around clinical performance measurement. In the area of maternal and reproductive health where I spend most of my time there are many "standard practices" which have little to no evidence for benefit in outcomes and take up an inordinate amount of time. As measures are developed and applied I have hopes that these poorly supported practices will fall under greater scrutiny. If measures are limited to those areas that have an evidence base then there will be a natural weaning of providers from practices that we hold near and dear but are of little value. While I am sure that there will be fits and starts this will be an important process and will actually push us to pay attention to more valuable practices. I also have hopes that this pressure will lead to greater emphasis on the evaluation of primary care clinical practices in areas where that is less common (contraception care, maternal care, etc) and the development of relevant measures. Finally, while time is always limited in medical care I would rather spend more time and address issues of value than spend less time (as we are now) and not address them. A critical underlying premise to this process is that physicians have been missing substantial areas of care. It may be necessary to increase patient-provider time, perhaps with return visits, to address all of these issues. Of course we will need measures of performance that can account for this time dimension as well as the development of technologies to track the care needs of particular patients. Competing interests: None declared |
|||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH |
TRACK: