SNF integration might discourage BPCI-A participation

Hospital-skilled nursing facility (SNF) integration might decrease incentives to participate in Medicare's Bundled Payments for Care Improvement Advanced (BPCI-A) program. However, the association is context-dependent, according to a study published in JAMA Network Open.

The cross-sectional study analyzed hospital participation in Medicare's Bundled Payments for Care Improvement Advanced (BPCI-A) program beginning with its launch in 2018.

Researchers matched each SNF-integrated hospital with two nonintegrated hospitals for four episode-specific analyses: hip and femur (HFP), major joint replacement of the lower extremity (MJRLE), sepsis and stroke.

Hospital-SNF integration was associated with a statistically significant 4.7-percentage-point reduction in BPCI-A participation in the MJRLE episode.

However, the study found that both hospital-SNF integration and BPCI-A participation were relatively uncommon in the first place, indicating that integration likely plays a modest role in the low uptake of BPCI-A.

There was no association between hospital-SNF integration and BPCI-A participation across the three other episodes.

The researchers hypothesized that the MJRLE finding might be "an artifact of multiple hypothesis testing."

However, another explanation could be increased variation in discharge to SNF for MJRLE compared to the other three episodes. Greater variation in discharge to SNFs suggests increased clinical discretion in the use of institutional post-acute care (PAC).

"Perhaps integrated hospitals perceived a greater risk of lost institutional PAC in this episode compared with other episodes," the researchers proposed.

Previous research from The Lewin Group underscores this possibility. The study found that hospitals participating in the MJRLE episode significantly reduced initial discharge to institutional PAC by 5% between October 2018 and December 2019. However, the researchers did not find significant decreases for most other hospital-initiated episodes.

"This suggests that integrated facilities might reasonably foresee greater risk of forgoing revenues in MJRLE compared with other episodes," the JAMA Network Open authors suggested.

They noted that additional factors might explain the lack of a more uniform association between hospital-SNF integration and bundled payment participation.

For example, hospitals might be making decisions that are perceived to be best for the hospital but not necessarily for the broader organization. However, this strategy would conflict with other research that suggests whole-of-enterprise decision-making.

For instance, SNF-integrated hospitals are more likely to refer inpatients with high expected profitability to their SNFs versus nonintegrated SNFs.

Other research has shown that greater hospital-SNF integration was linked to more intensive SNF care and higher Medicare payments, aligning with enterprise-wide profit maximization strategies.

Another potential explanation for the unclear relationship between hospital-SNF integration and BPCI-A participation is that integration might influence the intensity of efforts to reduce SNF utilization rather than the initial choice to participate. The study did not examine that possibility.

Hannah Nelson has been covering news related to health information technology and health data interoperability since 2020.