HCI3 Update From the Field: Avoiding the Flawed Numbers

Submitted by francois.debrantes@hci3.org on Monday, January 6, 2014 - 09:11

It's amazing how a number, however arrived at, if proposed by an "authority", has lasting power even when it's flawed – A few years ago, researchers at Dartmouth came up with what they felt was a reasonable minimum number of Medicare plan members to be included in an "Accountable Care Organization". That number was 5,000. Since then CMS has used it for its Shared Savings Program and commercial payers have used it (to the extent they even have 5,000 plan members with a provider organization) in their shared savings and other ACO contracts. Recently, the State of CT filed its Innovation Model grant request, targeting total cost of care payment schemes, and using – you guessed it – 5,000 patients as the target for accountability. As mentioned in a prior post, DeLia and colleagues published a paper in early 2012 that challenged the established number of 5,000. Their conclusions are quite clear and should have caused everyone to stop and consider how to establish a more optimal number for these types of payment models. Financial risk, as we learned in the Wall Street debacle of 2008 and the ensuing Great Recession, cannot be estimated lightly and its management requires an understanding of its certainty….or lack thereof. In other words if financial projections based on faulty math give you a positive outlook when the correct numbers are negative, the compounding effect of that surprise can have very damaging consequences.

What this means to you – Transferring financial risk from third party payers to health plan members and providers is an inevitable and logical progression towards a real health care marketplace. We need that risk transfer to be successful. That doesn't mean everyone ends up being a winner, but it does mean each party has a reasonable understanding of the potential for wins and losses, and can manage to those numbers. Over time, those that are better at managing to those numbers will succeed. But if the odds of winning and losing are flawed, if they are misrepresented and misunderstood, then they become unmanageable, and those who should be winning might lose and vice versa. The net result is that the good actors might get punished while bad ones get rewarded, and that spoils it for everyone. Sticking to a flawed number is a path of least resistance, but also one that is far more perilous than necessary. To help providers and payers, in collaboration with IMPAQ International, we developed an ACO calculator. Build on DeLia's research, it helps estimate the probability of wins and losses based on a few simple variables that any organization could supply: the number of plan members covered by the deal; the average cost of professional services and facilities over the past few years. The results, we hope, will help payers and providers better understand the uncertainty of the outcome and the likelihood that winners will win or lose. We all need payment innovation to work, but it won't if we stick to flawed numbers.