Men who got more of their daily calories from ultra-processed foods reported prostate cancer about 29% more often than men who ate the least, according to a study of 17,024 US men published August 7 in The American Journal of Medicine. Researchers at Florida Atlantic University’s Charles E. Schmidt College of Medicine pulled the numbers from the National Health and Nutrition Examination Survey, or NHANES, covering 2003 through 2023.
The catch sits in the design. NHANES asked men what they ate and whether they had ever been told they had prostate cancer, and the cancer count came from self-report rather than medical records.
Key takeaways
- Men in the top three quarters of ultra-processed food intake had about 29% higher odds of prostate cancer than men in the bottom quarter. There is about a 3 in 100 chance a gap that size showed up by luck alone (p = 0.031).
- After the researchers accounted for age, smoking, race and ethnicity, and poverty status, the increases were about 27%, 31%, and 34% depending on which intake groups were compared.
- The heaviest consumers alone showed the largest jump, 34%, but that one did not clear the bar for statistical significance.
What the study found
The team used two 24-hour dietary recalls and the NOVA classification system to work out what share of each man’s calories came from ultra-processed foods: soft drinks, packaged cereals and snacks, processed meats, and similar products. They then split the men into four equal-sized groups by that share.
Compared with the lowest group, men in the combined second, third, and fourth quarters had about a 29% higher risk (p = 0.031). Men in the combined top two quarters had about a 30% higher risk, with roughly a 3.5 in 100 chance of being coincidence (p = 0.035). Men in the top quarter on their own showed about a 31% higher risk, but that estimate could plausibly have been chance (p = 0.079).
After adjustment, the same three comparisons came out at about 27% (p = 0.045), 31% (p = 0.037), and 34% (p = 0.072). The authors attribute the weaker showing in the top group alone to the smaller number of prostate cancer cases there.
One methods choice deserves attention. The authors used 90% confidence intervals to test for significance “as harm was prespecified.” That is looser than the 95% intervals most medical journals expect, and it makes a borderline result easier to call real.
Dr. Kumar’s take
The headline treats this as a story about diet causing cancer. The data cannot carry that.
NHANES is a snapshot. A man’s two days of diet recall and his answer about a cancer diagnosis come from the same visit. Nothing here establishes that the ultra-processed eating came before the tumor.
Then there is detection. Prostate cancer is found largely because someone ordered a PSA test. Men who see a doctor regularly and engage with preventive care get screened more, and they also tend to eat less ultra-processed food. Adjusting for poverty status catches some of that, but it does not separate screening intensity from cancer incidence. Part of this signal may be who got tested rather than who got sick.
Note also that the university’s news release described the adjusted increases as ranging from 24% to 31%, while the published abstract lists 27%, 31%, and 34%. Reporters have been quoting the release.
None of this makes ultra-processed food harmless. The pattern shows up across too many outcomes to dismiss, including weaker attention and higher dementia risk and a 47% higher risk of heart disease. This particular study is a reason to fund a proper prospective study, not a reason to say packaged snacks give men prostate cancer. The senior author says the same thing, calling for “large-scale observational studies and randomized trials to adequately test the hypothesis.”
What it means for you
Do not change your prostate cancer screening plan because of this paper. Screening decisions rest on age, family history, race, and your own preferences about an ambiguous result, and none of that moved this week.
The reasonable move is the one that was already reasonable. Swapping packaged snacks and soft drinks for food you recognize is a good bet on general grounds. Whether it changes your prostate risk is an open question.