Study archive · page 38

A systematic review of 13 randomised trials found that regular resistance training improves all aspects of sleep, with the greatest benefit for sleep quality, and also improves anxiety and depression.

⚠What it shows: A credible review of a thin evidence base: just 13 trials, and the sleep benefit shrank when lifting was combined with aerobic work rather than done on its own. What a single session does to that night's sleep remains poorly studied and inconsistent.Systematic review · Sleep Med Rev

A skin-attached patch that pairs focused ultrasound with overnight brainwave recording, aimed at the subthalamic nucleus, increased REM duration by 4.6% and reduced REM latency by 24% in a 28-participant study.

⚠What it shows: One 28-participant registered study of a prototype device, reported by the engineering team that developed it and has filed a patent; the abstract gives no daytime outcomes, no safety data and no independent replication.Clinical trial · Nat Commun

Pooling 23 independent samples of healthy adults (N = 13,636), habitual expressive suppression, routinely masking outward emotion, showed a small but consistent association with poorer self-reported sleep quality (r =.14), and the link was stronger in studies focused on COVID-19-related stress.

⚠What it shows: A meta-analysis of 23 samples totalling 13,636 non-clinical adults, but built on zero-order correlations from self-report questionnaires, so it shows association only: suppression could follow poor sleep as easily as precede it, and the effect is small.Meta-analysis · BMC Psychol

A meta-analysis of 13 randomised controlled trials found community-based traditional Chinese exercises such as tai chi and Qigong reduce sleep disturbance in community-dwelling older adults; Qigong's effect was slightly larger but not significantly different from tai chi's.

⚠What it shows: Meta-analysis of 13 RCTs in community-dwelling older adults; the abstract reports no pooled effect sizes, the Qigong-versus-tai-chi difference was not significant, and applicability to younger people is untested.Meta-analysis · J Transcult Nurs

Digitally delivered cognitive behavioural therapy for insomnia reduced both insomnia and depression severity in a meta-analysis of seven randomised trials totalling 3,597 adults, with effects comparable to face-to-face CBT-I and generally maintained at follow-ups of 6 weeks to 6 months.

⚠What it shows: Meta-analysis of only seven RCTs; follow-up limited to 6 weeks to 6 months, and effects on depression were smaller than on insomnia.Meta-analysis · PeerJ

Folic acid supplementation lowered C-reactive protein by 0.59 mg/L on average versus placebo across 12 randomised trials, with higher doses linked to lower CRP. It had no significant effect on the inflammatory markers IL-6 or TNF-α.

⚠What it shows: Solid pooled evidence, but for a modest effect on a single marker: 12 randomised trials found CRP fell while two other inflammatory markers did not move. Whether a slightly lower CRP reading translates into better health is a question this analysis cannot answer.Meta-analysis · Nutrients

Across 15 randomised trials with 714 participants, resveratrol supplementation produced no significant change in liver enzymes: ALT, AST, GGT, alkaline phosphatase and bilirubin all held statistically steady.

⚠What it shows: Meta-analysis of 15 RCTs (714 adults); heterogeneity was high for some enzymes (I² up to 82.9%), and subgroup analyses hint that dose, duration and population status could matter, but the pooled result was null across the board.Meta-analysis · Int J Clin Pract

Across 22 randomised trials in people with type 2 diabetes, zinc supplementation reduced HbA1c by 0.47 points and fasting blood sugar by 23.32 mg/dl, and improved cholesterol, triglycerides and blood pressure. It also increased body weight by an average of 1.00 kg.

⚠What it shows: Dose-response meta-analysis of 22 RCTs (1,442 adults with type 2 diabetes); certainty rated moderate to high with no substantial publication bias, but the optimal form, dose and duration are unknown and weight rose slightly.Meta-analysis · Crit Rev Food Sci Nutr

A systematic review of 17 randomised trials suggests creatine adds to gains in strength, functional capacity and lean mass in healthy older adults, but only as continuous daily low-dose supplementation combined with at least 12 weeks of resistance training, with effects more consistent in older women.

⚠What it shows: Systematic review of 17 heterogeneous RCTs in healthy older adults without pooled effect sizes; no additive benefit was found for mental health outcomes, and optimal dosing and duration remain undetermined.Systematic review · J Geriatr Phys Ther

Across pooled data from 69 randomised trials plus one further RCT (3,413 participants), probiotic supplementation in athletes reduced the pro-inflammatory marker TNF-α and increased secretory IgA and IFN-γ, while IL-6, IL-8 and IL-10 showed no significant change.

⚠What it shows: Umbrella review and re-analysis of five meta-analyses plus one RCT (3,413 athletes); outcomes are blood and mucosal markers, not illness rates, and optimal strains and doses remain unresolved.Meta-analysis · PeerJ

Across 69 studies of real-life programmes promoting muscle-strengthening exercise in healthy adults, attendance was generally high and 70.0% of the randomised trials improved functional performance, but only nine studies measured whether participants actually took up strength training.

⚠What it shows: A systematic review of 69 studies (10 to 3507 participants each, 36 randomised trials) in healthy community-dwelling adults, but 52 studies were rated weak quality and only five strong, and evidence was conflicting for most outcomes beyond function.Systematic review · J Sci Med Sport

A meta-analysis of 13 randomised controlled trials found moderate-intensity continuous cardio does not consistently increase muscle size in healthy adults, with a pooled effect close to zero compared with not exercising.

⚠What it shows: Meta-analysis of 13 RCTs in healthy adults; pooled effect was near zero but confidence intervals were wide, and the authors flag considerable variability in protocols, populations and measurement sensitivity, so a small effect cannot be fully ruled out.Meta-analysis · Appl Physiol Nutr Metab

A network meta-analysis of 39 randomised trials in 2,585 overweight or obese university students found combined aerobic-plus-resistance training most effective for improving BMI, body weight and fat mass, while HIIT produced the biggest reduction in body fat percentage.

⚠What it shows: Network meta-analysis of 39 RCTs restricted to overweight and obese university students; searches relied on a single database platform, and findings may not generalise to other age groups or settings.Meta-analysis · PeerJ

Across 18 randomised trials involving 1,572 perimenopausal and postmenopausal women, mind-body therapies such as mindfulness, music therapy, dance and yoga produced moderate-to-large improvements in sleep quality, depression and anxiety compared with usual care or no intervention.

⚠What it shows: As strong as this area gets: a meta-analysis of 18 randomised trials in 1,572 women, with robust sensitivity checks and no sign of publication bias. The comparisons were usual care or nothing at all, so part of the benefit may come from the attention and routine rather than the practice itself.Meta-analysis · Front Public Health

A meta-analysis of nine trials in 527 postmenopausal women found whole body vibration treatments improved bone mineral density at the hip and spine compared with no intervention. When vibration was added to an exercise programme, the extra benefit was negligible.

⚠What it shows: Reasonably solid for the headline result: a meta-analysis pooling nine trials in 527 postmenopausal women. But the trials varied hugely in vibration settings, doses and posture, and the right level of stimulation has still not been defined.Meta-analysis · PLoS One

A meta-analysis of eight randomised trials in 812 menopausal women found melatonin, at doses of 1 mg to 5 mg for 3 to 12 months, modestly improved physical symptoms but did not improve sleep quality, overall menopausal symptoms, mood, anxiety or depression.

⚠What it shows: Meta-analysis of only eight RCTs with varying doses (1-5 mg) and durations (3-12 months); heterogeneity was high for several outcomes and the authors call for larger trials to confirm the findings.Meta-analysis · Menopause

In a systematic review of 14 randomised trials, exercise reduced symptoms of depression compared with no treatment and performed similarly to cognitive therapy. Every included trial had important methodological weaknesses, so the true size of the effect remains uncertain.

⚠What it shows: Treat the numbers loosely: all 14 pooled trials had serious flaws, most recruited community volunteers scored on a questionnaire rather than diagnosed patients, and effects looked bigger in trials with shorter follow-up. The reviewers themselves concluded that effectiveness could not be firmly determined.Meta-analysis · BMJ

Over a 10-month Antarctic overwintering at Concordia Station, 12 crew members reported progressively rising loneliness and conflict while cohesion and individual performance declined. More frequent close-range contact was associated with more conflict and paranoid thoughts, not with social support.

⚠What it shows: Observational study of a single 12-person crew over one 10-month mission; associations only, small sample, and findings from an extreme environment may not generalise to everyday settings.Study · Proc Natl Acad Sci U S A

In a meta-analysis of five randomised trials in 2614 Asian adults with overweight or obesity, once-weekly semaglutide 2.4 mg reduced body weight by 8.20% more than placebo, with smaller waist circumference, BMI and systolic blood pressure alongside mostly mild gastrointestinal side effects.

⚠What it shows: Pooled data from five placebo-controlled RCTs (2614 Asian adults with overweight or obesity), all rated low risk of bias, but heterogeneity was high for weight, waist and BMI (I2 = 84%, 84% and 66%) and long-term and real-world outcomes were not assessed.Meta-analysis · Diabetes Obes Metab

In a meta-analysis of six randomised trials in adults with overweight or obesity but without diabetes, tirzepatide cut body weight by 16.32% versus placebo, but roughly tripled the risk of nausea, diarrhoea and constipation, and more than doubled discontinuation due to adverse events.

⚠What it shows: Meta-analysis of six placebo-controlled RCTs in adults with overweight or obesity without diabetes; gastrointestinal side effects, serious GI events and drop-out due to adverse effects were all significantly higher than placebo.Meta-analysis · Obes Rev

A meta-analysis of 9 randomised controlled trials found watermelon consumption significantly lowered systolic blood pressure, total cholesterol and LDL cholesterol, but also significantly raised fasting blood sugar.

⚠What it shows: Meta-analysis of only 9 small RCTs; the authors describe the blood-pressure result as promising but explicitly call for longer, larger trials and urge caution on every outcome beyond blood pressure.Meta-analysis · Diabetes Res Clin Pract

A meta-analysis of 32 randomised trials found yoga improved cardiovascular risk factors versus non-exercise controls, including systolic blood pressure down 5.21 mmHg and LDL cholesterol down 12.14 mg/dl, but showed no significant advantage over conventional exercise.

⚠What it shows: A large pooled analysis pointing one way, but built on small, mixed trials of moderate quality, so the exact sizes of the improvements are soft. Yoga beat doing nothing across the board, yet it did not outperform ordinary exercise, and blood sugar measures did not improve.Meta-analysis · Eur J Prev Cardiol

In a meta-analysis of eight randomised trials, short-term exercise produced a large reduction in knee osteoarthritis pain overall (SMD -0.94), and non-weight-bearing strengthening had the biggest effect (-1.42) compared with weight-bearing strengthening (-0.70) and aerobic exercise (-0.45).

⚠What it shows: Pooled data from only eight randomised trials, all using programmes of more than three sessions a week for up to eight weeks, so this speaks to short-term pain relief only; the subgroup comparison is indirect and the review dates from 2013.Meta-analysis · Clin Rehabil

In a 30-year prospective cohort of 973 middle-aged Swedish men, alcohol consumers had lower odds of myocardial infarction (OR 0.57) and a median survival advantage of 1.3 years over non-consumers, yet beer specifically was linked to higher odds of heart attack (OR 1.70), and no association with stroke was found.

⚠What it shows: Observational cohort of 973 male Swedish automotive workers followed for 30 years; associations only, cannot establish causation, and findings may not generalise to women or other populations.Study · Scand J Prim Health Care