Medicare recipients who received care under the Comprehensive Care for Joint Replacement (CJR) bundle payment model were less likely to be discharged to a postacute care facility, according to new findings published online September 4 in JAMA.
Found in the same issue, a second study has quelled some of the fears related to the Bundled Payments for Care Improvement (BPCI) initiative for lower extremity joint replacement (LEJR) surgery, such as that it might lead hospitals to increase the overall volume of episodes paid for by Medicare or shift the case mix to lower-risk patients.
However, the findings of this study showed that the volume of procedures has remained much the same, and for the most part, the patient mix also has not changed.
In an accompanying editorial, Andrew M. Ryan, PhD, from the University of Michigan, Ann Arbor, called these findings “encouraging.”
“Given the increasing number of joint replacements at younger and younger ages, it is important that these procedures be a focus of cost and quality initiatives,” Ryan writes.
He adds that the results of the two studies “suggest that prospective payment to hospitals can be improved through common sense and straightforward bundled payment reforms. These programs should be continued and expanded, particularly for patients undergoing elective surgical procedures.”
Bundled payments are an alternative payment model, in that they reimburse providers for an entire episode of care in a single payment. With this payment model, providers and/or healthcare facilities are paid with a single payment for all services performed to treat a patient undergoing a specific episode of care. An “episode of care” is defined as the delivery of care for a specific condition or care that is delivered within a defined period.
The current Medicare model covers not only the hospitalization but also care after discharge for up to 90 days. Hospitals are assigned a target price for each clinical episode and can receive shared savings if spending falls under the target price. Conversely, they are also subject to penalties if spending exceeds that price.
There have been major challenges with bundled payments, such as dealing with costs that may be out of the provider’s control. In addition, there have been concerns that paying a fixed amount, regardless of the services needed, may lead hospitals to reduce necessary care or attempt to treat only healthier patients. Some fear that facilities may “cherry pick” those at lower risk on the basis of socioeconomic, racial, or clinical factors to improve their performance.
Two articles have now evaluated different aspects of the Medicare bundled payment programs.
Lower Rates of Postacute Care
In the first study, Amy Finkelstein, PhD, from the Massachusetts Institute of Technology and the National Bureau of Economic Research, both in Cambridge, Massachusetts, and colleagues examined the changes in discharges to institutional postacute care after hip or knee replacement surgery in 2016, during the first year that the Centers for Medicare & Medicaid Services (CMS) implemented CJR bundled payments.
The researchers drew the study cohort from a 5-year, mandatory participation, randomized trial conducted by CMS. The study randomly assigned eligible metropolitan statistical areas either to the CJR bundled payment model for LEJR episodes or to a control group. The current study is an interim analysis of these first-year data on LEJR episodes, from April 1, 2016, through December 31, 2016.
The analysis included a total of 67 CJR models and 121 controls, and the primary outcome was share of LEJR admissions who were discharged to institutional postacute care. Secondary outcomes included the number of days in institutional postacute care, discharges to other locations, and total Medicare spending.
The mean percentage of patients discharged to institutional postacute care was 2.9 percentage points lower (30.8% vs 33.7%; P = .005) in the CJR group; this difference was significant.
In the control group, the mean percentage of patients discharged to home without home healthcare was 2.6 percentage points higher (32.2% vs 34.8%; P = .14) in the CJR group, but the difference did not reach significance. Medicare spending for institutional postacute care was $307 lower ($3871 vs 3564; P = .04) in the CJR group, and this reached statistical significance.
The authors conclude that metropolitan statistical areas in the bundled care model had a significantly lower percentage of discharges to institutional postacute care, but there were no significant differences in total Medicare spending for each LEJR episode. “Further evaluation is needed as the program is more fully implemented,” they write.
Volume and Patient Mix Unchanged
In the second paper, Amol S. Navathe, MD, PhD, from Corporal Michael J. Crescenz VA Medical Center and the University of Pennsylvania, both in Philadelphia, and colleagues looked to see whether hospital BPCI participation for LEJR was associated with any changes in overall patient volume and/or in the type of cases.
Using Medicare claims data, the authors compared 131 markets (hospital referral regions) with at least one BPCI participant hospital (n = 322) and 175 markets with no participating hospitals (n = 1340). This extrapolated to a total cohort of 580,043 Medicare beneficiaries who received treatment before the BPCI initiative (January 2011-September 2013) and 462,161 who were treated afterward (October 2013-December 2015).
The authors note that participation in the BPCI initiative was not significantly associated with changes in the overall market-level volume. The mean quarterly market volume in nonparticipating markets increased 3.8%, going from 3.8 episodes per 1000 beneficiaries before BPCI to 3.9 episodes per 1000 beneficiaries after BPCI implementation. In the BPCI markets, the mean quarterly market volume similarly increased 4.4%, going from 3.6 episodes per 1000 beneficiaries before BPCI to 3.8 episodes per 1000 beneficiaries after its launch.
When looking at possible differences in case mix, BPCI participation was associated with differential changes in hospital-level case mix for only one factor (of 20 demographic, socioeconomic, clinical, and prior utilization factors); a small, statistically significant reduction in the likelihood of undergoing LEJR among beneficiaries with prior skilled nursing facility use (95% confidence interval, −0.96% to −0.10%; P = .01). There were no significant associations of BPCI participation with any of the other factors.
“These findings may provide reassurance regarding 2 potential unintended effects associated with bundled payments for [LEJR],” the authors conclude.
In his editorial, Ryan notes that “CMS is rightfully concerned that severity adjustment may become a means for hospitals to up-code severity and ‘game’ the program.”
“This tension between appropriately accounting for variation in hospital risk while minimizing the ability for hospitals to game performance measures is central in alternative payment models,” he writes, adding that one solution could be for CMS to incorporate more risk adjustment into these initiatives, along with a more aggressive approach to auditing hospital records.
Another question is how well the “successes” observed with bundled payment related to joint replacement can be extended to other surgical and medical diagnoses. “Because much of the savings in bundled payment programs have been concentrated among payments to institutional postacute care and clinical facilities, surgical conditions that do not have this spending profile may be more challenging to improve,” he says. “Episodes for medical conditions, especially those for exacerbations of chronic disease, may also be less amenable to improvement under bundled payment.”
Finkelstein’s study received research support from J-PAL North America and the National Institute on Aging. One coauthor reports that he is an investor in Dorsata Inc, a clinical pathway software startup, and a consultant to Sutter Health Inc. The remaining authors have disclosed no relevant financial relationships. Navathe’s study was supported by The Commonwealth Fund, and Navathe and several coauthors report relationships with industry. A complete list is available on the journal’s website. Ryan reports no outside funding and has disclosed no relevant financial relationships.