Post-acute care has provided critical savings for BPCI providers, who are still learning where specifically they can improve their results.


Sept. 19—Second-year results from the Bundled Payments for Care Improvement (BPCI) initiative showed savings for orthopedic surgery but not for cardiovascular surgery.

Medicare’s voluntary BPCI initiative offers bonus payments to participating hospitals when a given care episode is below a target price, and they can keep the bonus or share it with their partnering providers. However, the program will cut payments to participants when their episode payments exceed the target price.

“Early results are encouraging: orthopedic surgery bundles, in particular, have shown promising results on cost and quality in the first two years of the initiative,” Patrick Conway, MD, acting principal deputy administrator and chief medical officer for the Centers for Medicare & Medicaid Services (CMS), wrote in a blog post.

Early Results

The new report, prepared for CMS by the Lewin Group, covered BPCI episodes initiated between October 2013 and September 2014, which were generally upside-only. Downside risk was not added for all participants until October 2015.

BPCI’s joint replacement payment models attracted the most provider participation, including by 74 percent of providers in the popular Model 2 track. The Lewin study found standardized allowed Medicare payments for the hospitalization and 90-days post-discharge declined an estimated $864 more per episode for orthopedic surgery episodes initiated at BPCI hospitals—which tended to be large and urban—relative to a comparison group of hospitals.

Conway’s positive assessment of the findings was disputed by Elliott Fisher, MD, director of the Dartmouth Institute for Health Policy. Fisher noted several reasons to view the joint replacement results with caution, including his finding that hospitals in the comparison group had steeper declines in total spending if the primary endpoint was mean total joint replacement payments per hospital.

“It is thus too soon to tell whether the portion of the BPCI initiative focused on lower-extremity joint replacement is actually improving care and achieving savings for the Medicare program,” Fisher wrote in a JAMA editorial on the Lewin report’s joint replacement findings, which were published as a study in the same journal.

The Lewin report credited the larger reduction in Medicare payments for joint replacement patients in BPCI hospitals primarily to reduced payments for institutional post-acute care (PAC). Specific PAC use reductions included a decrease of 1.3 days in average skilled nursing facility (SNF) length of stay, compared to baseline. Additionally, institutional PAC use among beneficiaries who received PAC decreased from 64 percent at baseline to 57 percent during BPCI.

CMS cited reducing unnecessary PAC as key to early BPCI success.

A spring Avalere analysis had found 39 percent of total spending on hip and knee replacement episodes was tied to post-discharge care and hospital readmissions. Additionally, a Kaiser Family Foundation report cited the importance of reducing SNF utilization, reducing readmissions, and shifting utilization to clinically appropriate care settings.

Quality of care “appeared comparable” between orthopedic surgery patients in BPCI hospitals and those in a comparison group of hospitals, including similar readmission rates and emergency department visits.

Cardiovascular Results

During the first year of BPCI, the 30 hospitals in at least one cardiovascular surgery episode had “no statistically significant difference” in spending relative to a comparison group of hospitals, according to the report.

“Similar to CMS’s findings, we’ve seen those results go up and down depending on markets, and we’re still digging into what are the root causes,” Carter Paine, COO of naviHealth, one of the largest conveners for hospitals under BPCI, said in an interview.

Similarly, BPCI participant University of Rochester Medical Center (URMC) was unable to bend the cost curve or reduce readmissions among its congestive heart failure population, James Garnham, director of contract and payment innovation, noted at an April briefing in Washington, D.C.

Despite the poorer financial outcomes from the cardiovascular group compared with the joint replacement group, Paine emphasized the importance for hospitals of continuing to try to obtain savings in multiple areas due to the ongoing proliferation of mandatory CMS bundles.

Institutional PAC use also declined more among cardiovascular surgery patients treated at BPCI hospitals than in a comparison group, the report noted. The share discharged to an institutional PAC decreased from 55.1 percent to 44.2 percent, with smaller increases occurring among the comparison group.

“Monitoring the appropriateness of patients going into those facilities and then making sure they have an evidence-driven, personalized plan for their recovery and that they return to the community is key here,” Paine said about ways his provider partners obtained PAC savings.

Although the Lewin report found a statistically significant increase in BPCI cardiovascular surgery patient mortality during the study period, the authors noted that an examination of more recent data found no statistically significant change in mortality.

Program Implications

BPCI faced a steep drop-off between providers that initially expressed interest and those that actually chose to participate, but the program has since grown. During the period studied in the report, BPCI included 94 awardees and 227 episode initiators (any treating Medicare provider). Those numbers had increased to 305 awardees and 1,141 episode initiators as of July 1, 2016, according to a CMS spokesperson.

That increased participation will be important.

“Future evaluation reports will have greater ability to detect changes in payment and quality due to larger sample sizes and the recent growth in participation of the initiative, which generally is not reflected in this report,” Conway wrote.

The early drop-off in BPCI participation was cited by some health policy analysts as a possible reason for CMS to subsequently move aggressively into mandatory payment models, including the launch in the spring of the Comprehensive Care for Joint Replacement model. That model is based on BPCI and is mandatory for about 800 hospitals in 67 markets.

Some used the latest BPCI results to highlight the promise of voluntary models.

“The results indicate that, with only two years of experience, the bundled payment model holds promise to deliver higher-quality, more cost-effective care,” Mark Hiller, a vice president at Premier Performance Partners, said in a written statement. “We believe that CMS should continue to advance voluntary, rather than mandatory, options for further bundled payment episodes.”

In April, CMS announced it was extending BPCI, which was supposed to end this month, by two more years. The extension was pitched as a chance to obtain “a more robust and rigorous evaluation of the initiative.”


Rich Daly is a senior writer/editor in HFMA’s Washington, D.C., office. Follow Rich on Twitter: @rdalyhealthcare

Publication Date: Monday, September 19, 2016