| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Perspectives |
1 From the Department of Radiology, Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114. Received August 12, 2004; accepted August 17. Address correspondence to the author (e-mail: jthrall@partners.org).
Index terms: Perspectives • Radiology and radiologists, socioeconomic issues
Despite recent double-digit increases in health insurance premiums in many parts of the country and the aggregate expenditure of nearly 15% of gross domestic product on health care services in the United States (1), serious gaps remain between the ability of providers to deliver and the expectations of patients for ready access to care, good personal service, and high-quality outcomes (2,3). The magnitude of the gap in quality is sobering and, according to the available evidence in medical literature, affects all sectors of the health care system (2,3). Authors of a recent study (3) that covered 12 metropolitan areas throughout the United States found that patients received care that was consistent with quality indicators for 30 acute or chronic conditions only 54.9% of the time on average and for the least followed indicator as little as 10.5% of the time. The gaps found in the study (3) were not esoteric or necessarily hard to close; only 61% of appropriate candidates received aspirin after myocardial infarction despite evidence from clinical trials that demonstrated a 15% reduction in risk of death. Also, only 64.7% of patients received recommended care for hypertension despite known increased risks for heart disease, stroke, and death for people with long-term high blood pressure.
Fueled in part by recommendations in the Institute of Medicine’s landmark publication titled "Crossing the Quality Chasm" (4) and in part by encouragement from organizations such as the Leapfrog Group (5) and the Robert Wood Johnson Foundation (6), many payers, including commercial and not-for-profit insurance companies and the Centers for Medicare and Medicaid Services (CMS) (7), have initiated reimbursement programs designed to close the gaps in quality and service and to control costs through incentive payments. The new programs reward, or in some cases penalize, hospitals and physicians (7,8) on the basis of their ability to achieve contractual goals for a variety of issues in these broad areas. Goal-directed pay-for-performance programs are now estimated to encompass portions of the insurance coverage for over 30 million people (9), exclusive of the additional millions of people potentially covered by CMS programs.
Pay-for-Performance Financial Models
The Leapfrog Group (7) has collected summaries on over 100 pay-for-performance programs from around the United States, including some information about the financial models that determine the risks and rewards for providers. The most common financial model among these plans is a straight bonus system that rewards providers with additional payments for achieving stipulated performance targets. The issues encompassed by the performance targets are typically chosen by the payer but may be subject to negotiation.
While straight bonus systems do not place any provider revenue from existing contracts at risk, it is possible that over time such systems may be used by payers as the mechanism to determine incremental increases in reimbursement. If this happens, and it seems likely given the momentum in the pay-for-performance movement, providers will become more dependent on revenue from bonuses and may be pitted against each other to compete for them (8). Thus, providers should be thoughtful about the long-term implications of pay-for-performance bonus plans even if they appear at first to offer an unexpected windfall.
Placement of a percentage of contracted provider revenue directly at risk is the basis for the principal alternative to the bonus model in pay-for-performance contracts. According to the at-risk financial model, payers typically withhold, by contractual agreement, part of the otherwise expected reimbursement for each covered service. Providers can earn the funds at risk by meeting performance targets.
Key motivations for providers to enter into apparently unilateral and potentially disadvantageous risk arrangements with payers include the prospect of negotiating higher fees than would otherwise be possible without taking risk and the ability to use shared risk between physicians as a safe harbor from antitrust law in collective bargaining activities. As summarized by Choudhry and Brennan (10), the Department of Justice and the Federal Trade Commission have established conditions under which "physician network joint ventures" that take "substantial financial risk" can negotiate with managed health care organizations.
The ratio of bonus-based programs versus risk-based programs is about 4:1 among those summarized in the Leapfrog Group compendium (7). The information provided in the summaries does not allow the calculation of the total number of patients covered or the total dollars involved in either kind of plan or their respective ratios between these two plan types. Risk-based programs appear more suited to hospitals or larger groups of physicians than to solo practitioners, while bonus programs are being directed at all provider types among the plans summarized in the Leapfrog Group compendium.
The commitment of CMS (11) to the pay-for-performance concept is a telling indicator of its momentum. A CMS demonstration program (7) with the Premier consortium of 550 not-for-profit hospitals has earmarked $7 million a year to pay hospitals that meet criteria of superior care for the following five conditions: heart attack, heart failure, pneumonia, coronary artery bypass graft, and hip and knee replacements. Hospitals are judged competitively on the basis of extensive data submitted by them for Medicare patients with these conditions. The hospitals performing in the top 10% will receive a 2% bonus on the related Medicare payments, and the next 10% of hospitals, according to performance, will receive a 1% bonus. In this CMS pilot program, the lowest performing 10% of hospitals will eventually be penalized by 2% of the related revenues. Since the hospitals are competing with each other, there will inevitably be winners and losers, which offers a glimpse at one potential downside of pay-for-performance plans. Arguably, the leading hospitals receiving more funds will have a competitive advantage to stay on top, while laggards will have fewer resources and will risk falling further behind.
In the CMS program titled "Hospital Quality Incentive Data Initiative" (7,11), only hospitals that participate will receive full updates for rising Medicare inpatient costs. Nonparticipating hospitals will receive the cost adjustments minus 0.4%.
Through the Medicare Modernization Act (7,11), CMS has earmarked $50 million for fiscal year 2007 to make payments to physicians to assist them in acquiring hardware and software for ordering drugs electronically. On the basis of the number of patients covered alone, initiatives by CMS make pay-for-performance plans something that providers must learn to understand and manage.
Programs initiated by the recently founded not-for-profit organization Bridges to Excellence (12) are examples of straightforward bonus-based pay-for-performance incentive plans for physicians. Physicians working individually can earn $100 per patient per year in pay-for-performance incentives by meeting care objectives in eligible patients with diabetes (7,12) and $160 per patient for cardiac care. They can earn $55 per patient per year for implementing specific office-based practice improvements to reduce errors and improve quality. Patients must be eligible through special coverage by participating payers or employers. The latter includes an impressive list of major US corporations (13) such as General Electric, Proctor & Gamble, Verizon, UPS, and Ford. Physicians should interpret the involvement of these companies as another strong indicator of the momentum now propelling the pay-for-performance concept in the marketplace.
Risk-based programs work very differently than bonus programs and, as noted earlier, seem more suited to larger provider entities that have sufficient resources to manage the risks administratively and financially and that have sufficient size to negotiate effectively with payers. The Partners Healthcare System (PHS) is such an enterprise and is the parent organization of the Massachusetts General Hospital (Boston) and the Brigham and Women’s Hospital (Boston, Mass). PHS has actively pursued contracts with risk provisions to be able to represent all of its affiliated physicians under risk-based safe-harbor arrangements and to push insurance companies to reward higher quality and better utilization management with higher reimbursement.
According to the risk-based pay-for-performance contracts recently entered into by the PHS and its major nongovernmental payers, up to 10% of patient care revenue is withheld from the initial remittance for each covered service delivered by the member hospitals or affiliated physician groups. The withheld funds are placed into a series of risk pools earmarked for each hospital and provider group in the system. A negotiated percentage of each risk pool is assigned to each performance target. If an institution or provider group meets a particular target, the associated revenue is paid to it from the risk pool. If the target is not met, the money reverts to the insurance company. At the end of the annual contract period, all funds in the risk pool are either remitted to the providers or retained by the insurance plan, depending on provider performance in the respective target areas. A hospital or physician group can earn all, part, or none of the money at risk. The unknowns around these outcomes make budget development a challenge for both providers and payers.
Hospitals appear especially vulnerable to pay-for-performance plans that subtract dollars for missing targets versus those that add bonus dollars for achieving targets, because operating margins for hospitals are often modest or negative. For example, the PHS hospitals in Massachusetts, including the Massachusetts General Hospital, typically budget operating margins of just 3%, which is much higher than the expected average operating gain (14) for hospitals in the state. About 25% of care provided by Massachusetts General Hospital is now encompassed in pay-for-performance contracts, with the amount at risk in the range of 4%–5% of the total associated contract revenue. Thus, the aggregate revenue at risk in the pay-for-performance contracts is equal to a substantial percentage of the expected operating gain. Partly for this reason, the PHS hospitals in developing their budgets make the conservative assumption that they will not receive 100% of revenues at risk in pay-for-performance contracts. In any case, given the magnitude of the revenues involved, in the range of several tens of millions of dollars, a substantial decrease in income owing to failure to receive pay-for-performance incentive revenues will be increasingly difficult for the PHS hospitals to overcome.
Goals and Objectives in Pay-for-Performance Contracting
The goals of the architects of pay-for-performance plans are multifaceted but ultimately reflect the pressures facing the key players. Payers and providers are being pushed jointly by patients, employers, government agencies, and the public to improve quality and safety and to reduce unnecessary costs in the health care system. Thus, the relative emphasis between quality, service, cost, and administrative targets depends on the strategic outlook held by the parties involved in each contract about how best to set priorities and how best to use pay-for-performance financial incentives to achieve desired improvements.
Many pay-for-performance plans use quality criteria originally developed for other purposes like the Health Plan Employer Data and Information Set (HEDIS) program (7,15) of the National Committee for Quality Assurance. The quality criteria defined in the HEDIS program have been used in the managed care industry for many years as measures of health plan and provider performance. The major categories in HEDIS are (15) effectiveness of care, access to and availability of care, satisfaction with the experience of care, health plan stability, use of services, and health plan descriptive information. Within these broad categories, there are several dozen specific quality measures, such as the percentage of patients receiving immunizations or who are being screened for breast or cervical cancer.
The Integrated Healthcare Association (IHA) in California (16) was founded by six health plans having over 7 million commercial enrollees. IHA has undertaken pay-for-performance initiatives that illustrate the foregoing observations regarding setting of priorities and reuse of existing quality improvement tools. IHA will use HEDIS (7,16) criteria to evaluate some measures of provider performance. The distribution of payments will be 50% for clinical quality, 40% for patient experience, and 10% for investment in information technology (16), which indicates the relative importance of these issues from the perspective of the IHA member organizations.
The 40% share for patient experience in the IHA model is interesting, because high patient satisfaction is an important business goal of insurance companies to retain enrollees (16). Insurance companies are in intense competition with each other and are at risk for the behaviors of their contracted providers with respect to key service issues, including rapid access to appointments and overall patient satisfaction. IHA will use the Consumer Assessment Survey instrument (7) to determine patient satisfaction.
The 10% allotment for information technology by the IHA (7,16) is particularly noteworthy for its long-term implications. Incentives to improve practice infrastructure and implement electronic systems are also included (7,11,12) in the PHS, CMS, and Bridges to Excellence pay-for-performance plans. Use of electronic order entry systems has been documented in prospective studies to significantly reduce medication errors (17,18) and their associated costs. It is encouraging to see such wide support from payers to help providers improve practice support systems. Accomplished on a large enough scale, these improvements should lead to long-term sustainable gains in quality and safety.
If quality, service, and systems improvements are the happy public faces of the pay-for-performance movement, cost containment is the elephant in the room when payers and providers negotiate. The dialogue about how to simultaneously improve quality and reduce costs has been going on for decades without resolution. However, the pay-for-performance concept does provide a business model for insurance companies, hospitals, and physician groups to sharpen their focus on specific cost drivers and create incentives to better manage them.
Control of the utilization of expensive services such as inpatient admissions, high technology imaging, and pharmaceuticals is an obvious cost-containment strategy and a major feature of the PHS pay-for-performance contracts. By setting aggregate categoric targets for such services, PHS physicians and hospitals are left free to practice, with substantial latitude at the level of each individual patient care decision, but are held to overall targets set from a combination of general benchmark data and PHS historical data.
A major problem with utilization-management programs is achieving an average or representative patient population that is also large enough to even out statistical fluctuations in care requirements from year to year. Practices affiliated with academic medical centers tend to attract patient populations that require more services than are provided by community practices, and resource utilization in any small patient population can be skewed by a few outliers. These issues are difficult to correct. One year a physician group with a small patient base can look like care management geniuses, and the next year the group can appear completely inept on the basis of the frequency of illness and not their medical decision making. Nonetheless, utilization targets are embedded in the pay-for-performance concept. Any medical group or hospital contemplating such arrangements should look carefully at how the utilization targets are set, how their likely patient population compares to the nominal benchmark population, and what information systems are in place to help manage utilization.
A much more appealing pathway for trying to control costs is through the targeting of specific diseases known to require disproportionate resources and that are known to benefit from proved evidence-based management principles. For a number of chronic diseases, there are substantial data in the medical literature that demonstrate reduced costs, improved quality of life, and better outcomes when the diseases are managed prospectively through defined evidence-based programs of care. Diseases and conditions meeting these criteria include asthma, congestive heart failure, hypertension, and diabetes mellitus (3,6,19–21).
For diabetes, Bridges to Excellence (12) estimates that through its program, the employer or insurance plan can expect to realize average savings of up to $350 per patient per year even though physicians will be paid more in bonuses. Key issues include control of hypertension, blood lipids, and blood glucose levels as indicated with measurement of glycosylated hemoglobin. By assuming that the estimated cost savings materialize, financial incentives are aligned between payers and providers in this kind of pay-for-performance plan. Improvement in outcomes benefits the patients, their employers, and eventually the public. In this optimistic scenario, overall costs are reduced and quality improvement is "free."
Asthma offers another compelling example. Asthma is estimated to affect over 15 million people in the United States (19), with an increasing incidence. Findings of a number of studies (19) have shown that patient education, use of objective measures of respiration, allergy testing, reduction in exposure to tobacco smoke, and access to antiasthma drugs reduce hospital admission rates and thereby reduce costs of care and improve quality of life. Castro et al (19) found a 54% reduction in asthma-related admissions and a $12 924 reduction in annualized costs compared with those of controls in a group of patients randomized to a set of interventions that included adherence to the guidelines of the National Asthma Education and Prevention Program (21). This program of the National Heart, Lung, and Blood Institute has promulgated guidelines, developed in part by expert panels, for the diagnosis and treatment of patients with asthma.
Similar data are available for patients with congestive heart failure (20), where total costs of care are heavily influenced by readmission rates. Evidence-based care management is a proven way to reduce admissions (20) and makes congestive heart failure another obvious target for pay-for-performance plans.
Controlling costs and utilization through disease management programs is far preferable to transaction-based systems such as those that require physicians to do extra work to obtain prior approval for elements of care. The disease-management approach also allows some correction for the kinds of statistical fluctuations that defy the setting of rigid prospective utilization targets. That is, if a physician group has more patients with diabetes or patients with congestive heart failure, it should be judged by how well it cares for those patients and not by how it compares in utilization of resources with other physician groups with a different patient mix. It is again encouraging to see disease-management approaches included in pay-for-performance plans.
Radiology and Pay-for-Performance
The radiologic community will need to respond to the exigencies of the pay-for-performance era. Drug costs have skyrocketed in the past 10 years, diverting attention from radiology. This respite is over, and many insurance companies have instituted utilization controls (22) for high-cost imaging procedures that will undoubtedly find their way into pay-for-performance plans. It is likely that more payers will attempt to place controls on utilization of imaging in the future, especially higher cost and high-technology imaging.
Companies have been established that specialize in helping insurance companies control imaging utilization through a variety of mechanisms, including the definition of appropriateness criteria or rules for ordering imaging studies and the administration of prior approval systems. One such company (22) claims to manage radiology benefits for 12 million people and to save its insurance clients 20%–30% in the first year for high-technology imaging. Prior approval systems run by third parties are particularly onerous to clinicians. They increase the time, effort, and cost of arranging imaging studies for patients, without compensation for the extra work.
Radiologists who practice in hospitals and in multispecialty groups, including academic groups, will need to consider how best to participate in any pay-for-performance plans entered into by their organizations. PHS radiologists are working with clinical colleagues on ways to meet utilization targets for selected imaging studies in the at-risk contracts recently concluded with payers. As a negotiated alternative to having a commercial company control the order-approval process, PHS radiologists have developed decision support programs that provide point-of-care knowledge delivery of PHS-defined appropriateness criteria to clinicians ordering elective outpatient imaging studies. The decision support programs are designed to reduce overuse of imaging tests by better disseminating ordering criteria.
Concluding Observations
It is likely that pay-for-performance contracting will grow to encompass more payer plans and patients and become a major financial issue with which hospitals and physicians must come to beneficial equilibrium. The fundamental premise underlying pay-for-performance plans is that improved quality, better service, and reduced costs require changes in physician and hospital behaviors, and financial incentives will help achieve those changes. Whether pay-for-performance schemes will prove any better than previous attempts at "managing" care to achieve these ends is unknown, and there are substantial hurdles (8). Fragmentation of the health care system alone is a powerful negative force against pay-for-performance arrangements; providers have to deal inefficiently with differing requirements of multiple payers. Installation of electronic information systems is expensive and may not be financially viable unless enough of a physician’s practice is covered by incentives to do so. The limited funds available to pay hospitals and physicians may make pay-for-performance a zero-sum game, with winners and losers. The pay-for-performance concept may be ultimately self-defeating if some providers simply cannot compete, which will result in lower quality and service from them (8). Physicians may be discouraged from tackling tough cases or caring for sicker patients if cost and utilization targets are set incorrectly.
Nonetheless, the pay-for-performance movement is creating a powerful new business model for both payers and providers that goes beyond historic jousting about reimbursement or medical policy. Unlike capitation systems that simply transfer financial risk from insurance companies to providers, the pay-for-performance concept allows payers and providers to focus on improving a manageable number of the most important issues facing them at a time, among the major categories being quality, service, and cost. Payers and providers are under equal pressure to improve the health system and should now work creatively together to explore the potential opportunities presented by the pay-for-performance concept.
FOOTNOTES
Author stated no financial relationship to disclose.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  G. Avery and J. Schultz Regulation, Financial Incentives, and the Production of Quality American Journal of Medical Quality, August 1, 2007; 22(4): 265 - 273. [Abstract] [PDF]
|
|
|
|
 |
|
 S. B. Leichter Pay-for-Performance Contracts in Diabetes Care Clin. Diabetes, April 1, 2006; 24(2): 56 - 59. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
