Study archive · page 7

In a 12-week randomised trial, 150 Chinese adults aged 60-79 assigned to three weekly square-dance sessions improved more than health-education controls on the six-minute walk test, by 41.5 m, and on a timed up-and-go balance test. Blood pressure, depressive and anxiety symptoms, wellbeing and quality of life all shifted modestly in their favour, with 87% adherence and no serious adverse events.

⚠What it shows: Single 12-week randomised trial in 150 community-dwelling Chinese adults aged 60-79, single-blind with a health-education comparator; short duration, culturally specific activity and no follow-up beyond week 12.RCT · Front Public Health

Pooling 53 randomised trials in footballers, strength training ranked top for jump height, squat one-rep max, 5m, 10m and 30m sprints and agility, while endurance training ranked best for VO2max. The rankings come from probability scores across a network of trials, not head-to-head certainty.

⚠What it shows: Systematic review and network meta-analysis of 53 RCTs totalling 1,580 footballers; rankings rest on SUCRA probabilities rather than direct comparisons, trial quality and dosing varied, and findings apply to trained football players rather than general gym-goers.Systematic review · Front Physiol

In a 24-week feasibility trial with young Black women, adherence to supervised resistance sessions ran at 93.9% and 88.8% across the two groups — but collapsed to 14.3% and 15.4% once training became unsupervised.

⚠What it shows: Small feasibility RCT, n=27 young Black women, designed to test acceptability rather than health outcomes; needs larger efficacy trials.RCT · Int J Environ Res Public Health

Pooling 12 randomised trials in 1,050 women in the menopausal transition, conventional and mind-body exercise improved menopausal symptoms, health-related quality of life, anxiety and depressive symptoms compared with control conditions. Exploratory analyses hinted the size of the benefit depends on which type of exercise is used.

⚠What it shows: Meta-analysis of 12 RCTs (1,050 women) published between 2020 and 2025; pooled effects for symptoms and quality of life had very wide confidence intervals that only just excluded no effect, the individual trials were small and varied in exercise type, and subgroup comparisons by modality were exploratory only.Review · Eur J Obstet Gynecol Reprod Biol

In 56 automotive factory workers with non-specific chronic low back pain, adding a 12-week exercise programme to ergonomics training produced greater improvement in pain, functional capacity and sleep quality than ergonomics training alone. Fatigue severity did not differ between the groups.

⚠What it shows: A single randomised controlled trial in 56 automotive workers over 12 weeks; both groups improved, the between-group edge is reported without effect sizes in the abstract, one outcome (fatigue) was null, and results in one factory workforce may not transfer to other jobs or to older or sedentary populations.Study · J Occup Environ Med

Pooling 38 studies and 2314 participants, people with patellofemoral pain showed greater navicular drop — a marker of arch collapse — than pain-free controls, alongside greater rearfoot eversion in landing and step-down tasks. Clinical ankle dorsiflexion range, foot posture index and plantar pressure distribution did not differ.

⚠What it shows: Meta-analysis of predominantly case-control studies (42 reviewed, 38 pooled, 2314 participants); the navicular drop result rests on only four studies with 84 people per group, most other findings were rated very limited evidence, and cross-sectional comparisons cannot show whether foot mechanics cause knee pain or follow it.Review · Transl Sports Med

In up to 4,881 British adults aged 46-48, lifting weights at least twice a week was associated with a 3.99% lower total-to-HDL cholesterol ratio even after adjusting for device-measured moderate-to-vigorous activity and other lifestyle factors. Associations with blood pressure and HbA1c faded once smoking, alcohol, education, self-rated health and disability were accounted for.

⚠What it shows: Cross-sectional analysis of one birth cohort (up to 4,881 adults aged 46-48); resistance exercise frequency was self-reported, so no causal direction can be established and only one of three outcomes survived full adjustment.Study · Sport Sci Health

Among 863 adults in routine primary care, type 2 diabetes prevalence fell in steps across activity levels, from 29.7% in the least active to 8.9% in the most active. After adjustment, low activity was associated with 4.32 times the odds of documented type 2 diabetes compared with high activity, and moderate activity with 2.07 times the odds.

⚠What it shows: Cross-sectional observational study of 863 primary-care patients with activity estimated in routine physician interviews; no HbA1c, glucose, diet or medication data, so cause and direction cannot be established.Cohort · Medicina (Kaunas)

Pooling 53 studies covering 86,652 people, esports players did not differ from non-players on physical activity, BMI, sleep duration, anxiety, stress or grip strength, and showed slightly lower depression scores after sensitivity analysis. Non-professional players did carry 4.57 percentage points more body fat, and participation correlated with sedentary behaviour, BMI and gaming disorder.

⚠What it shows: Systematic review and meta-analysis of 53 studies (86,652 people), but the included studies were cross-sectional with moderate risk of bias and high heterogeneity, so nothing here shows direction of cause.Systematic review · Sports Med Open

In a pooled analysis of 12 randomised trials, combining aerobic and resistance training in overweight or obese adolescents lowered fasting insulin and blood pressure versus no exercise, and modestly beat aerobic exercise alone on fasting insulin, HOMA-IR and adiponectin. Fasting glucose, total cholesterol and triglycerides did not budge.

⚠What it shows: Meta-analysis of only 12 RCTs in overweight or obese adolescents; GRADE rated most outcomes low or very low quality, protocols varied widely, blinding was absent and studies clustered regionally, so effect sizes should be read cautiously.Review · Front Public Health

In 10 randomised trials with 599 adults with prediabetes or type 2 diabetes, time-restricted eating windows of 4-10 hours lowered fasting blood glucose by 0.30 mmol/l and body weight by 1.6 kg, but produced no significant overall change in HbA1c. HbA1c did fall modestly in the prediabetes subgroup only.

⚠What it shows: Meta-analysis of 10 RCTs totalling 599 adults, mean study length 4 months, graded moderate certainty; confidence intervals in several included trials were wide because samples were small, and effects on insulin sensitivity, beta cell function and continuous glucose monitoring were inconclusive.Systematic review · Diabetologia

In a randomised trial of German shift workers, four personalised online counselling messages did not improve sleep efficiency any more than simply completing weekly sleep diaries. Sleep efficiency, insomnia and depression symptoms improved in both groups, while daytime sleepiness improved in neither.

⚠What it shows: One small superiority RCT, intention-to-treat n=66 German shift workers, four weeks, waiting-list control that also kept sleep diaries; self-reported outcomes and the authors flag limited sample size and time frame.RCT · Sci Rep

In a meta-analysis of 19 cohort studies covering 9,460,018 pregnancies, sleep-disordered breathing was associated with higher odds of gestational hypertension (OR 1.50) and preeclampsia (OR 1.83), with obstructive sleep apnoea the strongest signal (preeclampsia OR 2.13). Insomnia and poor sleep quality on the Pittsburgh Sleep Quality Index were not significantly associated with hypertensive outcomes.

⚠What it shows: Meta-analysis of 19 observational cohort studies, so associations only and open to confounding by factors such as body weight; sleep disorders were defined and measured differently across studies and no trial has shown that treating them lowers hypertensive risk.Study · Eur J Obstet Gynecol Reprod Biol

In a randomised, double-blind trial of 44 healthy adults with poor sleep, hydrogen-rich jelly was no better than placebo on any sleep or anxiety measure. Only in exploratory analysis did people with fewer hydrogen-producing gut bacteria report greater improvement in sleep quality and mental stress.

⚠What it shows: One small randomised, double-blind, placebo-controlled trial in 44 healthy adults with poor sleep quality; the overall result was null and the microbiome finding is an exploratory subgroup interaction the authors themselves call a hypothesis needing confirmation in larger studies.RCT · Sci Rep

Across 38 placebo-controlled trials in 703 healthy adults, beetroot juice produced a small but significant improvement in VO2max/VO2peak (SMD 0.24). The gain was clearer in recreationally active people (SMD 0.38) and not clearly demonstrated in highly trained athletes (SMD 0.07).

⚠What it shows: Meta-analysis of 38 randomised and crossover trials, 703 healthy adults; effects are small, funnel-plot asymmetry suggested small-study bias and the pooled estimate shrank to 0.19 after adjustment. Certainty rated moderate.Systematic review · Front Nutr

In a crossover trial, 13 physically active men produced greater peak torque and relative peak power on isokinetic knee extension at 60 and 180 degrees per second after beetroot juice containing 800 mg of nitrate than after placebo. Nothing changed during the faster 240 degrees per second fatigue protocol, and muscle EMG and post-exercise lactate were unchanged.

⚠What it shows: One small randomised, double-blind, placebo-controlled crossover trial in 13 physically active young men using isokinetic dynamometry, not free-weight lifting; mechanisms unexplained and replication needed.RCT · Nutrients

Pooling 20 randomised trials in 745 college students, high-intensity interval training beat moderate-intensity continuous training for body weight (-1.23 kg), body fat percentage (-1.21%), waist circumference (-1.33 cm) and fat mass (-0.99 kg). It did not outperform steady cardio on cholesterol, triglycerides, fasting glucose or fasting insulin.

⚠What it shows: Meta-analysis of 20 RCTs, 745 healthy college students; differences are small in absolute terms, evidence quality was rated moderate to very low, and glucose and lipid outcomes were very low certainty.Meta-analysis · Front Endocrinol (Lausanne)

Pooling 12 randomised trials in adults with hypertension or elevated blood pressure, resistance training significantly lowered systolic pressure with a moderate effect and diastolic pressure with a smaller one. The authors caution it should be seen as complementary to medication, not a replacement.

⚠What it shows: Meta-analysis of 12 RCTs reporting standardised mean differences rather than mmHg, so the real-world size is unclear; without individual participant data the authors say they cannot show anyone shifted blood pressure category.Review · J Clin Med

A systematic review and meta-analysis in healthy adults links multi-session resistance exercise to improvements in global cognition, executive function, working memory and processing speed. Episodic memory showed no consistent benefit.

⚠What it shows: Meta-analysis with substantial between-study heterogeneity; interventions varied in duration, intensity and cognitive measures, and evidence in some domains was limited.Review · Ageing Res Rev

Older adults with mild cognitive impairment who practised tai chi and baduanjin three times a week for 12 weeks outscored controls by 2.04 points on the Montreal Cognitive Assessment and improved strength and mobility — though BMI and body fat didn't significantly change.

⚠What it shows: Single 12-week RCT in community-dwelling Chinese adults aged 60+; sample size not reported in the abstract and no long-term follow-up.Study · J Aging Phys Act

Young men taking protein with carbohydrate during their workouts gained no more muscle fibre size, lean mass or strength over 8 weeks than men taking carbohydrate alone with the same calories. The resistance training itself worked well in both groups, raising lean mass, strength, VO2max, fat oxidation and fibre cross-sectional area.

⚠What it shows: Small exploratory RCT: just 17 physically active young men (8 vs 9) over 8 weeks, which the authors say cannot exclude modest effects — a null result from an underpowered trial, not proof of no benefit.Study · Nutrients

A systematic review of 103 studies (9951 patients, 44 randomised trials) found that distending the shoulder capsule with fluid gave slightly more external rotation than a corticosteroid injection alone at 12 weeks, with no better pain or disability scores, and only marginal pain gains when added to physiotherapy. Technique varied enormously between clinics.

⚠What it shows: Systematic review with meta-analysis of 44 RCTs, but the significant findings were rated low or very low certainty with high heterogeneity (I2 up to 81.6%), and pooled techniques differed so widely that an optimal protocol could not be identified.Study · Br J Sports Med

Pooling 110 studies and 1,826 participants, blood lactate showed by far the largest acute post-exercise response (response ratio 4.43), ahead of IL-6, creatine kinase and LDH — but the authors stress none of these has been shown to be specific to fatigue.

⚠What it shows: Systematic review and random-effects meta-analysis of 110 articles, 129 experiments, 1,826 healthy participants, with very high heterogeneity (I² 87-98%); the attached LC-MS pilot involved only seven healthy young men and was exploratory.Review · Metabolites

In a small trial of 32 healthy older adults, some brain effects of a six-week probiotic — differences in resting-state functional connectivity and in anxiety and depression scores — were still detectable 4-6 weeks after intake stopped, while other changes had faded. The persistent differences appeared between the two probiotic formulations rather than against baseline.

⚠What it shows: Exploratory follow-up of one randomised, blinded, placebo-controlled trial in 32 self-reported healthy older adults (mean age 68.0 years); only 8 to 12 participants per group were analysed out of about 30 randomised each, brain function was an exploratory outcome of the parent study, and psychological measures were exploratory throughout.Study · Commun Med (Lond)