Lessons Learned From a National Pay-for-Performance Program
Bridges To Excellence, Inc. (BTE) is a not-for-profit 501(c)3 company that is led by a multi-stakeholder board of directors comprised of physicians, employers and health plans. Since its inception in 2002, the companys mission has been to create programs that reward physicians for delivering safe, timely, effective, efficient, equitable and patient-centered care.
BTE has previously published a comprehensive evaluation of its original four pilot sites that demonstrates the effectiveness of its programs:
- Incentives that reward physicians for adopting better systems of care result in physician practice re-engineering and adoption of health information technology;
- Incentives that reward physicians for delivering good outcomes to patients with diabetes result in physicians changing the way in which they practice care from reactive to proactive and in patients getting better care;
- Physicians that are recognized for adopting better systems of care and physicians that deliver better outcomes for patients with diabetes are more cost-efficient (on a severity and casemix adjusted basis) than physicians that are not recognized;
- Performance measures that focus on intermediate outcomes for patients with diabetes, hypertension, hyperlipidemia, coronary artery disease and cardiovascular disease, and measures of effective treatment protocols for patients with recent cardiac events hold the highest clinical and actuarial value of most measures of ambulatory care.
In addition to demonstrated effectiveness, BTEs programs have also yielded some significant lessons for policy makers, health plan leaders, employers and physicians, all of whom are actively looking at deploying different forms of pay for performance. These lessons have been used by BTE to identify critical ingredients that ensure the ongoing success of its programs and their growing adoption by communities across the United States.
Lessons Learned and Best Practices in Pay for Performance
- Use standard performance measures of clinical quality,
focusing mostly on intermediate outcomes derived from
medical chart reviews, not just claims.
When selecting performance measures for a program, employers and plans should always default to measures adopted by the National Quality Forum to ensure consistency with other programs such as the Ambulatory Care Quality Alliance. However, all measures are not created equal, and findings from experiments in pay for performance in England are consistent with BTEs that intermediate outcomes should be weighted more heavily than process measures. A study commissioned by BTE and performed by Towers Perrin, a large consulting firm, found that the actuarial value of intermediate outcomes (e.g., blood pressure) in patients with diabetes, coronary artery disease or cardiovascular disease were significantly higher than the actuarial value of process measures (e.g., the completion of a lipid profile). Figure 1 details the value of a set of standard measures of performance for patients with diabetes.
The savings are not fully additive, but the magnitude of
savings stemming from blood pressure control is significant
compared to the savings stemming from a nephropathy
assessment. Importantly, assessing the blood pressure control
of a patient can only be done by reviewing medical charts,
not by reviewing health plan claims. This underscores the
need to include clinical data from medical records in the assessment
of physician performance and, as a result, the importance
of adoption of electronic medical records to facilitate data
collection and reporting. - Provide physicians and physician practices with clearly
defined costs and benefits of the program, which helps
them determine the value of participating.
Practices or individual physicians that are asked to invest both time and money in that practice to improve their performance must have a good estimate of the benefits they will derive from that effort, as illustrated in Meredith Rosenthals recent article (JAMA, Oct. 2005, #294, p 1788). For example, a group of physicians in Delaware decided not to participate in a BTE implementation because they estimated that the benefits of meeting the threshold for the required performance measures exceeded the initial investment they would have to make to allow them to meet the performance measures. Physician practices are business entities, and like all business entities, they need to estimate the relative risk of an investment to determine its potential. Uncertainty in the potential benefits of a program increases its relative risk. - Use independent third-party organizations to measure
the performance of the physicians, reviewing the data
reported by these physicians from medical records in
their practice.
Physicians and physician practices that are asked to participate in a BTE implementation must perform a self-audit of their current performance. That self-audit is then sent to an independent third party (e.g., the National Committee for Quality Assurance) that reviews the data, scores it and determines if the physician meets the required performance threshold, while subjecting all to a random check. This effort, while time-consuming, creates the road map for change. For example, Dr. Lalude, a general internist practicing in Louisville, Ky., discovered through his self-assessment that his compliance with good standards of care for his patients with diabetes was 45 percent, far below the 80 percent threshold required for incentives. This discovery motivated him even more than the financial rewards to re-engineer his practice of care for those patients and the improvements achieved in under a year were significant, both in terms of quality of care for the patients and the cost of care. Additionally, instead of having his performance assessed by a payer who would only look at a portion of his patient panel, Dr. Laludes performance was assessed by the NCQA via a representative sample of his entire panel. - Bring together many payers and/or purchasers to make
rewards meaningful to physicians and practices.
In general, incentives must be greater than or equal to the cost of change, or, at the very least, sufficient to significantly contribute to the cost of change. For example, a multispecialty group practice in Albany, N.Y., was able to defray about a quarter of the $1 million investment in a new clinical information system due to the financial incentives received through BTE. The practices, whose care processes were re-engineered, achieved significant improvements in the management of patients with chronic conditions.
Figure 2 illustrates both points: (1) Overall there is a
statistically significant difference between the mean rewards
available for recognized physicians and nonrecognized
physicians in each pilot region (size patient count
is a proxy for size of bonus); (2) There is a significant
difference between the mean patient count for physicians
recognized for adopting new systems of care and the
patient count for physicians being recognized for diabetes
excellence only. With respect to this second point, the
average investment for a practice of three physicians to
adopt new clinical information and patient management
systems is $50,000. - Encourage physicians to adopt better systems of care,
including health information technology, to systematically
improve the delivery of care.
It is quite fascinating that the healthcare industry is the only one where the validity of adopting and using information systems and continuously improving processes to reduce waste and increase effectiveness is constantly questioned. However, recent research from the RAND Corporation published in Health Affairs in 2005 (24:5, pp. 1103-1117) shows unequivocally the impact that widespread adoption and use of health information technology including electronic medical records would have on the cost of care today. An independent physician organization with practices in the greater Albany, N.Y., region recently underwent a significant transformation as a result of its participation in Bridges To Excellence. To qualify for the more than $400,000 its practices have received to date from BTEs participating employers, Community Care Physicians (CCP) had to redesign the processes of care in most of its practice sites. It installed registries to track patients with chronic conditions, hired nurses and other physician extenders to monitor the patient population and identify patients whose chronic illness was not well managed. For the first time in its history, CCP and its practices were able to understand which chronic illnesses affected the majority of its 50,000-plus patients, and it could deploy specific care management protocols to ensure that these patients received evidence-based care. Internal monitoring of patient medical records showed that within a year after installing information systems and redesigning care processes, the average glycated hemoglobin of patients with diabetes decreased by 1.5 percent, a significant achievement. - Small practices need significant help in re-engineering, and
there are not many resources available to help them.
When BTE launched its pilots, many barriers to success were anticipated, but the lack of knowledge about and resources for small practice re-engineering were not among them, which caught purchasers by surprise.What quickly became clear was that even when financial rewards were sufficient, measures were well-accepted, and the practice had understood the need to modify its care practices to improve the quality of care, where no road maps, tool kits or consulting services were available. There are three types of resources that emerged from the pilots that are now being used in all BTE implementation areas across the United States:- Quality Improvement Organizations Every state has a Medicare contractor referred to as a QIO (www.ahqa.org) whose main function is to help in improving the quality of care for Medicare beneficiaries. Since most physicians deliver care to Medicare beneficiaries in addition to patients under the age of 65, QIOs can help almost all physicians and physician practices as part of their core mission. In particular, QIOs have tool kits for physicians on how to improve care for patients with diabetes or cardiac disease, as well as tool kits for how to adopt health information technology.
- Medical Specialty Societies The American College of Physicians (www.acponline.org) and the American Academy of Family Physicians (www.aafp.org) have created centers for the transformation of primary care and the creation of the medical home. Both organizations have state chapters that can help its members to leverage their tools and improve their processes of care.
- State and County Medical Societies These organizations represent physicians in the state or county and can act as conduits for knowledge sharing or other help. For example, the Jefferson County (Ky.) Medical Society helped physicians in Louisville to get grants to offset the cost of medical chart abstraction so they could measure their performance on diabetes care and participate in BTEs rewards.
- Organized groups, like independent physician associations,
require smaller per-physician bonuses than smaller practices.
In all BTE pilot markets, larger group practices have had the resources to re-engineer care that small practices found so difficult to obtain. Generally, they have physician leaders in the organization that are responsible for quality improvement and who share best practices among all the practices in the group. For example, Partners Community Health Care, Inc. in Massachusetts has had a dedicated BTE specialist in their central office that is a common resource shared among all Partners practices across the state. This specialist is fully versed in BTEs programs and can offer guidance to the practices on how to fill out applications for recognition and on what processes to adopt to meet the performance requirements. In addition, IPAs often act as a contracting intermediary with health plans and can create internal incentives within its practices to further emphasize certain aspects of cost and quality improvement. As a result, the net amount of rewards required by larger practices on a per-physician basis is lower than for smaller practices because the cost of transformation is also lower. - A focus on a single disease may limit program uptake
among primary care physicians.
Primary care physicians care for patients with many conditions; some acute, most chronic. However, patients with diabetes only represent a small portion of an internists practice and that volume of patients may not be sufficient to create a business case for fundamental care re-engineering. Dr. Lalude of Louisville, Ky., made some manual changes to his processes and has been able to better manage patients with diabetes, but he readily admits that the improvements have come for one condition, not all conditions. Moving from re-engineering care for one condition to multiple conditions requires the adoption and use of health information technology, which is an expensive investment. As seen above, the benefits of an intervention have to be proportional to its costs, and the benefits derived from bonuses on just one class of patients may not be enough to offset the significant costs of an electronic medical record. - Physicians that become recognized in BTEs programs are
happy to get more patients even those with chronic illness.
There were two commonly voiced arguments against pay for performance and Bridges To Excellences programs when they were initially launched. The first was that physicians would game the scoring by firing noncompliant patients. The second was that recognized physicians would become inundated with difficult-to-manage patients and that this influx would hurt their top-line revenue because they could no longer see other easier patients with low-acuity problems (e.g., colds). Neither of these arguments has stood the test of the pilots. As expected, re-engineered practices become much better at managing patients, and their ability to stabilize these patients actually opens up their practice to higher throughput. As such, the pilots have demonstrated that the same effects of process improvement that have been observed in industry can be achieved in medical care. In all BTE pilot markets, practices that have become recognized have asked purchasers to send them more patients, not fewer, and preferably patients with chronic conditions on which they can potentially earn a bonus. - Employers and plans should combine a pull (bonus) with
a push (steerage) to maximize the impact of a pay-forperformance
program among their plan members.
On average, after three years of implementing a BTE program, about 25 percent of the physicians or practices in a community become recognized.While that number varies according to all the factors listed above, the average has been fairly consistent to date. As such, for any employer or payer engaged in pay for performance, you really need to deploy two strategies to maximize the impact of the program. Since recognized physicians are more efficient than nonrecognized physicians, an employer or payer will have a direct financial benefit every time a patient switches from a nonrecognized physician to a recognized physician.
Encouraging employees and other plan members to change physicians can be done by having an aggressive communications campaign and deploying easy-to-use tools (like online directories) that can help to select a physician. During the pilot phase, many employers tested different forms of communications to encourage plan members to look up their physicians rating and to understand why it was important to select those that were recognized. However, even the best in class were unable to get more than 35 percent of their plan members to see recognized physicians. Other consumer-directed efforts suggest that differentiating co-pays or co-insurance levels has a significant effect on a plan members decision about where to seek care. This type of steerage mechanism is, in fact, what recognized physicians would like to see employers adopt, and employers should be encouraged to do so.
Pay-for-performance programs are appropriately being touted as a way of reversing the detrimental effects of fee-for-service payment and of encouraging physicians to improve the quality of care they deliver and, as such, of increasing the value of the dollars spent on care by American employers and employees. Our experience in Bridges To Excellence has indeed shown that quality and efficiency can go hand in hand when pay-for-performance programs have the right ingredients. There are 10 that have proven to have the strongest impact and are responsible for the tremendous uptake of BTE in all parts of the U.S. Congress, and the administration should take these into consideration as they increasingly apply pay for performance in the largest health insurance program in the country Medicare.
Additional details of Bridges To Excellence Pay-for-Performance studies may be found at www.bridgestoexcellence.org.
