Saahithi KaruturiHealth Care PolicyFall 2017Final Paper Bundled Payment Plan SystemIntroduction: The bundled payment plan system is being evaluated under the Affordable Care Act in the United States for participating providers. CMS has released evaluation reports for the programs, however the data is a small set and has proven to be fairly inconclusive. Only one of the bundled payment programs has concluded, and while savings have been shown, this experiment has not been replicated to show that consistent savings are possible. Additionally, the remainder of the programs are still ongoing, and do not end until next year. Thus far, the evaluation reports have shown statistically significant savings in only one or two clinical groups. While well-intentioned in terms of sharing financial risk with providers, benchmarking prices, and saving costs, the resulting impact may not be as intended or as successful as the creators behind the programs would have wanted them to be. Policy Overview:The Center for Medicare and Medicaid Innovation (CMMI), created by the Affordable Care Act, was designed to evaluate various payment and care delivery systems to reduce Medicare, Medicaid, and Children’s Health Insurance Plan (CHIP) expenses (CMS 2016). The CMMI has come up with the Bundled Payments for Care Improvement Initiative (BPCI) in an effort to improve the quality of health care, lower costs, and encourage better coordination of care (CMS 2016). There are four different payment models that participants in the BPCI initiative can choose from. The table below (Table 1.), obtained from the CMS website, outlines the criteria for each of the four payment models. Model 1Model 2Model 3Model 4EpisodeAll DRGs; all acute patientsSelected DRGs; hospital plus post-acute periodSelected DRGs; post-acute period onlySelected DRGs; hospital plus readmissionsServices included in the bundleAll Part A services paid as part of the MS-DRG paymentAll non-hospice Part A and B services during the initial inpatient stay, post-acute period and readmissionsAll non-hospice Part A and B services during the post-acute period and readmissionsAll non-hospice Part A and B services (including the hospital and physician) during initial inpatient stay and readmissionsPaymentRetrospectiveRetrospectiveRetrospectiveProspective Table 1.Model 1 began in April 2013 and ended in December 2016. Models 2, 3, and 4 were all divided into two different phases. In Phase 1, CMS uses this time to educate participants in the program and exchange data as well. In Phase 2, the participants assume the actual financial risk of participating in the program. The first participants began Phase 1 as of January 2013 and all participants were to enter phase 2 by July 2015. All participants enter signed agreements with CMS and thus assume financial risk entailed. Throughout the course of the agreement, the participant will communicate with CMS to assess whether the new payment and delivery system are working as intended, without compromising the quality of care to the patients. There are additional voluntary and mandatory bundled payment programs that have either begun in 2016 or have been delayed by CMS. The only voluntary bundled payment program is the Oncology Care Model in which oncology practices can receive bonus payments if they lower costs and maintain quality across Medicare Part A, B, and D for chemotherapy and other related care. This program began in July 2016. The remainder of the programs are mandatory and are the Comprehensive Care for Joint Replacement, which began in April 2016, the Episode Payments Model, and the Cardiac Rehabilitation Incentive Payment Model. The Episode Payments Model and the Cardiac Rehabilitation Incentive Payment Models have provisions that were introduced in 2017 to cancel them, that have yet to be voted on and thus any potential start date has been pushed to January 2018. The Comprehensive Care Joint Replacement program pays participants a set amount for the episodes covered and thus the financial gain would be to bring costs under the target price. Evidence of Policy Impact:Cost Impact: The Kaiser Family Foundation conducted an analysis of the annual evaluation report completed by CMS and aggregated the impact of each model on cost for Medicare. Table 2 in the Appendix details the findings. In summary, Model 1 has concluded and has shown an aggregate savings of $10 million over the project’s duration (Kaiser Family Foundation 2017). Model 2 showed only statistically significant Medicare savings for 1 of the 22 clinical groups that fall into this model’s umbrella (Kaiser Family Foundation 2017). The sole clinical group was the major joint replacement of the lower extremity group (Kaiser Family Foundation 2017). Similar to Model 2, Model 3 only showed statistically significant Medicare savings for 2 of the 14 clinical groups that fall into this model’s umbrella (Kaiser Family Foundation 2017). The two clinical groups are the major joint replacement of a lower extremity and the congestive heart failure groups (Kaiser Family Foundation 2017). For Model 4, there was no statistically significant Medicare savings for any clinical groups (Kaiser Family Foundation 2017). Based on the limited amount of data provided by the four models to date and a lot of the statistical insignificance of the findings, the cost savings that were anticipating by the CMMI seem to be inconclusive at best. It will remain to be seen whether over the next year, while Models 2, 3, 4 continue and eventually wrap up, if the data will become more conclusive and more Medicare savings will be realized over more clinical groups than it currently appears. It is worth noting that programs like the Oncology Care Model and the Comprehensive Care Joint Replacement both were started prior to conclusive results about bundled payment plans actually generating the intended savings for Medicare. Quality Impact: The CMS evaluation report for 2017, that was conducted by the Lewin group, assesses the quality of each Model. Model 2 did not see any statistical improvements of significance across most of the clinical groups, except two (The Lewin Group 2017). The report does conclude quality was consistent across for Model 2 since the inception of the program (The Lewin Group 2017). Appendix Table 2.Bundled Payment for Care Improvement (BPCI) Model 1Bundled Payment for Care Improvement (BPCI) Model 2Bundled Payment for Care Improvement (BPCI) Model 3Bundled Payment for Care Improvement (BPCI) Model 4Medicare Spending (Savings/Cost)Over the first two years, aggregate Medicare savings totaled $10 million relative to a control group—attributable mostly to discounted Medicare payments to hospitals. Notably, Medicare spending on care following hospitalizations was higher for BPCI episodes than the comparison group. Most hospital participants withdrew.In the first two years, relative to comparison groups, the Model 2 BPCI participants had statistically significant Medicare savings per 90-day episode for 1 of 23 clinical groups (major joint replacement of lower extremities—mostly hip and knee). No statistical difference in other 22 clinical groups.In the first two years, relative to comparison groups, Model 3 BPCI participants had statistically significant Medicare savings per 90-day episode for 2 of 14 clinical groups (major joint replacement of lower extremities—mostly hip and knee—and congestive heart failure). No statistical difference in other 12 clinical groups.In the first two years, there was no statistical difference in per-episode spending between Model 4 BPCI participants and comparison groups. Only two clinical groups had enough sample size for analysis.Aggregate Change in Medicare Spending vs. Comparison GroupAcross first two years, Medicare spending for BPCI group was –$10.5 million lower across all Medicare admissions, relative to comparison group; per-episode spending was not statistically different.Not reported.Not reported.Not reported.Major Joint Replacement of the Lower Extremity (e.g., hip/knee)Not reported by clinical category.Medicare spending for BPCI group was –$1,273 lower per 90-day episode, relative to a comparison group.1For episodes in which SNFs were the episode initiators, Medicare spending for BPCI group was –$2,568 lower per 90-day episode, relative to a comparison group.No statistically significant difference in Medicare spending between BPCI group and a comparison group in 90-day episodes.Congestive Heart FailureNot reported by clinical category.No statistical difference in Medicare spending between BPCI group and a comparison group in 90-day episodes.1For episodes in which home health agencies were the episode initiators, Medicare spending for BPCI group was –$970 lower per 90-day episode.No statistical difference in Medicare spending between BPCI group and a comparison group in 90-day episodes.All Other Clinical EpisodesNot reported by clinical category.Excluding hip/knee replacement group (separate row), no statistical difference in Medicare spending between BPCI group and a comparison group in 90-day episodes (or sample sizes too small to assess). Of the groups with enough sample size, 11 clinical episodes had non-significant lower Medicare spending; 11 had non-significant higher spending.1Excluding hip/knee replacement and congestive heart failure groups (separate rows), no statistical difference in Medicare spending between BPCI group and a comparison group in 90-day episodes (or sample size too small to assess). Of the groups with enough sample size, 9 of 10 SNF-initiator episodes had non-statistically significant lower SNF spending.No statistical difference in Medicare spending between BPCI group and a comparison group in 90-day episodes (or sample size too small to assess).Bibliography:Blumenthal, David , and David Squires. “The Promise and Pitfalls of Bundled Payments.” The Commonwealth Fund, The Commonwealth Fund, 7 Sept. 2016, www.commonwealthfund.org/publications/blog/2016/sep/bundled-payments.(Blumenthal & Squires 2016)CMS. “Bundled Payments for Care Improvement Initiative (BPCI) Fact Sheet.” Centers for Medicare & Medicaid Services, Centers for Medicare & Medicaid Services, 18 Apr. 2016, www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2015-Fact-sheets-items/2015-08-13-2.html.(CMS 2016)Daly, Rich. “Report Finds Mixed Savings Results for BPCI.” HFMA, HFMA, 19 Sept. 2017, www.hfma.org/Content.aspx?id=50078.(Daly 2017)Gronniger, Tim, et al. “How Should The Trump Administration Handle Medicare’s New Bundled Payment Programs?” Health Affairs, Health Affairs, 10 Apr. 2017, www.healthaffairs.org/do/10.1377/hblog20170410.059551/full/#one.(Gronniger, Fiedler, Patel, Adler, & Ginsburg 2017)Gruessner, Vera. “How CMS Alternative Payment Programs Impact Healthcare.” HealthPayerIntelligence, HealthPayerIntelligence, 25 Jan. 2017, healthpayerintelligence.com/news/how-cms-alternative-payment-programs-impact-healthcare(Gruessner 2017)Kaiser Family Foundation. “Side-by-Side Comparison: Medicare Bundled Payment Models.” The Henry J. Kaiser Family Foundation, The Henry J. Kaiser Family Foundation, 30 Nov. 2017, www.kff.org/interactive/side-by-side-comparison-medicare-bundled-payment-models/.(Kaiser Family Foundation 2017)Kivlahan, Coleen, et al. “Taking Risk: Early Results From Teaching Hospitals’… : Academic Medicine.” Academic Medicine, Journal of the Association of American Medical Colleges, July 2016, journals.lww.com/academicmedicine/Fulltext/2016/07000/Taking_Risk___Early_Results_From_Teaching.18.aspx.(Kivlahan, et al. 2016)The Lewin Group. CMS Bundled Payments for Care Improvement Initiative Models 24: Year 3 Evaluation & Monitoring Annual Report. Center for Medicare and Medicaid Services, Oct. 2017, downloads.cms.gov/files/cmmi/bpci-models2-4yr3evalrpt.pdf.(The Lewin Group 2017)Weeks, William and William B. Weeks. “The Unintended Consequences of Bundled Payments.” Annals of Internal Medicine, vol. 158, no. 1, n.d., pp. 62-U136. EBSCOhost, login.proxy.libraries.rutgers.edu/login?url=https://search-ebscohost-com.proxy.libraries.rutgers.edu/login.aspx?direct=true&db=edswsc&AN=000313119900010&site=eds-live.