September / October 2005
An Indispensable Element
for Quality Improvement
Walter McClure, one of the nation's most astute health policy analysts, is rightfully recognized as one of the principal architects of strategies to harness market forces as an instrument for healthcare system reform (Business Week, 1982; Iglehart, 1988). In McClure's provocative presentations, comprehensive reports, and thoughtful articles (McClure, 1976, 1978, 1992, 1995), one of his most basic, yet profound, arguments is that individuals and systems behave in the way they are incented to behave. McClure's reasoned application of the fundamentals of microeconomics remains relevant to all health system stakeholders concerned about quality improvement.
It is not enough to admonish care systems and caregivers to do better. Rules and regulations, policies and procedures, care process re-design, and information technology systems (such as electronic medical records) are necessary but not sufficient. Without financial incentives with significant rewards, those endeavoring to improve quality swim upstream against powerful, countervailing currents. Yet this is precisely what prevailing reimbursement policies tell providers to do.
Reimbursement stands as a strong barrier against quality improvement (Hagland, 2005; IOM, 2001). Most provider payment policies essentially pay for a process of care consisting of units of service defined by procedure codes. Reimbursement is a process-based, volume-driven mechanism. Providers and hospitals confront an incentive to maximize the units of services delivered. There is no direct relationship to or financial reward for adopting evidence-based best practices, restructuring care processes, improving patient safety, or achieving specific clinical and health status outcomes. Simply stated, if we want physicians and hospitals to improve quality, we have to reward them for doing so.
Health plans, insurers, purchasers and Medicare have started to launch new programs that financially reward physicians and other providers for achieving improvements in quality (Amundson & Himmelstein, 2004; Dewitt, 2004; Gutman, 2004; New York Times, 2005; Rosenthal, et. al, 2004). Approximately 39% of all health plans nationally have initiated some form of financial incentives for quality (Miller, 2004). Many are pilot programs, and most are still at an early stage of evolution. HealthPartners is a family of nonprofit, consumer-governed healthcare organizations that include a 630,000-member health plan (covering nearly one in four residents of the Minneapolis-St. Paul metropolitan area), the 450-bed Regions Hospital, and the HealthPartners Medical Group, with 637 physicians at 28 primary care and specialty care clinics. HealthPartners' efforts to financially reward providers date back to 1997 and follow many years of developing and implementing provider performance measurement and reporting. Its health plan has implemented two programs that align physician group and hospital incentives in support of quality improvement: recognition bonus awards that supplement regular reimbursement and integration of quality incentives within reimbursement itself.
The Outcomes Recognition Program (ORP), started in 1997, is a bonus program that offers special financial awards for primary care groups that meet plan-wide quality goals set annually by HealthPartners. Pay for Performance (PFP), initiated in 2001, integrates payment for quality directly into HealthPartners' reimbursement contracts with primary care groups, specialty groups, and hospitals. PFP pays physician groups and other providers for achieving individual quality improvement objectives established by mutual agreement with HealthPartners during reimbursement contract negotiations.
Bonus Awards for Quality Improvement
HealthPartners started its efforts to financially reward providers for quality with the Outcomes Recognition Program (ORP) for primary care physicians because they are on the front lines of patient care (Reed, 2002; Reece, 2003). ORP provides special recognition and bonus awards to primary care groups that achieve excellent results in promoting health and preventing disease, based on the health plan's enterprise-wide quality improvement goals (Solberg & Amundson, 1998). Yearly financial bonuses, calibrated against performance targets that HealthPartners assesses and adjusts annually, keep incentives aligned to sustain continuous quality improvement. Annual ORP targets are set at high but achievable levels. And in contrast to some health plans' new financial incentive projects, ORP is not a zero-sum game. All eligible medical groups that meet the targets are rewarded.
To be eligible for ORP for 2004, a primary care group had to care for at least 2,500 HealthPartners' members, operate under a risk contract, be in the health plan's provider network for a minimum of two years, and have patient chart availability of 90% or better. The 26 clinics eligible for ORP bonuses awarded in 2004 served more than 90% of the plan's members and could win financial rewards ranging from $90,000 to $295,000, depending on the size of their HealthPartners' patient bases and the number of ORP targets reached. The targets that HealthPartners sets are ambitious and challenging, so most winning physician groups hit only one or two targets. For 2004, 19 of the 26 eligible primary care groups won a total of 38 ORP awards and received a total of $656,000 in bonuses. Total annual ORP bonus awards have averaged $564,000 since the program began.
Comprehensive Clinical Measures
For ORP bonuses awarded in 2004, patient satisfaction accounted for 25% of the total possible bonus for each clinic. To attain that part of their potential ORP award, at least 55% of the clinic's surveyed HealthPartners' patients had to be "very satisfied." The remaining 75% of the total possible bonus was divided equally, 15% each, for 5 clinical quality targets: diabetes, coronary artery disease, preventive services, tobacco use cessation, and use of generic drugs.
ORP is unique because it rewards physician groups for achieving measurable improvements in clinical outcomes, not just for improving process. Importantly, ORP clinical targets are broad and comprehensive, not limited to only one or two measures per condition. They encompass all significant risk factors in each priority clinical area.
For instance, Optimal Diabetes Care requires at least 30% of a group's HealthPartners' adult (age 18 - 75) patients with diabetes to have all cardiovascular risk factors optimally managed. To meet this comprehensive quality standard, patients had to have blood sugar (HbA1c) levels less than or equal to 8%, LDL cholesterol less than 130 mg/dl, and blood pressure under 130/85 mmHg. Moreover, they had to be non-smokers, and those older than age 40 had to be taking aspirin daily. Three of the 26 eligible primary care group practices met this all-inclusive quality standard for ORP bonuses awarded in 2004.
The Optimal Coronary Artery Disease (CAD) target for adults also features multiple requirements. These include LDL cholesterol less than 130 mg/dl, blood pressure less than 140/90 mmHg, lipid control medication for those with LDL greater than or equal to 130, daily aspirin use, and documented non-smoking status. Sixty-five percent of a group's HealthPartners' patients had to be optimally managed for all these factors in order for a primary care group to hit the target. Last year, 14 groups met this demanding standard of care and earned ORP bonuses for CAD.
The Preventive Services target requires that 85% of a clinic's HealthPartners' patients be up-to-date on all age- and gender-specific preventive services. The preventive services goal is challenging because a member must be current on all appropriate preventive services to be counted. Seven of the 26 eligible primary care groups fulfilled this high standard for comprehensive preventive care for 2004 ORP bonus awards.
Many new financial incentive programs focus on only 1 or 2 clinical indicators per disease state. However, patient care cannot be effectively managed, and quality cannot be meaningfully improved, by addressing just 1 or 2 components of the care patients need. Such a limited approach to quality improvement will not generate substantial improvements in health outcomes.
ORP rewards physician groups with bonuses for demonstrating that they address all modifiable risk factors for particular chronic conditions and ensure provision of all relevant preventive services for general health maintenance. Optimally managing all risk factors is difficult but essential to achieve significant improvements in outcomes for patients with chronic conditions like diabetes and heart disease (Isham and Amundson, 2002). As primary care physicians increase the numbers of HealthPartners members who are effectively managed for all risk factors, and as they ensure that more members are up-to-date on all relevant elements of preventive care, they achieve real improvements in clinical outcomes and members' health status. Significantly, for HealthPartners' adult members with diabetes, rates of diabetic eye complications have decreased by 21%, heart attacks by 25%, and limb amputations by 55%.
Physician Perspectives on Bonus Awards
Twin Cities area primary care group practices that have participated in ORP over several years support the program and its quality improvement goals.
Karen MacKenzie, MD, is the medical director of North Suburban Family Physicians, a primary care group that consistently wins ORP bonuses. North Suburban won ORP awards in 2004 for Optimal CAD Care, Preventive Services, Tobacco Use Assessment and Cessation Assistance, and Patient Satisfaction. "The biggest value of ORP is that it provides focus for all quality improvement efforts," she says. "The secret to meeting ORP goals is to systematize tasks in routine work flow so that they become automatic."
According to Dawn Blomgren, MD, medical director of Northwest Family Physicians, another multi-year ORP bonus award winner, "the clinical areas that HealthPartners targets in ORP are important for quality improvement. By rewarding outcomes rather than looking just at process measures, the program encourages more experimentation to improve clinical performance. Being an ORP winner also enhances a group's reputation, making it easier for us to recruit top providers." In 2004, Northwest Family Physicians was recognized and awarded bonuses for meeting ORP goals for Optimal CAD Care, Tobacco Use Assessment and Cessation Assistance, Generic Drug Use, and Patient Satisfaction.
Crossroads Medical Centers was recognized in 2004 for Optimal CAD Care, Optimal Diabetes Care and Patient Satisfaction. Jean Ek, the group's executive director, comments: "ORP provides good opportunities to improve the quality of care. Our physicians see that meeting the goals will benefit their patients."
Integrated Financial Incentives
As noted, ORP supplements regular reimbursement with bonus awards. In 2001, HealthPartners set out to build upon ORP and launched Pay for Performance (PFP) to introduce financial incentives for quality as an integral element within standard reimbursement. PFP pays providers for quality within their reimbursement agreements with the health plan. PFP blends two elements of reimbursement, payment for quality and payment for process, into market-based reimbursement rates for primary care groups, specialty groups, and hospitals.
PFP focuses on quality measures that are objective, based on evidence-based guidelines, and consistent with existing quality indicators that reflect accepted standards of practice and prevailing quality improvement initiatives among provider organizations in the community. Thus, HealthPartners endeavors to not impose the burden and cost of new measurement requirements. To ensure fairness, PFP targets areas for quality improvement that physician group practices and hospitals can control. Most importantly, the strategy of PFP is to negotiate individual quality improvement goals that are relevant to specific provider groups and achievable for them. HealthPartners and the providers then chart pathways for sustained improvement over time.
For primary care, each priority specialty, and for hospitals, HealthPartners identifies several potential quality measures. A "menu" of quality measures is discussed, particular measures are selected, and specific performance levels and incentives are negotiated in providers' contracts with the health plan. Generally, a portion of HealthPartners' payments are set-aside and paid out at the end of each year, if providers reach their individual performance goals for quality improvement.
PFP began with one hospital and one cardiology group. In 2002, the program expanded to include primary care group practices, a range of specialties, and Minneapolis-St. Paul area hospitals whose contracts were up for renewal. With more than 1,400 specialty provider contracts, HealthPartners has centered PFP on high-volume medical groups in priority specialty areas. In 2004, PFP agreements were in effect with all high-volume cardiology and emergency medicine groups and with half of all orthopedics, ENT, and obstetrics and gynecology specialists.
In 2004, HealthPartners paid $10.6 million under PFP. This included approximately $5.6 million to primary care groups, $1 million to specialty groups, and $4 million to hospitals.
Negotiating Quality Improvement Objectives
For primary care, the PFP quality improvement menu includes a number of options that are consistent with ORP. Including ORP measures in PFP leverages and reinforces the efforts and progress of primary care groups involved in the bonus award program and reflects HealthPartners' commitment to consistency with current quality improvement efforts. One example is generic drug use, the percentage of HealthPartners' members with a drug benefit whose prescriptions are filled with an equivalent generic. The Mesaba Clinics care network selected this measure in PFP negotiations with HealthPartners and increased its use of generics from 47% in 2003, to 60% in 2004.
Similarly, Brainerd Medical Centers' use of generics increased from 43% to 54% from 2003 to 2004, so this primary care group also earned the negotiated quality component of its reimbursement under PFP.
Other primary care clinics examine their performance on ORP's Preventive Services composite measure and select specific elements for improvement. For instance, the Aspen Medical Groups targeted mammography for women aged 50-74 for PFP and negotiated an incremental improvement goal with HealthPartners. From 2003 to 2004, they boosted the number of eligible women who had a mammogram from 79% to 85%. The financial incentive of earning the negotiated portion of their payments dedicated to quality improvement on this measure focused and rewarded Aspen's efforts.
HealthPartners has many different PFP initiatives in place with its network hospitals. Implementation of performance goals with seven of HealthPartners' hospital partners resulted in more patients hospitalized with acute myocardial infarction (AMI) receiving high-quality care based on JCAHO best-practice measures. Baseline data was gathered during 2003, and improvement goals for 2004 were established in negotiations with individual hospitals. In 2004, every hospital met its goal.
Use of non-standard and confusing abbreviations in patient charts is a potential source of errors that compromise patient safety in hospitals. HealthPartners and six hospitals negotiated goals for decreasing the use of five dangerous abbreviations (such as "U" for units) in medications management. All six hospital partners successfully decreased the use of the prohibited abbreviations and earned their negotiated component of standard reimbursement.
Specialists' Experience with PFP
"In most managed care contracts, your financial rewards depend on the overall financial performance of the payer, not the performance of your own group practice and certainly not on quality improvement," says David Rothschiller, the executive director of St. Paul Heart Clinic, the largest independent cardiology group practice in the Minneapolis-St. Paul metropolitan area. "HealthPartners' Pay for Performance is distinctly different and directly links return of the set-aside to objective measurements of a group's performance on key components of care that its physicians can control."
In 2001, St. Paul Heart and HealthPartners established the group's baseline performance levels. They also set goals for 2002 for two measures from the cardiology menu for PFP: comprehensive assessment and management of coronary artery disease (CAD) risk factors and discharging AMI patients with appropriate beta blocker and aspirin medication regimens. Again, the CAD goals are consistent with the Optimal Coronary Artery Disease measures in ORP. St. Paul Heart's negotiated target rates for 2002 were 80% for CAD risk factor management and 65% for post-discharge AMI treatment. In 2002, the clinic earned its PFP set-asides, exceeding its goals by reaching 94% on the CAD measure and 97% on the AMI measure. For 2003, the clinic and HealthPartners negotiated improvement goals to push performance even closer to 100%.
When HealthPartners and St. Paul Heart began to discuss the clinic's 2004 PFP contract, they brainstormed a new initiative. Cardiologists share in the overall care of heart patients with primary care physicians. HealthPartners and St. Paul Heart examined the possibility of devising integrated treatment plans for every congestive heart failure (CHF) patient in collaboration with the referring primary care group practice. Alan J. Bank, MD, one of St. Paul Heart's medical directors, and Audrey Hansen, BSN, MA, PMP, the clinic's director of quality improvement, crafted a detailed proposal for discussion. Under PFP in 2004, St. Paul Heart developed and implemented a service agreement with one their referring primary care groups, outlining their respective responsibilities for the co-treatment of CHF patients. This service agreement includes their approach to communication, protocols for determining when treatment should be managed by the cardiologist and when the patient's primary care physician should take the lead, and the implementation of community based, best-practice guidelines.
Victor H. Tschida, MD is president of St. Paul Heart Clinic, clinical associate professor of medicine at the University of Minnesota, and past president of the Minnesota Chapter of the American College of Cardiology. "As we look to the future, and consider how financial rewards for quality could have a greater impact on the health of our patients, we would like to see other payers implement payment plans consistent with HealthPartners' model," he says. "That would be optimal and positively impact the entire community while reducing administrative costs for quality improvement efforts."
In negotiation with HealthPartners in 2002, Minnesota Orthopaedic Specialists, PA (MOS) decided to focus on best-practice guidelines for menisectomy arthroscopy and tobacco assessment. "These are areas where we had room for improvement," says Owen O'Neill, MD, president of the group practice. "Tobacco use had been removed from our intake form, so we wanted to address that."
For the menisectomy arthroscopy measure, MOS negotiated to increase the number of patients meeting care guidelines (appropriate symptoms and findings, a trial of conservative therapy before a decision to perform surgery, and post-surgical documentation of findings and outcomes) from 67% at baseline in 2002 to 80% in 2003. The group also agreed to try to boost the number of patients meeting tobacco use treatment standards from 6% to 40%. In 2003, MOS attained 83% on the menisectomy arthroscopy measure and 47% on the tobacco measure. For 2004 and 2005, MOS and HealthPartners agreed to similar targets in order to sustain improvement.
"It is a positive development if health plans reward group practices for performance and good patient care, and HealthPartners is the only health plan in our market taking the initiative to focus on outcomes in reimbursement," O'Neill says. "It is important to me and my colleagues that the criteria are relevant and reasonable and created with input from orthopedic specialists. We also appreciate that the incremental improvement goals center on factors we can control and are negotiated and not imposed."
In February 2005, the Medical Group Management Association (MGMA) issued a set of principles on financial incentives for quality (MGMA, 2005). What are some of these key principles, and are HealthPartners' programs consistent with them?
The primary goal of pay-for-performance programs must be improving health quality and safety. Focus is critical. HealthPartners' experience and success show that quality measures should be comprehensive and support progress toward improving outcomes in areas of care that are clinically important for patients.
Medical group practice participation in pay-for-performance programs must be voluntary. For HealthPartners, a major long-term goal of both ORP and PFP is to forge and strengthen partnerships with group practices and hospitals. Voluntary participation is essential to collaborative relationships. HealthPartners' programs demonstrate the value of balanced goal setting. Goals should be balanced between health plans' long range, enterprise-wide targets and incremental performance goals negotiated individually and voluntarily with physician groups and hospitals.
Practicing physicians and physician professional organizations must be involved in the design of pay-for-performance programs. Provider involvement in the formulation of quality standards and performance measures is essential. Providers help define HealthPartners' state-of-the art quality measures. Collaboration is indispensable for HealthPartners, especially in the negotiation of quality improvement goals and payments under PFP.
Physician performance measures used in pay-for-performance programs must be evidence-based, broadly accepted, clinically relevant, and continually updated and developed by practicing physicians. Application of evidence-based standards that are understood and accepted by the local medical community is critical to all of HealthPartners' care process transformation endeavors under its "Pursuing Perfection" initiative and its programs to establish financial rewards for quality. HealthPartners is a founding member of the Minneapolis-based Institute for Clinical Systems Improvement (ICSI), a nonprofit organization that supports collaboration on quality improvement among 45 Minnesota medical groups. ICSI's expert panels include practicing physicians. They undertake ongoing review of evidence-based guidelines and clinical research (Rank and Averbeck, 2005, Isham, 2004 and Aquilina, 2003).
Physicians must have the ability to review and correct performance data. HealthPartners upholds transparency as a value and goal and is committed to sharing performance data with physicians. For example, HealthPartners' "Balanced Scorecards for Specialists" assess and compare the clinical and business processes of specialty practices, their use of resources, the degree to which patients and referring physicians are satisfied with their performance, and their patient outcomes (Marr and Mullen, 2004). The scorecards are a valuable resource for specialty groups as they select areas for quality improvement under PFP. Physician groups have the opportunity to review their data before the comparative scorecards are published.
Pay-for-performance must reimburse physicians for any administrative burden for collecting and reporting data to payers. As previously discussed, HealthPartners is careful not to introduce the burden and cost of measurement requirements that are inconsistent with prevailing quality improvement efforts. Providers should be rewarded for actual improvements in quality. And whether bonuses or portions of routine reimbursement, the size of financial rewards must be sufficient to provide meaningful incentives and significant rewards.
Pay-for-performance programs must reward physician participation, including physician use of electronic health records and decision-support tools. HealthPartners' comprehensive measures establish an incentive for providers to develop patient registries and standardized care protocols, incorporate disease management approaches to chronic conditions in their practices, and adopt electronic health records.
Many health plans, insurers, and purchasers are now starting to formulate, test, and implement new models of reimbursement to reward measurable quality improvement. Aligning provider groups' financial incentives to support efforts to improve quality and advance patient safety is critical.
Going forward, stakeholders in quality improvement should remember Walter McClure's essential argument: Individuals and systems behave in the way they are incented to behave. Incentives must be aligned throughout all components of the healthcare system. Accordingly, if purchasers want health plans to measure and reward care systems and provider groups for quality improvement, they must incorporate financial incentives and rewards into the premiums they pay health plans. In turn, care systems and provider groups have to extend financial incentives and rewards they receive from health plans to individual physicians and other caregivers.
Babette A. Apland is senior vice president, health and care management and provider relations at HealthPartners. Gail M. Amundson, MD, FACP is HealthPartners' associate medical director for quality improvement. They may be contacted at firstname.lastname@example.org and email@example.com.
Amundson, G., Himmelstein, D. (2004). Pro & con: should physicians get paid for quality improvement? Rheumatology News, 3(1), 5.
Aquilina, D. (2003, March 3). Achieving and sustaining quality improvement. For the Record, 35.
Dewitt, P. (2004). The new incentive plan: payors and providers experiment with pay for performance. HealthLeaders, 7(3), 36-41.
Gutman, J., (Ed.). (2004). Case studies in health plan pay-for- performance programs. Washington, DC: Atlantic Health Information Services.
Hagland, M. (2005). Fast track to rewards. Healthcare Informatics, 22(3), 26-30.
Iglehart, J. (1988). Competition and the pursuit of quality: A conversation with Walter McClure. Health Affairs, 7(1), 79-87.
Institute of Medicine, Committee on Quality of Health Care in America. (2001). Crossing the quality chasm: A new health system for the 21st century. Washington, DC: National Academy Press.
Isham, G, and Amundson, G. (2002). A seven-step process model for quality improvement. Group Practice Journal, 51(9), 39.
Isham, G. (2004, October). Pursuing perfection to advance quality. The Quality Indicator, 3-4.
Kolata, G., Abelson, R. (2005, April 15). Flexing Medicare's muscles: A bonus for health, payable to the doctor. New York Times.
Marr, T and Mullen, D. (2004). Balanced scorecards for specialists: a tool for quality improvement. Minnesota Medicine, 87(4), 46-50.
McClure, W. (1976). The medical care system under national health insurance: Four models. Journal of Health Politics, Policy and Law, 1(1), 22.
McClure, W. (1978). On broadening the definition of and removing regulatory barriers to a competitive health care system. Journal of Health Politics, Policy and Law, 3(3), 303.
McClure, W. (1992). Health care system reform under the "Buy Right" strategy. Medical Group Management Journal, 39(2), 16-23, 76-80.
McClure, W. (1995). Good and bad models of market reform for managed care. Health Care Management, 2(1), 237-260.
Medical Group Management Association (MGMA) (2005). Principles for pay-for-performance and recommendations for medical group practices. Retrieved on May 3, 2005, from http://mgma.com/press/MGMApos-payforperformance.cfm.
Miller, J. (2004). Models for measuring: performance reward programs move forward, but few take off running. Managed Healthcare Executive, 14(2), 24-26.
Rank, B. and Averbeck, B. (2005). Advancing evidence-based medicine on the front lines of patient care. Group Practice Journal, 54(4), 24.
Reece, R. (2003, November). Program Links Rewards to Outcomes. The Quality Indicator, 13-15.
Reed, M. (2002, September). Plan rewards clinics for quality. The Quality Indicator, 13-15.
Rosenthal, M., Fernandoupulle, R., Song, H.R., & Landon, B. (2004). Paying for quality: Providers' incentives for quality improvement. Health Affairs, 23(2), 127-141.
Solberg L., Amundson G.›(1998). Taking advantage of the clinical setting for prevention. Best Practices in Managed Care, P. R. Kongstvedt & D. N. Plocher, (Eds.).›Gaithersburg, MD: Aspen Publishers, Inc., 45-53.
The spiraling costs of health care Rx: competition. (1982, February 8). Business Week, 58.