Should I get a full-body MRI scan each year?

I guess its time for me to weigh in on this whole-body MRI debate. While it may be new to the mind of the general public, the debate around screening using medical imaging for various use cases has raged in academic and policy circles for decades now. And, when I say raged, I mean like throwing chairs in meetings kind of rage because there are really strong arguments on both sides of the debates for the who, what, when, where, and why components of screening. What age should colonoscopies be recommended by the gastroenterology society and how often? What age should routine mammography be recommended? While these questions may sound trivial, their answers are fiercely contested by physicians, researchers, and public health leaders. This is because the recommendations often have bearing on insurance coverage and therefore access to care for millions of people. In some ways, the fierce disagreements that spur each side to do more research is a shining example of how research and the decisions that use it get better over time. But it is also an example of the bounds of empirical research because, at some point, with challenging questions, the decision becomes one of judgement, ethics, and cost.

While the breast cancer and colon cancer screening recommendations are one thing, the discourse around the full-body MRI scan for asymptomatic people is a long-running debate among radiologists, other physicians, epidemiologists, and public health professionals. And, again, the arguments are strong on both sides and the population-level evidence is weak.

Before we get into the MRI situation, lets define screening. The goal of screening is essentially to catch disease earlier than it might show up otherwise through the use of a screening mechanism given to a certain population of individuals. A nationally recommended screening program typically needs to have a good method of screening that is reliable and accurate, an important enough condition to catch, and evidence that screening will modify outcomes such as survival time or quality of life.

Medical imaging such as with a mammogram is often used in screening programs. We (again, the royal “we”) use mammogram-based screening to try to catch breast cancer early.

In recent years, the use of whole-body MRI scans as screening for the general public who are asymptomatic (no initial suspicion of a specific disease) has been popularized by companies offering the service, though the idea and service has been around for much longer. You have probably seen some social media posts or articles about the topic if you are a reader of Health Tech Happy Hour.

Whole-body MRI (WB-MRI) captures structural images of nearly the entire body, typically at least the head, neck, chest, abdomen, and pelvis, without ionizing radiation and often without contrast agents. Advances in sliding-table MRI platforms, multi-channel coils, parallel imaging, and whole-body diffusion-weighted imaging have compressed acquisition times to under an hour and improved the technology’s ability to flag malignancy. I am sure AI image reading will further enhance the utility of imaging for years to come.

For most of its history, radiologists and other physicians reserved the MRI for well-defined clinical problems, and the evidence base grew up around those uses.

Medicine deploys WB-MRI in targeted, evidence-backed ways. Current guidelines endorse it for early cancer detection in patients who carry strong genetic risk, with the strongest support in Li-Fraumeni syndrome, and for staging established malignancies such as multiple myeloma and metastatic prostate cancer. Dedicated reporting frameworks like MY-RADS and MET-RADS-P (we love acronyms!) standardize how radiologists document these surveillance scans. In each of these cases, the patient carries a meaningful pretest probability of disease (i.e., based on history, symptoms, lab tests, etc.), and the scan answers a specific clinical question.

That alignment between test, disease, and population is the foundation of “responsible screening”.

Consumer full-body MRI flips the public health approach to screening. Instead of scanning people who are likely to have disease, companies now market the scan to asymptomatic, average-risk adults who feel fine. The pitch is reassurance and proactive management of ones health risks. As consumers have become more attuned to health measurement with wearables and home laboratory tests, the consumer full-body MRI has emerged. One scan, many organs and body systems, hundreds of possible conditions surfaced before symptoms ever appear.

So, lets take an open-minded view of both sides of the debate and see what happens.

Prenuvo is the eight-hundred pound gorilla of full-body MRI scans. The company charges roughly $2,499 for a standard whole-body scan and layers on tiered annual memberships that run from about $1,199 for a focused scan and lab panel to nearly $5,000 for an executive package that adds brain and body-composition analysis.

Prenuvo operates roughly 21 U.S. clinic locations, and a full-body scan takes about an hour. Celebrity endorsements have hit the market hard. Kim Kardashian famously described the scan as “life-saving” in a 2023 social media post that triggered a measurable surge in public discussion, with one analysis finding that 38% of all Reddit posts about WB-MRI appeared in that single month.

Ezra, another company offering the scans, took a different approach. After Function Health acquired it in 2025, the combined company launched a 22-minute, AI-accelerated full-body scan for $499, bundled with a $365 annual lab-testing membership and available at roughly 100 locations nationwide.

The FDA cleared Ezra’s Flash AI model, which enhances image quality at faster scan speeds and, the company argues, makes annual imaging economically plausible for the first time. Standalone Ezra scans still start near $999, and higher tiers that add spine and skeletal imaging climb toward $3,999. Traditional radiology practices have entered the space as well. SimonMed promotes a longevity-oriented full-body scan near $899. It is not a stretch to consider that any idle time on an MRI machine may be filled with these scans.

Notably, none of these providers bill insurance, so patients pay entirely out of pocket. This is a cash-pay, luxury good.

First, early detection makes good logical sense to the average person. Catching a tumor or aneurysm before it sends you to the emergency department itself feels self-evidently good. It is really hard to talk people out of this one. Logically, why wouldn’t I want to know if I have a concerning issue? The arguments that use data and population-level research to dissuade the consumer are fighting an up-hill battle. On a basic level, full-body screening absolutely makes sense.

In one analysis of 118 asymptomatic subjects, WB-MRI detected two malignancies, a renal and an adrenal carcinoma, plus a precancerous pancreatic lesion, and roughly 5% of the group went on to surgery or medical treatment.

Second, many people feel disillusioned with primary care. Rushed visits, long waits, and a “wait and see” posture toward vague symptoms leave patients frustrated. A full-body MRI feels like a very proactive approach to health.

Because primary care ties testing to prevalence and deliberately avoids over-testing low-risk complaints, subtle symptoms like fatigue or bloating often get dismissed until they become severe or stick around for too long. Reddit users, according to a peer-reviewed analysis of posts, who pursued scans frequently cited prior negative experiences of delayed or inconclusive evaluations that pushed them toward self-referral and a sense of taking control of ones health.

Third, the excitement of advanced technology is something people are conditioned to like from popular culture and media. Americans also have a particular cultural predisposition for technological gains. AI-augmented scans and the language of “deep phenotyping” tap into a longevity and health-conscious culture that views personal data (e.g., from wearables) as both empowerment and clinically effective (even if it is not).

It is easy to think that if I wear my Apple Watch and collect hundreds of thousands of data points per year and get a full-body scan that I am carefully monitoring my health and that someone, somewhere might be able to use that data to help me if needed.

Fourth, the promise that a scan could save your life is emotionally powerful and heavily amplified on social media. Social media content about medical tests including WB-MRI skews overwhelmingly promotional, with limited mention of potential harms, which further inflates expectations. All it takes is a story of one person catching something that was asymptomatic, getting treatment, and saying that the ~$3,000 was well worth it to get others on board with the value of the scan.

Clinical risk assessment based on history and physical exam is imperfect, and directly scanning multiple organ systems could, in some circumstances, surface disease earlier than clinical suspicion alone would allow. This is essentially a “more data is more better” argument. It also meshes well with the logical and emotional views of the scans that drive consumer demand.

Early studies suggest high patient satisfaction from the scans, generally above 80% to 90%, and relatively low rates of psychological distress while awaiting results, in the range of 5% to 10%. At population scale, research cohorts such as the Study of Health in Pomerania and the UK Biobank have already generated genuine scientific value, linking imaging phenotypes to disease and enriching normal reference ranges in ways that pure clinical imaging cannot.

Patients themselves often describe positive experiences. Among Reddit users who actually underwent a scan, 85% spoke of it favorably, citing both clinically significant findings and the emotional reassurance of a clean result. For people living with health anxiety or nonspecific symptoms, a normal scan can deliver real comfort, and supporters argue that a single comprehensive exam may cost less than a string of separate targeted studies.

And, most importantly, for people with a real, lurking threat to their health, the scan may find it. The chance of doing so against ones life has indeterminant value.

The criticisms are substantial, and they explain why public health has not embraced the practice.

The core problem is based in Bayesian statistics (e.g., probability!). Screening works when a test is matched to a population with a reasonable probability of disease, whereas consumer WB-MRI applies a “shotgun approach” to people with very low average probability of a weird finding. Even a test with 95% sensitivity and specificity (which are two concepts that plague epidemiology students), applied to a condition with 0.5% prevalence, yields a predictive summary index close to zero. Therefore, most positives in this hypothetical circumstance will be false.

A systematic review of 12 studies covering 5,373 asymptomatic subjects found that 15% to 30% harbored at least one incidental abnormality (e.g., something weird on the scan), and the pooled prevalence of critical plus indeterminate findings reached 32.1% (e.g., 32.1% of scans produced something weird).

Among the “weird” findings that were checked, a portion proved false-positive, with a pooled proportion of about 16%. However, since only 12.6% of critical and indeterminate findings were ever verified through follow-up, resection, or additional testing, and no included study tracked long-term outcomes for negative exams; that number is likely much higher. The authors of the review conclude that providers should not offer WB-MRI for preventive screening outside a research setting. The field still lacks the outcomes data needed to show that scanning healthy people reduces morbidity or mortality, and cost-effectiveness remains unestablished.

For the more epidemiology inclined, lead-time and length biases can make earlier detection look like survival benefit even when it changes nothing about a patient’s outcome. Essentially, finding something sooner does not necessarily mean that the outcome is improved compared to when the disease would have been found without the scan.

Interpretation of the scans also demands subspecialty expertise that many settings lack. Turns out, that not all radiologists are created equal and they too have subspecialties (e.g., neuroradiology). One lawsuit, for example, followed a commercial scan that reportedly missed a 60% arterial stenosis in a patient who later suffered a stroke in the same territory. AI reading systems that are clinically validated against gold-standard care may help with this issue in the near future.

High out-of-pocket cash-pay pricing skews access toward the affluent and could divert scarce scanners and radiologists away from patients with medical indications. This is the public health practice argument where resource allocation needs to be taken into account. Long wait times for the use of scanners among patients with high probabilities of an actual disease may harm them. Although, I am sure that General Electric and Siemens would be happy to produce more MRI machines, so in theory the market could correct here if there is demand.

Professionals largely agree. In the social media analysis paper, 47% of posts opposed WB-MRI while only 25% favored it, and among radiologists who took a clear position, 97% opposed the practice.

The American College of Radiology does not recommend it for asymptomatic, average-risk individuals. They conclude their statement on the matter with, “The ACR will continue to monitor scientific studies concerning the utility of screening total body MRI.”

Full-body MRI screening is not going away. Patients who can pay and who see the value will not be dissuaded by scientific or epidemiological platitudes. The technology works, the scans keep getting faster and cheaper, and patient enthusiasm shows no sign of fading. Again, to an individual, the logic is sound.

Yet the evidence that scanning healthy people improves health remains limited, while the potential for harm through false positives and overtreatment is well documented. Such is the complex world of screening decisions and recommendations.

Only a large-scale randomized trial with long-term follow-up can definitively settle whether the practice helps.

Until then, I expect you will be hearing about this at the Thanksgiving table this year. Hopefully, this article helps you consider the different positions.