Summary: In a 12-week, double-blind, 2×2 factorial feeding trial of 70 adults with prediabetes (60 analysed, 15 per arm), consuming unsaturated fat at lunch rather than dinner was associated with improved insulin-sensitivity indices and lower postprandial insulin. The primary outcome, postprandial serum glucose, did not differ between groups, and the type of unsaturated fat (monounsaturated versus polyunsaturated) had no effect.
PICO Summary
| Element | Detail |
|---|---|
| Population | Seventy adults with prediabetes (mean age 57 years); 60 with complete faecal samples were analysed. Double-blind, randomised, controlled, 2×2 factorial feeding trial conducted in China (ChiCTR2100045645). |
| Intervention | Unsaturated fat intake timed to lunch, within a high-monounsaturated-fat or high-polyunsaturated-fat diet, over 12 weeks (n=15 per group). |
| Comparison | The same unsaturated fat intake timed to dinner (n=15 per group). There was no untimed or no-intervention control arm; the contrast is lunch versus dinner. |
| Outcome | Primary outcome (postprandial serum glucose): no significant difference between groups. Co-primary outcomes: unsaturated fat at lunch versus dinner improved insulin-sensitivity indices (P<0.05; reported for the Gutt and Stumvoll indices) and lowered postprandial insulin, while reductions in serum free saturated fatty acid did not reach significance (P>0.05). The timing effect was associated with shifts in gut microbiome composition and bile-acid metabolism and was independent of fat type. No effect estimates with 95% confidence intervals, absolute risk reduction, or NNT were reported in the abstract. |
Expert Commentary
This is a mechanistic, exploratory feeding trial, and its headline should be read with caution. The pre-specified primary outcome, postprandial serum glucose, was null: no significant between-group difference was detected. What was observed was a secondary signal that unsaturated fat eaten at lunch rather than dinner improved calculated insulin-sensitivity indices and lowered postprandial insulin, plausibly linked to gut microbiome and bile-acid changes. The framing is therefore one of hypothesis generation about meal timing, not proof that fat timing controls glycaemia. The most important limitation is statistical power: with only 15 participants per arm after attrition, the trial is small, multiple insulin-sensitivity indices were examined, and the risk of chance findings is real, so these results should be considered preliminary until replicated in a larger, adequately powered study with hard glycaemic endpoints. Can I use this with my patients? Not yet. The data do not support telling a person with prediabetes that shifting fat to lunch will improve their glucose, particularly when the primary glucose outcome did not move. The trial is single-country and tightly controlled through provided meals, which limits real-world generalisability, though the absence of declared industry or manufacturer sponsorship is reassuring. Larger, longer trials reporting effect sizes with confidence intervals and clinically meaningful glycaemic outcomes are needed before meal-timing of dietary fat enters prediabetes counselling.
References
Wei C, Xu X, Zhang J, et al. Timing of unsaturated fat intake improves insulin sensitivity via the gut microbiota-bile acid axis: a randomized controlled trial. Nat Commun. 2025;16(1):4211. doi:10.1038/s41467-025-58937-6