TrumpCare Would Result in a Lot of Preventable Deaths, And it is Not Uncivil to Say So
By 
We’ve addressed the claim that it’s a grave threat to Civil Discourse to accurately describe the effects of BCRA. Harold Pollack’s essay on the subject is a must-read:
This political framing is fairly ridiculous when you step back to think about it. Snatching coverage from millions of people will cause many to experience severe financial problems, declining mental health, and poor management of disabilities and chronic diseases. The human and policy arguments against doing so are hardly nullified if this only hurts people and makes them worse off without actually killing them.
However illogical this framing may be, it is a shrewd polemical gambit. Taking refuge in these specific statistical uncertainties allows conservatives to adopt the pose as dispassionate empiricists who just want to understand the data. Not coincidentally, it allows BCRA supporters to avoid confronting the bill’s brutal distributive realities, diverting attention from the many harms associated with losing health insurance that are more readily studied and thus more solidly grounded.
This framing also allows bill supporters to lambast liberals for uncivil fearmongering and for allegedly making hysterical claims that “MILLIONS WILL DIE!” I haven’t seen people saying that. But it’s easy enough to punch down on social media by finding hysterical-sounding people who exaggerate the numbers, rely on anecdotal data, or misinterpret observational studies.
[Avik Roy immolating his dignity on Twitter omitted]
There’s only—or at least—one problem here. Existing research, for all its flaws, indeed suggests that thousands of Americans will probably die needlessly every year if BCRA is passed. That’s not inflated rhetoric. That’s what any reasonable person would predict based on the available data.
My own favorite study examined the survival benefits of implementing Romneycare in Massachusetts. Other studies compare populations that for one reason or another gained access to expanded Medicaid to populations that did not. Not surprisingly, different studies of different insurance changes applied to different groups find diverse effects. The Massachusetts study suggests that one death is prevented for every 830 people newly insured. Writing at Vox, Ann Crawford-Roberts, Nichole Roxas, and Ichiro Kawachi note that Medicaid coverage thus rivals widely accepted clinical interventions such as screening colonoscopy. If that 1-in-830 finding generalizes BCRA, rendering 22 million people uninsured would imply about 20,000 deaths per year.
Now, existing studies might not fully generalize. Let’s say the Massachusetts study overstates the impact of BCRA by a factor of two. If so, BCRA would still cause around as many annual deaths as firearm homicides. Just as plausibly, the Massachusetts and Oregon studies could actually understate the survival benefits of Medicaid. Both states offer relatively generous medical and social services, including to the uninsured. In Massachusetts, the biggest and clearest mortality effects of Romneycare were found in the lowest-income counties. Medicaid expansion probably matters a lot more in rural West Virginia or Portland, Maine, than it does in southern Massachusetts or Portland, Oregon.
We have strong reason to believe that uninsuring millions of people would cause thousands of them to needlessly die every year. We can debate the magnitude of this effect and how we might measure it. But the burden of proof does not rest with liberals to defend this sensible proposition, nor to prove that the likely harms associated with BCRA will definitely come to pass. Rather, the burden resides with BCRA supporters to show why we can be confident that snatching coverage from low-income people will not bring serious harm. Bill supporters have not come close to meeting this standard.
This desperate gambit of TrumpCare apologists is both dishonest and a classic example of smarm in Scocca’s sense (“smarm is a kind of performance—an assumption of the forms of seriousness, of virtue, of constructiveness, without the substance. Smarm is concerned with appropriateness and with tone. Smarm disapproves.”) And in this context, it’s beyond reprehensible — the argument is literally “it’s worse to accuse people of favoring legislation that will kill people unnecessarily than it is to favor legislation that will kill people unnecessarily.”
In related news, the CBO has found that BCRA would reduce Medicaid expenditures by 35% by 2036. I can’t wait for the Times views-differ thinkpiece about whether this constitutes a “cut.” This is why apologists for TrumpCare are writing bullshit about how Medicaid doesn’t help people, and why it’s important to keep pointing out that the evidence says no such thing.