In case you hadn’t noticed the coincidence, Groundhog Day and the conclusion of the Centers for Medicare and Medicaid Innovation’s Innovation Challenge grant program are both coming very soon. While it might not be readily apparent, I think this is a sign.
For those of you who don’t favor light comedy, Groundhog Day is not only the day that Punxatawny Phil (or a similar groundhog) predicts the change of the seasons, but also the title of a Bill Murray/Andie MacDowell movie in which a misanthropic weather man is doomed to relive that very same day over and over again until he learns to get it (and life) right.
In health reform (like Groundhog Day), I am having a powerful sense of having lived this day before. And I want to be sure that this time we get it right.
I know that many people think that with the new Center for Medicare and Medicaid Innovation, its $10 billion war chest, and the entirety of the Affordable Care Act behind it, everything is different and we don’t need to look to the past for guidance. However, I don’t agree. The idea that we can improve the quality of care, improve population health, and lower per capita costs is not new, and it's not Don Berwick's. It has been tried over and over and over again with very limited success, particularly in changing the care of complex older adults in Medicare.
Before the Foundation decided to focus its efforts on Aging and Health, its prior program was Health Care Cost and Quality. This program, which actually furnished Dr. Berwick with his first grant in 1987 for the "National Demonstration Project," tried to build the infrastructure for the managed care/managed competition effort to reform health care in the early 1990s. The vertically integrated, capitated, managed care organizations of that era (and the members of the RWJF- and JAHF-sponsored National Chronic Care Consortium; see http://www.jhartfound.org/blog/?p=3514) shone as brightly as the Geisingers and Grand Junctions do today and then dimmed.
The new demonstrations that have gained so much attention--accountable care organizations and medical homes (in all their varieties)--have many predecessors, and the degree to which their lessons have been incorporated into current efforts is very unclear. A simple and recent case in point is the section 721 Chronic Care Improvement Pilots specified in the Medicare Modernization Act of 2003. The program as legislated by Congress sounded sensible—at-risk partial capitation payments to support disease management/case management services for high-risk/high-cost Medicare beneficiaries that would improve their health and lower the costs of their care.
Unfortunately, this program went beyond fiasco to a realm of disaster that usually leads to congressional investigations and prosecutions. As reported recently (2011) in a NEJM publication authored by RTI, CMS's outsourced evaluation contractors, the demonstration showed little benefit, didn't save money, and contributed virtually no lessons for future efforts. The hundreds of millions of dollars paid to the disease management contractors are lost. Several contractors went bankrupt, and the findings are (again) much, much more difficult to find than details of the supposedly secret Joint Strike Fighter.
What makes this much worse is that it was perfectly predictable. CCIP, which came to be known as Medicare Health Support, or MHS (because it sounded better; who wants to have a chronic disease?), used the same class of models that had been tried in the Medicare Coordinated Care Demonstration (MCCD) mandated by the Balanced Budget Act of 1997. The MCCD demonstration program started in 2001, and by the time MHS started in 2005 it was known by insiders to have at the best very mixed results. And yet nothing in MHS built on the emerging knowledge of MCCD.
My frustration with this troubling history bubbled up again when the Congressional Budget Office recently published a brief and not very helpful report on the history of CMS demonstrations. This report sheds little light on the internal management structure and limitations within CMS and Congress that have impeded learning across prior demonstrations. It concludes as many prior reports have (with the little empirical evidence we have) that new services and models must be closely coordinated with physician offices—the same conclusion as MCCD.
As far as I can tell, we go into the Innovation Challenge with little guidance from our prior 15-year history of demonstrations. If the history of demonstrations has taught us anything, it is that most innovations don't work. Plus, consider the infrastructure of the Innovation Challenge: its sudden announcement, to its short application timeline, to its last minute calls for ad-hoc reviewers. None of that gives me much confidence that the Innovation Challenge will be effectively testing the right models and asking the right questions to successfully redesign health care delivery. How can we take the next step and actually stop reinventing the flat tire if we’re not learning anything from previous efforts? We need to learn the lessons of those 15 years. We need systematic and careful mining of the data and literature; we need to refine a body of knowledge about what works and why so as to sharpen our hypotheses. Otherwise the 200 or so projects funded in the Innovation Challenge will be just so many random shots in the dark.
Just like Bill Murray in Groundhog Day, we have lived this day before. Can someone please tell me what we need to do to get to the next one?