Summary: In a crossover trial in type 2 diabetes where body weight was deliberately kept stable and diets matched for macronutrients and glycaemic load, a Palaeolithic diet and a standard diabetes diet produced no difference in HbA1c or fructosamine, suggesting food-group differences alone do not improve glycaemic control without weight loss.
PICO Summary
| Element | Detail |
|---|---|
| Population | 14 adults with type 2 diabetes and increased waist circumference; random-order crossover, two 4-week periods with 6-week washout. |
| Intervention | Palaeolithic diet excluding cereal grains, dairy, and legumes (fruit, vegetables, tubers, fish, shellfish, lean meat, nuts, eggs, olive oil); energy adjusted to keep weight stable. |
| Comparison | Diabetes diet including wholegrains, low-fat dairy, and legumes; matched for macronutrients and glycaemic load. |
| Outcome | No difference between diets in HbA1c or fructosamine. Weight was kept stable and the diets were successfully matched for macronutrient composition and glycaemic load, though not for fibre. The authors conclude characteristic food-group differences do not, by themselves, affect glycaemic control. |
Palaeolithic vs diabetes diet, weight-stable
RCT crossover · type 2 diabetes · 4 weeks
With weight held stable and diets matched for macronutrients and glycaemic load, the Palaeolithic and diabetes diets gave no difference in HbA1c or fructosamine. Food-group differences alone do not improve glycaemic control.
Expert Commentary
This is a clever mechanistic trial whose result is genuinely clarifying, and it must be reported as the negative finding it is rather than the positive one implied before. By holding body weight stable and matching the two diets for macronutrients and glycaemic load, the investigators isolated the question that usually gets confounded: do the characteristic food-group differences of a Palaeolithic versus a diabetes diet improve glycaemic control independent of weight loss? The answer was no, HbA1c and fructosamine did not differ between the diets. The honest interpretation is that much of the glycaemic benefit attributed to Palaeolithic-style eating in other studies probably operates through weight loss and the accompanying reduction in energy and carbohydrate, not through some special property of excluding grains, dairy, and legumes. The obvious limitations are the tiny sample of fourteen and the short four-week periods, and the diets were not matched for fibre. Can I use this with my patients? Yes, to temper dietary dogma. It supports telling patients that the key levers for glycaemic control are energy balance, weight, and overall carbohydrate quality, not adherence to a particular named diet, which lets me favour sustainable, individualised eating over restrictive food-group elimination.
References
Fontes-Villalba M, Fika-Hernando ML, Picazo Ó, et al. Randomised crossover controlled trial of dietary interventions for glycaemic control when body weight is kept stable. J Nutr Sci. 2025;14:e59. doi:10.1017/jns.2025.10028