Somewhere in America this morning, a healthy 34-year-old man with no symptoms and no family history is lying inside an MRI machine that no physician ordered, having paid roughly $2,500 out of pocket for the privilege. Somewhere else, a 52-year-old woman is skipping a mammogram that her insurance covers in full, because she read online that the compression spreads cancer.
Both of these people believe they are taking charge of their health. Both believe they are making the evidence-based choice. One of them has purchased a test with no institutional backing, and the other has declined one that decades of data support.
I have been writing about prevention for six years now (and teaching and preaching it for far longer), and I have made peace with the fact that vaccines will always come with backlash, and that the boring stuff that works (sleep, moving our bodies, nutritious food, social connection) will never trend because there is no product attached to it. What I did not expect was a double standard opening up inside the prevention conversation itself, running on a logic that is stranger than plain anti-science sentiment. Let’s discuss…
Direct-to-consumer whole-body MRI has gone from a curiosity to a category. Prenuvo has scanned over 100,000 people at around $2,499 a pop. Function Health, which acquired the competitor Ezra, has raised hundreds of millions in venture funding. Kim Kardashian has posted about it. Established radiology practices have followed the money in. SimonMed, one of the largest outpatient imaging chains in the country, launched a longevity division in January with scans running from about $900 to $2,200. Its homepage advertises screening for more than a thousand conditions in under an hour, with “no doctor’s referral needed.” The absence of a medical reason to be there is the product.
The evidence has not kept up with the marketing. No major medical society recommends whole-body MRI screening for people at average risk. Current guidelines reserve it for patients with specific genetic syndromes (Li-Fraumeni, for example, an inherited mutation which greatly increases a person’s cancer risk), where the odds of finding aggressive cancer are high enough to justify looking. In May, two radiologists from Michigan and Wisconsin published an editorial in JAMA titled “Elective MRI Screening of the General Public: Buyer Beware,” telling consumers plainly that the harms likely outweigh the benefits.
Look hard enough at any healthy body and you will find something.
A 2020 review of whole-body MRI in asymptomatic adults found that 95% of people scanned had some abnormal finding, roughly three in ten had something that needed further investigation, and about 1% turned out to have cancer. The JAMA authors add that a meaningful share of even those cancers are either slow-growing enough that they never would have hurt you or advanced enough that finding them early changes nothing. Everything else in that 95% is a cyst, a hemangioma, or an age-related change — findings a radiologist can usually name and dismiss on sight. Radiologists have a word for these: incidentalomas. Sometimes the incidentaloma is a shadow that nobody can interpret without more tests, and that’s where the harm can show up.
A nonspecific finding can send you into a workup with contrast, radiation, sometimes a biopsy, and a waiting period that can stretch for months or years before anyone can tell you what it was. One of the JAMA authors described how that process reorganizes your sense of yourself, so that you start living as a patient with a problem even when the finding eventually proves benign.
Earlier this month Andrew Huberman posted about whole-body MRI, making the argument almost everyone in this space makes. He acknowledged the cost and the false positives up front, then set both aside on the strength of what neurosurgeon friends had told him about operating on findings that surfaced because somebody got curious and bought a scan. Neurosurgeons replied in numbers to say this is not their experience. One faculty member at UCSF told USA TODAY that he has never operated on an incidental life-saving lesion found this way, that outside of unruptured aneurysms there is very little they operate on in a truly asymptomatic patient, and that while he would not claim it never happens, he believes it is rare.
A secondhand anecdote was enough to justify an unindicated imaging study, while in the same wellness ecosystem the seven randomized trials behind current mammography guidance get treated as insufficient. The standard of proof we demand scales inversely with how much we want the conclusion, and that tendency belongs to all of us (myself very much included) unless we build in some way to catch ourselves doing it.
There is also money involved. Several of the most prominent voices promoting these scans have paid relationships with the companies selling them (Huberman included), and those relationships tend to be buried in the fine print.
I understand why the scans are popular, because the stories are extraordinary and have some recognizable faces. Maria Menounos credits a whole-body scan with catching her pancreatic cancer. Last month, Joe Amabile from The Bachelor posted that his Prenuvo scan found a lesion, that follow-up imaging showed a blueberry-sized mass that appears to be a glioma, and that he was headed to Sloan Kettering for a craniotomy. He is 40, and I hope he is fine.
Stories like his are why the scans sell, and they are hard to argue with because arguing sounds heartless. But one person’s outcome cannot tell us whether the practice works for most of us. We do not get to see what would have happened if he had never booked it, and we never meet the people who booked the same scan and ended up in surgery for something that was never going to hurt them.
The Dwyane Wade story is interesting. When he first went public, Wade said the discovery came after urinary symptoms sent him to a physician. In a later interview, he credited the full-body scan. Both accounts can be true in sequence, but the retelling drifted toward the product and away from the symptom that prompted the visit. His mass was three centimeters, a detail that rarely travels with the story, and small kidney masses under four centimeters frequently go to active surveillance rather than surgery. Wade had 40% of his right kidney removed. It was cancer, and I am glad he is well. But the case most often cited for these scans could just as easily be an example of the harm their critics describe.
At the same moment we are celebrating scans that find things nobody needed to know about, mammography has landed on the naughty list in a growing corner of the wellness world. The claims circulating are that these tests were never safety-tested, that compression spreads cancer, that the radiation causes the disease it screens for. None of that is supported.
What people turn to instead is thermography, an infrared scan that measures heat patterns on the skin surface. The FDA has warned that no scientific evidence supports it as a standalone breast cancer screening tool.
The exaggerated claims about mammography aren’t entirely baseless, though. Buried in them is a documented scientific problem: overdiagnosis. And there are honest open questions inside mammography science that researchers argue about in public:
How much DCIS (ductal carcinoma in situ, a very early, noninvasive breast abnormality) that we find and treat would never have progressed to anything harmful.
Where the starting age should sit, at what interval, and how much of that changes with individual risk.
What supplemental imaging should look like for people with dense breast tissue, since dense tissue can hide tumors on a standard mammogram.
Those questions move, sometimes quickly. In 2024 the US Preventive Services Task Force lowered the recommended starting age for biennial mammography from 50 to 40, reversing its own 2016 guidance, and concluded that the evidence on supplemental ultrasound or MRI for dense breasts is still insufficient either way. The guidelines moved toward more mammography at the same moment a slice of the internet decided mammograms were dangerous.
The claims that mammography was never safety-tested, or that it can spread or cause cancer, are different kinds of claims than the questions above, and collapsing them together does a disservice to the people asking those questions in good faith. The overdiagnosis critique is also, by a wide margin, the strongest argument against whole-body MRI in average-risk adults. The wellness internet picked up the overdiagnosis critique and aimed it at the screening tests with the best evidence base while exempting the one it wanted to buy, and that inversion is the part I cannot get my head around.
We already ran this experiment. South Korea saw ultrasound-based thyroid cancer screening spread widely, and diagnoses rose fifteen-fold between 1993 and 2011 while mortality stayed flat. Yet tens of thousands of people developed complications from the resulting surgeries. That is what happens when you screen people who are unlikely to have the disease in the first place, and it is the best real-world case study we have for why thresholds exist at all.
I have no evidence that the people buying whole-body MRIs are the same people refusing mammograms, and I doubt the overlap is large in any individual. These are two currents running through the same wellness space, and I spent a while thinking of them as incoherence, the funhouse mirror version where reality is just warped. They are actually consistent with each other, once you stop assuming the disagreement is about screening.
The axis being sorted on is authority. A guideline is somebody else deciding where the benefit stops exceeding the harm for a population you happen to belong to. A whole-body MRI is something you buy with your own money, on your own timeline, because you wanted to know. If the operating principle is that no institution gets to set the terms of your relationship with your own body, then buying the unindicated scan and refusing the recommended colonoscopy are the same move pointed in two directions. They only look contradictory if you assume evidence is the variable. The sunscreen claims run on the same template, where the sun is natural and the SPF is the intervention, so the intervention becomes the suspect.
Underneath all of it, prevention’s success is a non-event. You cannot feel the cancer you did not get, or the outbreak that did not happen. The intervention is the only tangible thing in the sequence, so it absorbs all the causal attribution, including for outcomes it had nothing to do with. It is also why sleep and fiber and friendship will never trend. Their payoff is a subtraction from a counterfactual, and there is no way to post a photo of that.
Wellness skepticism did not start in 2020. Goop, raw milk, the vaccine fights of the 2000s, the entire alternative medicine economy, all of it predates the pandemic by decades. COVID did not create this dynamic, but I suspect it accelerated it, and it put a much larger population in the room. Multiple generations confronted their own mortality at the same time, and something like that does not resolve when an emergency declaration ends.
Psychologists have a framework for this called the terror management health model, which holds that thoughts of death push health behavior in different directions depending on how they surface. When mortality is consciously in front of you and you feel able to act on it, you act. When death thoughts run underneath awareness, decisions get driven instead by what the behavior says about you, which is how the same literature ends up explaining sun tanning.
Underneath is a fear of dying, and deciding for themselves is how people keep that fear quiet. That is where the authority instinct comes from. Seen that way, the two people I opened with stop looking like opposites: neither one is really trying to find anything out. He is buying reassurance, and she has found a reason not to look, and both are ways of handling the same fear. That also changes where my sympathy lands. I had been more forgiving of the guy in the Prenuvo tube, since his choice mostly costs him money and hers carries real risk, but if both are fear management, that asymmetry is harder to defend.
The ick is the same one I get about concierge medicine, where paying more moves you to the front of the line for care everyone needs and most people wait for. Whole-body MRI is that instinct extended into prevention itself. We now have a two-tier prevention economy. The tier with money buys imaging that no guideline supports. The tier without it struggles to access the screening that decades of data do support.
A radiologist at Johns Hopkins warned that popularizing these scans without medical indications diverts scarce capacity from a system that already lacks it. Every hour of magnet time spent reading a healthy 34-year-old’s incidentalomas is an hour not spent on someone with a diagnosis. And the people paying for that hour are, as the UCSF neurosurgeon pointed out, already the people looking after their health. They are the worried well, with insurance and a primary care doctor and the disposable income for an elective MRI. They are already the group least likely to have anything seriously wrong, which makes them the least likely to benefit from finding something and the most likely to be harmed by the cascade that follows. Meanwhile, the population-level prevention infrastructure that runs on public funding and reaches people who cannot pay for imaging like this, is being cut.
None of which is an argument that more prevention is always better, because prevention is not one thing. Whether a test is worth doing comes down to two questions: how strong the evidence behind it is, and how likely you are to have the thing it looks for. Colonoscopy at 45 clears both. A whole-body MRI for a healthy 34-year-old clears neither. That is a judgment only a clinician who knows your own history can make for you, which is why the ACS screening guidelines are a better place to start than anything with a checkout page.
The people who would benefit most from prevention are the ones who cannot get a colonoscopy scheduled within six months, who are choosing between the mammogram copay and groceries, who have not seen a primary care physician in four years. No amount of luxury imaging touches them. Fixing access does.
Stay Curious,
Unbiased Science
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.