Monday, 16 March 2015 15:09

Value-Based Purchasing: Should We Slow Down?

Written by


As we inaugurate this newsletter, the volume-to-value movement is picking up what seems to be unstoppable momentum. 

  • On Jan. 1, the Centers for Medicare & Medicaid Services (CMS) Hospital Value-Based Purchasing (VBP) program added its fourth domain, efficiency, and its physician value-based payment modifier went live for some practices with 100 or more “EPs” (physicians, physician assistants, nurse practitioners, and other advance practice providers).
  • On Jan. 25, the U.S. Department of Health and Human Services (HHS) announced an initiative to increase the percentage of Medicare payments that are linked to quality and value, citing the intent to move to 85 percent by 2016 and to 90 percent by 2018.
  • Two days after the CMS announcement, a private-sector “healthcare transformation task force” announced that its provider and payor members – six of the nation’s top 15 health systems and four of the top 25 health insurers – have committed to put 75 percent of their business into value-based arrangements by 2020.  
  • New accountable care organizations (ACOs) are forming almost faster than they can be counted. According to the aforementioned task force announcement, there are currently 721 ACOs: 404 that accept only public-sector contracts, 214 that accept only private-sector contracts, 103 that have both public and private contracts, and 21 that have no external contracts.

But a note of caution: in this age of “evidence-based medicine,” one would assume that our commitment to move VBP forward quickly is the result of solid, consistent achievement in VBP pilots, demonstrations, and operational programs over the past decade or so. Actually, though, the evidence about these projects and programs is surprisingly equivocal.  

  • The most ambitious assessment of VBP’s performance to date was conducted by RAND Corporation last year. RAND reviewed, in depth, the results of 103 studies of pay-for-performance (P4P) programs, 45 studies of ACOs, and 58 studies of bundled-payment arrangements. RAND’s review found that for many of these programs, there was no evidence of significant improvement in costs, quality, or satisfaction, and that “where observed, improvements were typically modest.”
  • “Disappointingly mixed” is the verdict of Indiana University School of Medicine professor Aaron E. Carroll, whose literature-review essay was published in the New York Times in July 2014. He summarized those VBP programs that haven’t worked as follows: “sometimes it’s because providers don’t change the way they practice medicine; sometimes it’s because even when they do, outcomes don’t really improve.” 
  • First-year performance data of the Medicare Shared Savings Program (MSSP) for ACOs, published by CMS in November 2014, also revealed mixed results and modest overall impact. About 54 percent (118/220) of the ACOs participating saved money for Medicare, but only about half of those saved enough to share in the savings themselves. The other 102 ACOs lost money. Most of the gains and losses were small –  between 0 and 2.5 percent of revenue. 
  • In CMS’s pioneer ACO program, only 19 of the original 32 participants remain – all of whose past performance and infrastructure were strong enough to convince themselves and CMS that they could take on downside risk as well as rewards for gains. 

Of course, a lack of compelling evidence that success is certain is not about to stop VBP programs from moving forward. But as data and experience accumulate, we are beginning to find that some approaches to VBP may have more positive impact than others. To increase the odds that VBP will succeed, here are some fundamental questions that policymakers, purchasers, payors, providers, and patients should be asking about VBP programs. 

  • Are the incentives meaningful enough? Provider perceptions of fairness may be more important than actual dollar amounts. One study of Medicaid P4P programs suggested that behavior and outcomes improved only in those programs for which the providers thought that the incentive payments were worth the effort to obtain. Continuity of programs and incentives is also important. One of the few completed longitudinal studies of VBP showed more improvement to incented quality metrics at 260 hospitals in a one-year large demonstration project than in 780 hospitals not in the program – but five years later, the differences between the groups had disappeared. 

In designing incentives, it’s also important to recognize that in the United States, most VBP programs as currently designed, both public and private, are “zero sum games.” That is, all of the money awarded to winners in the program have to come from penalties assessed against the losers. So, large increases to the winners’ pool could seriously underfund the losers – the very institutions that may need to make investments to improve quality.

Prev Next »

Last modified on Tuesday, 24 March 2015 16:42

Philip Nathanson is the President of Nathanson Consulting LLC.  Phil has held quality leadership positions at CMS, Aetna, and NCQA. His consulting clients include hospital systems, HIM firms and biotech companies. His articles on quality and healthcare management have appeared in Becker’s Hospital Review, H&HN Online, Topics in Healthcare Financing and other journals.