Many's the time we have pointed out the problem with putting too much faith in epidemiological associations — that is, thinking that an association between two factors in people's lives means that one causes the other. Take beta-carotene, for example. Epidemiological associations suggested that beta-carotene, a precursor to vitamin A, might protect people from lung cancer. But after the surge in sales of beta-carotene supplements had taken off, the results of a randomized, placebo-controlled trial suggested just the opposite — that beta-carotene supplementation seemed to actually increase the risk of lung cancer. Oops.
The problem here wasn't that the earlier studies weren't done correctly, it was that they were taken to mean a causal link between beta-carotene and cancer before the appropriate placebo-controlled trial was performed. So there's nothing wrong with looking at associations — they point the way to more definitive studies. Yet in spite of this, the scientific literature is rife with overly enthusiastic interpretations of epidemiological associations — and this latest one on potatoes and high blood pressure is a case in point.
A group of researchers, led by Dr. Lea Borgi from Brigham and Women’s Hospital in Boston, used data from three large prospective cohort studies — the Nurses' Health Study I and II, and the Health Professionals Followup Study. In all, data from over 180,000 individuals were evaluated. The main outcome measure was new cases of hypertension (HTN) that had been diagnosed by an individual's health care provider.
The participants' intake of white potatoes — baked, mashed, boiled, French fried, or as potato chips — was quantified from periodic questionnaires. The results indicated that compared to people who reported eating less than one serving per months, those who ate four or more servings per week had an increased risk of developing HTN. But these increases, while statistically significant for four of the five preparation methods, were pretty small. The increases were 11 and 17 percent for baked, boiled, or mashed potatoes considered as a group, and for French fries respectively. Interestingly, there was no increased risk seen with consumption of potato chips — indeed the HTN risk was slightly lower in those (by about 3 percent) than that of the comparison group for participants in the Health Professionals Followup Study.
The researchers suggested their results might be due to the high glycemic load of potatoes — that is, that the fact that white potato consumption can raise blood sugar, which then might cause dysfunction in arterial linings, leading to increased blood pressure. But isn't this just another way of suggesting that there's a causal relationship between potatoes and HTN?
In an accompanying editorial, Dr.s M.F. Harris from the University of New South Wales and Dr. R.A. Laws from Deakin University in Melbourne, Australia, point out that, "We seem to be constantly assailed by media reports of studies and (sometimes conflicting) recommendations about the risk posed by specific dietary foods."
Without negating the importance of diet to health, the editorialists emphasize the complexity of food intake behavior, and conclude that "studies that examine associations between various dietary patterns and risk of disease provide more useful insights for both policy makers and practitioners than does a focus on individual foods or nutrients."
We couldn't agree more — and would add that those that show a greater degree of increased risk would deserve more consideration.