For many years Medicare Advantage plans were accused of cherry-picking healthy enrollees and research suggested that Medicare was paying the MA plans more for a beneficiary than the same beneficiary would cost in fee-for-service Medicare. In 2004, Medicare began using a risk score based on beneficiary health status to adjust payments to the plans. Understandably, the MA plans began to pay more attention to the completeness of coding for beneficiary conditions and diseases, which would result in higher payments. So the next countermove was for Congress and CMS to begin to reduce payments for “coding intensity”, much as has been done for hospitals under DRGs and for other at-risk providers, and to stop including certain oft-abused diagnostic categories in the risk calculation. A study published in the Medicare and Medicaid Research Review examines the recent experience with coding practices of the MA plans. (HHS Study) FFS coding can itself be poor and inconsistent and not reflect the reality of the beneficiary’s health needs. So careful analysis is needed to balance FFS and MA costs. And the whole rationale for MA plans is that they will reduce spending while at least maintaining quality, so the cost to Medicare for the same beneficiary should be lower in MA. The research validated the rapid growth in coding intensity for MA plans versus FFS Medicare, with MA risk scores increasing from 88% of FFS ones in 2004 to 102% in 2013. In addition to coding intensity, other possible reasons for the change include a change in the health status of beneficiaries who are enrolling in MA and MA beneficiaries getting sicker faster than those in FFS. But looking at other data suggested to the researchers that these alternative explanations at most account for a small part of the coding difference. And it appears that there is substantial variation across MA plans in how aggressively they code, which would not likely be due to beneficiary health differences.
At this point you have to ask if the current reimbursement methodology should even be maintained. Apparently not all plans engage in “complete” coding activities, so those that don’t are being treated unfairly when there reimbursement is reduced. And all MA plans now almost have to engage in these coding efforts just to keep up, and this costs money which might be better spent on health care or real quality improvement programs. Since it does not appear that MA plans are preferentially enrolling healthier members at this point, some of the rationale for the risk-adjustment system is gone. A pure competitive bidding payment methodology might be best. CMS could make a payment based solely on the lowest bid in each market.