Newtown, CT – February 27, 2016
A new report from the Kaiser Family Foundation strikes an important cautionary tone – in it, researchers examined the spread and evidence of the most prominent alternative payment models deployed by Medicare, and conclude that for many the evidence is sparse, and not as encouraging as some headlines suggest. Two conclusions are particularly striking. The first is that the evaluation of each model is being done separately, failing to determine the potential interactions between them, even though we know that most of these models are complementary and are being deployed simultaneously in markets. The second is that despite the need to better understand the evidence, Medicare is rushing to spread some of these APMs. Of the four models in BPCI, only model 2 seems to be effective and that shouldn’t come as a surprise. In Medicare, the majority of the savings for acute care episodes comes from post acute care. As such, models 1 and 4, which exclude post-acute care, aren’t proving to be very beneficial. And model 3, which only focuses on post-acute care, isn’t panning out either. Again, not a big surprise. In fact, I vividly remember a meeting on bundled payment designs that was organized before the launch of BPCI by our friends at Brandeis. Representatives of the post-acute care lobby were at the table and strongly arguing for a model that would allow them to control the bundle. CMS bent to the lobby even though all of us were scratching our heads, trying to figure out how that made any sense. As it happens, it didn’t and doesn’t. And therein lies the real lesson.
What this means to you – we’ve said it before and we’ll say it again: It’s the Design Stupid!!! If you design a payment program that doesn’t make much sense, there’s a pretty good chance it won’t succeed. Kaiser’s report, unfortunately, fails to make this essential point and some will inevitably draw the wrong conclusions. And in its rush to meet artificial and self-imposed targets for the spread of APMs, CMS is also ignoring the design flaws in some of its programs and simply perpetuating them and spreading them farther. Interestingly, another report, just released by the DC-based Learning Action Network, makes several recommendations on how to design and implement bundled payments for joint replacements. Contrarily to other LAN reports, this one is light on dictates and instead focuses on bringing forth various approaches from field experiments. Importantly, it highlights all that CMS is doing wrong in its Comprehensive Joint Replacement program and, to an extent, the BPCI. We’re pretty sure no one in the current administration will pay much attention to these recommendations, preferring to continue down its own misguided path, but perhaps the next administration will care to listen to those who are working hard to make APMs fair, operational, and successful. And that’s exactly what’s happening in many places around the country, if not in Washington DC. Kaiser’s report concludes: “CMS’s commitment to “rapid cycle” improvements in the demonstrations means that CMS is able to refine the payment models as they are being implemented in light of early results and unforeseen circumstances.” The reality, however, is that CMS doesn’t seem committed to or capable of rapid cycle improvements, and has so far failed to refine its new payment models in light of early results, other field experiments and feedback from participants. So just remember, it’s not because Medicare’s models aren’t working that APMs don’t work. They do when they’re well designed.
Francois de Brantes