The Weekly Sift

Up until now I’ve been unwilling to make this claim,
but not because I didn’t believe it.

From the beginning, it seemed like common sense to me: Losing health insurance increases your risk of dying. Uninsured people get less care, and medical care saves lives, so lack of care logically would cost lives.

Big-picture statistics backed up that intution: Other industrialized English-speaking countries provide universal healthcare, and people live longer there. (Life-expectancy-at-birth: Australia 82.15 years, Canada 81.76, United Kingdom 80.54, United States 79.68.) None of those countries is an exact duplicate of the US, but is Canada so different that its people should live two years longer, or is their healthcare system just better than ours?

I knew that people have denied this. Back in May, Republican Congressman Raul Labrador bluntly stated “Nobody dies because they don’t have access to health care.” During his 2012 campaign, Mitt Romney pointed to emergency rooms and asserted that everyone gets life-saving care when they really need it. “If someone has a heart attack, they don’t sit in their apartment and die.”

But that argument didn’t impress me: Yes, the uninsured get life-saving care when they’re in car accidents or having heart attacks, but a lot of the treatable things that kill people more slowly, like high blood pressure or diabetes, aren’t emergencies. And while an ER might take out the tumor that’s blocking your intestine and threatening to kill you in a matter of hours, it won’t provide the follow-up chemotherapy or radiation that you’ll need if you plan to keep on living for more than a few months.

So I kept being tempted to say that TrumpCare would kill people. Other people have: Democratic politicians like Senator Bernie Sanders and Congressman Ruben Gallego, journalists like ThinkProgress‘ Ian Millhiser, and doctors like Christy Duan and Andrew Goldstein. They based their claims on solid scientific studies like this one and this one.

But every time I got ready to repeat that claim, I’d google “lack of health insurance kills people” and run into articles claiming to prove the opposite or just debunk the idea that we know one way or the other. Chasing the links in those articles always led me to different scientific studies, like this one or this one.

In other words, it looked like one of those my-bubble-versus-your-bubble arguments that I try to stay out of. Liberals cherry-pick the studies they want to believe, conservatives do the same, and we all talk past each other. Yes, I think of myself as a liberal, but my true allegiance is to the reality-based community. Like Fox Mulder, I believe that the truth is out there, and I would rather find it than just go on believing whatever I’m inclined to believe anyway.

So it’s been on my to-do list for months to devote some serious time to this issue, until I could feel confident that I really understood what is actually known. But given how much hard work would be involved and the possibility that I still might not arrive at a clear conclusion, that project never rose to the top of my stack. So I never boldly wrote, “TrumpCare will kill people.”

Fortunately, people better equipped than myself have taken the challenge on. Benjamin Sommers, Atul Gawande, and Katherine Baicker recently published an article in The New England Journal of Medicine acknowledging the controversy and comparing the studies quoted by each side. Weighing it all, they came to this conclusion:

The body of evidence summarized here indicates that coverage expansions significantly increase patients’ access to care and use of preventive care, primary care, chronic illness treatment, medications, and surgery. These increases appear to produce significant, multifaceted, and nuanced benefits to health. Some benefits may manifest in earlier detection of disease, some in better medication adherence and management of chronic conditions, and some in the psychological well-being born of knowing one can afford care when one gets sick. Such modest but cumulative changes — which one of us has called “the heroism of incremental care” — may not occur for everyone and may not happen quickly. But the evidence suggests that they do occur, and that some of these changes will ultimately help tens of thousands of people live longer lives. Conversely, the data suggest that policies that reduce coverage will produce significant harms to health, particularly among people with lower incomes and chronic conditions.

If the name Atul Gawande rings a bell, it’s probably because (in addition to being a doctor and public health researcher) he’s the author of popular books like Complications, Better, and Being Mortal. He also writes about health issues for The New Yorker, making him that rare researcher who’s able to popularize his own work, as he did this week in “How the Senate’s Health-Care Bill Threatens the Nation’s Health“.

To understand how the Senate Republicans’ health-care bill would affect people’s actual health, the first thing you have to understand is that incremental care — regular, ongoing care as opposed to heroic, emergency care — is the greatest source of value in modern medicine. There is clear evidence that people who get sufficient incremental care enjoy better prevention, earlier diagnosis and management of urgent conditions, better control of chronic illnesses, and longer life spans.

… Insurance expansions have made people more likely to get primary and preventive care, chronic-illness care, and needed medications — including cancer screenings, diabetes and blood-pressure medicines, depression treatment, and surgery for cancer before it is too late.

These improvements in care help explain why people who have health insurance are twenty-five per cent more likely to report being in good or excellent health. It also explains why they become less likely to die. Proper health care saves lives, and the magnitude of the reduction in deaths increases over time.

… Conservatives often take a narrow view of the value of health insurance: they focus on catastrophic events such as emergencies and sudden, high-cost illnesses. But the path of life isn’t one of steady health punctuated by brief crises. Most of us accumulate costly, often chronic health issues as we age. These issues can often be delayed, managed, and controlled if we have good health care — and can’t be if we don’t.

The incremental nature of most medical interventions — the drugs I take to keep my cholesterol low might or might not prevent a heart attack in 2030 — explains why the life-saving effect of insurance is hard to find in many studies, especially ones that only examine a few years. (Sometimes a decrease in mortality is noticed, but isn’t reported as a conclusion because the difference detectable within the time frame of the study isn’t statistically significant yet.) For the health crises that threaten to kill you in short order, Mitt Romney is right: The ER will help you whether you are insured or not. (You may have to go bankrupt when their bill comes, but that’s a different issue.)

But emergency care is far from the only way that medical care saves lives. Having watched both my parents grow old and die, I understand that many — perhaps most — deaths in this era aren’t caused by a sudden crisis out of the blue. Instead, dozens of problems that are not immediately life-threatening have a way of building on each other until people get encircled by them. A sudden crisis may kill you, but only because you have gradually lost all your room to maneuver. One problem limits your mobility, another makes it hard to sleep or enjoy food, your long-time interests and activities become hard to maintain, you become feeble, and then you get depressed and stop even trying to regain your lost abilities. Whether that encirclement happens to you at 50 or at 90 depends largely on what kind of care you get.

The difficulty of measuring these kinds of outcomes and attributing them to specific causes means that precise estimates of the number of such deaths should be taken with a grain of salt. An article in Vox on Wednesday claimed that 208,500 additional people might die over the next ten years if the Senate TrumpCare bill passes. That’s speculative, as the authors acknowledge. Maybe it will only be 50,000 people, maybe 400,000. Current research isn’t sharp enough to be precise.

But people will die, probably quite a large number of them.