Study archive · page 28

Astaxanthin taken alongside regular training enhanced fat oxidation and improved physical performance in a meta-analysis of 11 trials with 346 healthy people, with the clearest gains in aerobic exercise, at doses of 20 mg or more, and beyond 12 weeks of use. Effects on cognitive accuracy were marginal and reaction time did not change.

⚠What it shows: Early days: just 11 trials with 346 healthy people in total, and the fatigue signal rests on only two studies, so these estimates could shift a lot once bigger trials arrive.Meta-analysis · Biol Res Nurs

Omega-3 supplementation modestly improved triglycerides, total cholesterol, blood pressure and several inflammatory markers in people with metabolic syndrome and related heart conditions across 48 trials with 8489 participants, while LDL cholesterol was unchanged. Doses of 2 g per day or less looked most beneficial overall.

⚠What it shows: Dependable direction, modest size: results pooled from 48 trials with 8489 people who already had metabolic syndrome or heart disease, so the findings may not transfer to healthy adults. The average improvements were real but small.Meta-analysis · Crit Rev Food Sci Nutr

Pooled results from 23 randomised trials in 401 healthy adults found beetroot juice did not significantly improve 10 m or 20 m sprint times, countermovement jump, peak or mean power, perceived exertion, heart rate or blood lactate. Time to peak power and handgrip strength did improve, but the certainty of evidence was low.

⚠What it shows: Take it as a caution rather than a verdict: results pooled from 23 trials in 401 healthy adults, with the evidence rated low to very low certainty for most outcomes, so both the nulls and the small wins could shift with better trials. Data on women were also thin.Meta-analysis · Nutrients

Across 26 randomised, placebo-controlled crossover trials, an acute dose of caffeine gave female team-sport athletes small improvements in repeated sprints, agility, sprint speed, anaerobic power and jumping, and modestly reduced how hard exercise felt. Evidence for sport-specific skills and cognition remained too thin to call.

⚠What it shows: Fairly solid for a small effect: results pooled from 26 trials where each athlete tried both caffeine and a dummy dose in random order. Effects on skills and thinking rest on too few studies to trust, and the data cannot yet say anything about the best dose or timing.Meta-analysis · Nutrients

Across 78 randomised trials, magnesium supplements modestly lowered blood pressure (2.50 mmHg systolic, 1.58 diastolic), fasting glucose, LDL cholesterol, triglycerides and body weight, and slightly raised HDL. The effects were small and the reviewers say their clinical relevance remains uncertain.

⚠What it shows: About as strong as supplement evidence gets, but the payoff is small: results pooled from 78 randomised trials show shifts of a few points at most on each marker. Evidence for doses of 500 mg a day or more and courses beyond 25 weeks is still thin, so the upper end remains untested.Meta-analysis · Nutrients

Pooled evidence from 20 studies found omega-3 supplementation had no effect on adults' total sleep duration, sleep latency, sleep efficiency, sleep quality or insomnia severity. Benefits appeared only in infants' sleep organisation and in children with clinical-level sleep problems.

⚠What it shows: About as clear as the current evidence gets: results pooled from 12 trials and 8 studies that followed people over time found nothing for adult sleep on any measure tested. The included studies varied in who they tested and how, and the authors still want more robust trials before calling it settled.Meta-analysis · Nutr Rev

Cognitive behavioural therapy for insomnia delivered through mobile apps reduced insomnia symptoms and improved sleep quality in a meta-analysis of 16 trials with 2146 adults, with benefits lasting up to six months. Programmes with all five CBT-I components, running six weeks or longer and delivered by healthcare professionals, showed the strongest results.

⚠What it shows: Solid evidence: results pooled from 16 trials with 2146 adults all pointed the same way, and the benefit held at follow-up. The trials only tracked people for up to six months, so nobody yet knows whether the improvement lasts beyond that.Meta-analysis · J Clin Nurs

Pooled data from 15 Chilean studies covering 13,157 people put 25% of adults at moderate-to-high risk of obstructive sleep apnoea, based mostly on screening questionnaires rather than objective sleep tests.

⚠What it shows: Solid as a screening estimate, thin as a diagnosis: results pooled from 15 studies of 13,157 people relied mostly on questionnaires, with only three using overnight breathing recordings. The studies also disagreed with each other strongly, so the true figure could sit anywhere between roughly 16% and 34%.Meta-analysis · Rev Med Chil

Pooled results from 22 randomised trials in 1816 people suggest holographic scraping, a traditional Chinese skin-scraping therapy, reduced insomnia, anxiety and depression scores; only 2 trials reported on adverse effects, so its safety remains unclear.

⚠What it shows: Interesting but far from settled: results pooled from 22 trials in 1816 people looked good on symptom scores, yet the picture for measured sleep quality was mixed. Only 2 of the trials reported side effects, so safety is simply unknown.Meta-analysis · Medicine (Baltimore)

High-intensity interval training improved every component of metabolic syndrome, taking an average 4.12 cm off waist circumference and lowering blood pressure, triglycerides and fasting glucose across 23 trials with 1374 adults. Protocols with under 15 minutes of high-intensity work per session matched longer ones.

⚠What it shows: Solid evidence: results pooled from 23 trials with 1374 middle-aged and older adults with metabolic syndrome. HIIT beat doing nothing across the board but was comparable to traditional steady cardio, so the win here is time efficiency rather than superiority.Meta-analysis · Br J Sports Med

Pooled data from 21 randomised trials and 11 studies that followed people over time, 312,645 participants in all, linked lifestyle interventions to lower risks of major cardiovascular events, death from any cause, cardiovascular death and stroke. Combined diet and exercise programmes showed the greatest benefit, and the authors conclude the protection is comparable to or greater than drug therapy.

⚠What it shows: Weight it carefully: the pooling mixed 21 trials with 11 studies that simply followed people over time, 312,645 in all, so lifestyle and drugs were rarely compared head to head. The claim that behaviour change matches medication is the authors' reading across different study designs rather than a direct contest.Study · J Cardiovasc Nurs

In pooled data from nine studies covering 480,479 people, regular stair climbing was associated with a 39% lower risk of cardiovascular death and a 24% lower risk of death from any cause.

⚠What it shows: Encouraging but not proof: most of these studies followed people rather than testing them, so stair climbers may simply have been healthier to begin with. The pooling covered nine studies and 480,479 people.Systematic review · Am J Cardiovasc Drugs

In a small study of 16 runners covering 800m, pushing a stroller reduced speed and stride length, yet energetic cost and heart rate were not significantly different from running unencumbered. The pushing method used also significantly changed speed and stride length.

⚠What it shows: Treat this as an early snapshot: just 16 runners over a single 800m effort at speeds they chose themselves, so it says little about long runs, hard sessions or injury risk.Study · PLoS One

A review pooling 18 randomised trials with 1365 older adults found regular resistance training improved working, immediate and short-term memory, but had no significant effect on verbal or delayed memory.

⚠What it shows: Reasonably solid but not the final word: results pooled from 18 trials where people were assigned at random, covering 1365 older adults, showed benefits for some memory types only. The reviewers themselves flag weaknesses in the underlying trials, so treat the exact sizes loosely.Systematic review · Exp Gerontol

Exercise, antidepressants and their combination reduced depressive symptoms to a similar degree in adults with non-severe depression across 21 trials with 2551 participants, and all three beat control conditions. People assigned to exercise were more likely to drop out than those taking antidepressants.

⚠What it shows: Strong within its lane: results pooled from 21 trials with 2551 adults found no difference between the treatments, but it only covers non-severe depression. Exercisers quit more often than people on medication, which matters even more outside a supervised study.Meta-analysis · Br J Sports Med

A review of 34 studies in adults with chronic pain found that pain catastrophizing, the tendency to dwell on and magnify pain, and feelings of helplessness partly explained how pain intensity feeds depressive symptoms. Sleep quality, pain self-efficacy and pain interference showed early signs of playing a role too.

⚠What it shows: Treat this as an early signal: most of the 34 studies measured everything at a single point in time, so they cannot show what causes what, and the key pooled estimate leaned heavily on one study. The pain-to-depression analyses covered 13,587 adults.Review · J Clin Med

Exercising at a consistent time and as part of a consistent routine is linked with stronger exercise habits and more actual activity, according to pooled data from 14 studies of 6069 people. Where you train, who you train with and your mood showed links only to habit strength, not to the activity itself.

⚠What it shows: Worth acting on but modest: the links come from results pooled across 14 studies of 6069 people, and they are small. Because researchers followed people rather than changing their schedules, it cannot prove that a fixed slot causes the habit.Meta-analysis · Appl Psychol Health Well Being

Pooled evidence finds no statistically significant increase in depression, anxiety or suicidal thoughts among people taking semaglutide compared with placebo or other treatments. The authors stress this is an absence of detected risk in noisy data, not definitive proof of safety.

⚠What it shows: Reassuring rather than final: the pooled sources, from proper trials to voluntary side-effect reports, disagreed with each other sharply and some carried real risk of bias. That messiness means a modest effect in either direction cannot be ruled out.Meta-analysis · Clin Obes

In pooled data from four trials of 5425 adults with overweight or obesity, weekly cagrilintide cut body weight by 6.08% versus placebo, and the CagriSema combination with semaglutide cut it by 5.98% while also trimming waist size by 10.91 cm and improving blood sugar. Both treatments modestly increased adverse events.

⚠What it shows: Promising but young: just four trials with 5425 people in total, and their results varied a lot from one to the next. Longer studies are still needed before anyone knows how safe and durable these drugs are.Study · Ann Med Surg (Lond)

Acid-reflux drugs may double as mild glucose lowerers: across 15 studies of 2,038 people, proton pump inhibitor use was linked with a 5.32 mg/dL fall in fasting blood glucose, with the biggest apparent benefit in poorly controlled type 2 diabetes. HbA1c results were inconsistent and insulin resistance did not change.

⚠What it shows: Intriguing but shaky: 15 mostly small studies totalling 2,038 people, mixing proper trials with weaker designs, and the review itself was never registered in advance. Treat the exact size of the glucose drop loosely.Review · J Midlife Health

In a pooled analysis of 8 trials with 549 adults with metabolic syndrome, eating walnuts modestly lowered blood triglycerides but did not change cholesterol, blood sugar, insulin, inflammation, body weight, waist size or blood pressure.

⚠What it shows: Solid as far as it goes: results pooled from 8 trials in 549 adults with metabolic syndrome, but the trials were few and fairly small. A modest triglyceride drop is the only effect that showed up reliably, so bigger benefits can be neither ruled in nor out yet.Meta-analysis · Pharmacol Res

Pooling 10 trials with 1254 people with pre-diabetes and features of metabolic syndrome, weight loss plans built around meal replacements beat conventional food-based diets by 1.38 kg of extra weight loss, with small added reductions in LDL cholesterol and systolic blood pressure.

⚠What it shows: Believable but modest: results pooled from 10 trials in 1254 people with pre-diabetes, and the authors themselves rate the certainty of the evidence low to moderate because the trials were imprecise and sometimes disagreed with each other.Meta-analysis · Obes Rev

In pooled results from 15 studies of 448 older adults averaging 67.8 years, heavy loads around 80% of one-rep max tended to build slightly more strength than moderate loads around 45%, but the gap shrank when total work was matched, and muscle growth was small either way.

⚠What it shows: Solid pooled evidence with one clear message: across 15 studies of 448 people averaging 67.8 years, lighter weights lifted for enough repetitions built substantial strength too. The trials only compared roughly 80% against 45% loads, so the middle ground between them is untested.Meta-analysis · Scand J Med Sci Sports

Probiotics, prebiotics and synbiotics each modestly reduced depression and anxiety scores compared with placebo across 72 randomised trials, and probiotics also improved sleep quality. Effects varied by population, intervention duration and probiotic type.

⚠What it shows: Encouraging but wobbly: results pooled from 72 trials showed modest benefits, yet the studies disagreed with each other a great deal and many were of limited quality, so the true effect may be smaller than it looks. The authors themselves call for larger, better trials with longer follow-up.Meta-analysis · BMC Psychiatry