The Projected Death Count from Trump’s One Big Beautiful Bill

Yves here. Most eyes are on the ICE protests in Los Angeles, which are spreading to other big cities. A side benefit of Trump’s show of force is the way it takes attention and political energy away from the attack on ordinary citizens’ welfare embodies in his “big beautiful bill” This post tallies the expected impact of cuts to Medicaid, Medicare, and the Affordable Care Act. It is not widely understood that over time, Medicaid has become an all-purpose residual medical insurance program. So cuts to Medicaid rend what had become a complicated but kinda-sorta working and overall very important stopgap.

By Sara Talpos, a contributing editor at Undark. Originally published at Undark

Last week, the growing tension between President Donald Trump and Elon Musk exploded into public view when Musk took aim at the One Big Beautiful Bill Act. The legislation, which calls for sweeping changes to taxes and spending, is currently before the Senate. Somewhat lost in the hullabaloo was a new analysis by researchers at the University of Pennsylvania and Yale University. They found that the bill’s provisions for public health insurance — Medicaid, Medicare, and the Affordable Care Act — are likely to lead to more than 51,000 extra deaths each year.

That analysis, the researchers wrote, originated at the request of two Democratic senators, Ron Wyden and Bernie Sanders. But the goal was to provide a nonpartisan assessment, said Eric Roberts, a member of the research team and a health economist at the University of Pennsylvania’s Perelman School of Medicine. People of all political stripes are covered by Medicaid, he noted, which “touches almost all lives at some point in someone’s life: if at birth, if at the end of life, or in the middle of life.”

To arrive at their projections, the researchers worked with a body of scientific literature that documents the relationship between public insurance usage and death rates, and analyzed how the bill’s public health insurance provisions — which include work requirements and the expiration of certain tax credits, among others — could lead to increased mortality.

Undark spoke with Roberts by phone. The interview has been edited for length and clarity.

Undark: The Congressional Budget Office has estimated that the bill would lead to roughly 7.7 million people losing insurance coverage by 2034. Why did they place that estimate a decade out? And will it take time for this new bill, if passed, to fully influence people’s access to coverage?

Eric Roberts: Normally, forecasts are for 10-year periods when budget planning is done. So the scoring of bills includes a 10-year horizon, which is why they go through 2034. The CBO [also] reports annual estimates of coverage loss, but as a high-level summary, they tend to quote the effect at 10 years in terms of cumulative coverage loss.

The effects that they project do grow a bit over time. That’s due to the expectation that it would take time for states to phase in Medicaid work requirements, for example, which is one of the major forecasted drivers of coverage loss.

UD: Why would Medicaid work requirements result in coverage loss?

ER: The proposal is for so-called able-bodied, working-age adults who are not engaged in caregiving for another individual, who are not pregnant, who are not attending school, that they have to demonstrate that they’re currently working or are seeking work in order to maintain Medicaid coverage.

This was tried in Arkansas in the last Trump administration. The findings are really that fewer people actually started working because of this requirement, that many people who were subject to this requirement were already engaged in work or met one of the exemptions, for example, for attending school or providing caregiving.

But what the work requirements did was increase the administrative burden of demonstrating continued eligibility for Medicaid. People would not file forms on time or correctly. Apparently, the website that individuals were required to use to provide certification of work was very hard to use on mobile devices. A lot of people fell through the cracks simply because they were navigating a cumbersome reporting requirement.

So the forecast is that people will lose Medicaid, not necessarily because they don’t qualify, but because additional paperwork often just causes people to fall through the cracks.

UD: Your group’s analysis projects that this loss of health care coverage [due to work requirements] will lead to more than 10,000 deaths per year. Can you walk me through the work that was done to arrive at this number?

ER: We used a number of different studies that have estimated the effects of Medicaid on mortality. Because of the phased expansion of Medicaid — including under the Affordable Care Act — in the last decade, we actually have very good evidence on the effects that Medicaid expansion has on population mortality.

Most of the research leverages the fact that some states have opted not to participate in expansion, so you kind of have this control group that allows us to estimate mortality trends in the absence of expansion. We can compare trends in expansion states to those that didn’t expand to come up with an estimate of the net reduction in deaths that was attributable to expansion.

Those estimates suggest that Medicaid expansion reduces mortality generally on the order of magnitude of about three to four deaths per 1,000, which sounds quite small until you think about the baseline mortality rate in this population, which is about 1 percent, or about 10 deaths per 1,000. So reduction of three to four deaths per 1,000 is a 30 to 40 percent relative reduction in deaths on a baseline mortality rate.

What we said is, essentially, if we now remove the benefit of getting Medicaid from those populations, they would be expected to experience an average of a 30 to 40 percent relative increase in mortality. And so we aggregated that projected effect across age and demographic groups to arrive at that overall estimate. 

UD: The new bill doesn’t extend certain tax credits for people who receive health insurance under the Affordable Care Act. If the bill is passed as is, when will these tax credits expire? And why would their expiration lead to an additional 8,811 deaths?

What you’re referring to are enhanced tax credits that were part of the American Rescue Plan Act, or ARPA, passed in 2021. The Affordable Care Act created, as you know, a system of individual marketplace coverage for people who couldn’t get affordable coverage through their employer, or who didn’t qualify for Medicaid, and provided premium subsidies in the form of advanced payment tax credits to individuals to obtain health insurance.

So essentially, the IRS paid their premium in advance and treated it like a tax credit at the end of the year. So this helped a lot of people get affordable coverage who would not otherwise have gotten it.

The ARPA legislation bumped up those tax credit subsidies quite a lot, and enrollment in the marketplaces has reached an all-time high. We attribute that partly to the fact that coverage has become more affordable for people.

Removing those enhanced tax credit subsidies, which are due to expire this year unless they are extended, would likely result in fewer individuals signing up for marketplace coverage. There is very good evidence that enrolling in marketplace coverage has similarly protective effects against mortality.

UD: How confident are you in these figures? Is there any reason to think that things will turn out differently than projected?

ER: Projections are always hard. They are forecasts of the future, and they rely on certain assumptions — that individuals who are dropped from coverage, for example, would not pick up alternative coverage. What I would say is that the mortality estimates that we are using to make these projections, though, are based on rigorous evidence that has been reproduced by multiple independent studies. There are multiple independent studies that show an effect of Medicaid on mortality, and multiple studies that show effects of drug coverage, for example, on mortality.

There is some variability around those estimates, but I think we can say pretty confidently that large reductions in public insurance coverage will have clear population health effects — and that that has been independently demonstrated by multiple studies.

UD: Republicans’ trust in scientists dropped noticeably during the pandemic. If you could speak to this group of Americans, what would you tell them about your team’s analysis to reassure them that it was conducted fairly and with their best interests in mind?

ER: Much of our research is taxpayer funded, and we approach that with a great deal of responsibility. It’s funded not for Democrats, not for Republicans only, but for the benefit of Americans. Personally, I’m a beneficiary of public insurance. I had Medicaid when I was a kid, and so I view the work that I do as, in some ways, a return for societal investment in people like me.

I hope that our audience includes people of all political persuasions because we’re not trying to do research for one particular ideology.