French Fries, but Not Baked or Boiled Potatoes, Tracked With Higher Diabetes Risk

CATEGORY: FOODS & NUTRIENTS

Every three weekly servings of French fries were linked to a 20% higher risk of developing type 2 diabetes—while baked, boiled, or mashed potatoes showed no significant link—in a study that followed 205,107 US adults for more than three decades, published in The BMJ.¹ Type 2 diabetes is the common form of diabetes in which the body gradually loses its ability to keep blood sugar in a healthy range. The results suggest that for this everyday food, how the potato is prepared may matter more than the potato itself.

22,299 Diagnoses Across 5.1 Million Person-Years

Researchers followed participants in three long-running US projects—the Nurses’ Health Study, the Nurses’ Health Study II, and the Health Professionals Follow-up Study—all free of diabetes when they joined. Every four years, participants filled out detailed food questionnaires that asked, among many other things, how often they ate potatoes and in what form: French fries, or baked, boiled, and mashed. Across more than 5.1 million person-years of follow-up, 22,299 participants developed type 2 diabetes, with diagnoses confirmed through follow-up medical questionnaires. The analysis accounted for body weight, physical activity, smoking, and other features of each person’s diet.

Whole Grains Instead of Fries, 19% Lower Rate

The overall link between potatoes and diabetes was modest—a 5% higher rate for every three weekly servings of any kind of potato—but it was driven almost entirely by fries. Each three weekly servings of French fries was associated with a 20% higher rate of type 2 diabetes, while baked, boiled, or mashed potatoes showed essentially no association. The comparison food mattered too. In substitution models, having whole grains in place of three weekly servings of fries was associated with a 19% lower diabetes rate, while whole grains in place of potatoes overall was associated with an 8% lower rate. Replacing potatoes with white rice, by contrast, was associated with a higher rate. A companion meta-analysis—a study pooling results from many studies—of 13 cohorts with 587,081 people found the same pattern, with fried potatoes linked to a 16% higher rate.¹ Those 13 cohorts covered 587,081 people and 43,471 diagnoses of type 2 diabetes.¹

Deep Frying and Fast-Food Meals May Explain It

This was an observational study, so it shows an association, not proof that fries cause diabetes—people who eat fries often may differ in other ways that questionnaires cannot fully capture. Still, the authors note plausible explanations: fries are deep-fried and more calorie-dense than boiled or baked potatoes, and they often travel with fast-food meals. The white-rice result is a reminder that what a food replaces shapes the picture.

Fried Potatoes Over Once Weekly, 12% Higher Risk

A pooled analysis of seven US studies following 105,531 adults found no link between potatoes overall and type 2 diabetes, but about 12% higher risk among people who ate fried potatoes more than once a week.² A 2025 systematic review pooling 51 observational studies also linked fried potatoes to 16% higher diabetes risk—though it tied non-fried potatoes to a small 5% increase, so the contrast between fries and other potatoes may be less absolute.³

Key Takeaways

• In 205,107 US adults followed for more than 30 years, every three weekly servings of French fries was linked to a 20% higher risk of type 2 diabetes; baked, boiled, or mashed potatoes showed no significant association.¹
• In substitution models, whole grains in place of three weekly servings of fries were associated with a 19% lower diabetes rate—while white rice in place of potatoes was associated with a higher rate.
• The study is observational—association, not causation—but a pooled analysis of 13 cohorts covering 587,081 people showed the same pattern for fried potatoes.

References

1. Mousavi SM, Gu X, Imamura F, et al. Total and specific potato intake and risk of type 2 diabetes: results from three US cohort studies and a substitution meta-analysis of prospective cohorts. BMJ. 2025;390:e082121. link
2. Djousse L, Zhou X, Lim J, et al. Potato consumption and risk of type 2 diabetes mellitus: a harmonized analysis of 7 prospective cohorts. The Journal of Nutrition. 2024;154(10):3079–3087. link.
3. Su Y, Liu X, Jiang B, et al. Potato intake and the risk of overweight/obesity, hypertension, diabetes, and cardiovascular disease: a systematic review and meta-analysis of observational studies. Nutrition Reviews. 2025;83(3):466–478. link.