Every Extra 100 Grams of Ultra-Processed Food Daily Tied to Higher Disease Risk in an 8.8 Million-Person Study

A pooled analysis of 51 prospective cohort studies covering 8,819,894 adults across the Americas, Europe, Asia and Oceania has found that higher intake of ultra-processed foods is associated with a meaningfully greater risk of cardiovascular disease, cancer, diabetes, digestive disease, depression, anxiety and death from any cause. The analysis, published in Family Medicine and Community Health on 22 September, a BMJ Group journal, is summarised in Medical News Today's coverage.

The scale of the analysis

The researchers, whose work was supported by the Natural Science Foundation of China and a Jiangxi provincial fund, searched the published literature for cohort studies that tracked ultra-processed food intake against later health outcomes, ultimately combining 51 studies with follow-up periods ranging from two to 32 years. Participants' average ages at enrolment ranged from 23 to 77, and their diets were assessed using food frequency questionnaires, 24-hour dietary recall, or full dietary history, all standardised to a common measure of ultra-processed food intake in 50-gram servings so that results across very different studies could be compared directly.

What counted as a meaningfully higher risk

Comparing people with the highest reported ultra-processed food intake against those with the lowest, the analysis found a 24 per cent higher risk of cardiovascular events, a 12 per cent higher risk of cancer, a 23 per cent higher risk of obesity or being overweight, a 24 per cent higher risk of metabolic syndrome or diabetes, a 27 per cent higher risk of depression or anxiety, a 31 per cent higher risk of digestive disease, an 11 per cent higher risk of high blood pressure, and an 18 per cent higher risk of death from any cause. Expressed per additional 100 grams of ultra-processed food eaten each day, the increases were smaller but still consistent: 14 per cent higher cardiovascular risk, 4 per cent higher cancer risk, 3 per cent higher risk of death, and 2 per cent higher risk of metabolic syndrome or diabetes.

What ultra-processed actually covers

The category is broader than the sugary drinks and packaged snacks most people picture first, and typically includes reformulated breads, flavoured yoghurts, reconstituted meat products, instant noodles and ready meals built substantially from industrially extracted ingredients and additives rather than whole foods, following the classification system most cohort studies in this analysis relied on. That breadth is part of why the researchers report associations across such a wide range of conditions, from heart disease to mood disorders, rather than a single narrow outcome.

What the researchers are careful to flag

The authors acknowledge real limitations: dietary intake in nearly all of the underlying studies relied on self-report, which is an imprecise instrument at the best of times, the analysis excluded anyone under 18, and it remains genuinely difficult to fully separate a diet high in ultra-processed food from the broader lifestyle patterns, income, activity levels, sleep, that often travel alongside it. An association this consistent across 51 independently run studies and nearly nine million people is nonetheless a stronger signal than any single cohort could provide on its own, even if it stops short of proving that the processing itself, rather than something correlated with it, is the direct cause.

What it does not tell you to do differently tomorrow

The analysis does not identify a safe threshold below which ultra-processed food carries no added risk, nor does it distinguish between an occasional convenience meal and a diet built primarily around processed products, a nuance that per-100-gram risk figures inevitably flatten. What it adds to a large existing body of research is scale and consistency, a dose-response pattern across multiple health outcomes drawn from cohorts followed for as long as three decades, which is the kind of evidence public health guidance eventually has to reckon with even as more mechanistic research works out exactly why processing itself appears to matter.