Final Evaluation of CMS Bundled Payment Program

By October 23, 2018Commentary

CMS has engaged in several bundled payment initiatives over the last few years.  As is usually the case, these new reimbursement methods are required to have periodic outside evaluations.  The final, five-year, evaluation of the bundled payment program has been released by the Lewin Group.   (Lewin Report)   The initiative had four models covering 48 clinical episodes that providers could choose to participate in.  This report covered the last three models; of those 2 and 3 had retrospective true-ups against a target, with the conveners of the episode sharing in some upside and downside performance and 4 was a true prospective payment.  The average episode participant selected 8 clinical episodes and 90% of the episodes were in model 2, although there were more participants in model 3.  Joint replacement was by far the most common episode.  A number of providers who began participating in the bundled payments eventually withdrew.  Very few people participated in model 4.  Medicare spending declined for most of the clinical episodes and that decline was statistically significant for about half of the episodes, but Medicare did not save money after considering the reconciliation payments to providers; in fact, Medicare had a net increase in spending of about $300 million.  Less use of post-acute care was the primary reason for the reduction in spending.  Quality of care as measured by mortality, readmissions and ER use was roughly equivalent between patients treated under bundled payments and those treated outside the bundles.  There were not significant health status differences between beneficiaries in and out of the bundled payments, suggesting that there was not much, if any, steering of patients to a bundled reimbursement, or to traditional Medicare payments.  Providers with more resources and higher baseline spending tended to be more likely to participate in a bundled payment.  There was slightly less satisfaction with care reported by patients cared for under the bundled payment.  Overall, bundled payments don’t look like a bad idea but at least as structured for the initiative, didn’t save money.  If quality isn’t better and they don’t save money, you have to ask why you would use bundled payments.  But the design might be changed and bundles might be made mandatory, as they have for some joint replacements, and there could be significant savings.

Leave a comment