Fixing Health Care's Value Deficit: The Problem of Unexplained Geographic Variation (Part 1)

03/05/2010

---

By BRUCE CUTTER, MD Simply living in different locations can explain wide differences in cost and quality of care. We are all witness to a massive value deficit—watching in alarm as health care costs inexorably climb out of control and quality stagnates. A primary cause of this deficit is unexplained geographic variation in costs, where, for the same level of care, one region varies substantially from another. Geographic variation, though, is finally attracting the attention it deserves, and stimulating important discussion of potential solutions.   This article, appearing in two parts, reviews the evidence for unexplained variation, outlines a potential solution to the problem, and discusses potential business opportunities. Studies going back over a decade, particularly those of Elliot Fisher and colleagues at the Dartmouth Atlas, lay out a convincing case for the presence of considerable unexplained and unwarranted variation in health care costs and quality in the US, and the lack of correlation of cost with quality. In 1999, the Dartmouth Atlas reported on geographic variations in cost and quality in various regions of the United States.[1] This work was extended in a publication in 2003 in the Annals of Internal Medicine.[2] In a recent New England Journal of Medicine ("NEJM") Perspective, Sutherland et al update the Dartmouth Atlas data and provide evidence that most of the variation in cost—over 70 percent—in different geographic areas is unexplained by any factor other than geography. Health status, poverty, income and other factors explain less than 30 percent of the variation.[3] When it comes to much of health care—and thereby health care costs—“geography is destiny”. These publications paint an impressive picture of an uneven health care system in costs and quality. Despite a system that is supposed to feature “evidence-based care” as a bedrock principle, simply living in a different locations can explain widely different care. As a result, many experts now highlight unexplained variation as a key driver of health care's value deficit—that is, worsening quality at higher costs.   Atul Gawande captures the essence of the Dartmouth Atlas findings in a recent article in the New Yorker. He bases his comments on visits and interviews with various health care professionals in two Texas towns, which have dramatically different health care costs—differences that cannot be explained other than by geography.[4] In another NEJM Perspective, Fisher et al again review their data documenting unexplained variations in cost and quality and illustrate how decreasing variation and adopting the practices of low cost providers would by itself go a long way towards slowing the growth in health care costs.[5] Specifically, the authors estimate that based on the Medicare trustees’ report on projected revenues and Part A and B spending, Medicare will be $660 billion in the red by 2023. If annual spending growth (per capita) were reduced from 3.5 percent (the national average) to 2.4 percent (the rate in San Francisco, one of the lower but certainly not the lowest cost area), the Medicare estimated balance would be a positive $758 billion. The cumulative change is $1.42 trillion. Clearly, relatively small absolute changes (although not so small in percentage terms) can have a major impact over time on health care transformation and sustainability. Explanations What is driving this unexplained variation? And what, if anything, can we do about it? Understanding the origin of the problem provides us the opportunity to develop solutions. An article published in Health Affairs provides a window onto the issue.[6] The authors construct a series of hypothetical, standardized clinical vignettes and present them to a variety of primary care physicians from different geographic areas in order to see what degree of variation exists, and attempt to understand the drivers of variation. They design the vignettes to address discretionary decision-making about medical interventions. The results are consistent with the Dartmouth Atlas data. When it comes to interventions, with strong supporting evidence, there is not much difference between high and low-spending regions when recommending specific clinical interventions. When it comes to gray areas, which are abundant in medicine, the authors discover that regional variation in health care spending strongly correlates with local physicians' tendency to recommend discretionary interventions.   Patients are seen more frequently, more screening tests of uncertain benefit ordered, and more resource-intense interventions recommended. There are more visits, more referrals, more hospital admissions, more intensive care unit admissions, less discussion of palliative care issues, and so on. Variation, they conclude, is likely the result of local medical culture differences. The 2006 and 2008 editions of the Dartmouth Atlas both document wide and unwarranted geographic variation in the quality and amount of health care delivered to Medicare recipients with severe chronic illness. The analysis features care delivered in academic medical centers, including those institutions that, by reputation, are among the most prestigious institutions in the country.[7] It exposes substantial variation in a variety of measures of care intensity including physician visits, hospitalizations, and use of intensive care that depended on local “culture”, a local medical “signature”. The Atlas also provides data documenting what its authors have termed “supply sensitive care”. This study shows strong correlation between intensity of care in hospitals and regions and the supply of medical resources. In other words, available hospital beds lead to hospital admissions, physician supply stimulates physician visits, more CT scanners lead to more scans, and so on. Although patient level research is not available, the Dartmouth Atlas researchers show no meaningful correlation between greater utilization of medical care and better outcomes at population levels. The picture emerging illustrates vast unexplained and unwarranted geographic differences in health care intensity and spending—differences not correlated with quality and outcomes, and driven by availability of resources and local medical cultures. Understanding that medical care is largely physician-driven, we can conclude that change efforts must take into account and modify local physician practices (or physician “culture”) if we can ever hope to alter the game fundamentally. How to accomplish this under national health care reform is no small challenge. Physician culture “Culture” is a complex topic, and detailed consideration is beyond the scope of this article. It cannot be ignored however, and is likely key to successful transformation of the US health care system. Culture, by definition, is the complex interaction of factors both intrinsic and extrinsic to the individual, and, in medicine at least, significantly impacted by geography and also by training and specialty. If we are going to fundamentally change the game in medicine, we will need to understand and address the importance of culture. The studies referenced above would indicate that when it comes to variation in practice patterns and spending, action and culture are largely local. Solutions will need to incorporate this. Having the evidence at hand—documented unexplained geographic variation in health care costs, the factors driving this, and the importance of this unexplained variation in the quality, efficiency and cost of health care—we should now be able to address two fundamental questions: Are there real-world examples that have the potential to “change the game”? Can we decrease unexplained variation, and do so in a way that will reverse health care's value deficit—better care quality at less cost? The Dartmouth Atlas, for one, does not offer, at the level of an individual or group of patients, the right treatment or the greatest value—it just documents at a population level the impressive variation in costs and quality that exist along with the lack of correlation between them. How, then, can we match quality/outcomes/costs and assess value at every level, from the individual patient to the population? And can we drive these changes in a reasonable time frame? Medicine is broken, and needs a revolution to fix it.[8] The current non-system is unsustainable and unacceptable both in terms of quality and costs. The tremendous unexplained geographic variation in health care costs, and associated lack of bang for our buck not only signify our problems, but also point to potential solutions. We need solutions that: effectively address and deal with the problem of unexplained variation maintain/enhance quality and outcomes while controlling costs embrace and make a reality the Institute of Medicine’s principles of quality are applicable to the real world of medicine as it is in fact organized and delivered, and critically, have the ability to change provider and payer culture and the game itself Experience and data in medical oncology, an expensive, emotional, and complicated area of medicine, suggest a potential solution, one that meets all the above criteria and has the potential to “change the game”. The operational term for this solution is clinical pathways. Properly developed and enacted, with appropriate provider accountability and compensation model change, pathways have the potential to be an important part of the solution.   Part Two will consider how clinical pathways could drive the transformation of oncology in a way that is evidence-based, patient-centered, and value-focused. Developed and executed correctly, pathways can be an important solution to the value deficit in medicine.   [1] Wennberg JE, Cooper MM, eds. The quality of medical care in the United States: a report on the Medicare program: the Dartmouth Atlas of Health Care. Hanover, NH: Dartmouth Medical School, 1999 (Accessed February 10th, 2010 at http://dartmouthatlas.org/atlases/99Atlas.pdf) [2] Fisher ES, Wennberg DE, Stukel TA, Gottlieb DJ, Lucas FL, Pinder EL. The implications of regional variations in Medicare spending. 2. Health outcomes and satisfaction with care. Ann Intern Med 2003;138:288-98 [3] Sutherland JM, Fisher ES, Skinner JS. Getting Past Denial-The High Cost of Health Care in the United States. N Eng J Med 2009; 361:1227-1230 [4] Atul Gawande. The Cost Conundrum: What a Texas town can teach us about health care. The New Yorker, Annals of Medicine. June 1st, 2009 [5] Fisher ES, Bynum JP, Skinner JS. Slowing the Growth of Health Care Costs-Lessons from Regional Variation. N Eng J Med 2009; 360: 849-855 [6] Sirovich B, Gallagher PM, Wennberg DE, Fisher ES. Discretionary Decision Making By Primary Care Physicians And The Cost of U.S. Health Care. Health Affairs 2008; 27: 813-823 [7] Tracking the Care of Patients with Severe Chronic Illness: The Dartmouth Atlas of Health Care, 2008 edition [8] Neubauer MA, Hoverman J, Kolodziej M et al. Cost-Effectiveness of Evidence-Based Treatment Guidelines for the Treatment of Non-Small-Cell Lung Cancer in the Community Setting. Journal of Oncology Practice 2010; 6: 1-7 ☆ ☆ ☆ Stephen S. S. Hyde is the author of "Cured! The Insider’s Handbook for Health Care Reform". He is a former public company CEO and board member of numerous health care companies. A successful entrepreneur, consultant, actuary, insurance expert, medical group CEO, federal managed care regulator, and Harvard MBA, he has nearly 40 years of experience in virtually every major aspect of health regulation, insurance, and care delivery. He is also the author of Prescription Drugs for Half Price or Less. Read Stephen’s latest comments at HydeonHealthCare.com. © Lyceum Associates, Inc.  All rights reserved.

Please contribute additional comments and thoughts to the GatherSmart® Forum