How a Medicare Payment Fix Could Worsen Disparities​

Medicare Advantage (MA) plans, which now cover 30 million people, are under increased scrutiny for essentially overcharging the government using a common tactic called upcoding.

It works like this: The federal government pays MA plans in line with the number of diagnoses a patient has. When a plan falsely claims that its beneficiaries have more medical issues than they actually have, the plan is upcoding to get more money from the government. 

This is a big deal because the practice does not help beneficiaries and costs taxpayers billions of dollars. And those costs could grow as MA becomes increasingly popular. 

The government, through newly enforced audits, is cracking down on plans that upcode. 

Changing the way plans get paid 

One solution proposed by former CMS officials is to change how private Medicare plans get paid. For example, rather than paying them based on a patient’s diagnoses, Medicare could instead pay plans based on factors that can represent clinical complexity but are harder to misreport, such as “social risk factors.”

It is well documented that health care needs increase when members have more social risk factors. Variables such as race, educational attainment and area deprivation index (ADI) – a measure of socioeconomic disadvantage in an individual’s neighborhood – can be confirmed outside the clinical environment and are thus less likely to be upcoded.

A recent paper in Health Affairs by Michael McWilliams and colleagues tests this proposed change to see what impact it could have on payments to MA plans. They find that while this alternative payment approach may be harder to manipulate, it could introduce a different problem: exacerbating disparities. 

The problem is this: Medicare uses historic patterns of health care use and spending to determine its payment rates to MA plans. And MA beneficiaries with more social risk are less likely to receive needed health services and therefore have lower historical spending. Specifically, the authors found that Medicare spending was:

• $574 lower for Black beneficiaries and $1,462 lower for Hispanic beneficiaries than for white beneficiaries.
• $85 lower for beneficiaries with less than a high school diploma than for those with more education in the same county.
• $68 lower for beneficiaries per 30 point increase in ADI of their neighborhood.

Thus, changing to a payment system based on social risk factors could paradoxically decrease MA payments by predicting that less federal spending is required. 

This has implications for equity. If MA plans are paid less by the government for caring for disadvantaged patients, the plans may be more incentivized to enroll healthier, advantaged ones.

Certain patients use less care 

The authors’ findings are surprising. It’s often assumed that people with higher risk factors are in poorer health and therefore consume more health care, leading to higher government spending on those patient populations. 

But, for a number of reasons, some patients with social risk factors may use less health care, causing the government to spend less. 

Black beneficiaries, for example, are less likely to trust their providers and may face discrimination when trying to access care, reducing their interactions with the health care system. And living in a poorer neighborhood could mean less job flexibility, internet access or transportation to make appointments. 

Factors such as those have the authors questioning whether basing MA payments on social risk factors would promote less federal spending on these patients, even though they often have worse health outcomes.

There’s no question upcoding by MA plans is an expensive and growing problem. But, as this paper demonstrates, proposed payment fixes may present unintended consequences of their own.