January 26, 2024
Earlier this month, the Medicare Payment Advisory Commission (MedPAC) held a meeting to discuss its annual reports (the full versions of which are available only to commissioners) on the status of the Medicare program, including its report on the status of Medicare Advantage (MA). Things got a little heated at the meeting because of a long-standing dispute in the health policy space about the relative merits of MA compared to Medicare fee-for-service (FFS), and specifically over the report’s finding that MA is costlier and privileges healthier patients. Let’s examine what the MA report said and the controversy surrounding it.
MedPAC’s MA status report (the slides for which are available here) claims that the program had higher payments compared to FFS, linking the projected $82-billion difference in 2023 to MA’s greater coding intensity and favorable selection. Coding intensity is the number of diagnostic codes applied to a given patient and patient population (the greater the number of diagnostic codes per patient, the greater the coding intensity), while favorable selection occurs when a greater proportion of healthier-than-expected patients are enrolled in one type of plan over another, which leaves other plans to handle the costlier burden of sicker patients. In this case, MA having healthier-than-expected, and thus less expensive, patients means FFS would have sicker, more expensive patients.
Notably, MA has a higher coding intensity than FFS. MedPAC projects MA’s risk scores, a measure of how sick a patient is, to be 20.1 percent higher than if the beneficiaries were instead enrolled in FFS, resulting in a projected $47 billion difference in 2023 between MA and FFS spending. MedPAC concluded that chart reviews and health risk assessments (HRAs) account for about half of overall MA coding intensity: Essentially, plans discovered more illnesses in patients because they looked for them. Moreover, MedPAC stated that previous analyses showed risk scores grew 6 percent faster in the first year of MA enrollment than FFS, and 2 percent faster in the second year.
MedPAC determines favorable selection by measuring whether spending for MA enrollees is systematically lower than its risk scores predict, though it acknowledges that this is difficult to measure directly. Favorable selection can lead to higher costs because of the payment structure for MA plans. In a nutshell, at the beginning of every year, the Centers for Medicare and Medicaid Services (CMS) gives plans a lump-sum payment that is determined using data from prior years to predict how sick (and thus, costly) a plan’s population will be. If the population ends up being healthier and using fewer services than expected, the plan spends less on care and gets to keep the difference, meaning CMS spends more than it would have if those enrollees were in FFS. MedPAC claims that prior authorization and narrower networks – which are used by MA plans for cost control but can potentially limit care options for a patient – as well as potentially higher cost sharing than FFS (if an FFS beneficiary does not have a Medigap plan), may cause sicker beneficiaries who need more care to self-select out of MA plans. Favorable selection is believed to be responsible for a projected $32 billion in greater spending for MA than would have happened if those enrollees were in FFS. Overall, after coding intensity and favorable selection are accounted for, MedPAC projects MA payments to be 123 percent of FFS spending in 2024.
One reason this is drawing attention is that MA was intended to save money over FFS. But as noted at the start, not everyone agrees with MedPAC on how much more MA spends. A 2021 Milliman report found that MA’s per-member, per-month cost to the federal government was $943, versus $949 for FFS beneficiaries. The coding intensity (which should not be confused with the upcoding scandals that have rightly received backlash) differences based on chart reviews and HRAs simply indicate that more thorough examinations of patients find more issues. (This author believes that identifying what’s wrong with people is one of the main purposes of health care.) If FFS looked harder, it would probably see its coding intensity increase.
MA plans are also paid more by the federal government if their patients are sicker – so it’s the government itself incentivizing them to provide more thorough care to patients. According to a report commissioned by Better Medicare Alliance, a trade group of MA insurers, this increased identifying of health issues has led to a 43 percent lower rate of avoidable hospitalizations and faster speeds for diagnosis, specifically diagnosing Type 2 diabetes an average of five months earlier compared to FFS beneficiaries. As for favorable selection, 52 percent of MA beneficiaries have less than $25,000 in annual income, while MA has higher proportions of Black, Hispanic, and Asian/Pacific Islander beneficiaries than FFS. Both low-income and minority populations have historically worse health outcomes than wealthier and White populations. Given these factors, one would expect MA plans to have sicker patients, not “surprisingly healthier” ones.
Perhaps the most important sign that MA provides more comprehensive care than FFS is the simplest: 52 percent of all Medicare beneficiaries are enrolled in MA plans in 2024, following a steady increase since 2010’s 26 percent. The hits on MA – coding intensity and favorable selection – may point to greater flaws in FFS that are driving increasing numbers of seniors to choose MA. If the intent is to accurately compare these Medicare programs, MedPAC must take into account all of the factors noted above. Meanwhile, MA’s opponents would far better serve the public if they focused on how to make FFS more attractive.