Medicare Advantage (MA), Medicare’s popular system of private health plans, is in the hot seat.
During a recent congressional hearing, members and witnesses expressed concern about MA plans’ efforts to secure higher reimbursements.
Meanwhile, the Center for Medicare and Medicaid Services, the agency that runs Medicare, announced an annual audit of every health plan—and the Department of Justice opened an investigation into UnitedHealth Group, the largest MA insurer, for business practices that result in improper payments.
Broadly speaking, many Washington officials believe that Medicare Advantage plans are “gaming” the current system at taxpayer expense. But the question lawmakers should be asking is, “What’s wrong with the current system that it enables, and even encourages, gaming?”
One clue is that the current system for determining payments to Medicare Advantage plans is complex. The more complicated something is, the more ways it can (and probably will) go wrong.
In this case, the complexity results mainly from Medicare’s “risk adjustment” methodology, which tries to project how much enrollees are expected to cost and set plan payments accordingly.
That kind of prospective risk adjustment methodology can work reasonably well in the aggregate—provided it relies on a limited number of variables supported by statistically valid data.
For instance, using Medicare’s large claims datasets, actuaries can calculate the difference in medical spending between, say, the average 70-year-old and the average 80-year-old. The same can be done for other basic variables such as, sex and geographic location.
However, the current MA risk adjustment methodology tries to go further and predict each enrollee’s future medical costs. So Medicare requires plans to submit data on the health status of the beneficiaries they enroll using diagnostic codes for a wide variety of conditions, ranging from diabetes to depression to high blood pressure. That not only enormously increases complexity—it creates new problems.
For instance, with MA the more diagnostic codes submitted for a beneficiary, the more Medicare pays the plan. Yet for a traditional Medicare patient, the doctor’s incentive is to submit only the minimum codes needed for payment. These contrary incentives bias the resulting data to portray MA patients as sicker than they really are, while simultaneously portraying traditional Medicare patients as being healthier than they actually are—thus making comparisons between MA and traditional Medicare unreliable.
Another problem is that the current system skews plan payments toward patients with chronic conditions while doing a poor job of adjusting payments for the occurrence of expensive acute conditions like cancer.
The solution is to split the existing risk adjustment system into two parts.
The first part would prospectively adjust plan payments using a few well-documented characteristics that correlate with variations in medical spending, such as, age, sex, institutional status and location.
The second part would be a retrospective risk transfer pool through which the MA plans in each state or region could collectively adjust for any maldistribution of claims costs.
While the plan actuaries would collectively design the details, the basic effect would be to redistribute funds from plans with more than the average share of low-cost enrollees to plans with more than the average share of high-cost enrollees. Thus, plans that enrolled a larger share of sick or expensive beneficiaries would be made whole with payments that more accurately reflect their real costs.
As we explain in our book Modernizing Medicare: Harnessing the Power of Consumer Choice and Competition (Johns Hopkins University Press, 2023), such a change would not only eliminate the guesswork and possible gaming of the system but also eliminate unnecessary costs to the taxpayer. It’s just common sense.
As we explain in our book Modernizing Medicare: Harnessing the Power of Consumer Choice and Competition (Johns Hopkins University Press, 2023), such a change would not only eliminate the guesswork and possible gaming of the system but also eliminate unnecessary costs to the taxpayer. It’s just common sense.
Robert E. Moffit, PhD, is a Senior Research Fellow at the Heritage Foundation. Edmund F. Haislmaier is the Preston A. Wells Jr. Senior Research Fellow in Heritage’s Center for Health and Welfare Policy.
