Young men from Kilimanjaro villages swapped their home cooking for fried chicken, chips and white rice for two weeks. A blood test then showed their plasma could make about a fifth more of the main clotting enzyme

In the villages on the slopes of Kilimanjaro, the heritage diet is built on green vegetables, beans, plantains, cassava and taro, and whole grains such as millet and sorghum. Down in town, many young men eat what researchers call a Western-style diet. Fried chicken, chips, white rice, macaroni, pancakes, refined maize meal.

A team from KCMC University in Moshi and Radboud University Medical Center in the Netherlands ran an experiment around that divide. It recruited 77 healthy men aged 20 to 40. Men from the villages who normally ate the heritage diet were switched to Western-style meals for two weeks. Men from town were switched the other way. A third group kept their usual town diet but drank about a litre a day of Mbege, a fermented drink made from bananas and finger millet, for one week. Meals in the switch groups were cooked centrally and eaten under supervision, three times a day.

The main results came out in Nature Medicine in April 2025. The switch to Western food pushed the men's blood towards a more inflammatory state; the switch to the heritage diet and the banana drink mostly did the opposite. That was the main finding.

The new paper, published online by the American Heart Association journal Arteriosclerosis, Thrombosis, and Vascular Biology on 8 October 2026, looks at the same stored plasma for something else: clotting. Blood clots by producing an enzyme called thrombin, and lab tests can measure how much thrombin a plasma sample is capable of making once clotting is set off. That total is called the endogenous thrombin potential. In research it is read as a measure of how primed the blood is to clot.

In the men who went from village food to Western food, that potential rose. Compared with the change in the small group of men who stayed on their usual diet, it was about 19% higher after the two weeks, when clotting was triggered at the stronger of the two settings the lab used. The statistical margin is thin: the lower end of the confidence interval sits at 1%, and the authors report the p value of 0.037 as nominal, meaning not corrected for the many other things they measured.

Prothrombin, the raw material thrombin is made from, also went up inside that group, from 98.0% to 104.5% of the reference level. That rise survived the correction for multiple tests. But when it was set against the men who did not change their diet, it was no longer clearly different. Something about those two weeks may have nudged it, or it may have drifted on its own.

Going the other way, from Western to heritage food, did little that was consistent.

The banana drink produced the oddest picture. Men drinking Mbege had lower levels of several clotting proteins afterwards, including fibrinogen and factor VIII, and also of antithrombin, one of the body's own brakes on clotting. Yet their thrombin potential at the weaker trigger went up. Those changes pull in different directions, and with no control group in that arm, there is no way to say how much of it was the drink.

The authors are frank about the limits, and they are real. Clotting was a secondary outcome with no ranking of which result mattered most, so they call the whole analysis exploratory. There were only five control men in each switch group. Food was not measured out, so the men ate as much as they wanted. In the original paper, the group switched to Western food put on a median of 2.6 kg in two weeks. A change in clotting could come from what they ate, how much they ate, or the weight itself; this design cannot pull those apart.

And thrombin potential in a test tube is not a clot in a vein. No one in this trial had a thrombosis, and two weeks says nothing about what years of a diet would do. We could read only the summary of the new paper, not the full text, so we have not seen the full list of clotting measures or the size of the other changes.

What the study does add is a mechanism worth a proper test. Diet is usually linked to heart attacks and strokes through cholesterol, blood pressure and weight, all of which take months or years to shift. Here is a hint that blood's readiness to clot may respond within a fortnight. If a larger trial with matched calories finds the same, it would be one more reason the speed of the move from traditional to fast food across much of the world matters.