ACOs: A Concept in Motion
While health-policy experts debate the potential of accountable-care organizations (ACOs) to address the problem of cost in US health care, a panel held on July 22, 2011, at the RBMA Executive Education Program in Scottsdale, Arizona, reveals that with prototype ACOs already in play and health-care systems assessing their ACO needs, radiology practices should waste no time in securing a seat at the table. Otherwise, they risk being left on the sidelines, with limited options. The panel, “Setting Up an ACO: The Good, the Bad, and the Ugly,” included Palmer C. Evans, MD, senior advisor, Southern Arizona Accountable Care Organization (SAACO) in Tucson; Brent Hardaway, vice president, Premier Inc (Charlotte, North Carolina); and W. Kenneth Davis Jr, JD, partner, Katten Muchin Rosenman LLP (Chicago, Illinois). I also was a member of the panel. The Prototype Evans presented first on the ACO built by the 612-bed Tucson Medical Center (TMC) before the term gained currency. TMC shares the Tucson market with the Carondelet Health Network division of Ascension Health (St Louis, Missouri) and the University of Arizona Medical Center, among others; in this region, some systems are attempting to purchase additional practices, Evans notes. TMC is not pursuing that strategy, preferring to confederate independent practices. The seeds for the ACO were planted in 2007, when TMC discovered that it was going to have to partner differently with its physicians in order to succeed. One indicator was supply costs. TMC’s primary orthopedic practice was ordering more hips and knees from one major orthopedic vendor than any other practice in the country was, but it was paying list price. “In talking with the CEO, we decided the best place to start was in co-management,” Evans says. Beginning with orthopedics (and later adding neurosciences, cardiac imaging, and cardiovascular surgery), TMC announced each service-line partnership as an LLC, with 75% owned by the physicians and 25% owned by the hospital. “What that does is get collaboration on a different level,” he says. For starters, TMC saved $13 million on orthopedic supplies within 18 months. Late in 2008, TMC discovered that one of hospital’s primary-care groups had entered into a gain-sharing agreement with UnitedHealthcare (Minnetonka, Minnesota) that would result in shared savings if chronic-care patients could be kept out of the hospital. On notice to expect reduced admissions from a loyal referrer—but understanding that this was the right thing for patients—TMC reached out to UnitedHealthcare Western Region. It discovered that Tucson—with few major employers and many small businesses—was a tough market for UnitedHealthcare, and it was interested in working with a system in which physicians and the hospital were working toward better-quality care, lower cost, and higher efficiency. It had also launched seven patient-centered medical-home projects in Arizona, four of which were associated with TMC. Brookings Comes Calling When the Engelberg Center for Health Care Reform/Brookings Institution (Washington, DC) approached TMC to become one of three pilot sites for the Brookings/Dartmouth collaborative ACO model late in 2009, the hospital already had experience with medical homes and in collaborating with primary-care physicians to manage chronic-care patients. “That was the first time I heard the word ACO,” Evans recalls. Before accepting the offer, the board had to be convinced that TMC could reduce hospital admissions, improve population health, and remain viable. “Our board—which is a community board—as stewards of the community resource decided this was the best thing to do,” Evans says. SAACO is 80% physician owned and 20% hospital owned. With the aims of local accountability for cost, standardized performance/quality measures, and payment incentives for provider collaboration, SAACO formed a physician steering committee in early 2010 made up primarily of four primary-care physicians, a hospitalist, a cardiologist, and a general surgeon. SAACO subsequently formed a management-services organization (MSO) that will serve as a steering committee for the Medicare ACO that it plans to build. “The purpose of the ACO is to collect shared savings, distribute them, and measure physician quality,” Evans explains. “The MSO will be the engine that will drive the ACO.” The MSO, a commercial enterprise run by a team from some of the participant physician groups, is responsible for providing an administrative infrastructure for coordinating IT and case/disease management, implementing medical homes for primary-care physicians who want to participate, and coordinating communications. Meeting the Quality Mark TMC also has been gaining experience with quality metrics through its work as a pilot site for the Brookings Institution’s starter set of claims-based measures to assess quality of care (and determine payments) in the CMS ACO program. Of the 65 measures that must be met by Medicare ACOs beginning in January 2012, Evans estimates that his group could meet 52. “There is some question as to whether anyone can get to 65,” he notes. Evans is enthusiastic about some of the technology that will facilitate providing physicians with information about their quality performance. After reviewing three systems, TMC chose one (owned by UnitedHealthcare) that provides prospective predictive modeling of the likelihood that a patient will end up in the hospital, as well as some retrospective tools. Although TMC was an early supporter of the state health information exchange (HIE), Evans says that it was not moving quickly enough to meet the ACO’s needs, so it is developing its own private HIE to aggregate hospital and physician electronic medical records, PACS, and other health information systems in use throughout the ACO. It also will have the ability to tap into the state HIE. Evans says that an ACO can reduce spending by implementing initiatives that have been proven to work: efficient use of nurse practitioners and physician assistants, extreme care coordination, aligned provider incentives, chronic-disease management, the use of data and collaboration, health-risk assessments, evidence-based medicine, and accountability. “You know that 20% to 30% of patients are readmitted to the hospital within 30 days of discharge,” he says. “Medicare is going to start penalizing us for those readmissions.” Evans emphasizes the urgency of reducing costs: In 1965, when Medicare began, the average lifespan, after age 65, was four years; now, it’s 20 years. There were 10 taxpayers per beneficiary; now, there are three. There were fewer than 10 million beneficiaries; now, there are more than 44 million. By the time the last baby boomer turns 65, the over-65 population will be nearly double what it is today, and by 2050, one person in five will be 65 or older. “Tell me we don’t have a problem,” Evans says. The Proposed Rule: A New Kind of Provider Beginning with a discussion of the statutory roots of the ACO in the Patient Protection and Affordable Care Act (PPACA), I presented on the CMS proposed rule, which engendered a firestorm of criticism earlier this year. At press, CMS had sent the final rule to the Office of Management and Budget, preliminary to its publication in the Federal Register. The program is set to begin on January 1, 2012. The PPACA introduced the ACO, in March 2010, as a legal provider organization that is directly responsible for many Medicare services and that can ensure patient access to the rest. The law establishes a broad framework for ACO provider participation and the principles for sharing the savings. It makes clear that participation is voluntary, but it is a permanent program. While the law is vague, it makes eminently clear that an ACO must be patient centered, and its language strives to ensure maximum freedoms and rights for patients (perhaps to deflect charges that ACOs are reminiscent of 1990s-style managed care). It also specifies that a patient need not receive all care from the ACO to which his or her primary-care physician is attached. The law specifies that CMS payments must be linked directly to the quality of care. It includes shared savings and provides for a mechanism to distribute those savings, both to promote accountability for a patient population and to encourage investment in the infrastructure to support redesigned care processes. Eligible participants are physicians and other providers; networks of individual practices, partnerships, or joint ventures between hospitals and ACO professionals; and other groups of suppliers deemed appropriate by the DHHS. The law also stipulates that ACOs make a three-year commitment and provide enough primary-care physicians to care for their assigned populations (5,000 patients or more). Five Aspects In March 2011, the proposed rule—including proposals from the Federal Trade Commission, the IRS, and the Department of Justice—was issued to a hail of criticism. Many critics took exception to patients being assigned to an ACO based on use of a primary-care physician, internist, or geriatrician, thereby excluding specialists and the multispecialty groups that participated in the Medicare Physician Group Practice Demonstration. Critics also objected to having patients assigned to an ACO at the end of the year, after care has been delivered. Organization: An ACO’s structure must include an integrated organization that invests in quality improvement and cost containment, with dedicated physician leaders at the helm. The ACO also must possess the requisite health IT for data aggregation and feedback and must have experience with non-Medicare payor initiatives, particularly managed care. Reporting: Not only are ACOs required to report on many aspects of patient care, but they also must be able to process and analyze quarterly reports from CMS. At the outset, an ACO must include, in its application, its plans for promoting evidence-based medicine, beneficiary engagement, quality and cost measures, and coordination of care. The ACO will be measured against these self-developed standards. CMS proposes making this a two-way flow of information by reporting quarterly with financial data, quality-performance scores, and metrics on assigned populations. At the start of the first year, the agency also proposes making health data on its historically assigned patient population available to the ACO. ACOs will be permitted to request beneficiary-identifiable claims for beneficiaries who have received service from the ACO in that performance year. Patient centeredness: Government establishes itself as a patient advocate by mandating that patients have a seat on the ACO governing body, by linking payment to quality scores, and requiring a patient-experience survey. ACOs are required to report their quality scores, prices, and shared savings or losses publicly. The amount of shared savings permitted is capped to discourage withholding care. Quality performance: ACOs will be assessed on 65 quality measures in five domains: patient–caregiver experience, care coordination, patient safety, preventive health, and at-risk population health. Benchmarks will be distributed at the start of the first year, and failure to meet the performance standards in any domain could result in warnings or termination. In the first year, however, ACOs will be judged solely on the ability to report properly, and—provided sufficient cost targets are met—will receive 50% of shared savings based on complete and accurate reporting. This is to encourage investment in the IT infrastructure that will be required to monitor and coordinate care. No quality measures are radiology specific, but Physician Quality Reporting System measures will be synced with the shared-savings program for all group-practice ACO participants and suppliers. All eligible providers, including radiology groups, can qualify to earn 0.5% of the ACO eligible providers’ total estimated allowed charges (under the Medicare Part B Physician Fee Schedule) during the first performance period. Shared savings: The good news is that ACO providers will receive payment under the same fee-for-service program to which they are accustomed, in addition to shared savings, if earned. The rule proposes two levels of risk; the greater the risk, the greater the potential shared savings. In track 1, savings are reconciled annually, in the first two years, using a one-sided approach that involves no risk. In the third year, the program transitions to the two-sided method, in which the ACO agrees to share losses. The sharing rate for this method is 50% of savings beyond a threshold calculated as 2% of the benchmark. Track 1 shared savings are capped at 7.5% of an ACO’s benchmark. Track 2 features the two-sided risk-reward method, intended for ACOs with more experience in managing risk. They will be eligible for higher sharing rate of 60% of savings beyond the threshold, capped at 10% of the ACO’s benchmark. Losses for ACOs participating in track 2 will be capped at 5% in year 1, 7.5% in year 2, and 10% in year 3. CMS will retain 25% of an ACO’s shared savings to cover potential downstream losses. CMS anticipates that 75 to 150 ACOs will participate in the program; an estimated $800 million of shared savings will be distributed in the first three years. Imaging Services: Developing Partnerships in Accountable Care The largest health-care systems are not waiting for the fine print. Hardaway, reporting on the capabilities framework that the group-purchasing organization has developed to help member systems develop ACOs, also notes that imaging has a significant role in these initiatives. As health-care organizations move from fee-for-service payment to accountable care, they must develop competence in six core components. The first is a people-centered foundation that ensures that ACO population members not only receive the necessary care, but also believe that they can get the care that they need to improve their own health. The second is what Premier calls the health home and others call the patient-centered medical home, which functions as the coordinator of care, from a primary-care perspective, for all care delivered by the system. The third component is a high-value network, which contains all providers of services offered in addition to the health home (including radiology providers). A high-value network ensures the highest quality at the most efficient cost. The fourth element is accountable-care leadership, the legal organization that provides structure and governance. “One of the things that is an absolute necessity is that it is physician governed,” Hardaway says. The fifth part is population health data management. “Right now, hospitals and health systems only have good data about what is going on within the four walls of that system,” he says. “We must have a broader view of the information available.” The sixth component is payor partnership. “We learned, in the 1990s, that providers were terrible at taking insurance risk, and we still need to have some partner who is able to do this with us and for us,” he says. “The payors are the only ones that have all of the data that we need to manage the system, so we have to work together in order to bring all of these pieces together.” The ability to measure effectiveness is the foundation underpinning all six core components of accountable care, Hardaway says. “You can’t do this without measurements,” he advises. “You can’t show that you are making improvements without measurement.” Assessing Readiness Premier’s capabilities framework is based on the six core competencies, which are dissected into 150 operating activities. In its assessment of 60 health-care systems, Premier has found that the average score is 20%. “What we see is that no one is doing very well, and that is to be expected,” Hardaway says. “This is all brand new. In order to do well in the system we have now, you do not need most of the capabilities needed to do well in the system of the future.”
Figure. Four types of accountable-care participants; baseline participation is unlikely to be sustainable;adapted,withpermission,fromPremierInc.Hardaway outlines the implications of the high-value network for radiologists and other specialists. The network represents a wide range of clinical providers and facilities supporting primary-care practices. A question that Premier is frequently asked is whether a health-care organization should limit its network of providers as it develops an ACO. The answer is no—not immediately. “Currently, our reimbursement system is such that you don’t want to limit the number of folks in a network,” Hardaway advises. “What you want to say is, ‘Here are the criteria that you are going to be judged by over the next 18 to 24 months. Do you agree to those in the contract you sign with the accountable-care delivery system?’ It’s an all-comers strategy at the outset, but it’s a narrowing of that network as we go forward.” Another attribute of the high-value network will be the development of evidence-based care models. Hardaway emphasizes that providers have to be involved in developing these care models. The Imaging Partner Premier recommends that health-care organizations interested in developing an ACO look for five capabilities in their imaging partners: The first is resource-management expertise. “Imaging providers must have the capabilities to manage resources extremely well in this system of care,” Hardaway says. Second, imaging providers must be transparent in their quality and cost measurements. Third, the imaging provider must possess a robust imaging-informatics infrastructure. Fourth, radiology and consulting services must be immediately available. Fifth, radiology providers must have the ability to optimize the patient experience. This includes having the tools available to manage resources and to make cost/quality information available not only to partners in the delivery system, but to all patients. Hardaway concludes by describing the different models being created for partnerships between ACOs and their imaging providers. “The first one is baseline, and we do not think it is sustainable, going forward,” Hardaway says, “but there are some others that we see occurring (or that we think will occur).” Apart from the baseline model, Premier envisions three other models (see figure): the contractor, an imaging provider of higher-quality/acuteness credentialed services on a contract basis; the partner, a coordinator of imaging services among owned and network facilities; and the driver, an imaging provider that would accept capitated payments and quality requirements. “The difference between this system and what we saw in the 1990s is that there are quality and patient-satisfaction hurdles that have to be met or overcome prior to being able to share in savings—or, if you are taking capitation, before you would be eligible for any bonus,” he concludes. What Should You Be Doing? W. Kenneth Davis Jr, JD, entreats radiology practices to get involved in reforming health care, offering a summary of strategic steps that practices and imaging centers should consider adopting. “You’ve got to get involved; it’s not much more complicated than that,” he says. “What it is ultimately about is, first and foremost, cutting costs, in the long term, for Medicare.” Davis emphasizes urgency by pointing out that ACOs must enter into a three-year contract with Medicare, and that the ACO cannot add participants (operators, founders, or owners) during that three year agreement period. It can terminate participants, but cannot add them. Providers and suppliers can be added through contracting, but they will have no hand in governance. “What that suggests is that you need a seat at the table as soon as possible,” he says. Drawing on an ACR® white paper,¹ Davis outlines several fundamental concepts that practices should keep in mind when engaging with ACOs: Seek to be paid on a fee-for-service basis. While payment appears to be moving toward the adoption of risk, radiologists need to be careful that the ACO does not seek to offload risk to the practice (if other participants have no incentive to manage utilization, for example). If an ACO is capitated, one economic incentive against overutilization would be to pay the radiologists within the ACO on a fee-for-service basis. “Believe me, those other participants in the ACO will have a strong incentive to manage utilization,” Davis says. Take the reins of utilization management, which will be the crux of the ACO. The ACR suggests deploying order-entry systems and decision support. Be the party responsible for managing all imaging within the ACO, and try to be compensated for doing so. If a practice is responsible for utilization management in imaging, it also makes sense (from operational, clinical, and financial standpoints) for the practice to be involved in the day-to-day operational management of the technical component. Be prepared to operate under capitation. If that happens, it is critical that for appropriate utilization management to be built into the ACO; radiologists should be involved in that, as well as in the technical implementation of computerized radiology order entry and decision-support software. “You want to make sure that if you are being compensated on a capitated basis, there is appropriate risk management,” Davis says. Improve relations with your hospitals. Some radiologists are going to become employees, and if that happens, it is best to be employed by someone with whom you have a positive relationship. Davis is surprised that the ACR would acknowledge that some radiologists will become hospital employees. Address referring-physician conflict of interest. While the ACR suggests doing away with self-referral, Davis recommends focusing instead on appropriateness criteria. “Talking too much about self-referral can become self-defeating,” he says. “The economic conflict of interest is causing inappropriate levels of utilization, and you need to address that.” Consider consolidation; this is a very fluid, moving environment. An ACO is likely to have more than one radiology group involved, and there is the possibility, especially with large radiology groups, that you will be involved in more than one ACO—yet another reason for further consolidation of radiology groups. Adhere to rigorous quality standards and make that high quality known, Davis urges radiologists. “You cannot go wrong by selling quality. That is one of the absolutes,” he says. “Quite candidly, the risk radiologists have is the potential for being viewed as nothing more than a cost center in the ACO.” Recognize the different roles that imaging centers and hospital-based practices play in the ACO. Sell the ACO on the fact that outpatient imaging centers fundamentally are less expensive to operate than hospital imaging departments. Create proper alignments within the ACO. Align with competing imaging providers, referral sources, and self-referring physicians. The risk of not doing so, Davis says, is that in an environment that is designed to take risk, you will fail. Coordinate the use and flow of images and information within the care environment. Radiologists have led the way in moving images and information, and Davis says that they need to put that on steroids. Integrating with the electronic health record and adding decision support to order-entry systems will put radiology in a better position, in IT terms, to operate the ACO, he adds. In conclusion, Davis recommends that radiologists think like investment bankers. ”That is not to diminish the clinical aspects,” he says. “Think like an investment banker both clinically and economically: Where are the value propositions? How is patient care going to flow? How is the money going to flow? If you can follow that line through the entire process, you are going to identify problems you can fix up front, and you will identify upside propositions.” Cheryl Proval is editor of Radiology Business Journal and vice president, publishing, imagingBiz, Tustin, California.
Figure. Four types of accountable-care participants; baseline participation is unlikely to be sustainable;adapted,withpermission,fromPremierInc.