Federal and state policy makers are. Many Large Medical Groups Will Need To Acquire New Skills And Tools To Be Ready For Payment Reform

Transcription

1 doi: /hlthaff HEALTH AFFAIRS 31, NO. 9 (2012): Project HOPE The People-to-People Health Foundation, Inc. By Robert Mechanic and Darren E. Zinner Many Large Medical Groups Will Need To Acquire New Skills And Tools To Be Ready For Payment Reform Robert Mechanic (mechanic@ brandeis.edu) is a senior fellow at the Heller School for Social Policy and Management, Brandeis University, in Waltham, Massachusetts. Darren E. Zinner is a social scientist at the Schneider Institute for Health Policy and a senior lecturer at the Heller School, Brandeis University. ABSTRACT Federal and state policy makers are now experimenting with programs that hold health systems accountable for delivering care under predetermined budgets to help control health care spending. To assess how well prepared medical groups are to participate in these arrangements, we surveyed twenty-one large, multispecialty groups. We evaluated their participation in risk contracts such as capitation and the degree of operational support associated with these arrangements. On average, about 25 percent of the surveyed groups patient care revenue stemmed from global capitation contracts and 9 percent from partial capitation or shared risk contracts. Groups with a larger share of revenue from risk contracts were more likely than others to have salaried physicians, advanced data management capabilities, preferred relationships with efficient specialists, and formal programs to coordinate care for high-risk patients. Our findings suggest that medical groups that lack risk contracting experience may need to develop new competencies and infrastructure to successfully navigate federal payment reform programs, including information systems that track performance and support clinicians in delivering good care; physician-level reward systems that are aligned with organizational goals; sound physician leadership; and an organizational commitment to supporting performance improvement. The difficulty of implementing these changes in complex health care organizations should not be underestimated. Federal and state policy makers are experimenting with new payment models that seek to reward health care organizations for slowing the growth in spending while delivering high-quality care. The Centers for Medicare and Medicaid Services (CMS) has announced that thirty-two organizations, covering about 860,000 Medicare beneficiaries, entered the Pioneer Accountable Care Organization program on January 1, These organizations are responsible for managing a designated population of Medicare beneficiaries under predetermined budgets for three years and will share both savings and losses with the federal government. In April 2012 CMS announced the selection of the first twenty-seven organizations that are participating in the Medicare Shared Savings Program, and in July 2012 it announced the selection of another eighty-nine organizations to participate. 2 These organizations will also operate under global budgets but will not be responsible for paying back financial losses in the first three program years. Many health economists and policy analysts believe that new payment models like these are essential for addressing the unsustainable growth in health care spending. 3 Yet federal 1984 Health Affairs September : 9

2 officials made Medicare s new payment initiatives like the Shared Savings Program voluntary, in part because many health care organizations are not yet prepared to participate. If policy makers want to expand such initiatives more broadly across Medicare and Medicaid, they will need to understand the organizational characteristics, strategies, and operating models that are associated with success. They will also need to use this knowledge to support organizational learning across the provider community. In the 1990s numerous providers signed risk contracts primarily global capitation arrangements in which they received fixed per member per month payments to care for defined groups of health plan members. Yet many of these providers were unprepared to manage either financial risk or clinical care and lost substantial sums of money under these contracts. 4 By 2000 most payer and provider organizations had shed capitation contracts and returned to fee-for-service arrangements. 5 As a result, very few organizations have recent experience operating under contracts that hold them accountable for cost and quality, such as Medicare s Pioneer Accountable Care Organization and Shared Savings Programs. Yet some medical groups never ceased operating under risk arrangements like capitation. We wanted to assess the extent to which integrated delivery systems and large multispecialty groups already participate in capitation or other riskbased payment arrangements. We also wanted to determine whether groups that have a large proportion of their revenues linked to predetermined budgets differ in their structure and operational processes from similar groups that predominantly receive fee-for service payments. We surveyed twenty-five physician groups affiliated with the Council of Accountable Physician Practices, whose members include some of the largest multispecialty groups in the United States. The survey is intended to provide a snapshot of medical groups readiness for payment reform and to identify areas that groups must address to advance their preparedness. Study Data And Methods Data With the assistance of the Council of Accountable Physician Practices, we identified the CEO or chief medical officer of the twentyfive affiliated medical groups. In the summer of 2011, we sent each representative a ten-page questionnaire that asked for details about his or her group s organizational structure; financial and utilization statistics; physician compensation policies; information management capabilities; distribution of 2010 patient revenue by type of reimbursement model (fee-for-service, pay-for-performance, episode payment, partial capitation, shared savings, shared risk, and global capitation); and progress implementing performance improvement strategies. We received responses from twenty-one groups, for an 84 percent response rate. The survey instrument is available in the online Appendix. 6 From June to October 2011, we conducted onehour follow-up interviews with the senior executives at each group to obtain additional details about the groups history, local market environment, organizational strategies, and interest in expanding their risk contract volume. An advisory committee of medical directors affiliated with the Council of Accountable Physician Practices reviewed our survey instrument, analysis plan, and study findings. Methods One aim of the survey was to assess whether there was an association between the proportion of revenues that groups received from risk-based contracts and their operating characteristics. We defined risk-based contracts as those that paid groups based on full and partial capitation, as well as shared risk and shared savings methods. And we divided the groups into two categories of equal size: fee-for-service (0 34 percent risk-based contract revenue) and risk-based ( percent risk-based contract revenue; see the discussion of this point in the Limitations section). Approximately half of the surveyed medical groups treat patients who are enrolled in a closely affiliated health plan that is owned by the medical group or that is part of a larger health system that the group also belongs to. For example, a medical group in Geisinger Health System in Pennsylvania would treat patients in Geisinger Health Plan. Some of these plans pay the affiliated medical groups on a fee-for-service basis, while other plans use global capitation. The executives we interviewed said that regardless of the method of payment between the affiliated entities, they viewed this revenue as capitation because the overall organization is at risk. Consequently, we classified payments from these closely affiliated health plans as risk-based payments. Limitations Our study has a few important limitations. First, our sample of medical groups is not nationally representative. Groups in the Council of Accountable Physician Practices are larger than typical medical groups, employ their own physicians, and in many cases have histories as prepaid group practices that is, they either were health maintenance organizations or essentially operated under capitation for years. 7 Second, our survey has a relatively small sample size, which limits our ability to analyze sub- September : 9 Health Affairs 1985

3 categories of groups. Third, we interviewed only one or two senior managers from each group, and the information they provided may not fully represent the organization as a whole. Fourth, our classification of organizations into fee-for-service or risk-based groups is not based on any preexisting definition. Whether 34 percent is an appropriate upper level for defining fee-for-service groups and whether 45 percent is an appropriate lower bound for risk-based groups are topics for further research. However, our interviews with executives support the premise that there is a tipping point at which the operating approach of organizations begins to change. The executives suggested that these figures reflect a reasonable range. We conducted several sensitivity analyses with cutoff points ranging from 20 to 50 percent, with essentially similar results. Despite these limitations, this study is broadly relevant to policy makers, payers, and provider organizations interested in payment reform because it details a range of operational processes and strategies that have been adopted by riskbased medical groups. Other health care organizations will have to consider similar approaches as they prepare to enter into new payment arrangements, whether or not they are comparable in structure to the medical groups in our survey. Characteristics Of Surveyed Groups The surveyed medical groups varied in size, geographic location, and degree of integration with other health care facilities (Exhibit 1). For example, the groups operated in fourteen states and nineteen metropolitan regions. Overall, the surveyed groups were more likely than physician groups nationally to be affiliated with health plans and hospitals. They were also substantially larger: Nearly two-thirds of the groups in our survey employed more than 440 physicians. James Robinson and colleagues estimated that fewer than 9 percent of the roughly 900 US medical groups with at least twenty doctors had that number of physicians. 8 Thus, roughly 16 percent of the nation s very large groups were included in our survey. The service capacity of the groups also varied considerably. On average, about one-third of employed physicians were in primary care, although some groups had up to 50 percent primary care physicians. Twelve groups (57 percent) belonged to delivery systems that included Exhibit 1 Surveyed Medical Groups, By Location, Size, And Structure Name Location FTE primary care physicians FTE specialists a Affiliated hospitals Atrius Health Eastern MA No Austin Regional Clinic Austin, TX No Billings Clinic MT, northern WY Yes Dean Health System Southern WI Yes Everett Clinic Northwestern WA No Geisinger Health System Central and northeastern PA Yes Group Health Physicians Seattle Tacoma, WA Yes HealthCare Partners Los Angeles, Southern CA No HealthPartners Medical Group Minneapolis St. Paul, MN Yes Henry Ford Medical Group Detroit, MI Yes Intermountain Medical Group UT, southeastern ID Yes Marshfield Clinic Central, western, and northern WI Yes Mayo Clinic Health System MN, IA, WI Yes Oschner Health System New Orleans Baton Rouge, LA No Palo Alto Medical Foundation Oakland San Jose San Mateo, CA No Permanente Medical Group National b 5,240 11, Yes Reliant Medical Group c Central MA No Scott and White Healthcare Central TX Yes Sharp-Rees-Stealy Medical Group San Diego, Southern CA Yes Virginia Mason Medical Group Seattle, WA No Wenatchee Valley Medical Center Central WA No Affiliated health plan Median d Yes Average d % e SOURCE Self-reported medical group survey. NOTE FTE is full-time equivalent. a Includes hospitalists. b The Permanente Medical Group is managed through the following eight regions: Northern California, Southern California, Colorado, Georgia, Hawaii, Mid-Atlantic (Maryland, Virginia, and District of Columbia), Northwest (Northwest Oregon and Southwest Washington), and Ohio. c Formerly the Fallon Clinic. d Not applicable. e 57.1 percent of practices have an affiliated health plan Health Affairs September : 9

4 hospitals, and four groups were affiliated with more than ten hospitals (Exhibit 1). Twelve groups (57 percent) had affiliated health plans. Only five groups (24 percent) were affiliated with neither hospitals nor health plans. Study Results Payment Mix And Extent Of Risk Contracting On average, the surveyed groups received 57 percent of their 2010 patient revenue from commercial insurers, 32 percent from Medicare, 7 percent from Medicaid, and 4 percent from other sources. 9 Their managed care revenue defined as payments from commercial, Medicaid, and Medicare health maintenance organizations was 44 percent of total revenue, but this varied widely across groups (median: 29 percent; interquartile range: percent). Fee-for-service payments represented 54 percent of 2010 patient revenues (unweighted average), excluding fee-for-service payments from affiliated health plans (Exhibit 2). Only three groups did not have any fee-for-service contracts. Capitation revenue averaged 30 percent (38 percent, when fee-for-service revenue from affiliated health plans was included). This is three to four times the proportion reported in national surveys of medical groups. 8 Two-thirds of the groups in our survey reported having at least one partial or global capitation contract. Shared savings and shared risk contracts represented less than 5 percent of their patient revenue in Only one group reported revenue from episode payments. 10 For the median group, 47 percent of its revenue was from risk-based contracts (interquartile range: percent; see the Methods section). Individual groups ranged from zero to 100 percent risk. Operating Characteristics We divided the groups into two categories fee-for-service and risk-based as described above. The fee-forservice groups received 88 percent of their revenue, on average, from fee-for-service contracts; half of the groups in this category received 95 percent or more. The risk-based groups received 71 percent of their revenue from risk contracts, on average. Nine of the eleven risk-based groups had more than half of their total revenue at risk, and two of these nine received all of their revenue from global capitation (data not shown). Risk-based groups were less likely than fee-forservice groups to have affiliated hospitals but more likely to have affiliated health plans (Exhibit 3). The fee-for-service groups that did have affiliated plans reported receiving a smaller share of revenue from those plans than did riskbased groups with affiliated plans (data not shown). Physician Compensation The differences in physician compensation practices among the groups appear to be associated with their contracting structure. 11 Nine of the ten fee-forservice groups based the majority ( percent) of primary care physician compensation on productivity (Exhibit 3). 12 The practices of riskbased groups were more dichotomous. Five of these groups paid 80 percent or more of primary care physician compensation as salary, whereas five paid 80 percent or more of compensation based on productivity (data not shown). We also asked groups about the extent to which physician compensation was based on objective measures of quality, satisfaction, or efficiency. The groups generally reported low rates of performance-based compensation for primary care physicians (median: 5 percent; interquartile range: 2 10 percent) and specialists (median: 2 percent; interquartile range: 0 6 percent). The risk-based groups paid a slightly higher share of compensation based on performance measures 5 percent for specialists and 12 per- Exhibit 2 Payment Mix For Surveyed Medical Groups, 2010 Average size of contracts as percentage of total patient revenue SOURCE Authors analysis of self-reported medical group survey data. NOTES Percentages are unweighted averages. Fee-for-service is a model in which individual services are paid based on a predetermined fee schedule. Pay-for-performance is a fee-for-service model in which at least 10 percent of physician payment is contingent on meeting specific cost, quality, or patient satisfaction targets. Shared savings is a model with a per member per month budget target, in which the payer shares any savings below the budget target with the medical group. Shared risk is a model with a per member per month budget target, in which the payer shares both savings below the budget target and losses above the target with the medical group. Global capitation is a model with payment linked to a per member per month budget target for all medical services, in which the medical group receives total payments equal to the target and bears full risk for expenses in excess of the target. Partial capitation is a model with payment linked to a per member per month budget target for specific categories of service, such as professional services or primary care services, in which the medical group receives total payments equal to the target for designated services and bears full risk for expenses in excess of the target. Other revenue includes payments from self-pay patients and payments for charity care. September : 9 Health Affairs 1987

5 Exhibit 3 Characteristics Of Surveyed Medical Groups, By Percentage Of Risk Contracting Characteristic Health system characteristic Fee-for-service groups (n = 10) Average number of FTE physicians a b Has affiliated hospital 80% 45% Has affiliated health plan Physician compensation a Primary care physician Productivity c Salary 8 39 Performance metrics d 3 12 Other 4 3 Specialist Productivity c Salary Performance metrics d 2 5 Other 4 1 Data management e Shared electronic health record Results management f Computerized order entry f Data warehouse and analytic software Patient disease registries Practice variation analysis Reminders g Automated drug warnings Management strategies h Strategies to reduce avoidable admissions and readmissions Strategies to reduce network leakage i Preferred relations with efficient hospitals and specialists Management programs for high-risk patients Strategies to reduce variation for episodes of care Patient engagement initiatives Risk-based groups (n = 11) SOURCE Authors analysis of self-reported medical group survey data. NOTES As explained in the text, we divided the surveyed groups into two categories of equal size fee-for-service (0 34 percent participation in contracts that produced risk-based revenue) and riskbased ( percent participation). FTE is full-time employee. a Calculated using unweighted averages of reported amounts. b Excluding the Permanente Medical Group because its number of primary care physicians (5,240) is larger than those of the other groups combined. c Including service volume, measured by relative value units reflecting the time, intensity, and complexity of different services. d Including quality, efficiency, or patient satisfaction targets. e Percent of groups reporting "fully implemented." f Includes laboratory and radiology tests. g Physician and patient reminders about interventions and screenings. h Percent of groups reporting "far along." i Network leakage refers to patients receiving medical care outside of the group s principal service network. cent for primary care physicians (Exhibit 3) although when we removed a single group, the primary care average fell to 7 percent (data not shown). These results indicate that most groups relied on salary or productivity as the basis of compensation and had taken only limited steps toward compensating physicians for objective measures of quality, patient satisfaction, or efficiency. Information Management We asked the groups whether they had fully implemented, partially implemented, or not implemented specific electronic information management tools. Eighty-five percent of the groups overall reported that they had fully implemented electronic health records. But there were large differences in the deployment of analytic tools to support patient management. All of the risk-based groups reported that they had fully implemented systemwide electronic data warehouses and analytic software for performance measurement, compared with only one of the ten fee-for-service groups (Exhibit 3). Approximately two-thirds of the risk-based groups reported that they had developed patient registries and were analyzing practice variation within the group, whereas only 10 percent of the fee-for service groups had fully developed these capabilities. Risk-based groups also reported greater use of tools to help clinicians and their 1988 Health Affairs September : 9

6 staff deliver more effective clinical care, including automated drug warnings, and physician and patient reminders for interventions and screenings. 13 Management Strategies Most of the groups reported engaging in a range of management strategies. All but two groups reported using a formal management method such as Lean production or Six Sigma for performance improvement, and many reported high rates of participation in these initiatives by physicians and staff members. Sixteen of the twenty-one groups reported that they were far along in implementing processes to reduce avoidable hospital admissions and readmissions. Such practices are consistent with efforts to improve quality as well as the federal government s current focus on reducing avoidable readmissions. A much higher percentage of risk-based groups than fee-for-service groups reported that they were far along in implementing programs to reduce network leakage, or referrals of patients outside the group s network of providers (Exhibit 3). Risk-based groups were also more likely to have developed preferred relationships with efficient specialists and hospitals, with the goal of directing an increased volume of referrals to providers who have demonstrated the ability to deliver high-quality care and contain costs. This reflects the fact that risk-based groups are financially accountable for cost and quality even when their patients are treated by nonaffiliated providers, while fee-for-service groups are not. Risk-based groups had also implemented management programs for high-risk patients at nearly twice the rate of the fee-for-service groups (Exhibit 3). Neither category reported much progress in implementing patient engagement initiatives. Future Risk Contracting Expectations We sent the groups a survey addendum in the fall of 2011 and asked them to predict how their reimbursement contracts were likely to change by 2013 (a copy of the survey addendum can be found in the online Appendix). 6 Several groups would not make quantitative estimates, and two fully integrated groups will, by definition, experience little change in payment structure. We combined the estimates of the thirteen other groups. Collectively, they predicted that their fee-for-service revenue would decline by eighteen percentage points, with compensating growth in revenue from shared savings programs, pay-for-performance contracts, and global capitation (Exhibit 4). The fee-for-service groups predicted a more rapid decline in fee-for-service revenue (a reduction of twenty-six percentage points) that would be primarily replaced by shared savings (an increase of fourteen percentage points). More than half of all groups, including 75 percent of the feefor-service groups, predicted that they would make changes to physicians pay within the next two years to align incentives with these new contracts (data not shown). We asked the groups about the challenges they face in increasing their revenue from risk-based contracts. We classified these responses into five categories and asked the groups to rate their importance in the survey addendum. The biggest perceived challenge concerned the need to expand their internal processes to analyze, evaluate, and manage the care they deliver. This was especially true for fee-for-service groups, which collectively saw the need to improve their data management systems (75 percent reported that this was very important ) and care management systems (88 percent said that this was very important ). Many groups also identified payer capacity as a limiting factor, noting that health plans in their market have a limited willingness or capacity to offer risk contracts (36 percent considered this very important, 21 percent somewhat important ). Other groups cited the challenge of establishing risk contracts under preferred-provider arrangements, which dominate many local markets (29 percent said this was very important, 43 percent somewhat important ). In this regard, groups worried about taking financial risk in products that give patients unlimited choice in where they receive care. Contrary to our preliminary hypothesis, the senior managers of these groups did not believe that their physicians would oppose these new contracts (29 percent indicated that physician resistance was somewhat important ; none ranked it as very important ). Discussion The Council of Accountable Physician Practices medical groups that we surveyed are not characteristic of American medicine. They are large, established, multispecialty groups that employ physicians, use sophisticated electronic health record systems, and have well-defined cultures that emphasize collaboration among medical team members. These groups deliver high-quality patient care, and many health policy makers and policy analysts view them as prototypes for accountable care organizations. 14 Despite their strong organizational capabilities and their commitment to the principles of accountable care, we found that fee-for-service September : 9 Health Affairs 1989

7 Exhibit 4 Expected Payment Mix For Surveyed Medical Groups In 2013 Fee-for-service Fee-for-service with pay-for-performance Episode-based payments Shared savings Shared risk Partial capitation Global capitation Other Percent SOURCE Authors analysis of self-reported survey. NOTES Includes data from thirteen of the twentyone surveyed groups. Descriptions of the payment models are presented in Exhibit 2. For a definition of "episode-based payments," see Note 10 in text. was still the dominant form of payment for these groups in Nonetheless, more than half of the groups received 45 percent or more of their revenue from some form of risk arrangement. Decisions to accept this level of risk are based on each organization s history, operating philosophy, and readiness, as well as the willingness of local health plans to offer such contracts. 15 Eight of the eleven risk-based groups in our survey have closely affiliated health plans. Many of these plans provide data, analytic support, and actuarial services to help manage their groups risk contracts. Most US medical groups do not own health plans, nor do they have comparable collaborative relationships with insurers. Our survey also suggests that most groups with limited risk-contracting experience have not invested in the costly data and analytic infrastructure necessary to manage these contracts effectively. Some health plans are developing collaborative models that include data analysis and consultative support to help providers succeed under risk contracts. 16 But because most providers do not have experience managing populations and lack the needed infrastructure, private and government payers will have to expand the use of such collaborative programs if they expect to accelerate the growth of accountable care models. Thirteen of the groups in our survey five of the eleven risk-based groups and eight of the ten fee-for-service groups also have affiliated hospitals. For risk-based medical groups, there are important advantages to having closely aligned hospital partners, such as capital to fund data and care management infrastructure and enhanced opportunities to manage patient care across the continuum of services. But risk contracting requires groups to manage hospital services as a cost center rather than a revenue center. The requisite cultural shift is extremely difficult for hospital managers to achieve, and it will probably pose an insurmountable barrier for some health care organizations. Views Of The Future Senior executives whom we interviewed said that they expected the market to evolve away from fee-for-service reimbursement. Eleven out of sixteen respondents said that they were actively exploring new payment models. These executives predicted that the largest immediate change would be an expansion of shared savings arrangements. This estimate reflects the implementation of Medicare s Shared Savings Program as well as the growing prevalence of private payers offering such arrangements. 17 But the pace of predicted change varies according to each group s level of experience and local market conditions. Less experienced groups generally preferred a gradual expansion of pay-for-performance and shared savings contracts. In contrast, some experienced groups in our survey envisioned moving their entire business toward a global capitation model. These groups include four that have already joined Medicare s Pioneer Accountable Care Organization program. 18 Not all of the groups we surveyed wanted to expand their risk contracting portfolio. Some expressed skepticism that payers would work with them fairly and transparently. Others were concerned that payment models like the Medicare Shared Savings Program would penalize efficient providers that have already altered their practice style to achieve lower costs. To the extent that future capitation, shared savings, or bundled payment rates are based on current spending profiles, the most efficient groups believe that they face more risk and less opportunity than their less efficient counterparts. Medical groups and integrated delivery systems view payment reform with both optimism and concern. Their apprehension reflects the difficulty of implementing change in complex health care organizations, a reality that is greatly underappreciated by policy makers and pundits. For example, implementing evidence-based clinical protocols, reducing practice pattern variation, and changing physician compensation to align it with performance-based contracts all typically require multiyear efforts to engage stakeholders and build a consensus for change. 19 These groups believe that health care organizations face a challenging financial future. However, they also recognize that payment reform offers a competitive advantage for groups that 1990 Health Affairs September : 9

8 are capable of ongoing performance improvement. Conclusion With a growing national focus on payment reform, health care organizations may feel pressured to enter new payment arrangements before they are ready. Taking on substantial risk without a road map could lead some on a voyage back to the 1990s. Our survey and interviews suggest that the following are key elements of a readiness road map: information systems that track performance and support clinicians in delivering good care; physician-level reward systems that are aligned with organizational goals; sound physician leadership; and an organizational commitment to support performance improvement. These elements will not simply materialize with payment reforms. It is incumbent on policy makers, public and private payers, physician associations, and successful medical groups to encourage accountable care through leadership, information, and targeted investments in performance improvement infrastructure. This work was supported by a grant from the Commonwealth Fund. The authors thank Jay Crosson, Nancy Taylor, and the many Council of Accountable Physician Practices members who provided invaluable insights into the workings of their respective organizations. NOTES 1 HHS.gov [Internet]. Washington (DC): Department of Health and Human Services. Press release, Affordable Care Act helps 32 health systems improve care for patients, saving up to $1.1 billion; 2011 Dec 19 [cited 2012 Aug 1]. Available from: pres/12/ a.html 2 HHS.gov [Internet]. Washington (DC): Department of Health and Human Services. Press release, HHS announces 89 new accountable care organizations; 2012 Jul 9 [cited 2012 Aug 1]. Available from: pres/07/ a.html 3 Fisher ES, McClellan MB, Bertko J, Lieberman SM, Lee JJ, Lewis JL, et al. Fostering accountable health care: moving forward in Medicare. Health Aff (Millwood). 2009;28(2): w DOI: /hlthaff.28.2.w Robinson JC. Physician organization in California: crisis and opportunity. Health Aff (Millwood). 2001;20(4): Bazzoli GJ, Lee SY, Alexander JA. Managed care arrangements of health networks and systems: a review of the 1999 experience. J Ambul Care Manage. 2003;26(3): To access the Appendix, click on the Appendix link in the box to the right of the article online. 7 Council of Accountable Physician Practices. About CAPP: our beliefs [Internet]. Alexandria (VA): CAPP; [last updated 2011 Sep 9; cited 2012 Aug 1]. Available from: Robinson JC, Casalino LP, Gillies RR, Rittenhouse DR, Shortell SS, Fernandes-Taylor S. Financial incentives, quality improvement programs, and the adoption of clinical information technology. Med Care. 2009;47(4): Because the groups varied in size, the percentages reported for patient revenue and payment method were calculated using unweighted averages. In other words, each group s individual value was given equal weight when calculating the overall average rather than giving each group a weight based on its size. 10 Episode-based payment combines payments for all relevant services (for example, hospitalization, physician services, postacute care) into a single amount for specific episodes of clinical care, such as total joint replacement or congestive heart failure, over a defined period of time. 11 Robinson JC, Shortell SM, Rittenhouse DR, Fernandes- Taylor S, Gillies RR, Casalino LP. Quality-based payment for medical groups and individual physicians. Inquiry. 2009;46(2): Nine of ten fee-for-service groups paid specialists at least 65 percent based on productivity; seven of ten paid them at least 88 percent based on productivity. 13 Casalino L, Gillies RR, Shortell SM, Schmittdiel JA, Bodenheimer T, Robinson JC, et al. External incentives, information technology, and organized processes to improve health care quality for patients with chronic diseases. JAMA. 2003; 289(4): Weeks WB, Gottlieb DJ, Nyweide DJ, Sutherland JM, Bynum J, Casalino LP, et al. Higher health care quality and bigger savings found at large multispecialty medical groups. Health Aff (Millwood). 2010; 29(5): Bodenheimer T, Wang MC, Rundall TG, Shortell SM, Gillies RR, Oswald N, et al. What are the facilitators and barriers in physician organizations use of care management processes? Jt Comm J Qual Saf. 2004;30(9): Chernew ME, Mechanic RE, Landon BE, Safran DG. Private-payer innovation in Massachusetts: the alternative quality contract. Health Aff (Millwood). 2011;30(1): Higgins A, Stewart K, Dawson K, Bocchino C. Early lessons from accountable care models in the private sector: partnerships between health plans and providers. Health Aff (Millwood). 2011;30(9): These groups are Atrius Health, Austin Regional Clinic (as a partner in the Seton Health Alliance), HealthCare Partners, and Sharp- Rees-Stealy Medical Group (as part of Sharp Healthcare). 19 Kenney C. Transforming health care: Virginia Mason Medical Center s pursuit of the perfect patient experience. New York (NY): Productivity Press; September : 9 Health Affairs 1991

9 ABOUT THE AUTHORS: ROBERT MECHANIC & DARREN E. ZINNER Robert Mechanic is a senior fellow at the Heller School for Social Policy and Management, Brandeis University. In this month s Health Affairs, Robert Mechanic and Darren Zinner report on their evaluation of how well prepared medical groups are to participate in arrangements in which they would agree to deliver care within predetermined budgets. The authors survey of groups that are members of the Council of Accountable Physician Practices showed that even many of these large, multispecialty groups lack risk contracting experience and will need to develop new competencies and infrastructure to get ready for the new federal payment reform programs. Mechanic is a senior fellow at the Heller School for Social Policy and Management at Brandeis University, and executive director of the Health Industry Forum, a national program devoted to developing strategies to improve the quality and effectiveness of US health care. His research focuses on health care payment systems and the adaptation of organizations to new payment models. Previously, Mechanic was a senior health care analyst with Forrester Research and a senior vice president of the Massachusetts Hospital Association. From 1988 to 1998 he was a consultant and vice president with the Lewin Group, a Washington, D.C. based health care consulting firm, where he focused on hospital finance, state health policy, and health reform. He is a trustee of Atrius Health, a 1,000-physician multispecialty group practice. Mechanic earned an MBA in finance from the University of Pennsylvania. Darren E. Zinner is a senior lecturer at the Heller School, Brandeis University. Zinner is a social scientist at the Schneider Institute for Health Policy and a senior lecturer at the Heller School for Social Policy and Management, Brandeis University. He is also a senior member of Brandeis University s Health Industry Forum. Zinner s research focuses on the management and structure of the US scientific enterprise, especially on how funding for and the organization of scientific teams affects productivity and outcomes. He earned his doctorate in health policy from Harvard University and a master s degree in technology and policy from the Massachusetts Institute of Technology Health Affairs September : 9