Study archive · page 22

After rapid weight loss on very-low-calorie diets, meal replacements helped people keep off 3.9 kg more and anti-obesity drugs 3.5 kg more than controls in pooled trials, while high-protein diets preserved 1.5 kg more. Exercise programmes and dietary supplements showed no significant benefit for keeping the weight off.

⚠What it shows: Strong for what it covers: results pooled from 20 trials with 3,017 people who first lost around 12.3 kg on very-low-calorie diets. The exercise evidence came from only five arms, so a benefit cannot be ruled out, but none was shown here.Meta-analysis · Am J Clin Nutr

Pooling 23 observational studies covering 181 405 people, higher fibre intake was associated with slightly lower depressive and anxiety scores, but 10 randomised trials of fibre supplements in 740 people found no benefit over placebo for either outcome.

⚠What it shows: Strong evidence that the supplement version does not deliver: results pooled from 23 studies that followed 181 405 people plus 10 trials in 740 people. Only one of those trials included people with diagnosed depression, so fibre pills remain untested where they might matter most.Meta-analysis · Nutr Rev

Pooling nine randomised trials in 540 people with obesity, intermittent fasting produced no greater reduction in body weight or BMI than conventional control diets at any time point, and no difference emerged between alternate day fasting and time restricted eating.

⚠What it shows: Reasonably solid on the headline point: results pooled from nine trials lasting longer than two months in 540 people with a BMI above 30, though the authors rated the quality of evidence low to moderate. Data on mood and well-being were too thin to pool at all.Meta-analysis · Nutr Metab Cardiovasc Dis

A meta-analysis of nine randomised trials in 899 Chinese adults found intermittent fasting reduced body weight, body fat, insulin resistance, LDL cholesterol and triglycerides compared with control groups, while waist circumference, fasting glucose and blood pressure did not change significantly.

⚠What it shows: Reasonably solid for the short term: results pooled from nine trials in 899 Chinese adults agree on weight and body fat, but the trials were brief and mostly tested time-restricted eating, so long-term effects and other fasting styles remain open questions.Meta-analysis · Nutrients

Pooled trials show whey protein combined with exercise raises muscle protein synthesis in healthy adults, with a dose-dependent effect across 10 to 60 g and rises of up to 2.5-fold when 20-40 g is taken 45 minutes before multiple-set resistance exercise.

⚠What it shows: Solid enough to take seriously: results pooled from 21 trials in healthy adults, with muscle building measured directly in biopsy samples. A bigger synthesis response over a few hours is not the same as more muscle after months of training, so the long-term payoff is still unproven.Meta-analysis · Nutrients

Extra protein on its own did not significantly improve lean mass or strength in adults over 50, in a pooled analysis of 28 randomised trials. Only when paired with resistance exercise did higher protein intake add limb lean mass and handgrip strength, with the clearest lean mass gains in sarcopenic or frail groups.

⚠What it shows: Strong evidence with a clear split: results pooled from 28 trials in adults aged 50 and over show protein only pays off alongside lifting. The trials disagreed with each other considerably and most carried some risk of bias, so the true size of the benefit is uncertain.Meta-analysis · Am J Clin Nutr

Across 18 randomised trials in adults with overweight or obesity, intermittent fasting reduced weight, BMI, fat mass, waist circumference, daily energy intake and diastolic blood pressure compared with unrestricted eating, yet it did not improve sleep quality.

⚠What it shows: Solid on the weight and blood pressure findings: results pooled from 18 trials comparing fasting schedules with unrestricted eating in adults with overweight or obesity. Sleep was only a secondary measure in most of the studies, so trials built to test sleep directly are still needed.Meta-analysis · Sleep Med Rev

Health qigong improved self-reported sleep quality in pooled randomised trials involving 1,147 adults with and without health conditions, though the effect shrank to small once one outlier study was removed.

⚠What it shows: Promising rather than proven: results pooled from 13 trials with 1,147 adults, but the trials disagreed sharply and the benefit turned small after removing one outlier. Half the studies compared qigong with doing nothing, so expectation effects were not ruled out.Meta-analysis · Complement Ther Med

Traditional Chinese exercise improved sleep quality across 20 pooled randomised trials, cutting Pittsburgh Sleep Quality Index scores by 2.24 points on average, with tai chi outperforming qigong. Several groups, including college students, stroke patients and people with episodic migraine, saw no significant benefit.

⚠What it shows: Encouraging but uneven: results pooled from 20 trials showed better sleep in groups including older adults and people with insomnia, but no benefit for students, stroke patients or breast cancer patients. The studies disagreed with each other a lot, so the authors themselves say to read the findings cautiously.Meta-analysis · Medicine (Baltimore)

Pooling 24 randomised trials in nonelderly people, a network meta-analysis found resistance training was the most effective drug-free intervention for improving sleep quality, ahead of general physical activity, nutritional interventions and other forms of exercise.

⚠What it shows: A credible ranking rather than a final answer: it pools 24 trials of different drug-free sleep strategies in younger and middle-aged people and compares them indirectly, since most were never tested head to head. The abstract does not say how large the benefit was, only that lifting came out on top.Meta-analysis · PLoS One

Across 12 randomised trials of 526 people with obstructive sleep apnoea, exercise training cut the apnoea-hypopnoea index by 7.08 events per hour, reduced daytime sleepiness and improved fitness, and it worked regardless of baseline severity or body mass index.

⚠What it shows: Reasonably strong: results pooled from 12 trials in 526 people with sleep apnoea, all assigned at random to exercise or control, and the benefits held across severities and body sizes. The individual trials were small, and people already using CPAP consistently gained fitness but saw no extra drop in apnoea events or sleepiness.Meta-analysis · J Clin Sleep Med

Plant-based diets followed for at least 14 days improved HOMA-IR, a marker of insulin resistance, and lowered fasting insulin in adults with overweight or obesity, across pooled randomised trials. Single studies in people with prediabetes and type 2 diabetes found no such differences.

⚠What it shows: Encouraging but short-term: results pooled from eight trials with 716 adults, and the clear gains were confined to people with overweight or obesity. Some diets lasted only a fortnight, so nobody knows whether the improvements persist over months or years.Meta-analysis · Nutrients

In pooled data from seven randomised trials with 8,069 overweight or obese participants, the combination drug CagriSema produced 13.99 kg more weight loss than placebo, 9.24 kg more than cagrilintide alone and 7.58 kg more than semaglutide alone, with mainly gastrointestinal side effects and no increase in serious adverse events.

⚠What it shows: Strong short-term evidence: results pooled from seven trials in 8,069 overweight or obese adults, showing consistent extra weight loss beyond either drug alone. Long-term safety and whether the loss lasts are still open questions, and the blood sugar results were mixed between trials.Review · Naunyn Schmiedebergs Arch Pharmacol

Pooling 35 randomised trials in 2752 adults with overweight or obesity, exercise training raised adiponectin and maximal oxygen uptake by 5.26 mL/kg/min, and lowered fasting insulin, fasting glucose, insulin resistance and triglycerides, while inflammation markers such as CRP, IL-6 and TNF-α did not change significantly.

⚠What it shows: Strong on the metabolic wins, weaker on inflammation: results pooled from 35 trials in 2752 adults with overweight or obesity. The classic inflammation markers did not move, and the authors call for longer studies to confirm the picture.Meta-analysis · J Sports Med Phys Fitness

Across 12 randomised trials, cinnamon supplementation lowered fasting blood sugar by 7.72 mg/dL and improved insulin resistance and cholesterol in women with PCOS, but the average weight loss was just 0.47 kg.

⚠What it shows: Encouraging but shaky: results pooled from 12 trials in women with PCOS point the same way on most markers, yet the trials disagreed sharply with each other on blood sugar, so the true size of the effect is uncertain. The average weight change was under half a kilogram, so treat this as a nudge rather than a treatment.Meta-analysis · Eur J Obstet Gynecol Reprod Biol

In a meta-analysis of 15 randomised trials covering 916 women with polycystic ovary syndrome, polyphenol supplements modestly reduced serum insulin, luteinising hormone and BMI compared with placebo, while testosterone levels did not change.

⚠What it shows: Real but modest: results pooled from 15 trials in 916 women, and the BMI change in particular was tiny. Testosterone, a key driver of PCOS symptoms, did not move, so this is a supporting act rather than a stand-alone treatment.Meta-analysis · Eur J Obstet Gynecol Reprod Biol

Daily iron supplements reduced anaemia and iron deficiency, raised haemoglobin, improved maximal and submaximal exercise performance and reduced fatigue in menstruating women, across 67 trials of 8506 women. The trade-off: constipation and loose stools were each roughly twice as common on iron.

⚠What it shows: About as trustworthy as supplement evidence gets: results pooled from 67 trials in 8506 menstruating women, with the haemoglobin gain rated high quality. Most trials reported too little detail to judge their rigour, and only 10 were considered reliably run, so some of the secondary findings carry less weight.Meta-analysis · Cochrane Database Syst Rev

Green tea supplementation reduced body weight by an average of 1.78 kg and BMI by 0.65 kg/m² in a pooled analysis of randomised trials. Waist circumference only fell significantly at doses of at least 800 mg a day or in trials shorter than 12 weeks.

⚠What it shows: Believable but modest: results pooled from randomised trials in people with obesity, though the individual studies disagreed with each other and the dose patterns were inconsistent. The average saving was under two kilograms, so this is a nudge, not a transformation.Meta-analysis · Phytother Res

Whey came top of six protein sources for boosting muscle mass, handgrip strength and walking speed in middle-aged and older adults doing resistance training, in a network analysis pooling 78 randomised trials with 5272 participants.

⚠What it shows: A strong signal from results pooled across 78 trials in 5272 middle-aged and older adults, ranking whey against milk, casein, meat, soy and peanut protein. Many participants were hospitalised or living with health conditions, so the ranking may not transfer neatly to healthy trained lifters.Meta-analysis · Nutrients

A meta-analysis of 11 randomised trials found aerobic exercise improved erectile function scores by 2.8 points on average versus non-exercising controls, with bigger gains, up to 4.9 points, in men with more severe erectile dysfunction.

⚠What it shows: Reliable as far as it goes: pooled results from 11 trials where men were assigned at random to aerobic exercise or to no exercise. The trials covered quite different groups of men, which makes it harder to say exactly who can expect how much benefit.Meta-analysis · J Sex Med

In a meta-analysis of six randomised trials, aerobic exercise training lowered the inflammatory marker TNF-α in male smokers, but pooled C-reactive protein levels did not fall significantly.

⚠What it shows: Thin evidence, treat gently: only six trials qualified, just two measured TNF-α, and every study enrolled only men or only women, so the TNF-α result rests on male-only data. The authors themselves say more research is needed before firm conclusions.Meta-analysis · Respir Med

Exercise training lowered the inflammatory markers CRP and IL-6 in older adults, and only the combination of exercise with protein supplementation also reduced TNF-α, in a pooled analysis of randomised trials. The authors conclude combined interventions show the greatest benefit against inflammaging.

⚠What it shows: A credible pooled result from randomised trials in older adults, and the pattern across three inflammatory markers is consistent. The authors themselves call for larger trials and more specific populations before anyone treats this as a settled recipe.Meta-analysis · Nutr Rev

A meta-analysis of 29 randomised trials in 2,049 adults found supervised aerobic exercise training produced modest improvements in attention and processing speed, executive function and memory, while effects on working memory were less consistent.

⚠What it shows: A real but small effect: results pooled from 29 trials in 2,049 adults doing supervised aerobic training for longer than a month. Every cognitive gain was modest, so this supports exercise for brain health without making it a miracle.Meta-analysis · Psychosom Med

Across 54 randomised trials in 3390 people with type 2 diabetes, gut microbiome-targeted therapies such as probiotics modestly improved blood sugar control, lowering HbA1c by 0.28% and fasting glucose by 7.97 mg/dL, and reduced inflammatory markers.

⚠What it shows: Modest but consistent: results pooled from 54 trials in 3390 people with type 2 diabetes mostly point the same way, though the trials disagreed with each other and few followed people for long, so lasting benefit is unproven.Meta-analysis · J Clin Endocrinol Metab