A randomized pilot trial at the Durham VA Medical Center put 64 older male veterans with obesity and prediabetes on the same calorie-restricted diet and the same light exercise program, then split them by protein dose: 0.8 g/kg/day (the RDA) versus 1.4 g/kg/day delivered as roughly 30 g of high-quality protein per meal. After six months both groups had lost meaningful weight, but the higher-protein group gave up about half as much fat-free mass (0.8 kg versus 1.5 kg) and gained nearly twice as much physical function on the Short Physical Performance Battery (1.8 points versus 1.0). Grip strength and six-minute walk distance also favored the higher-protein arm, most clearly at three months.
Losing weight after 65 is a trade. Fat comes off, and so does muscle. In conventional weight-loss programs, a quarter or more of everything shed is fat-free mass, which is exactly the tissue an older person cannot afford to lose. The question this trial asks is whether you can rig the trade in your favor simply by changing how much protein sits on the plate.
The researchers, based at the Durham Veterans Affairs Medical Center and Duke University, recruited 64 male veterans over 55 with class II obesity and prediabetes, a group that clinical trials routinely under-recruit. Everyone got the same calorie cut, the same weekly nutrition classes, and the same three 30-minute sessions of light band-based exercise. The only planned difference was protein: the recommended dietary allowance of 0.8 g per kg of body weight in one arm, and 1.4 g per kg spread across meals at roughly 30 g a meal in the other. To make that realistic rather than aspirational, the study handed participants the food, including lean pork, chicken strips, whey powder, and protein drinks.
Six months later the scale told a boring story. Both groups lost roughly 7 kg. What differed was the composition of that loss. In the higher-protein arm, fat-free mass accounted for about 11 percent of the weight lost. In the RDA arm, it accounted for about 21 percent. Same weight off, twice the muscle cost.
Function tracked composition. The Short Physical Performance Battery, a walk-balance-and-stand-up test that predicts disability and mortality in older adults, rose by 1.8 points in the higher-protein group and 1.0 in the control group. Grip strength rose 3.0 kg versus 1.2 kg. At the three-month mark the higher-protein men were walking 63 metres further in six minutes, against 34 metres for the controls, though that gap had closed by month six as the control group caught up.
One hoped-for benefit did not appear. HbA1c sat at 5.9 percent in both arms at every timepoint. Higher protein did not blunt glycemic control, which was the safety question, but it did not improve it either.
The catch is size. The pandemic shut the trial down for 19 months and enrollment never recovered. Only 12 control-arm men supplied six-month data. The authors are candid that this is now a pilot, and the fair reading is a plausible signal awaiting a properly powered trial.
Actionable Insights
The message is narrow: if you are an older adult deliberately losing weight, protein intake changes what you lose, not how much you lose.
Scale of the effect. Both groups dropped about 7 kg in six months. The higher-protein group lost 0.8 kg of lean tissue doing it; the RDA group lost 1.5 kg. As a share of total weight lost, that is 11 percent versus 21 percent. On the physical performance test, the higher-protein group improved 1.8 points versus 1.0. Clinicians treat a 1-point change on that 12-point scale as meaningful for real-world mobility, so the 0.8-point gap is near a clinically noticeable difference on its own. Grip strength improved 6.8 percent versus 2.8 percent.
Effect size in standardized terms. Cohen’s d ran 0.48 to 0.68 for the muscle and function outcomes, conventionally moderate. Loosely translated, a randomly chosen higher-protein participant beat a randomly chosen control about 63 to 68 percent of the time, against 50 percent if protein did nothing.
The dose. About 1.4 g protein per kg of current body weight, roughly 159 g a day at these body weights, split into meals of at least 30 g rather than loaded into dinner. Distribution appears to matter as much as total.
Context and Source
- Open Access Paper: Influence of Protein Intake during Obesity Reduction on Physical Function and Body Habitus in Older Males with Prediabetes: A Pilot Study
- Institution: GRECC, Durham VA Health Care System and Duke University Medical Center, Durham, North Carolina, United States
- Country: United States
- Journal: Current Developments in Nutrition, volume 10 (2026), article 109460. Published by Elsevier on behalf of the American Society for Nutrition
- Impact evaluation. The impact score of this journal is 4.2 (2025 Journal Citation Reports, released June 2026; Scopus CiteScore is also 4.2), evaluated against a typical high-end range of 0 to 60+ for top general science, therefore this is a Low-to-Medium impact journal.