The first two years of a mandatory payment bundle for joint replacements saved Medicare more than 2% annually, but much of it came through reduced payments to post-acute providers.
Reviewing Medicare Part A and Part B fee-for-service claims for 815 hospitals assigned to the bundle format, researchers found 90-day spending dropped by about $582 per episode. Participating hospitals created savings by cutting post-acute spending by an average of 5.5%, with payments to skilled nursing operators down 4.5% from the previous year’s baseline.
The study of the lower extremity Comprehensive Care for Joint Replacement, or CJR, model was published online first in JAMA Internal Medicine on Monday. It was conducted by members of Harvard’s business and medical schools and Boston-based software analytics firm Avant-garde Health.
The cost reduction appeared to be the only significant change brought on by the bundle’s 2016 adoption.
The researchers evaluated levels of spending, quality of care, volume of episodes and patient characteristics and found no detectable changes in hospital length of stay, readmissions, complications, 30- or 90-day mortality, or volume.
The findings, which showed a 22.9% drop in spending on inpatient rehab facilities, echoed earlier evidence that hospitals had decreased use of institutional post-acute care in the program’s first nine months.
If a joint patient went to a nursing home, the researchers found providers shortened stays by 1.5 days.
Under CJR, hospitals that beat a cost target established by the Centers for Medicare & Medicaid Services were given an incentive payment the first year. In the program’s second, they also assumed downside risk for performing below the target.
Bundled payments “may serve as a useful model for policy efforts to change clinicians’ and facilities’ behavior without harming quality,” the researchers reported.
Notably, sending patients home to recover more often wasn’t associated with longer length of hospital stay, higher 90-day readmission rates or higher mortality risk.