Reversal on daily aspirin shows medicine's weak spot

There's a pattern to medical reversals that can help explain this week's seeming U-turn on that age-old advice to take an aspirin a day to prevent heart attacks and strokes. Evidence had actually been building for some time that this might do a lot of people more harm than good.

This latest news shouldn't serve as an indictment of medicine but as a warning to be skeptical of certain kinds of recommendations. That includes any medical intervention aimed at healthy people — especially treatments that aren't backed by multiple controlled clinical trials.

Doctors sometimes recommend untested interventions because they think they're taking a better-safe-than-sorry approach. The problem is they often get the safe and sorry sides of the equation mixed up. They did that for years when they thought they should wait until there was proof dietary fat was safe before taking it off the list of foods that were bad for you. They should have recognized that it would have been safer to let people keep eating the way they'd done for millenniums, and not changed anything without reasonably strong evidence.

I had a conversation recently for a podcast episode on this topic with physician Gilbert Welch, who is author of "Less Medicine, More Health." "I'm a conventionally trained physician, and I believe that medical care can do a lot of good — in particular for people who are sick and injured," he said. "It's hard to make a well person better, but it's not that hard for us to make them worse." This sounds weirdly obvious, and yet it doesn't reflect the way our medical system works.

The podcast wasn't about aspirin but about all the tests and screenings foisted on people who are feeling fine. Some will get diagnosed with cancer. Some will get treated with surgery or chemotherapy, and most will assume the treatment saved their lives. But there's evidence that this isn't necessarily the case, and that for many kinds of screening, harms may outweigh benefits.

If you're going to accept any medication or intervention as a healthy person, it should therefore be backed by strong evidence. We have that for COVID-19 vaccines — massive clinical trials and then large-scale data from Israel. Those showed that harms were possible, but were rare enough that the protective benefits far outweighed the risks.

We also have large clinical trials comparing aspirin with placebos, and they show enough risk to tilt the balance against daily use for healthy people who have not previously had a heart attack or stroke.

One such trial, called ARRIVE, followed more than 12,000 people from seven countries over more than six years. Another called ASPREE followed nearly 20,000 people in the U.S. and Australia. Both reached similar conclusions: There was no measurable benefit and some measurable risk of dangerous bleeding including brain hemorrhage and gastrointestinal hemorrhage. And more people died in the aspirin section of the APSREE trial than the placebo part over the study period.

I asked Gilbert Welch about some questionable practices I'd experienced with my own doctor. I doubt they had any record of my getting a COVID-19 shot last spring at a vaccination center, but they never called to ask about this important preventive measure. Instead I got several calls about getting a checkup, with the threat that they might not be able to give me the referrals I needed for my insurance to pay if I got sick or hurt.

So they had me there. I did it, and I walked out with a prescription for a mammogram, though I have no symptoms, risk factors or family history of breast cancer. Annual mammograms remain embedded in standard practice despite growing evidence they can do more harm than good in many women.

Welch said cancer screening is lucrative to medical centers, and primary-care doctors often get ratings or reviews from their medical system that hinge on coaxing patients to submit to these screening tests. It may be more in the physician's interest than in mine.

Another excellent book on what's wrong with medicine is "Overkill" by Paul Offit. It has an eye-opening chapter on why you might want to think twice about your physician's advice to take vitamins. Offit has long been a leading advocate for vaccines — so he's not against treating the healthy as long as the treatment has been through the rigors of scientific testing. The vitamin chapter, some of which ran as a New York Times piece, cites several studies where the group that took vitamins got cancer more often than a placebo group.

Medicine is not exactly a science — it's a combination of things- with some science thrown in the mix. There's a reason that we have a science-based medicine movement but nobody needs a science-based physics movement. Medicine is also based on tradition, intuition, compassion and various other ingredients. That's all useful for healing the sick and injured.

Sometimes you can't wait for more than circumstantial evidence when you're dealing with a deadly disease, such as AIDS in its early days. Then it might be best to throw untested treatments at the problem if there's a hypothetical possibility they would work and no hope otherwise. But maybe instead of always thinking better safe than sorry, doctors and their patients could consider the adage that if it isn't broken, don't fix it.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners. Faye Flam is a Bloomberg Opinion columnist and host of the podcast "Follow the Science." She has written for the Economist, the New York Times, the Washington Post, Psychology Today, Science and other publications.