Gaming the System: How Major Insurers Are Able to Extract Big Profits from Medicare Advantage
March 15, 2023
John McCracken, PhD
In 2022 Medicare Advantage (MA), the private plan alternative to traditional Medicare, enrolled more than 28 million people, representing 48 percent of the eligible Medicare population and $427 billion of Medicare spending. By 2030, MA is expected to insure over 60 percent of American seniors.
The program was originally meant to save taxpayers money, but it never has. Instead, Medicare Advantage has become rife with abuse, with large private insurers taking advantage of the rules to overcharge the government. The federal government is presently suing four of the five largest players—UnitedHealth, Humana, Elevance and Kaiser—accusing them of Medicare fraud, and a fifth—Aetna—recently told investors its practices were also being investigated by the Department of Justice. Collectively, these five insurers control two-thirds of the MA market.
Major insurers have been attracted to Medicare Advantage because it is very profitable; the gross profit margin—the average amount by which premium income exceeds claims cost—is about double the gross margins for plans in their group and individual markets. Evidence from government audits, fraud lawsuits and academic research has shown that a significant source of these profits has been the result of overstating the health problems of their members to collect extra payments.
MA Risk Adjustment
MA plans submit bids to CMS for a standard beneficiary, but the actual monthly payment for each plan member will increase or decrease based on that individual’s risk score. Generating higher risk scores for MA members can significantly increase plan revenues.
In addition to a patient’s documented clinical record, CMS allows either a retrospective chart review or a health risk assessment (HRA) to be used as a basis for an individual’s risk score. There is growing evidence that these two alternatives have been leveraged inappropriately to “upcode” diagnoses and increase risk payments without the beneficiaries receiving the care indicated for those diagnoses.
There is no requirement that the results of either a chart review or a HRA be linked to a patient’s previous medical service record. Almost immediately, MA companies saw ways to exploit that situation. HRAs can now be conducted by visits to beneficiaries’ homes by nurses employed by third-party vendors. One company, Mobile Medical Examination Services, is accused of pushing its employees to document a range of diagnoses, including some — vertebral fractures, pneumonia and cancer — they lacked the equipment to detect.
Chart reviews are increasingly performed using specialized software and artificial intelligence to identify and reclassify conditions with risk codes linked to higher payments, even if those conditions were long dormant and had nothing to do with the patient’s current medical care. Moreover, chart reviews can them be submitted without identifying and providing the specific services associated with the diagnoses. The sicker a patient appears to be, the higher the risk-adjusted payment to the MA plan for that member, even though the plan is not required to provide the member treatment for the reported condition.
Unsupported risk-adjustment payments have been a major driver of improper payments in the MA program. In 2017, seven major insurers generated more than 90 percent of their risk adjustment payments from diagnoses that were reported only on unlinked chart reviews and/or in-home HRAs. For the industry as a whole, 40 percent of risk payments were based on diagnoses that MA companies reported on unlinked chart reviews.
The Medicare Payment Advisory Committee (MedPAC) has documented approximately $140 billion in risk adjustment overpayments over the past 12 years. According to some health services research, MA will cost Medicare over $600 billion more in the next 8 years than would have been the case if coding intensity were set to the empirically justified level.
The most common allegation against the MA companies has been that they do not correct potentially invalid diagnoses even after becoming aware of them. For example, in a 2020 lawsuit the Justice Department claimed that thousands of inaccurate diagnoses at Anthem (now Elevance) were not deleted. According to the lawsuit, a finance executive calculated that eliminating the inaccurate diagnoses would have reduced the company’s 2017 earnings from reviewing medical charts by $86 million, or 72 percent.
Are MA Enrollees Sicker Than Traditional Medicare?
Since 2004, payments to MA plans have been as high as 17 percent in 2009 to a recent average of 4-5 percent above the amount traditional Medicare would have spent for the same beneficiaries On one hand, MA plans seem to be providing extra services that help their enrollees manage their care. On the other hand, recent studies suggest that enrollees are no sicker than those in traditional Medicare.
The average MA risk score in 2006 was 98 percent of the FFS average, but in 2017 (the latest year with publicly available data) average MA risk was an estimated 112 percent of FFS. Three other measures of risk, however, show much smaller increases in MA acuity over that time period:
- Standardized mortality rates for MA enrollees changed little over the period. They have held steady at about 90 percent of the mortality rate of FFS beneficiaries, which can be explained by the fact that that many of the sicker MA enrollees have transferred back to traditional Medicare.
- The relative risk of MA measured with a risk score based on prescription drug utilization changed very little from 2008 to 2014.
- Relative MA risk measured using demographic characteristics also changed very little from 2009 to 2019.
Further, the relative risk scores for the MA population using a combination of all three of these measures are consistently below 1.0. This finding is similar to findings from other studies and provides support for the proposition that MA members appear to be sicker not because they actually are sicker, but rather because diagnostic information is reported more aggressively by MA plans.
In response to the upcoding evidence, Congress required CMS to adjust risk scores down 3.4 percent beginning in 2010 and 5.9 percent in 2018 and beyond. The CMS administrator has the authority to further rein in costs by implementing adjustments larger than the statutory minimum. No administrator has chosen to do so, however, perhaps because of industry pressure and/or potential pushback from Congress. Further, MedPAC has recommended that diagnoses reported in home-based health risk assessments not be used in constructing risk scores. This too, has not yet been done.
Quality Ratings
Medicare uses a five-star rating system to inform the public about the quality of private Medicare Advantage plans. The stars are based on more than 40 measures that track whether enrollees are receiving appropriate preventive care and how well their chronic diseases are controlled.
Plans with four or more stars can typically receive bonuses 5 percent higher than projected local spending in the traditional fee-for-service program. High-performing plans in urban counties where fee-for-service spending is low and MA enrollment is high do even better; they are eligible for a bonus of up to as much as 10 percent.
The Star Quality Bonus program is not an accurate indicator of quality, however, because star ratings are reported at the contract rather than the plan level. Most MA contracts include multiple plans, which may have different benefits and serve different geographic areas. As has been well documented, insurers lump together poor-performing plans with better-performing plans to earn the quality bonus payments for the entire group, masking the performance of low-quality plans and maximizing the bonus money received.
The quality bonus program in Medicare Advantage is expensive. Since 2015, it has paid out $49.4 billion in additional plan payments. These payouts dwarf those in pay-for-performance programs of traditional Medicare and commercial insurers; some plans earn many times what can be earned in incentive programs in traditional Medicare.
Opinions differ about whether MA enrollees experience better care and outcomes than those in traditional Medicare, but the weight of evidence is that they do not. MedPAC has concluded that, because of poor quality reporting by MA plans, the commission’s yearly reports to Congress can no longer provide an accurate description of the quality of care in Medicare Advantage.
Going Forward
The federal government has accused five of the biggest Medicare Advantage insurers, representing more than two-thirds of the market, to have submitted inflated bills. And four of the five have faced Department of Justice lawsuits alleging that efforts to over-diagnose their customers crossed the line into fraud.
Last year, the Justice Department’s civil division listed Medicare Advantage as one of its top areas of fraud recovery. The MA landscape is changing, albeit slowly, but under escalating federal and Congressional pressure, eventually these excess profits will be wrung out of the system.
John McCracken is Clinical Professor of Healthcare Leadership and Management in the Jindal School of Management, The University of Texas at Dallas.