 Research Insights Payment and Delivery Reform: Can Implementation Keep Up with Policy? Summary Introduction Payment and delivery reform are essential to containing costs in In the 1990s, employers and insurers pushed to remake the U.S. the U.S. health care system while addressing the need for improved health system in the image of integrated delivery organizations like quality. These efforts are directed at both the demand-side (patients Kaiser Permanente. The effort stalled in the face of consumer and and employers) and the supply-side (payers, physicians, clinical provider resistance, but concerns about cost and quality continued organizations), and the success of future reform relies on the to grow. In the aftermath, attention shifted to the development evaluation of current and prior experiments to improve value. and assessment of tools that could help produce results like those However, most of this experimentation has been piecemeal and observed in integrated organizations, but without a convulsive voluntary, embedded in widely varied local environments usually destabilization of the existing system. characterized by competitive insurance, hospital, and medical service markets. So, while the Affordable Care Act of 2010 seeks to promote These tools included guideline- or education-based strategies, the most promising innovations, little is known generally about what payment reform, performance-based payment, and coordination the most effective supply-side strategies are, how ready payers and and integration of the delivery system. During this time, research providers are to implement changes, or what policymakers should establishing correlations between treatment pathways and patient expect from ACA initiatives. outcomes flourished, making it possible to define preferred standards of practice. Aided by slow but steady growth in the This paper summarizes key points from an expert panel use of electronic health records, collection of data on provider AcademyHealth convened to identify how the knowledge performance increased, paving the way for new provider payment from existing research can inform policy development and incentives tied to performance data. Recognition of the importance implementation in this area. The meeting discussion covered a range of care coordination led to revived interest in organizational and of supply-side strategies to improve value, including: Accountable financing innovations like bundled payment, medical homes, and Care Organizations (ACOs), bundled, capitated, and performance- most recently accountable care organizations. based payment, and guideline- or education-based initiatives. During the meeting, the need for more precise performance Comprehensive research on the effectiveness of pay-for- measurement and better data emerged as key issues for successful performance and disease management strategies, for example, payment reform efforts. Participants also noted that prior delivery has been difficult to conduct because of the small scale of most system reform has largely occurred in receptive environments that programs, their diverse settings, and the myriad of potentially were well-situated to implementing needed changes. Future research confounding factors surrounding them. Many efforts are of recent should focus on the potential for success from these efforts in more origin, creating further difficulties for researchers who have only typical provider settings. limited data to work with, and must in many cases rely on insights from payers and providers from their front-line experience in implementing payment and delivery system changes. Payment and Delivery Reform: Can Implementation Keep Up with Policy? The varied experiments to-date provide a unique opportunity to But results of a forerunner program, the Premier Hospital Quality look at current and prior efforts to improve value and evaluate their Incentive Demonstration, were disappointing. Furthermore, a success in order to inform future implementation. In December, recent study projecting impacts of the VBP program estimated that 2012, AcademyHealth’s Research Insights project brought together two-thirds of participating hospitals would see payment changes of analysts with key public sector experts and leaders from the payer less than 1 percent.1 These results raise several questions that were community, clinical organizations, and the medical profession to reiterated by participants in the AcademyHealth meeting. How large review existing evidence and to assess research needs. Key questions a reward or penalty is needed to impact provider performance? addressed at the meeting included: More importantly, will payers have to spend so much to get providers’ attention that they lose money on such projects? Using • What is known about how payers, clinical organizations, and penalties to fund rewards is logical but difficult in practice given physicians are responding to external incentives to improve the need to balance P4P incentives between organizations already value? performing at a high level and others that are less accomplished but • What are the challenges or barriers to implementing promising improving. strategies? Private Initiatives • Is there evidence from existing research or current experience As a public, national program, Medicare is constrained by a need on the effectiveness of particular strategies employed by payers, for uniformity across regions and populations. Private payers are clinical organizations, and physicians? freer to shape their initiatives to specific market conditions. Large This brief presents a summary of the December meeting. Because private payers have invested heavily in electronic information the session was “off-the-record,” this document is intended to systems to guide market-specific strategies. Like integrated delivery convey the general content of the meeting without attributing organizations that use their own data systems to research treatment specific comments to particular participants. The discussion was effectiveness, some large private payers can tailor provider contracts informed by existing research, though neither it nor this brief to reflect the specific needs of their covered populations and the incorporates a systematic review of the literature on supply-side readiness of local providers for performance incentives or the strategies to improve value. We incorporate a bibliography of assumption of risk and responsibility for care quality, patient important current literature on the topic at the end of the brief, a outcomes, and costs. subset of which is specifically referenced in the text. One such effort is UnitedHealth Group’s Premium Physician The Payer Perspective Designation Program, which began in 2005 and involves about Initiatives by payers to improve the value of the health services 250,000 physicians in 41 states. Using a large suite of measures they buy are increasingly common, as are efforts to collaborate developed by the National Quality Forum and the National with providers in implementing new payment strategies. Since Committee for Quality Assurance, UnitedHealth uses claims data the effectiveness of incentives is dependent on the share of a to rate contracting physicians’ quality by determining whether their provider’s patients that a given payer represents, much of what is patients receive care that conforms to evidence-based standards. known about the effectiveness of new payment strategies comes Those whose performance scores rank in the top 25 percent receive from large payers like Medicare and a handful of private insurers a quality “star” and are eligible for incentive payments. who count their members by the millions. In a few instances, multipayer initiatives have been launched to achieve comparable For purposes of this discussion, what is perhaps most notable about leverage. But multipayer collaborations are hedged by antitrust the program is what its very robust aggregation of data reveals about rules and competitive realities. variations in cost within and across regions and markets. For some chronic conditions, there were ten- and twenty-fold differences in Thus, Medicare has often been a leader in payment innovation. cost between the 10th and 90th percentiles of participating physicians, The program has been a pioneer in quality reporting, often seen all of whom had first to receive a quality designation to be rated on as a necessary first step toward value-driven payment. Under the efficiency. Cost differences for major procedures were significant, Affordable Care Act, Medicare launched the most ambitious pay- but not as large.2 for-performance (P4P) plan yet, the Hospital Inpatient Value-Based Purchasing (VBP) Program, in late 2012. The program will involve For UnitedHealth, these cost differences represent important more than 3,000 hospitals and have $850 million in quality bonuses opportunities for improvement and savings. But these findings to distribute in its first years. also have cautionary implications for other efforts to improve performance on quality and cost. One is that higher measured 2 Payment and Delivery Reform: Can Implementation Keep Up with Policy? quality does not necessarily translate into lower costs. Another is Payers have this experience to remember as they seek to encourage that with such large differences in performance, bringing a majority provider organizations such as medical homes and accountable of providers toward a reasonable norm is not likely to happen care organizations to take responsibility for the overall health of quickly. Meeting participants observed readiness varies widely, and their attributed service populations – on a budget. Risk sharing, policymakers need to set realistic expectations for the pace of the a well-travelled concept that has at times been referred to as “risk “change wave.” corridors”, has recently emerged as a testable middle ground between fee-for-service and fully capitated payment. Limitations Not all observers are satisfied with the adequacy of the measures on As an example, Blue Cross Blue Shield of Massachusetts, with which most value-based payment projects are based. There are few about a 50-percent share of the state’s private insurance market, measures that capture diagnostic accuracy, waste, overuse, surgical initiated its Alternative Quality Contract in 2009, with 11 provider outcomes, or the management of severely ill and complex patients organizations signed up by 2010. The approach is similar to that with multiple co-occurring conditions. Challenges exist as well for of the Medicare Shared Savings Program and payment design for the construction of reliable and consistent measures of episode accountable care organizations under the Affordable Care Act. The costs on which the value equation must also rest.3 key to the shared risk approach is prospective estimation of an enrolled populations’ expected yearly costs, based on past spending So after a decade of experimentation, some inherent limitations of of individuals in the group, their demographic characteristics, and performance-based incentives are apparent. First and foremost, they their health status. The contracting organization shares savings with are generally designed to fit together with fee-for-service payment the payer if actual costs are less than the estimate, and shares the and therefore can do little, per se, to displace a payment mechanism extra cost if spending is greater. widely believed to be responsible for the excesses of the U.S. health system. Further, many payers lack enough market presence to The concept is simple but requires significant technical capabilities influence providers strongly. The sensitivity of providers to rewards —especially in data management—and a willingness to work is difficult to observe and calibrate. Existing performance measures collaboratively between payers and providers. Detailed patient data are incomplete. In fragmented markets, payers will have difficulty must be marshaled to make accurate estimates of future costs, the assessing the performance of small provider organizations with too notoriously difficult challenge for any system of risk adjustment. few patients for statistically valid evaluation. Where P4P has been in Movement of patients in and out of provider groups must be tracked place over a period of years, measured differences in performance and adjusted for. Measurement and attribution can be especially among providers may flatten out and blunt payment differentials. complicated when shared risk enrollees are only part of a provider Robust data capture and careful, creative research will be ever more in organization’s patient panel. Accounting for care received outside the demand to guide payers through such uncertainties. participating organization represents an additional challenge. Shared Risk A survey of 27 payer and provider organizations who have To repeat, the limitations of P4P programs reflect the limitations participated in shared risk programs paints a picture of an arduous of fee-for-service payment as a vehicle for promoting value. negotiation process, stretching out over years in some cases, to The 1990s demonstrated mixed results in scaling up the use of arrive at agreements in which both payers and providers are satisfied capitated payment associated with exemplary integrated delivery that their interests are served and protected. “Each design choice organizations. In general, it was evident that while well-managed exhibits tensions between conflicting goals and interests,” a report provider organizations could control their costs per case, they on the survey concludes. “Payers’ desire to protect themselves from lacked the actuarial expertise and access to capital with which overpayment by imposing statistically defensible confidence limits on insurers negotiate the random incidence of disease and mischance. payment thresholds conflicts with the notion that nascent programs Health maintenance organizations tended to use per-capita may need to offer first-dollar incentives to attract early adopters.”5 payment in limited ways. Some large provider organizations were paid full capitation, but more often HMOs capitated just An analysis of the first two years’ experience with the Alternative hospitals or just physicians, and in the latter instance more often Quality Contract in Massachusetts found that spending was primary care physicians than specialists.4 Despite a brief spate reduced by 2 percent relative to a matched comparison group. of enthusiasm for provider-based health plans, few of these were Quality scores also improved, but it is not clear that the savings successful or lasting. were associated with improved clinical performance. Some reduction in utilization was observed, but gains were made primarily by shifting procedures, tests, and imaging to lower- 3 Payment and Delivery Reform: Can Implementation Keep Up with Policy? priced facilities. Moreover, the analysis found that with the The Patient Perspective* infrastructure support paid to participants by Blue Cross Blue Payers may also direct incentive strategies toward consumers. Shield, along with the budget savings retained by them, the Disease management (DM) programs began to develop more insurer’s total costs probably exceeded its savings.6 than a decade ago to educate chronic illness patients about how to manage their condition and adhere to their medica- A Multipayer Approach tion regimes. The effectiveness of the DM programs has been A counterfactual example illustrates in more detail the challenges difficult to ascertain. Program data has often been held back of taking payment innovation to scale in competitive markets for proprietary reasons. Controlled trials that can isolate DM where no single payer can dictate change. With the state acting program effects from comorbidities and behavioral factors are as convener, Maryland initiated an unusual Multipayer Patient operationally daunting and likely to be prohibitively expensive. Centered Medical Home program in 2011. A unique infrastructure existed in the state’s Health Care Cost Commission, which Recent years have also seen renewed interest in workplace administers Maryland’s all-payer rate-setting system and maintains wellness programs to incentivize illness management and an all-payer claims database. The state’s five largest private payers, healthy behaviors. Again, many confounding variables have Medicaid, and state and federal employee groups participated. The attenuated research findings on the effectiveness of these program made infrastructure contributions to 52 geographically- programs, as firms seek to determine their return on investment. It is both enlightening and challenging that behavioral dispersed practices of varying size, with 250,000 attributed economists have entered the conversation about patient patients. Modest savings were achieved in the early months, but incentives. They bring a revitalized understanding of how people measurement challenges were encountered in the course of meeting make decisions that promises new incentive approaches, but the program’s quality and NCQA certification goals. which also calls into question some current practices. The enabling conditions on which the Maryland program was More integrative approaches to leveraging patients’ energies built can only be imagined in most other settings. Multipayer are moving forward under the rubric of patient “engagement” rate alignment and data flows, a culture of collaboration, an or “activation.” Some early studies show promise for patient organizational framework for program implementation, and a engagement strategies, but results again are difficult to legislatively-mandated evaluation component are all assets that disentangle in complex behavioral environments. The consensus other multipayer endeavors must, for the most part, build from among exponents is that patient engagement efforts need multi- scratch with sustained and vigorous effort.7 channel reinforcement and a committed effort from provider organizations. So payers may choose to invest in them but for the most part cannot themselves control their implementation. The AcademyHealth discussants identified several aspects of recent payer innovations that will need scrutiny. One is how * Although demand-side or patient strategies were not a topic of discussion during the December providers are responding in terms of organizing care processes, 2012 Research Insights meeting, patient/consumer engagement strategies will be the primary agenda item in a related but separate Research Insights meeting to take place in June 2013. managing population health, coordinating across care sites, and the like. How are consumers responding to changes in cost- The Provider Perspective sharing and patient engagement efforts? What are the effects of provider consolidation on local markets? Clinical Organizations Payer initiatives are generally voluntary for clinical organizations Based on all these experiences with payment reform, one and tend to attract provider groups that already have some question looms above all others in assessing the prospects for experience with P4P, risk contracting, preferred referral networks, transformation: Can new payment and delivery models that have and developed information technology systems. With the achieved modest success in conducive environments replicate interest that Medicare and many large private insurers now their accomplishments across a national landscape marked by have in contracting with relatively large-scale accountable care fragmentation and fractiousness? Evaluating the progress of diverse organizations, providers will be under increased pressure to experiments across varied environments was one of the research participate or risk losing patients. Physicians and hospitals will be, challenges most often noted by meeting participants. to some extent, thrown together without necessarily having long experience of working closely together, although the premise on which the ACO concept is built is that there are naturally occurring constellations of providers defined by which hospitals community physicians send their patients to most often. 4 Payment and Delivery Reform: Can Implementation Keep Up with Policy? Financial incentives may bring physicians to the table with leadership, strong payer-provider relationships, and experience with hospitals and diverse specialties. But most preliminary research performance-based payment.10 But, as with Maryland’s multipayer suggests that aligning responsibilities and finances with new experiment, such conditions are the exception rather than the rule. partners and implementing improved clinical functioning is a slow and stressful process that requires determined leadership Intensive organizational commitment and a strong supporting and, ultimately, culture change. The hope for ACOs is that they environment are the hallmarks of other early efforts at care will mimic the performance of integrated organizations that have transformation. In many experiments, bilateral arrangements often succeeded in improving quality and curbing costs. But in between large payers and providers show promise when both the 1990s, it proved to be difficult to overcome the legacy of a parties commit to working out the complexities of new payment fragmented system in which generalist physicians, specialists, arrangements. The University of North Carolina Health Care nurses, and hospitals pursued separate agendas. System and Blue Cross and Blue Shield of North Carolina formed a jointly-owned enterprise to deliver team-based care to a subset of Stimulated in part by interest in patient-centered medical homes, chronically-ill BCBS patients, staffed by pharmacists, nutritionists, some small organizations have achieved modest successes in behavioral therapists, and case managers as well as physicians and implementing care teams, adopting electronic information systems, nurses. Reimbursement to the practice is based on both a standard, and expanding the roles of nurses, physician assistants, and medical fee-for-service claims payment and a shared savings payment that assistants. Medical and nursing schools have increased efforts to compares the attributed group’s total medical costs relative to a foster interprofessional education, although often in the face of matched comparison group. This is supported by a robust, same- frustrating institutional inertia and territoriality. day data-sharing platform. Some recent research suggests that the difficulties these efforts have Even with two such dominant partners—BCBS has a North encountered may be due in part to unexplored depths of social and Carolina market share just above 50 percent—Carolina Advanced behavioral factors. Sensitivity and communications training, for Health started up on a limited scale. Other provider groups have example, might fail to change physicians’ disinclination to work expressed interest in participating. UNC’s Shep Center and the collegially with nurses if physicians identify themselves and their RAND Corporation are assisting with evaluations that will help to peers as authorities to whom nurses are meant to be subordinate. determine if the model can be expanded.11 Of particular interest Physicians may also resist the efforts of the management of clinical will be seeing how viable interprofessional teams will be in rural organizations to impose new workflows and clinical processes, even areas often marked by small physician practices, health workforce if logically designed, if they perceive management as less educated shortages, and lagging information technology resources. and less competent than themselves, or suspect that efficiency measures undermine medical objectives. In such instances, financial All along the learning curve, there is a need for understanding incentives may fail to gain traction.8 the success factors for organizations that aspire to improve their clinical and financial performance. Here there is a notable absence A study of four start-up ACOs participating in a learning of magic bullets. A recent survey of large multispecialty medical collaborative jointly facilitated by Dartmouth University and the groups that are arguably the organizations best positioned to do Brookings Institution describes a deliberate strategy of harnessing well in a shared-risk environment—all are members of the Council social identity, rather than threatening it: of Accountable Physician Practices—found that only sustained experience with risk contracting seemed to produce the capabilities An independent practice association preserved members’ needed for transforming care. cherished value of autonomy by emphasizing coordination, not “integration”; a medical group promoted integration The study compared groups with a relatively low share of revenue within its employed core, but not with affiliates; a hospital, from shared risk and full or partial capitation (less than 34 percent) engaging community physicians who mistrusted integrated to others with a higher share (more than 45 percent). Among all the systems, reimagined integration as an equal partnership; an surveyed groups, the average was three to four times greater than integrated delivery system advanced its careful journey toward in a national survey. Key characteristics of the groups doing more intergroup consensus by presenting the ACO as a cultural, not risk contracting differed markedly from the comparison groups, structural, change.9 even though the latter had much more experience than the national average. The risk-based practices had almost five times as many All four sites employed a shared savings model and worked salaried physicians; more than two times more use of computerized collaboratively with motivated insurance partners seeking value order entry; ten times more data warehousing and analytic software; enhancement. Success factors were identified as: committed 5 Payment and Delivery Reform: Can Implementation Keep Up with Policy? three times more disease registries; seven times more practice In some cases, their patient panels are too small to generate variation analysis; and twice as many preferred relations with statistically meaningful quality measurement. specialists. “There is a tipping point at which the operating approach of organizations begins to change,” the study authors concluded. Many of the underlying barriers to influencing physician practice emerged early in the drive to promote evidence-based medicine, Conversely, the authors noted a drop-off in risk-based contracting practice guidelines, performance measurement, and the aspirational after the disappointments of the 1990s and a widespread lack of science of quality improvement. In some cases, physicians rebelled mature capabilities as a result. Partnerships between physicians openly against the promulgation of evidence-based standards by the and hospitals as envisioned for ACOs have the potential to U.S. Agency for Health Care Policy and Research (now the Agency improve medical groups’ access to capital and information for Healthcare Research and Quality) and its patient outcome technology, and better manage care coordination across sites research teams, the PORTs, in the 1980s and ‘90s. Those skirmishes of care. But such relationships are perennially challenged by were layered over a contentious history of tensions originating conflicting imperatives, as hospitals seek to maximize admissions with the advent of third-party payment in the 1930s, which was while care managers seek to reduce them. And despite increased widely perceived by physicians at the time as an imminent threat to efforts at collaboration between payers and providers, a long professional sovereignty. Medicare’s statutory ban on interference history of adversarial relations breeds skepticism, while high- with the practice of medicine was a carry-over from the terms performing groups that emerging incentives may disadvantage medicine extracted from private insurers from the ‘30s onward. them relative to lesser organizations that will feast on low-hanging fruit. “The difficulty of implementing these changes in complex An authoritative literature review in 1999 summarized the health care organizations should not be underestimated.”12 difficulties with practice standards and associated measurement issues. The findings present a complex picture that goes even Individual Physicians beyond this legacy of bilateral tensions. Knowledge barriers were Research on individual physician responses to performance- identified as important in many surveys, often with reference based payment and risk sharing suggest that many are still on to an explosion of clinical research and journal articles, which the lower rungs of the “ladder of maturity” that leads to care many physicians simply did not have enough time to read. The transformation. In principle, most now accept the notion of literature on barriers to guideline adherence found a wide range value-driven payment and have at least some experience with of evidence on how frequently non-adherence is associated with performance measurement. But incentive payments alone are disagreement with guidelines. Physicians may interpret existing not generally considered to be sufficient motive for thorough- evidence differently, worry about patient risk, or reject the premises going behavior change. Some say they have already achieved of standardization, among other explanations. Some physicians many of the quality targets common to P4P programs and may don’t believe recommended changes will make a difference or have difficulty making further improvement, although guideline achieve promised outcomes. Some are bound by inertia; others feel adherence continues to fall below expectations in the eyes of many constrained by the time, effort, and financial resources they can payers and policymakers. Many are skeptical about the validity spare for improvements, or by factors beyond their control such as and importance of commonly-used measures, and about whether referral outcomes and patient behavior.13 quality improvement is more likely to be cost-reducing than cost- increasing. They want a greater role in designing practice-change Even within larger groups, where supports are available, individual models, and they worry about the impact of externally imposed physicians may not be engaged with organizational strategies to changes on their relationships with their patients. improve quality scores or control spending. The change agenda may be perceived as externally-imposed – the concern of a practice Another daunting obstacle is practice size. While employment of manager, not a physician working at the sharp edge of care. At one doctors by hospitals is increasing, most recent growth in average large, sophisticated independent practice association in Western practice size has been in single-specialty groups that may do New York, physicians in the mid-2000s balked at productivity little to facilitate care coordination. One third of all physicians rewards based on a ratio of actual to expected costs. They resented were still in solo or two-physician practices in 2004-2005; and the judgmental character of the system, and the fact that it did in 2009, two thirds of all office visits were to practices of five or not recognize that quality improvement and cost reduction don’t fewer physicians, according to the National Ambulatory Medical always go hand in hand. Global measures did not capture important Care Survey. Significant investments in information technology specialty- and condition-specific services, and physicians were held and ancillary staff are often beyond the means of these smaller accountable for events beyond their control. The outcome was a practices, leaving them without important tools to control costs. redesign and creation of a more granular—if also labor-intensive— 6 Payment and Delivery Reform: Can Implementation Keep Up with Policy? system for identifying outlier costs, and improved receptivity from centers to assist primary care providers in the adoption of health group members.14 Another study of the same group found that information technology tools, under the American Recovery and some of its member physicians saw value in the improvement Reinvestment Act of 2009. Modeled on state agricultural extension system, agreed with its goals, and responded cooperatively, programs that have helped farmers keep up with agricultural demonstrating the variability of physician views even within the science and best practices, the RECs are joint resource pools like same practice.15 Maryland’s learning collaboratives, designed especially for small and individual practices, with an emphasis on those serving The difficulty of controlling patient behavior is a persistent theme disadvantaged communities. in studies of the effectiveness of practice change programs, and reflects a dimension of patient-physician relations that may Based on the experience of larger organizations, most believe be difficult to reach with many change strategies. A study of a that health IT has the potential to provide clinical decision hypertension control program involving more than 80 physicians as support, enable e-prescribing and access to test results, maintain well as other non-physician clinicians at 12 Veterans Administration patient registries, and facilitate communication with patients – clinics found that financial incentive seemed to have no impact on all proven tools for improving care and efficiency. The centers provider behavior. Clinicians believed that outcomes depended encourage participation from universities with informatics and on patient behavior more than anything they did themselves. health professions programs; hospital, health center, and provider “We need mamas,” one frustrated physician complained. “If the networks; Medicare quality improvement organizations; public VA could find someone to make the patients take their medicine, health agencies; and professional organizations. They are available like a dorm mother or something, the numbers would look a lot to help physicians with vendor and product selection, technical better.”16 A study of regional, multi-stakeholder quality-reporting assistance with IT adoption, management expertise, workflow collaboratives found that consumers and plans valued report redesign, and meeting workforce needs. cards on individual physicians, but that the physicians themselves were resistant because they perceived that they were being held With many centers still in their formative stages, there is apparently accountable for patient behavior that was beyond their control.17 little research on their progress, and they were not discussed at the AcademyHealth meeting. A forerunner program, though, the A consensus of recent research seems to support the view that Primary Care Information Project in New York City, has been in multiple interventions and multi-dimensional support systems are operation since 2007. The project subsidizes software costs for needed to drive aggregate physician performance toward a tipping eligible practices in underserved neighborhoods, and provides point where payment and delivery of care can be fundamentally technical assistance and “coaching” after that. changed. As observed elsewhere in this report, payers, hospitals, large practices, and community institutions including universities To date more than 3,300 physicians in 600 practices have enrolled, and government can help supply an enabling infrastructure for and the PCIP was designated a regional extension center in 2011. individual physicians and small practices. An analysis of quality scores for a sample of enrolled practices reached a sobering conclusion. No significant association was Participants in the AcademyHealth meeting observed that recent found between participation and quality until after 24 months; and research on individual physicians’ performance and behavior no improvement was found for groups that received only one to often takes note of the localized and context-specific nature of three technical assistance visits. Significant quality improvement their challenges and conditions. The physicians’ own age, training, did occur in practices that received eight or more such visits. In employment status, and experience are important influences. other words, results do not come quickly or easily.18 The sociodemographic and epidemiological make-up of patient populations may vary widely across and within communities. Conclusion: Where to Invest? The balance of consolidation and competition among insurers, Research on the progress of payment and delivery innovations hospitals, and specialists may impinge on the operation of small is in critical demand from policymakers and stakeholders. The independent practices. Affordable Care Act placed a large bet on accountable care organizations, but without robust participation of providers it The variability of local environments—“riotously pluralistic,” will be difficult to achieve its goals. Many physicians and clinical as one respected academic has described it—poses a particular organizations may sit on the sidelines to see how early adopters challenge for large-scale policy interventions. One example that is fare in terms of financial and clinical results. They also have relevant for small practices was the allocation of some $644 million concerns about being regimented by payers or being forced into for establishment of a network of about 60 regional extension uncomfortable relationships with other providers. 7 Payment and Delivery Reform: Can Implementation Keep Up with Policy? Because of their worries about being held accountable for outcomes research-publications/find-rwjf-research/2010/12/moving-payment- that depend on patient behavior, physicians need to know—as from-volume-to-value0.html do payers—whether emerging patient engagement strategies can successfully improve chronic illness management and primary Cabana, MD, Rand, CS, Powe, NR, et al. Why don’t physicians follow prevention. Payers need to understand where they can achieve clinical practice guidelines? A framework for improvement. JAMA. positive returns on investment with incentives for providers and 1999;282(15):1458-69. patients, and how to cull meaningful data on quality and cost. Davis DA, Thomson MA, Oxman AD, Haynes B. Changing physician Robust data flow is essential for both payers and providers to performance: A systematic review of the effect of continuing medical manage cost and performance, but is difficult to achieve in education strategies. JAMA. 1995;274(9):700-705. fragmented markets. Payer-provider collaboratives can help meet this goal, but implementation of such arrangements in the absence Dudley RA. Pay-for-performance research: how to learn of fortuitous enabling circumstances remains to be proven viable. what clinicians and policy makers need to know. JAMA. On the other hand, ACOs may stimulate further consolidation of 2005;294(14):1821-3. provider groups, with the attendant dangers of stifling competition and choice, and driving up prices. Ellis P, Sandy LG, Larson AL, and Stevens SL. Wide variation in episode costs within a commercially insured population highlights The research agenda is thus open-ended. AcademyHealth meeting potential to improve the efficiency of care. Health Aff (Millwood). participants seemed to concur that prioritizing research investments 2012;31(9):2084-2093. should focus on identifying the most promising pathways to the “tipping point.” There will be many options. Frolich, A, Talavera, JA, Broadhead, P, Dudley, RA. A behavioral model of clinician responses to incentives to improve quality. Health About the Author Policy. 2007;80(1):179-93. Rob Cunningham is an independent writer in Washington, D.C. Gold, M. Accountable Care Organizations: Will they deliver? About AcademyHealth Mathematica Policy Research, Inc. 2010;1-17. Available online: AcademyHealth is a leading national organization serving the fields http://www.mathematica-mpr.com/publications/pdfs/health/ of health services and policy research and the professionals who account_ care_orgs_ brief.pdf produce and use this important work. Together with our members, we offer programs and services that support the development and Hoff TJ. The physician as worker: what it means and why now? use of rigorous, relevant and timely evidence to increase the quality, Health Care Manage Rev. 2001;26(4):53-70. accessibility, and value of health care, to reduce disparities, and to improve health. A trusted broker of information, AcademyHealth Greene RA, Beckman HB, and Mahoney T. Beyond the efficiency brings stakeholders together to address the current and future needs index: finding a better way to reduce overuse and increase efficiency of an evolving health system, inform health policy, and trans-late in physician care. Health Aff (Millwood). 2008;27(4):w250-9. evidence into action. For additional publications and resources, visit www.academyhealth.org Hoff, TJ. Practice under pressure: Primary care physicians and their medicine in the Twenty-First Century. Piscataway: Rutgers Selected Bibliography University Press; 2010. Audet AM, Kenward K, Patel S, Joshi MS. Hospitals on the path to accountable care: highlights from a 2011 national survey of hospital Hoff TJ, Weller W, DePuccio M. The patient-centered medical readiness to participate in an accountable care organization. Issue home: A review of research. Med Care Res Rev. 2012;69(6):619-44. Brief (Commonwealth Fund). 2012;22:1-12. Hysong SJ, Simpson K, Pietz K, Sorelle R, Broussard Smitham K, Baker DW, Qaseem A, Reynolds PP, Gardner LA, Schneider EC. Design Petersen LA. Financial incentives and physician commitment to and use of performance measures to decrease low-value services and guideline-recommended hypertension management. Am J Manag achieve cost-conscious care. Ann Intern Med. 2012;158(1):55-9. Care. 2012;18(10):e378-91. Berenson RA. Moving payment from volume to value: What role James J. Pay-for-Performance. New payment systems reward doctors for performance measurement? Robert Wood Johnson Foundation/ and hospitals for improving the quality of care, but studies to date Urban Institute. 2010;1-5. Available at http://www.rwjf.org/en/ show mixed results. Health Aff (Millwood) Health Policy Brief. 8 Payment and Delivery Reform: Can Implementation Keep Up with Policy? October 11, 2012. Available online: http://www.healthaffairs.org/ Robinson JC. Theory and practice in the design of physician healthpolicybriefs/brief.php?brief_id=78 payment incentives. Milbank Q. 2001;79(2):149-77. Korda K, Eldridge GN. Payment incentives and integrated care Rosenthal MB, Frank RG. What is the empirical basis for paying for delivery: Levers for health system reform and cost containment. quality in health care?. Med Care Res Rev. 2006;63(2):135-57. Inquiry. 2011;48:277-287. Rosenthal TC. The medical home: growing evidence to support Kreindler SA, Larson BK, Wu FM, et al. Interpretations of a new approach to primary care. J Am Board Fam Med. integration in early accountable care organizations. Milbank Q. 2008;21(5):427-40. 2012;90(3):457-83. Ryan AM, Bishop TF, Shih S, and Casalino, LP. Small physician Kreindler SA, Dowd DA, Dana Star N, Gottschalk T. Silos and practices in New York needed sustained help to realize gains social identity: the social identity approach as a framework for in quality from use of electronic health records. Health Aff understanding and overcoming divisions in health care. Milbank Q. (Millwood). 2013;32(1):53-9. 2012;90(2):347-74. Smith M.A. et al. Public reporting helped drive quality Larson BK, Van citters AD, Kreindler SA, et al. Insights from improvement in outpatient diabetes care among Wisconsin transformations under way at four Brookings-Dartmouth physician groups. Health Aff (Millwood). 2012;31:570-77. accountable care organization pilot sites. Health Aff (Millwood). 2012;31(11):2395-406. Song Z, Safran DG, Landon BE, et al. The ‘Alternative Quality Contract,’ based on a global budget, lowered medical spending and Lee GM, Kleinman K, Soumerai SB, et al. Effect of nonpayment improved quality. Health Aff (Millwood). 2012;31(8):1885-94. for preventable infections in U.S. hospitals. N Engl J Med. 2012;367(15):1428-37. Takach M. About half of the States are implementing patient- centered medical homes for their Medicaid populations. Health Aff Mechanic RE, Altman SH. Payment reform options: episode (Millwood). 2012;31(11):2432-40. payment is a good place to start. Health Aff (Millwood). 2009;28(2):w262-71. Town, R, Wholey DR, Kralewski J, Dowd B. Assessing the influence of incentives on physicians and medical groups. Med Care Res Rev. Mechanic R and Zinner DE. Many large medical groups will need 2004;61(3):80S-118S. to acquire new skills and tools to be ready for payment reform. Health Aff (Millwood). 2012;31(9):1984-91. Volpp KG, Loewenstein G, Asch DA. Choosing wisely: low- value services, utilization, and patient cost sharing. JAMA. Miller HD. From volume to value: better ways to pay for health 2012;308(16):1635-6. care. Health Aff (Millwood). 28(5):1418-28. Weissman JS, Bailit M, D’Andrea G, and Rosenthal MB. The design O’Malley AS, Bond AM, and Berenson RA. Rising hospital and application of shared savings programs: Lessons from early employment of physicians: Better quality, higher costs? Center for adopters. Health Aff (Millwood). 2012;31:1959-1968. Studying Health System Change. 2011;136:1-4. Werner RM and Dudley RA. Medicare’s new hospital value-based Owens DK, Qaseem A, Chou R, Shekelle P. High-value, cost- purchasing program is likely to have only a small impact on conscious health care: concepts for clinicians to evaluate the hospital payments. Health Aff (Millwood). 2012;31(9):1932-39. benefits, harms, and costs of medical interventions. Ann Intern Med. 2011;154(3):174-80. Young GJ, Beckman H, Baker E. Financial incentives, professional values and performance: A study of pay-for-performance in a Qaseem A, Alguire P, Dallas P, et al. Appropriate use of screening professional organization. J Organiz Behav. 2012;33(7):964-83. and diagnostic tests to foster high-value, cost-conscious care. Ann Intern Med. 2012;156(2):147-9. Young, G. Multi-stakeholder regional collaboratives have been key drivers of public reporting but now face challenges. Health Aff (Millwood). 2012;31(3):578-84. 9 Payment and Delivery Reform: Can Implementation Keep Up with Policy? Endnotes 10.Larson BK, Van citters AD, Kreindler SA, et al. Insights from transformations 1. Werner RM and Dudley RA. Medicare’s new hospital value-based purchasing under way at four Brookings-Dartmouth accountable care organization pilot program is likely to have only a small impact on hospital payments. Health Aff sites. Health Aff (Millwood). 2012;31(11):2395-406. (Millwood). 2012;31(9):1932-39. 11.See http://www.carolinaadvancedhealth.org/ 2. Ellis P, Sandy LG, Larson AL, and Stevens SL. Wide variation in episode costs 12.Mechanic R and Zinner DE. Many large medical groups will need to acquire within a commercially insured population highlights potential to improve the new skills and tools to be ready for payment reform. Health Aff (Millwood). efficiency of care. Health Aff (Millwood). 2012;31(9):2084-2093. 2012;31(9):1984-91. 3. Berenson RA. Moving payment from volume to value: What role for perfor- 13.Cabana, MD, Rand, CS, Powe, NR, et al. Why don’t physicians follow mance measurement? Robert Wood Johnson Foundation/Urban Institute. clinical practice guidelines? A framework for improvement. JAMA. 2010;1-5. Available at http://www.rwjf.org/en/research-publications/find-rwjf- 1999;282(15):1458-69. research/2010/12/moving-payment-from-volume-to-value0.html 14.Greene RA, Beckman HB, and Mahoney T. Beyond the efficiency index: find- 4. Gold, M. Accountable Care Organizations: Will they deliver? Mathematica ing a better way to reduce overuse and increase efficiency in physician care. Policy Research, Inc. 2010;1-17. Available online: http://www.mathematica-mpr. Health Aff (Millwood). 2008;27(4):w250-9. com/publications/pdfs/health/account_care_orgs_brief.pdf 15.Young GJ, Beckman H, Baker E. Financial incentives, professional values and 5. Weissman JS, Bailit M, D’Andrea G, and Rosenthal MB. The design and ap- performance: A study of pay-for-performance in a professional organization. J plication of shared savings programs: Lessons from early adopters. Health Aff Organiz Behav. 2012;33(7):964-83. (Millwood). 2012;31:1959-1968. 16.Hysong SJ, Simpson K, Pietz K, Sorelle R, Broussard Smitham K, Petersen LA. 6. Song Z, Safran DG, Landon BE, et al. The ‘Alternative Quality Contract,’ based Financial incentives and physician commitment to guideline-recommended on a global budget, lowered medical spending and improved quality. Health hypertension management. Am J Manag Care. 2012;18(10):e378-91. Aff (Millwood). 2012;31(8):1885-94. 17.Young, G. Multi-stakeholder regional collaboratives have been key driv- 7. See http://mhcc.maryland.gov/pcmh/ ers of public reporting but now face challenges. Health Aff (Millwood). 8. Kreindler SA, Dowd DA, Dana Star N, Gottschalk T. Silos and social identity: 2012;31(3):578-84. the social identity approach as a framework for understanding and overcom- ing divisions in health care. Milbank Q. 2012;90(2):347-74. 18.Ryan AM, Bishop TF, Shih S, and Casalino, LP. Small physician practices in New York needed sustained help to realize gains in quality from use of elec- 9. Kreindler SA, Larson BK, Wu FM, et al. Interpretations of integration in early ac- tronic health records. Health Aff (Millwood). 2013;32(1):53-9. countable care organizations. Milbank Q. 2012;90(3):457-83. 10