Early, small, conflicting, or animal-only. Not actionable alone.
In trials of women past menopause, creatine's gain in lean mass is under a pound on a scan that counts water, and three of the five trials behind that figure found no gain at all. The leg-press gain this page once reported is not in the trials it was said to come from.
A 2026 review found nine trials that measured what bovine colostrum does to body fat, and its summary credits one with a reduction: 0.4 points on 60 grams a day. The whey group in that trial fell 0.4 points too. The gut finding rests on five small trials in healthy volunteers.
Krill oil is sold on better absorption. Some trials do find more omega-3 in the blood on krill oil, the newest among them. None shows it does more for you: of 64 pooled trials, four compared the two oils, and no difference in blood fats held up.
Seven trials found waists three inches smaller with a light device. Nine of the ten trials in the review used lasers, eight added the device to a diet or an exercise programme, and none tested a home LED panel. Against a sham, the gap on the scale was about a pound.
Collagen improved skin hydration and elasticity in the randomised trials that two overlapping reviews pooled. A company paid for, supplied or staffed at least 21 of the 26, the trials' results differ in size far more than chance would explain, and 95% of participants were women. We rate it Preliminary.
A 2025 scoping review found ten places called Blue Zones. Three are well characterised and two of those are fading; Rugao gets more people to 90 but fewer to 100, and Loma Linda was never studied as one.
Finnish men who took 4–7 saunas a week had a 63% lower adjusted risk of sudden cardiac death, a figure resting on 201 men and 10 deaths. Then 20 randomised trials of heat, most of them not Finnish sauna, found almost nothing.
A month-long trial put box breathing against two other breathing techniques and meditation. All three breathing groups got the same small lift after each session, and smaller studies find the square tying with the alternatives more often than beating them.
Cortisol falls in the pooled trials, most of them paid for or supplied by the extract makers. Whether people feel less stressed is contested. Sleep is the best-supported claim, testosterone rises in men but not women, and liver injury is very rare but can be severe.
The trial behind the 2024 headlines was retracted in September 2025, four days after a meta-analysis pooled it, and a year on that meta-analysis still carries no correction. What survives: a real but modest effect on blood sugar after meals, and about 2 lb of weight across twelve short trials, most of it back within a month of stopping in the one trial that checked.