Relative Medical Risk of MA and FFS Medicare Beneficiaries

While the Medicare Advantage program is an undoubted success from the beneficiary perspective, with over a third of them now enrolled in an MA plan and many receiving extra benefits and reduced cost-sharing; there are ongoing concerns about the payment method used by CMS for the plans.  CMS reimburses on a health risk-adjusted basis, so inevitably the plans have an incentive to pay close attention to the diagnostic coding related to the health of their members.  The more diagnoses, the more you get paid.  CMS in turn has attempted to address these concerns by a “coding intensity” payment reduction which currently is around 6%.  A study carried by Health Services Research attempts a new analysis of the relative health of MA and FFS Medicare beneficiaries.    (HSR Article)   The authors used data from 2008 to 2015 on Part D drug plan enrollment and drug use to construct a predictive total health spending model and compare those predictions between MA and FFS beneficiaries.   Based on their risk groups and associated predictions, MA enrollees had about 91% of the prevalence of the top 20 diagnostic groups compared to FFS patients in 2015.  And their expected spending was about 7% lower for the year.  The risk of the FFS population was relatively constant from 2008 to 2015, but the risk of the MA population increased by about 3.5 percentage points over that time, suggesting a convergence of MA and FFS beneficiary health risk, which is what you would expect as the proportion of all beneficiaries in MA continues to grow rapidly.  Over the same period, however, the risk reported to CMS by the health plans increased by almost 10%.  The reported health risk is growing faster than the coding adjustment applied by CMS to payments.  Since MA plans are paid over $200 billion a year, the implication is that CMS is overpaying them by billions of dollars.

Coding intensity is definitely higher in the MA plans, for the obvious reasons relating to the financial incentive created by CMS.  To the extent that CMS can accurately identify the difference in coding between MA and FFS, payments should be reduced accordingly.  But this is more complex than it sounds.  Consider an alternative approach to understand this complexity.  Let’s go try to make the coding for FFS patients more comprehensive and accurate and then let’s ask ourselves why the providers treating these patients aren’t identifying and potentially aren’t addressing all the uncoded health issues.  If they were, what would that cost the Medicare program?  We also need to consider that MA plans are almost certainly doing a much better job of addressing beneficiaries’ health needs and improving their health.  Over time, that has significant benefits for CMS and the patients.  If MA plans can treat the same beneficiary more effectively and more efficiently, should they be penalized for doing so.  Personally, I think the bigger issue to address in Medicare Advantage reimbursement is why CMS is still doing regional benchmarking and why they are limiting the competitive nature of the bidding process.  CMS should have a pure competitive bid process and its base payment should be set at the lowest bid by a plan that has a track record of acceptable quality, period.  Anyone who wants to charge more than that will have to persuade beneficiaries why they should pay more.