Perspective
Paying for Performance — Risks and Recommendations
N Engl J Med 2006; 355:1845-1847November 2, 2006
- Article
Fee-for-service payment and a high degree of autonomy have long been defining characteristics of physician practice in the United States. And for the past 40 years, the Medicare program has protected — and largely reinforced — this traditional professional model. But change appears to be imminent.
The past decade has brought compelling evidence of serious gaps in the quality of medical care. The increased availability of reliable measures of the technical quality of care in both hospital and ambulatory settings has proved that it is feasible to measure quality — and highlighted a remarkable variability in performance. Rising costs are threatening budgets in both the public and private sectors and the affordability of health insurance. Both public and private payers are demanding increased accountability. And many observers believe that financial incentives provide the best leverage for modifying providers' behavior.
As a result, private payers (under pressure from purchasers) and the Medicare program (under pressure from Congress) have been experimenting with approaches to rewarding improved performance. Congress recently called on the Centers for Medicare and Medicaid Services (CMS) to implement a pay-for-performance system for hospitals, and there is strong interest in expanding such programs to individual physicians.
The rush to adopt pay-for-performance programs, however, has raised a number of concerns among physicians and policy analysts. Foremost among these concerns are the following:
What is the underlying goal? Although the ostensible goal of pay-for-performance programs is to improve the quality of care, many physicians fear that “efficiency” (i.e., the cost of care) will be the only focus — a fear exacerbated by the use of such measures to select physicians for inclusion in preferred provider networks.
Are the measures adequate? A closely related concern is that current performance measures (see table
Selected Quality Measures for Ambulatory Care.)1 have serious limitations. Physicians recognize that good care — especially for the frail elderly and patients with multiple chronic illnesses — often requires a careful balancing of risks, benefits, and patients' preferences, not rigid adherence to clinical guidelines. Performance measurement at the level of the individual physician faces serious technical challenges, ranging from attribution (most patients with serious illnesses receive care from multiple physicians) to inadequate sample sizes (few physicians have enough patients for accurate measurement). And a measurement system that focuses on individual physicians or specific institutional sites (hospitals and nursing homes) risks reinforcing the fragmentation and lack of coordination that are perhaps the greatest failing of the current delivery system.Is implementation feasible? Even for hospitals and large medical groups, the investments required to collect the measures for current public-reporting initiatives are substantial. Physicians in small-office practices — still the vast majority — will have to collect the data by hand through a review of medical records or make major investments in electronic health records. Audits will be required to ensure accuracy, and the costs will be high.
Will rewards be sufficient? Whether the potential bonuses will be sufficient to compensate for the costs of data collection or to motivate change is unknown. If the pool of funds for provider payments is fixed, then a policy that creates large rewards for some could require substantial cuts in payments for others. If only physicians achieving a high level of performance receive payments, those for whom such a level is out of reach will not bother to try.
Could there be unintended consequences? Perhaps the greatest fear is that implementation of pay for performance could cause more harm than good. For instance, unless physicians are firmly convinced that risk adjustment is sufficient, they could decide that the easiest way to achieve high scores is to avoid sick or challenging patients (those who need them the most); systems serving the disadvantaged could see their revenues fall (undermining our tattered safety-net programs); and the emphasis on financial incentives could further undermine morale and the core professional value of altruism that is already threatened by the increasing commercialization of medicine.
These concerns, among others, are discussed in a report recently released by the Institute of Medicine (IOM) called “Rewarding Provider Performance: Aligning Incentives in Medicare.”2 The IOM committee, on which I served, acknowledged these challenges but strongly recommended moving forward with pay for performance as one element of a comprehensive strategy to improve the quality of care. Each of the aforementioned concerns was addressed in our recommendations.
The committee pointed to the current payment system as an important cause of the fragmentation, complexity, unreliability, and waste of the current delivery system. We called on the CMS and Congress to begin implementing pay for performance as a means to learn how to modify the payment system to foster higher performance and encourage systemwide and comprehensive improvement.
We emphasized that measures and rewards of performance should target multiple dimensions of care, initially including measures of technical quality, patient-centered care, and efficiency but moving toward longitudinal and health-outcome measures as soon as it is feasible. Because most Medicare beneficiaries receive care from multiple physicians in diverse institutional settings, the committee highlighted the need for measures and rewards that foster shared accountability and coordination of care. Given the current threats to primary care, we recommended allowing Medicare beneficiaries to identify their regular care providers, and then rewarding such providers.
Although we called on the CMS to mandate that all Medicare institutional providers participate in pay-for-performance programs as soon as possible, we recognized the serious challenges faced by physicians in small-office practices and recommended that participation by physicians be voluntary at first. Special efforts will be required to assist providers in developing data-collection systems and adopting the electronic health records required to support quality improvement, performance measurement, and payment reform.
We recognized that substantial additional funding will not be forthcoming in the current fiscal environment and called for the creation of modest funding pools derived from current payments. Annual changes in payment rates would be a likely source, so that some providers could see little or no increase in fees. All providers, however, should be able to receive performance rewards if payments are targeted to multiple dimensions of care and both improvement and achievement are rewarded.
Finally — and perhaps most important — the committee (recognizing that the evidence underlying pay for performance is weak and that unintended effects are possible) called on the federal government to implement pay for performance in the context of an effective monitoring and evaluation system that assesses early experiences (in order to identify potential harms and take corrective action), evaluates the approach's impact broadly (on all aspects of care), and identifies and disseminates information on how best to improve performance.
Unfortunately, much of the current work in performance measurement and pay for performance seems to ignore one or more of the commonsense principles outlined by the IOM committee. The technical quality measures still reflect a tiny segment of clinical practice. The efficiency measures used or under discussion target discrete diagnoses and episodes (which can make fragmented care by multiple providers appear “efficient”), rather than the longitudinal costs and outcomes of care (which would reward comprehensive, coordinated care by single providers). Much of the focus appears to remain on the measurement of individual performance of physicians — a daunting technical and administrative challenge if implemented nationally. And little attention is being devoted to designing or building a comprehensive evaluation framework that would allow us to learn from our inevitable mistakes.
The shift from autonomy to accountability and from fee-for-service practice to new methods of payment appears inevitable. Accountability for performance on the basis of evidence is now the watchword for clinical services. We would be wise to apply a similar standard to the implementation of our health policy reforms.
An interview with Dr. Bruce Weinstein, medical director for Compass Medical Group, can be heard at www.nejm.org.
An interview with Dr. Fisher and Dr. Karen Davis on pay for performance, Medicare, and the IOM report can be heard at www.nejm.org.
Source Information
Dr. Fisher is a professor of medicine at Dartmouth Medical School, Hanover, NH, and the VA Outcomes Group, White River Junction, VT.
- References
References
1
The Ambulatory Care Quality Alliance recommended starter set: clinical performance measures for ambulatory care. Rockville, MD: Agency for Healthcare Research and Quality, 2004. (Accessed October 12, 2006, at http://www.ahrq.gov/qual/aqastart.htm.)
2
Institute of Medicine. Rewarding provider performance: aligning incentives in Medicare. September 2006. (Accessed October 12, 2006, at http://www.iom.edu/CMS/3809/19805/37232.aspx.)
- Citing Articles (30)
Citing Articles
1
Vahé A. Kazandjian, Allison Lipitz-Snyderman. (2011) HIT or Miss: the application of health care information technology to managing uncertainty in clinical decision making. Journal of Evaluation in Clinical Practice 17:6, 1108-1113
CrossRef2
Adam A. Powell, Katie M. White, Melissa R. Partin, Krysten Halek, Jon B. Christianson, Brian Neil, Sylvia J. Hysong, Edwin J. Zarling, Hanna E. Bloomfield. (2011) Unintended Consequences of Implementing a National Performance Measurement System into Local Practice. Journal of General Internal Medicine
CrossRef3
Carolyn T. Thorpe, Grace E. Flood, Sally A. Kraft, Christine M. Everett, Maureen A. Smith. (2011) Effect of Patient Selection Method on Provider Group Performance Estimates. Medical Care 49:8, 780-785
CrossRef4
Karen W. Hoover, Mary Butler, Kimberly Workowski, Felix Carpio, Stephen Follansbee, Beau Gratzer, Brad Hare, Barbara Johnston, John L. Theodore, Michael Wohlfeiler, Guoyu Tao, John T. Brooks, Terence Chorba, Kathleen Irwin, Charlotte K. Kent. (2010) STD Screening of HIV-Infected MSM in HIV Clinics. Sexually Transmitted Diseases 37:12, 771-776
CrossRef5
Rodney Johnson, Chitra Dinakar. (2010) Pediatric Pay-for-performance in Asthma: Who Pays?. Current Allergy and Asthma Reports 10:6, 405-410
CrossRef6
Peter J. Pronovost, Christine A. Goeschel. (2010) Viewing Health Care Delivery as Science: Challenges, Benefits, and Policy Implications. Health Services Research 45:5p2, 1508-1522
CrossRef7
Jose M. Martin-Moreno, Jose Luis Alfonso-Sanchez, Meggan Harris, Beatriz Gonzalez Lopez-Valcarcel. (2010) The effects of the financial crisis on primary prevention of cancer. European Journal of Cancer 46:14, 2525-2533
CrossRef8
Song, Yunjie, Skinner, Jonathan, Bynum, Julie, Sutherland, Jason, Wennberg, John E., Fisher, Elliott S., . (2010) Regional Variations in Diagnostic Practices. New England Journal of Medicine 363:1, 45-53
Full Text9
James C. Stevens. (2010) Pay for Performance and the Physicians Quality Reporting Initiative in Neurologic Practice. Neurologic Clinics 28:2, 505-516
CrossRef10
Hugh Gravelle, Matt Sutton, Ada Ma. (2010) Doctor Behaviour under a Pay for Performance Contract: Treating, Cheating and Case Finding?. The Economic Journal 120:542, F129-F156
CrossRef11
J. Frank Wharam, Michael K. Paasche-Orlow, Neil J. Farber, Christine Sinsky, Lisa Rucker, Kimberly J. Rask, M. Kathleen Figaro, Clarence Braddock, Michael J. Barry, Daniel P. Sulmasy. (2009) High Quality Care and Ethical Pay-for-Performance: A Society of General Internal Medicine Policy Analysis. Journal of General Internal Medicine 24:7, 854-859
CrossRef12
Leila Khan, Scott Mincemoyer, Robert A. Gabbay. (2009) Diabetes Registries: Where We Are and Where Are We Headed?. Diabetes Technology & Therapeutics 11:4, 255-262
CrossRef13
Sherrie H. Kaplan, John L. Griffith, Lori L. Price, L Gregory Pawlson, Sheldon Greenfield. (2009) Improving the Reliability of Physician Performance Assessment. Medical Care 47:4, 378-387
CrossRef14
Kevin C. Chung, Ashwin N. Ram. (2009) Evidence-Based Medicine: The Fourth Revolution in American Medicine?. Plastic and Reconstructive Surgery 123:1, 389-398
CrossRef15
Jeffrey R Curtis, Jonathan D Adachi, Kenneth G Saag. (2009) Bridging the Osteoporosis Quality Chasm. Journal of Bone and Mineral Research 24:1, 3-7
CrossRef16
Narinder Kapur. (2009) On the pursuit of clinical excellence. Clinical Governance: An International Journal 14:1, 24-37
CrossRef17
Joseph G. Ouslander. (2008) Paying for Performance in Nursing Homes: Don't Throw the Baby Out with the Bathwater. Journal of the American Geriatrics Society 56:10, 1959-1962
CrossRef18
Karen Hoover, Guoyu Tao, Charlotte Kent. (2008) Low Rates of Both Asymptomatic Chlamydia Screening and Diagnostic Testing of Women in U.S. Outpatient Clinics. Obstetrics & Gynecology 112:4, 891-898
CrossRef19
Doran, Tim, Fullwood, Catherine, Reeves, David, Gravelle, Hugh, Roland, Martin, . (2008) Exclusion of Patients from Pay-for-Performance Targets by English Physicians. New England Journal of Medicine 359:3, 274-284
Full Text20
Karen Hoover, Guoyu Tao. (2008) Missed Opportunities for Chlamydia Screening of Young Women in the United States. Obstetrics & Gynecology 111:5, 1097-1102
CrossRef21
N. Debacker, F. Nobels, H. Vandenberghe, P. Van Crombrugge, A. Scheen, V. Van Casteren. (2008) Organization of a quality-assurance project in all Belgian multidisciplinary diabetes centres treating insulin-treated diabetes patients: 5 years’ experience. Diabetic Medicine 25:2, 179-185
CrossRef22
Leslie Spry. (2008) Building the Chronic Kidney Disease Management Team. Advances in Chronic Kidney Disease 15:1, 29-36
CrossRef23
Elliott S. Fisher. (2007) 2007 Robert and Alma Moreton Lecture: Pay for Performance: More Than Rearranging the Deck Chairs?. Journal of the American College of Radiology 4:12, 879-885
CrossRef24
David A. Johnson, . (2007) Pay for Performance: ACG Guide for Physicians. The American Journal of Gastroenterology 102:10, 2119-2122
CrossRef25
Michael Schatz, Michael Blaiss, George Green, Donald Aaronson. (2007) Pay for performance for the allergist-immunologist: Potential promise and problems. Journal of Allergy and Clinical Immunology 120:4, 769-775
CrossRef26
KAREN DAVIS, STUART GUTERMAN. (2007) Rewarding Excellence and Efficiency in Medicare Payments. The Milbank Quarterly 85:3, 449-468
CrossRef27
D. Seidenwurm, P. Turski, J. Barr, J. Connors, M. Lev, S. Mukherji, E. Russell. (2007) Performance Measures in Neuroradiology. American Journal of Neuroradiology 28:8, 1435-1438
CrossRef28
Atul Kumar, Alissa Mark, Ravi Singh, LiMing Dong. (2007) The jury is still out on ERCP-related volume outcomes. Gastrointestinal Endoscopy 65:4, 739
CrossRef29
Carl Nathan. (2007) Aligning pharmaceutical innovation with medical need. Nature Medicine 13:3, 304-308
CrossRef30
B.H. Thiers. (2007) Pay for Performance in Commercial HMOs. Yearbook of Dermatology and Dermatologic Surgery 2007, 277-278
CrossRef
