A Bipartisan Approach to Better Care and Smarter Spending for Elderly Adults with Advanced Chronic Illness.
Last December, the Chronic Care Working Group (CCWG), a bipartisan collaboration of Senate Finance Committee members chaired by Senators Isakson (R-GA) and Warner (D-VA), released its policy options document.1 The CCWG had a formidable charge—to “analyze current law, discuss alternative policy options, and develop bipartisan legislative solutions” that increased care coordination across settings, streamlined Medicare payment while aligning incentives, facilitated high-quality care, produced better outcomes, and reduced growth in Medicare spending. The requirement that proposals be budget neutral or cost saving narrowed the choices. After reviewing more than 530 comments, the CCWG met with 80 stakeholder groups, ultimately recommending 24 options, including two that focused on improving the infrastructure of chronic care with better quality measures and more-accurate measures of cost for payment and computing shared savings. Of the proposals, 16 were for enhanced chronic care, with 11 focused on capitated payments (e.g., Medicare Advantage (MA)) or shared savings (accountable care organizations (ACOs), Independence at Home (IAH)) (Table 1). Despite government efforts to promote systems of care that feature cost containment strategies such as capitation (MA) or shared savings (ACOs), more than half of the 49 million Medicare beneficiaries continue to receive care from fee-for-service (Medicare A and B) providers, without spending constraints. Proposals to enhance care within traditional fee-for-service (FFS) Medicare were limited to new billing codes for high-intensity chronic care management, counseling regarding prognosis of serious illnesses, and limiting constraints on telemedicine for specific patient groups. There is wide recognition that Medicare spending is concentrated in beneficiaries with advanced chronic illness.2 Nearly half of FFS spending is for 5% of beneficiaries, the 1.7 million with annual spending exceeding $54,000 per person, compared with $9,900 for the average Medicare beneficiary. Recognizing where spending is concentrated is necessary but is only the first step toward creating targeted policy to advance high-quality care and reduce costs. Of the 34 demonstrations that CMS formally evaluated over the past 15 years,3 only two were substantially successful in reducing costs and increasing quality.4, 5 The main lessons from these demonstrations are not surprising. There is value in targeting high-cost beneficiaries and developing trusting human relationships where engaged patients and providers work toward common goals; there is significantly less cost savings and quality improvement in efforts where providers work by themselves, focus work on telephonic management, are disconnected from primary care, or focus merely on technology-assisted data acquisition and algorithm-driven care plans.4 Two notably successful FFS programs—Health Quality Partners and Massachusetts General Hospital (MGH)—relied on administrative fees to pay for provider time and additional personnel needed for managing complex patients. Both programs relied on engagement between care managers, patients, and providers; the MGH model embedded care managers in primary care practices. In contrast, patient-centered medical homes, despite a similar conceptual design, have had an uneven record in controlling costs, in part because of a less-targeted focus.6, 7 Several of CCWG's proposals favor expanding MA plan options and converting special needs plans (SNPs) to permanent status. These managed care organizations receive risk-adjusted capitation for each beneficiary, providing a strong incentive to actively manage their chronically ill members. A few provider–insurer organizations have proven effective at managing some of Medicare's frailest beneficiaries under capitation. Programs of All-inclusive Care for the Elderly (PACE), which typically operates small programs for 100 to 1,200 enrollees, now covering 34,000 frail elderly adults eligible for nursing facility care, and the Commonwealth Care Alliance dually eligible managed care organization (MCO) in Massachusetts are strong examples of effective care under a capitation model.8, 9 Although these organizations receive a risk-adjusted, capitated payment, the important feature is that providers and patients develop and manage the care plan in small groups. Another approach that has shown promise for these populations are Medicare shared savings programs (MSSPs), such as ACOs, which are also strongly incentivized, with rewards contingent on improving upon a risk-adjusted baseline cost target for assigned beneficiaries. CMS' evaluation of ACOs suggested that they saved $424 per beneficiary in their first year and $257 per beneficiary over 2 years.10 Several CCWG recommendations seek to make ACOs and other shared savings plans more accountable (e.g., by allowing prospective attribution by patients) and more flexible (e.g., broadening options to include nonmedical support services and relieving restrictions, such as the homebound status requirement for home health care.) The CCWG logic is clear—with a budget constraint, fewer of the rules to control overuse observed under FFS payment are needed, rules which may have perverse effects on outcomes and costs. Topping the CCWG's list of options is Independence at Home (IAH), a 3-year demonstration (2012–15) that Congress unanimously extended for 2 years in June 2015.11 The favorable initial published results are consistent with those of the successful Department of Veterans Affairs national Home Based Primary Care (HBPC) program12 and other published HBPC models.13-15 IAH rests on three precepts: focus on the most frail, immobile, and costly beneficiaries; use mobile interdisciplinary provider teams led by physicians and nurse practitioners responsible for their care and accountable across the health spectrum; and align incentives. IAH's novel approach to “shared savings with discipline” uses eligibility criteria identifiable in CMS administrative data,1 so that the savings baseline can be accurately calibrated. It allocates the first 5% of savings to CMS, after which there is an opportunity for shared savings, and IAH disqualifies practices that fail to achieve savings or to meet three of six quality metrics after 2 years. CMS found that IAH practices saved more than $25 million during the first performance year—an average of $3,070 per beneficiary—while delivering high-quality care. With 8,400 frail elderly adults enrolled, Year 1 savings were 7.7%, with nine of 17 (53%) practices achieving the 5% savings threshold (range $2,676–13,068/patient-year). For those nine practices, the average savings rate was 17%, with the top-performing practice reducing costs by 32%. Successful practices were broadly representative, coming from medium and large academic centers, and private for-profit and nonprofit corporate and community practices. Although IAH programs managed fewer than 1% of the number of individuals in the original 32 ACOs (Pioneer ACOs), IAH generated nearly 9% of the combined first year CMS savings of Pioneer ACOs and IAH.10 Unlike the two-sided risk in the Pioneer ACO model, in IAH, discipline comes from being disqualified and thus unable to share in substantial achievable savings. Using the 2012 Medicare 5% beneficiary file, it was estimated that 2.2 million Medicare beneficiaries meet IAH criteria,1 6.2% of all FFS beneficiaries, with 40% of the 5% most-expensive beneficiaries in 2012. Of all Medicare FFS beneficiaries, these individuals represented 23% of the deaths, 43% of Medicare readmissions, 37% of new long-term institutional stays, and 28% of FFS spending, validating the targeting criteria. Over the next decade, if home-based primary care through IAH reached only 35% of this population, maintaining the first year demonstration experience, it is likely that it would generate savings exceeding $35 billion, with CMS retaining more than $17 billion,16 which is within 3% of the estimated savings from applying VA's HBPC experience to 35% of the IAH-eligible Medicare population. The congruence of effectiveness between VA's national program and the CMS demonstration gives support to the estimate. Two of the CCWG's recommendations are crucial for cost containment in FFS Medicare, applicable to ACOs and IAH: developing more-accurate risk adjustment models for the highest-cost beneficiaries and developing appropriate quality metrics, with a focus on patient experience. The current system of risk assessment for ACOs relies on hierarchical condition categories (HCCs) to develop cost benchmarks.17 Analyses of beneficiaries similar to those qualified for IAH18 have confirmed other reports19 that current HCC models, calibrated on the entire Medicare population, substantially underpredict costs for a selected population, with mortality four times as great, costs five times as high, and more than half of beneficiaries joining a practice being clinically unstable at enrollment. HCC predicted costs also have wide geographic variation whereas frail beneficiaries are more similar cost-wise between low- and high-cost counties. In PACE, a combination of PACE rates, the PACE HCC model, and PACE frailty adjusters have effectively addressed these phenomena.8, 14 Similar recalibration of the HCC model to account for mortality risk, frailty, and local variability among individuals meeting IAH qualifying criteria is needed while incorporating a time-dependent clinical instability factor.18 Quality metrics offer a balancing force for shared savings incentives. In contrast to the large suite of measures for ACOs, the IAH demonstration used six quality measures to moderate savings shares: soliciting patient preferences; improving care transitions through home visits within 48 hours; including medication reconciliation after a hospitalization or emergency department visit; and validated, care-sensitive usage metrics (30-day readmissions, ambulatory care–sensitive hospitalizations, and emergency department visits). The National Home-Based Primary and Palliative Care Network is field testing additional metrics, particularly ones focused on patient and caregiver experiences.20 Effective, efficient chronic illness management requires making decisions about goals and value. The potential for shared savings to provide the needed constraint in FFS Medicare should foster the transition from volume to value while keeping the provider–patient relationship at the center, giving people care that they want and need. The CCWG's lead option, converting IAH to a Medicare benefit, will advance that transition, achieving the CCWG's goals of coordinating care across sites, aligning incentives, and improving outcomes while lowering costs. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. Drs. Kinosian, Taler, and Boling are participants in the IAH Demonstration, as are the other 14 members of the IAH Learning Collaborative Writing Group. Drs. Kinosian, Taler, and Boling have provided testimony to CMS and the Senate Finance Committee's Chronic Care Working Group, as well as MedPAC, on the Independence at Home demonstration. Dr. Kinosian is a board member of the American Academy of Home Care Medicine; Drs. Taler and Boling are past board members. Author Contributions: Kinosian: design, analysis, manuscript preparation. Boling, Taler: design, manuscript preparation. Sponsor's Role: NA.
Read more
6