Early success—similar to results of a previous voluntary bundled payment program—surprised reviewers and could spur more mandatory models, an adviser says.


Sept. 14—Hospitals in Medicare’s first mandatory bundled payment model succeeded in cutting Medicare gross spending in their first year, according to a new analysis. The results come as Trump administration executives say they plan to introduce more mandatory models.

The Comprehensive Care for Joint Replacement (CJR) model required most hospitals in 67 geographic areas to accept a bundled payment for all services involved in joint replacement and 60 days of post-op care.

Although hospitals and advocacy groups were nervous about requiring facilities without experience in bundled payment to participate, participating hospitals’ average Medicare payments for the episodes decreased 3.3 percent more than payments to hospitals in a control group. That reduction occurred over just nine months, since CJR launched April 1, 2016, and the first year ended with CY16.

“Our results indicate that CJR participant hospitals successfully responded to the financial incentives of the CJR model during the first performance year,” the Lewin Group wrote in an assessment conducted for the Centers for Medicare & Medicaid Services (CMS).

In fact, the positive early results surprised the Lewin assessors.

“Possibly the most notable outcome during the first CJR model performance year was that statistically significant changes in utilization and payments occurred so quickly,” they wrote. “With approximately nine months of implementation, the CJR model resulted in outcomes that are consistent with what has been achieved in other bundled payment initiatives.”

Medicare had found that CJR’s precursor model, the voluntary Bundled Payments for Care Improvement (BPCI) initiative, was associated with a 3.8 percent decrease in per-episode spending and had stable-to-improved quality.

“It was reaffirming with what we were seeing with our clients,” said Darcie Hurteau, a director for DataGen, which advises hospitals on CJR.

One caveat is that the Lewin estimates do not account for bonus payments Medicare made to CJR participant hospitals or for penalty payments received from the hospitals, and do not represent total savings to the Medicare program.

Average Medicare payments for CJR episodes decreased by $1,775, from $27,314 to $25,539 (6.5 percent). In the control group, average joint-replacement payments decreased by $865, from $26,980 to $26,115 (3.2 percent).

In other results, average payments in CJR decreased by:

  • 7.7 percent for elective joint procedures
  • 2 percent for fracture cases
  • 6.8 percent in high-payment regions
  • 5.8 percent in low-payment regions

How They Did It

The savings primarily stemmed from reduced use of institutional post-acute care (PAC) providers, according to the Lewin Group. However, quality of care was maintained, as indicated by claims-based quality measures.

“The incentive is to reduce the utilization where it makes sense, not necessarily across the board,” Hurteau said in an interview.

Other ways to generate savings, based on interviews with some participating hospitals as cited in the report, included planning earlier, educating patients about discharge to less-intensive PAC settings, developing preferred provider networks, and coordinating with PAC providers.

The report noted that an earlier start to discharge planning “better set patients’ expectations about discharge destinations.”

For elective episodes, savings were driven by decreased use of skilled nursing facilities and inpatient rehabilitation facilities (IRFs), and lower Part B payments. Meanwhile, fracture savings stemmed from relative decreases in IRF use and readmission payments, the authors wrote.

One hospital official told the authors that the organization used knowledge and skills gained from BPCI, including by creating an outpatient rehabilitation program and defining best practices for outpatient care.

As measures of quality of care, no statistically significant changes occurred in readmission rates, emergency department visits, or mortality for elective or fracture CJR episodes relative to patients in the control group.

Some observers were concerned that the mandatory model would encourage hospitals to select elective joint replacement patients whom they projected to have lower costs or better outcomes than average.

However, “there were no indications that the CJR patient population was healthier in the intervention period than in the baseline, relative to the control group population,” the Lewin report stated.

Looking Forward

The Trump administration amended the program for its third year, reducing the number of geographic areas where most hospitals must participate from 67 to just the 34 highest payment areas. Hospital participation in the 33 other regions became voluntary for the final three years.

“Many of the hospitals that subsequently left CJR did so because the rules also had been changed so hospitals’ savings target was no longer their own past performance but a regional benchmark,” Hurteau said.

Many remaining participants are hospitals in the higher-cost regions, “so there are opportunities for them to improve efficiencies,” Hurteau said.

The leaders of CMS and the U.S. Department of Health and Human Services (HHS) have changed since the CJR program was curtailed in 2017, and the new leaders have repeatedly emphasized that mandatory payment models will have a role in the future.

“In some cases, as I’ve said before, that is going to mean mandatory models from CMMI [the Center for Medicare & Medicaid Innovation] and other mandatory reforms,” Alex Azar, secretary of HHS, said at a Sept. 6 health policy event. “Requiring participation can be necessary to determine whether a model really works, but it may also be necessary to meet what we see as an urgent need for reform.”

Although Hurteau is not supportive of mandatory models, results like the early CJR savings could spur more of them, she said. A key to their success is allowing for an initial period without payment penalties, giving providers time to ramp up their efforts. CJR had such an on-ramp, but the upcoming voluntary BPCI Advanced (BPCIA) does not.

Based on results from Hurteau’s client hospitals, she expects Medicare savings from CJR to increase after the year that was analyzed in the Lewin report.

The Lewin authors said hospital representatives they interviewed that had previous bundled payment experience “felt they were prepared to identify areas for improvement and implement care redesign changes to succeed under the CJR model.” That could indicate which hospitals will opt to participate in BPCIA.

CMS is finishing enrollment in BPCIA, which is scheduled to start in 2019.

Hurteau has heard divergent interest in BPCIA from hospitals, with some unimpressed by its “completely different target structure,” she said. Despite their reservations, some hospitals may try BPCIA because CMS is allowing them to drop out after March if early savings don’t materialize.


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

Publication Date: Monday, September 17, 2018