Potatoes and Diabetes: It’s Complicated

Does the link between white potatoes and diabetes extend to non-fried potatoes without butter or sour cream?

The trouble for white potatoes began in 2006, when the Harvard Nurses’ Health Study, which had followed the diets and diseases of tens of thousands of women for 20 years, found that greater potato intake was associated with a greater likelihood of getting type 2 diabetes. However, of the hundred or so pounds of potatoes Americans eat every year, most are in the deep-fried forms of potato chips, french fries, or other processed products. What happened when they looked specifically at mashed or baked potatoes? They found the same link with diabetes. Okay, but what might potato eaters eat more of? Maybe I should rephrase that: What might meat-and-potatoes people eat more of? Indeed, people who ate more potatoes ate more meat, and we know that animal protein may be associated with increased diabetes risk. But the researchers tried to statistically adjust for that and still found increased risk with potatoes.

Well, what do people put on baked and mashed potatoes? Butter and sour cream. Again, the researchers tried to adjust for other dietary factors like these as well as effectively looking at the ratio between plant and animal fats and whether potato eaters drank more soda or maybe skimped on other vegetables. Yet, still, there seemed to be this association between potatoes and diabetes.

Okay, but that was just one study. By 2015, Harvard researchers had also looked into other cohorts, including the all-male Health Professionals Follow-up Study to complement the all-female Nurses’ studies, and they continued to find a small increased diabetes risk associated with baked, boiled, or mashed potatoes, though french fries do indeed appear nearly five times worse. The authors concluded that potatoes are considered to be a healthy vegetable in dietary guidelines, but the current evidence “casts serious doubts on this classification.” Walter Willett, the chair of Harvard’s nutrition department at the time, went a step further, suggesting potatoes should be siloed up there with candy, as you can see below and at 2:18 in my video Do Potatoes Increase the Risk of Diabetes?.

A meta-analysis of potato consumption and risk of type 2 diabetes published in 2018 combined all six of the prospective studies that had been done to date, and the researchers found about a 20% increase in diabetes risk associated with each serving of potatoes a day, concluding “[l]ong-term high consumption of potato…may be strongly associated with increased risk of diabetes.” But, again, the great majority of the potatoes consumed were fried, and we know deep-fried foods contain all sorts of nasty things, like advanced glycation end-products. The researchers weren’t able to assess french fries versus non-fried potatoes. Even just three servings of fries a week is associated with nearly 20% greater risk of type 2 diabetes, whereas there was only a tiny associated risk with potatoes in general, and that included the fries mixed in.

The world’s largest manufacturer of frozen french fries took issue with this conclusion. Claiming to make one in three fries eaten on planet Earth to the tune of billions of dollars, the company has the money to fund reviews to cast doubt on the science. One review said that the scientific literature should be read with caution because the impact of potatoes on disease risk factors may depend on the foods they’re grouped with as part of a dietary pattern. Indeed, they do have an actual point. Observational studies can never prove cause and effect, and maybe potato consumption—even baked potato consumption—may just be a marker for an unhealthy diet in general. As much as researchers try to adjust for these other factors, as the journal of the Potato Association of America is quick to remind us, it’s not possible to separate the effects of potatoes and fries from the effects of the overall crappy Standard American Diet.

Is there a country where potato consumption is associated with a healthy diet? If potato consumption was still associated with diabetes there, then that would be concerning. Enter a seventh study, but this time out of Iran, where most potato consumption is of boiled potatoes. In fact, those who ate potatoes had the healthiest diets and ate the most whole plant foods—fruits, vegetables, legumes, and whole grains. And though the researchers tried to tease out those other dietary factors, those eating the most boiled potatoes had only half the odds of developing diabetes. This supports the notion that it may be hard to completely separate out just the potatoes. The bottom line, this systematic review concluded, is that we really don’t have “convincing evidence” that the intake of potatoes in general is linked to type 2 diabetes, but we should still probably hold the fries.

Doctor’s Note

This is the first in a five-part series on potatoes. Stay tuned for:

Interested in a sampling of diabetes videos? Check out the related posts below.