Study archive · page 9

In healthy adults aged 60 and over, resistance training lowered the inflammatory marker C-reactive protein and improved leg strength and six-minute walk performance, according to a pooled analysis of 19 trials with 728 participants. Two other inflammatory markers, TNF-α and IL-6, did not change significantly.

⚠What it shows: Fairly trustworthy on the big picture: results pooled from 19 trials in 728 healthy over-60s, all of which assigned people to lifting or a comparison group at random. The catch is that only one of three inflammation markers moved, and the trials used different programmes, so the best way to train for this effect is still unknown.Meta-analysis · Arch Gerontol Geriatr

Pooled results from 14 randomised trials in 684 adults with type 2 diabetes found elastic-band resistance training lowered HbA1c and fasting glucose with moderate-certainty evidence, and improved handgrip strength. Effects on walking ability, sit-to-stand performance and muscle mass remained uncertain.

⚠What it shows: Fairly trustworthy on the blood sugar and grip results: pooled from 14 trials in 684 adults with type 2 diabetes, and the findings held up when any single trial was removed. Side effects were patchily reported, and the walking and muscle-mass results were too inconsistent to lean on.Meta-analysis · Front Public Health

Results pooled from six randomised trials in 221 older adults with sarcopenic obesity found resistance training improved Timed Up and Go times by 2.23 seconds and Short Physical Performance Battery scores by 0.93 points, with no serious adverse events reported.

⚠What it shows: Promising but not settled: results pooled from just six trials totalling 221 people, and the reviewers themselves rated the certainty of the evidence low to very low. The gains were consistent across trials and attendance was high, which counts for something.Review · J Exerc Sci Fit

Pooling seven studies of 8,922 postmenopausal women, those with higher protein intakes had significantly greater handgrip strength than those eating less protein, though the difference was small. They also had a lower body mass index, with no difference in age between the groups.

⚠What it shows: Suggestive rather than proof: results pooled from seven studies of 8,922 postmenopausal women, rated low or moderate risk of bias, and the grip difference was small. They compared women who already ate differently rather than assigning anyone a diet, so activity levels and general health could be doing part of the work.Study · Menopause

Pooled individual data from 10 randomised trials in 677 people of Black or South Asian descent living at latitudes of 40 degrees north or higher suggest that keeping winter blood levels above the 25 nmol/L deficiency threshold takes 23.9 to 24.1 µg of vitamin D a day, substantially more than the equivalent estimates for White individuals and above what several agencies currently recommend.

⚠What it shows: Strong for estimating intake needs, unproven for health outcomes: pooled individual data from 10 trials in 677 Black and South Asian children and adults living at northern latitudes. The targets are blood levels rather than demonstrated benefits, and the authors say much more work is needed in these populations.Meta-analysis · Eur J Nutr

In the first review of randomised trials testing the Mediterranean diet against other diets for cognition, only 8 of 66 individual cognitive outcomes favoured the Mediterranean diet while 2 favoured the controls, and there was no benefit for incident cognitive impairment or dementia, a far weaker picture than observational studies suggest.

⚠What it shows: The fair reading is that the trial evidence is thin, not that the diet fails: only 5 trials covering 1888 people have tested this, the prescribed diets varied so much that results could not be pooled, and the strongest trial did show gains in memory and global function. A verdict either way is premature.Systematic review · Am J Clin Nutr

Pooled data from randomised trials of orforglipron, an oral GLP-1 pill, show gut side effects climb with dose over 26 weeks: at 45 mg the odds of nausea and of vomiting were more than 11 times those on placebo, and the odds of quitting over gut symptoms were about 10 times higher, yet no dose raised pancreatitis risk.

⚠What it shows: Solid on side effects but only over 26 weeks: pooled results from placebo-controlled trials across five doses in adults with and without type 2 diabetes. Longer-term safety, and how the pill compares directly with injectable GLP-1 drugs, remain open questions.Meta-analysis · Endocrinol Diabetes Metab

Pooling 47 meta-analyses of randomised trials covering 47,720 people, probiotic and synbiotic supplements delivered significant reductions in fasting glucose, fasting insulin, insulin resistance and long-term blood sugar (HbA1c), with the clearest benefit in people whose glucose control was already impaired.

⚠What it shows: As broad as nutrition evidence gets: results pooled from 47 separate pooled analyses covering 47,720 trial participants, with the clearest benefit in people whose blood sugar control was already impaired. The underlying analyses disagreed with each other a fair amount, and two authors work for a major food and drinks company.Systematic review · Crit Rev Food Sci Nutr

Pooled results from 51 crossover studies in healthy adults show a single bout of continuous aerobic exercise lasting at least 30 minutes lowered insulin, raised glucagon, and lowered blood glucose only when performed after a meal, not in the fasted state.

⚠What it shows: Solid but with a wide spread: results pooled from 51 small studies in healthy adults, fewer than 400 people per outcome, and the studies disagreed with each other a lot, so the exact size of the drop is uncertain. The glucagon finding rests on just 5 studies and 47 people.Meta-analysis · Sports Med

Pooling 43 trials covering 3843 perinatal women, mind-body exercise such as yoga and Pilates and aerobic exercise both eased depression symptoms; mind-body forms worked best after the birth, aerobic exercise did most for women already showing depressive symptoms, and a meaningful improvement appeared at around 100 MET-minutes of movement a week.

⚠What it shows: About as solid as this topic gets: results pooled from 43 trials of 3843 women, all assigned to exercise or comparison groups at random. The trials differed in what they tested and how, so the exact ranking of Pilates over yoga is less certain than the overall finding that movement helps.Meta-analysis · Int J Gynaecol Obstet

A systematic review of six randomised trials found group-based exercise programmes moderately improved postpartum women's self-reported physical activity and psychological wellbeing in the first two years after birth, but did not shift objectively measured activity, though only one trial used objective measures.

⚠What it shows: Encouraging but thin: a sweep of all the studies on this turned up just six trials in healthy postpartum women, and the activity gains showed up only in what women reported themselves. The single trial that used activity monitors found no change, so the true effect on movement is uncertain.Systematic review · BMC Womens Health

Pooled results from 18 randomised trials in 1,099 postpartum women found breast massage increased milk volume and reduced engorgement, pain and anxiety, alongside higher serum prolactin. Complete breastfeeding rates did not differ from control care.

⚠What it shows: A consistent direction of benefit across 18 trials in 1,099 new mothers, though the studies disagreed with each other a lot on how big the effect was, and massage techniques varied. It made no difference to the rate of complete breastfeeding.Meta-analysis · Breastfeed Med

Pooled results from eight randomised controlled trials suggest that wearing an N95 or FFP2 mask during aerobic exercise has no significant effect on heart rate, breathing rate, blood oxygen saturation or blood pressure in healthy people. Findings on how hard the exercise felt were mixed.

⚠What it shows: Reassuring, and built on the better kind of trial: results pooled from eight studies where healthy adults were assigned at random found no meaningful change in any measured vital sign. It says nothing about people with heart or lung conditions, and the trials could not agree on whether masked exercise feels harder.Meta-analysis · J Bras Pneumol

Pooled results from 17 randomised trials in 1,755 people, all conducted in China, linked the traditional exercise Ba Duan Jin to lower blood pressure, improved cholesterol, less anxiety and better sleep. In hypertensive patients it was also associated with roughly half the risk of death from any cause over 20-30 years of long-term follow-up.

⚠What it shows: Treat the survival claim cautiously: every one of the 17 pooled trials was run in China, and a roughly halved death risk from a gentle exercise routine would be a remarkable result if it held up elsewhere. The blood pressure and cholesterol improvements are more modest and more believable.Meta-analysis · Front Public Health

A pooled analysis of 39 studies covering 234 306 people with atopic dermatitis and 10 935 459 comparison individuals found a significantly higher risk of depression in adults, and the same raised risk in children and adolescents. Findings varied substantially between the individual studies.

⚠What it shows: Consistent in direction, vague on size: results pooled from 39 studies of 234 306 people with atopic dermatitis against nearly eleven million comparison individuals, but the studies disagreed substantially and no single risk figure is reported. They tracked people rather than testing anything, so this cannot say which condition drives the other.Meta-analysis · J Paediatr Child Health

In pooled results from six randomised trials in 241 adults with atopic dermatitis, probiotics significantly lowered overall SCORAD severity scores and improved quality of life. Itch severity, skin severity, a standard dermatology life-quality index and immune markers showed no significant change.

⚠What it shows: A mixed picture worth holding loosely: only six trials with 241 adults, the studies disagreed with each other heavily, and the itch itself did not improve even as overall severity scores fell. None of the trials reported side-effect data.Meta-analysis · J Health Popul Nutr

A systematic review of three randomised trials in 154 community-dwelling older adults found that setting device-specific goals with nurses or physicians, and simply raising awareness of being monitored, made older adults more likely to keep using wearable trackers.

⚠What it shows: Thin but tidy evidence: a sweep of all the studies on this found just three trials totalling 154 older adults, average age 65, so the winning strategies rest on a small pool. All three trials were judged low risk of bias, which helps, but bigger tests are still needed.Systematic review · J Med Internet Res

Across 16 randomised trials in 2,296 healthy older adults, tablet apps offering activity recommendations produced the biggest rise in overall daily physical activity, while wearable trackers paired with motivational step goals produced the largest gains in daily steps.

⚠What it shows: A ranking of approaches, not a product test: results pooled from 16 trials in 2,296 healthy older people, compared both directly and indirectly. The programmes varied widely, so the winner here is a type of intervention, not any specific app or tracker.Meta-analysis · Public Health

A sweep of nine randomised trials in athletes found caffeine's effects on sleep inconsistent, varying with dose, timing and habitual use. The reviewers nonetheless conclude that insomnia after ingestion is likely, and that poorer sleep could undermine the recovery athletes depend on.

⚠What it shows: Treat this as a caution rather than a verdict: only nine trials in athletes made the cut, and their results disagreed with each other on sleep quality, duration and awakenings. How much caffeine costs you likely depends on the dose, the timing and how accustomed to it you are.Systematic review · Clin Nutr ESPEN

Pooled data from 17 controlled trials found nitrate supplementation, most often as beetroot juice or sodium nitrate at 300 to 600 mg, gave a moderate improvement in time to exhaustion tests. Effects on time trials and graded exercise tests were small and not statistically significant, and benefits appeared more often in inactive to recreationally active people and with several days of loading.

⚠What it shows: Reliable for one narrow thing: results pooled from 17 controlled trials show nitrate helps people last longer at a fixed effort, while the improvements in race-style time trials were small enough to be chance. Benefits also showed up mostly in inactive to recreationally active people, so trained athletes may see less.Meta-analysis · Int J Sport Nutr Exerc Metab

Across 13 prospective cohorts totalling 193 126 people, those in the highest category of ideal cardiovascular health metrics had roughly half the all-cause mortality risk of the lowest category, along with markedly lower cardiovascular death, cardiovascular disease and stroke risk. Each one point increase in ideal metrics was linked to 11% lower all-cause mortality and 19% lower cardiovascular mortality.

⚠What it shows: A strong and consistent pattern, but not proof of cause: results pooled from 13 studies that followed 193 126 people rather than testing changes on them. People who score well on these metrics differ in many other ways, so part of the gap will belong to things the studies could not measure.Meta-analysis · Clin Cardiol

Pooled results from 22 trials in 1255 college students with depressive symptoms suggest exercise may meaningfully reduce depression scores, with possible benefits for anxiety and physiological markers too, though the certainty of the evidence is low.

⚠What it shows: Promising but shaky: results pooled from 22 trials in 1255 students all point the same way, yet the underlying studies were weak enough that the authors rate their own confidence in the depression finding as low, and even lower for the biological markers. The true effect could be a good deal smaller than the headline figures suggest.Review · Acta Psychol (Amst)

Pooling 36 studies of 498 nonacclimatised people exercising at 26°C or hotter, a single session immediately raised both pro- and anti-inflammatory blood markers, with IL-6 still elevated two hours later and higher peak core temperature amplifying some of the responses.

⚠What it shows: Consistent but noisy: results pooled from 36 studies covering 498 people who were not acclimatised to heat, 86% of them men. The studies disagreed with each other enough that the authors themselves call predictions from their model possibly unreliable.Study · Temperature (Austin)

Pooled results from 15 trials in 7053 previously healthy adults and children found vitamin D3 supplements did not significantly reduce respiratory tract infections; the 6% risk reduction observed was not statistically significant.

⚠What it shows: A serious sweep of the trial evidence, but with limits: results pooled from 15 trials in 7053 people, and the trials disagreed with each other while overall quality was low. It also could not properly test whether people who are genuinely deficient in vitamin D fare differently.Meta-analysis · PLoS One