Study archive · page 2

Across 12 studies of 859 team-sport athletes, adding blood flow restriction to resistance training produced small additional gains in muscle hypertrophy (SMD 0.32) and strength (SMD 0.42) compared with resistance training alone. Jumping and sprinting performance showed no statistically significant difference.

⚠What it shows: Meta-analysis of 12 studies in 859 team-sport athletes; effects were small, heterogeneity was low but protocols varied, and the review cannot say which cuff pressures, loads or programme lengths drive the result.Study · Front Physiol

In a dose-response meta-analysis of 84 trials (3008 non-athletes, 1059 athletes) with clinically confirmed hamstring shortening, static stretching improved range of motion versus no intervention (active knee extension 9.03°, passive knee extension 8.44°, straight leg raise 12.11°), while proprioceptive neuromuscular facilitation gained 10.10° on active knee extension and held a small 3.18° advantage over static stretching. A brief 1.5-minute single session did not exceed the 12° minimal detectable change, whereas repeated stretching over at least 1 week totalling 70 minutes was more likely to.

⚠What it shows: Meta-analysis of 84 randomised and crossover trials in healthy people with clinically confirmed hamstring shortening, but most studies carried high or some-concerns risk of bias and certainty ranged from very low to high; the dose-response curves are modelled, not tested head-to-head.Review · Ann Phys Rehabil Med

An exploratory multilevel meta-analysis of 11 trials in 303 generally healthy, mostly young adults found that adding intermittent fasting — usually time-restricted eating — to resistance training produced broadly similar adaptations to resistance training alone (g = 0.09, 95% CI −0.23 to 0.40). There were no clear differences in maximal strength, local muscular endurance, hypertrophy-related outcomes or explosive and high-speed performance.

⚠What it shows: Exploratory meta-analysis of just 11 randomised or crossover trials totalling 303 mostly young participants of varied training status, with interventions running from 4 weeks to 12 months; the confidence interval spans zero in both directions, so it shows no clear difference rather than proven equivalence.Review · Front Nutr

Pooling 60 controlled trials in postmenopausal women, resistance training of at least four weeks significantly lowered systolic (g=0.68) and mean blood pressure (g=0.68), with smaller reductions in resting heart rate (g=0.30) and diastolic pressure (g=0.37). Echocardiographic studies reported no adverse structural changes to the heart.

⚠What it shows: Systematic review and meta-analysis of 60 randomised and non-randomised controlled trials in postmenopausal women; effects on heart rate and diastolic pressure were small, and the vascular and heart-rate-variability findings were only described narratively, not pooled.Study · J Appl Physiol (1985)

A systematic review of 11 studies in 176 athletes and physically active people reported that pomegranate supplementation improved whole-body strength, feelings of vitality, muscle fatigue and soreness, vessel diameter and blood flow, and antioxidant markers, with faster recovery of post-training muscle-damage markers. Doses ranged from 50 ml to 500 ml over 7 days to 2 months.

⚠What it shows: Narrative systematic review of 11 studies published 2010-2018 with just 176 participants in total, 155 of them male; no meta-analysis was performed, and supplement form, dose (50 ml-500 ml) and duration (7 days-2 months) varied so widely that results cannot be pooled or a dose inferred.Systematic review · Int J Vitam Nutr Res

Across 19 randomised trials with 732 participants using caffeine doses of 2 to 9 mg/kg, performance benefits were most consistent in people carrying the CYP1A2 AA genotype, smaller in AC carriers, and null or even detrimental in those with the CC genotype — though the studies did not all agree.

⚠What it shows: Systematic review of 19 RCTs in 732 athletes and physically active people published 2012-2024, with no meta-analysis and no consensus between studies; 84% of the included trials were rated as raising 'some concerns' for risk of bias and only two were low risk.Systematic review · Nutr Res

Pooling 48 randomised placebo-controlled trials (689 healthy adults), low caffeine doses of ≤3 mg·kg-1 shortened aerobic time-trial completion time (SMD -0.27), while moderate doses of 4-6 mg·kg-1 produced a larger and more consistent effect (SMD -0.52). No eligible trial reporting time-to-completion tested doses above 6 mg·kg-1.

⚠What it shows: Meta-analysis of 48 randomised placebo-controlled trials in healthy adults aged 18-59 using acute oral anhydrous caffeine, but only 689 participants in total, effects pooled as standardised mean differences across varied time-trial protocols, and no trials of doses above 6 mg·kg-1 met the inclusion criteria, so the top of the dose range remains untested for this outcome.Meta-analysis · Nutrients

In a meta-analysis of 12 randomised trials (459 older adults with sarcopenia), adding branched-chain amino acid-rich supplements to resistance training improved only skeletal muscle index (SMD 0.337, p = 0.029). Grip strength, walking speed, chair-stand performance, SPPB score, knee extension strength, total lean mass and fat mass showed no significant advantage.

⚠What it shows: Meta-analysis of 12 randomised trials with only 459 participants in total, all older adults with sarcopenia; the one positive outcome had a confidence interval reaching close to zero and several functional measures pointed in inconsistent directions, so this cannot be read across to younger or healthy trainees.Meta-analysis · Arch Gerontol Geriatr

Pooling 51 echocardiographic studies of 74 632 people with hypertension, 36.6% had left ventricular hypertrophy; men were less likely than women to have it (OR 0.62), and among those affected concentric hypertrophy was almost twice as likely as eccentric (OR 1.94).

⚠What it shows: Large meta-analysis of 51 observational echocardiographic studies in 74 632 hypertensive patients; prevalence estimates depend on how each study defined hypertrophy and on differing clinic populations, and this design shows association, not cause.Study · J Hypertens

In a systematic review of randomised trials in working-age adults followed for a year or more, lifestyle interventions covering diet, exercise, smoking cessation and alcohol reduction cut morbidity and mortality in secondary prevention, and multifactorial programmes also lowered cholesterol; in primary prevention, risk factors fell most efficiently when several behaviours were targeted at once. Effect sizes were heterogeneous with wide confidence intervals.

⚠What it shows: Systematic review of randomised trials — 21 single-factor and 21 multifactorial interventions, all with at least one year of follow-up in working-age adults — but published in 2000, with heterogeneous effect sizes, wide confidence intervals and inconsistent descriptions of what each intervention actually involved.Meta-analysis · Ann Med

Pooling trials of at least 4 weeks, all types of strength training significantly lowered both systolic and diastolic blood pressure; in people not taking medication, isometric exercise reduced systolic pressure more than dynamic exercise, while effects were no longer significant in programmes running beyond 20 weeks or training fewer than three times per week.

⚠What it shows: Systematic review with meta-analysis limited to English-language studies of at least 4 weeks from two databases; the abstract does not report the number of trials or participants, populations and training types were mixed, and the 'optimal dose' conclusion comes from subgroup comparisons rather than a head-to-head trial.Meta-analysis · J Hypertens

In a meta-analysis of eight acute randomised crossover trials in 159 sedentary adults at metabolic risk, brief 'exercise snacks' that interrupt sitting reduced post-meal glucose iAUC by 5.69 mmol·h/L versus uninterrupted sitting. Heterogeneity was high (I² = 85%), the prediction interval crossed zero, and the insulin reduction was not statistically significant.

⚠What it shows: Only eight acute laboratory crossover trials, 159 sedentary adults at metabolic risk, with the glucose result resting on five trials; heterogeneity was very high, the prediction interval crossed the null, and no long-term outcomes were measured.Meta-analysis · Front Endocrinol (Lausanne)

A network meta-analysis of 24 phase 3 trials lasting at least 52 weeks found liraglutide, semaglutide and tirzepatide all reduced body weight, waist circumference and BMI versus placebo, with tirzepatide 15 mg ranked highest and liraglutide 3.0 mg producing the smallest weight loss. Semaglutide 7.2 mg ranked above semaglutide 2.4 mg on every efficacy endpoint, but the average incremental benefit was around 2-3 percentage points.

⚠What it shows: Network meta-analysis of 24 phase 3 RCTs in adults with BMI of 25 or above; most comparisons between doses are indirect rather than head-to-head, heterogeneity for efficacy outcomes was considerable, and few between-dose differences reached statistical significance. Says nothing about people outside these trial populations.Study · J Endocrinol Invest

A network meta-analysis of 29 randomised trials in 1,301 middle-aged and older adults with type 2 diabetes found high-intensity resistance training produced the largest HbA1c reduction versus usual care (MD -0.62), ranking top of 10 exercise interventions, while moderate-intensity resistance training led on fasting glucose and high-intensity aerobic training drove the biggest gains in peak oxygen uptake. No intervention significantly reduced resting heart rate.

⚠What it shows: Network meta-analysis of 29 RCTs and 1,301 participants with type 2 diabetes; only the HbA1c result for high-intensity resistance training reached moderate certainty, the fasting glucose and blood pressure networks were rated low or very low, and the VO2peak network was sparse and described as exploratory.Meta-analysis · Front Public Health

In a meta-analysis of 22 diet-and-exercise interventions, overweight participants lost 1.1 kg less over a year than those with class-I obesity and 1.5 kg less than those with class-II obesity — but percentage weight change did not differ significantly across BMI classes.

⚠What it shows: Meta-analysis of 22 interventions restricted to healthy Caucasian adults with BMI 25-40; absolute between-class differences were small (1.1-1.5 kg) and findings may not generalise to other ethnic groups or higher BMIs.Meta-analysis · Int J Behav Med

In a meta-analysis of randomised trials in resistance-training adults with a mean age of at least 60 years, 10-15 g of milk protein was sufficient to augment gains in lean body mass, while four out of five studies of whey protein at the same or higher doses showed a negative effect versus control supplementation (-0.49 kg, 95% CI -0.69 to -0.29). Improvements in muscle strength appeared only at protein doses of 22 g or more.

⚠What it shows: Twenty studies were reviewed and 17 pooled, all in resistance-training older adults with a mean age of at least 60; the striking whey result rests on just five studies, the review is a between-study comparison rather than a head-to-head trial, and the authors note that adequate daily protein intake dilutes any supplement effect.Meta-analysis · Nutrients

In a meta-analysis of 36 observational studies of university students, higher sedentary behaviour was significantly associated with greater depression risk (pooled OR 1.60, 95% CI 1.26 to 2.04), while the apparently protective association for higher physical activity (OR 0.53, 95% CI 0.27 to 1.07) did not reach statistical significance. Pooled correlations between activity level and depressive symptoms were inverse but extremely heterogeneous.

⚠What it shows: Meta-analysis of 36 mostly cross-sectional observational studies in university students (30,307 people in the correlation analysis), with substantial to extreme heterogeneity (I2 94.4% and 99.2%), samples skewed towards China, and only 2 studies behind the sedentary estimate. Association only — it cannot show which comes first.Review · Front Psychol

Aerobic exercise programmes that raised older adults' cardiorespiratory fitness by roughly 14% coincided with improvements in cognitive function, with the largest effects on motor function (effect size 1.17) and auditory attention (0.50). Most individual comparisons, however, showed no significant effect, and the review could not confirm the fitness gain itself drives the cognitive benefit.

⚠What it shows: Cochrane review of 11 RCTs in adults over 55 without cognitive impairment (search to December 2005); the majority of comparisons were non-significant, and the data are insufficient to show cardiovascular fitness is the mechanism.Meta-analysis · Cochrane Database Syst Rev

In a meta-analysis of 16 randomised trials, online pelvic floor muscle training reduced postpartum urinary incontinence incidence (RR 0.56, 95% CI 0.38 to 0.82) and modestly eased symptom severity (SMD -0.25), while in older women severity did not significantly improve (SMD -0.12) and the only gain was better adherence. Quality of life did not differ significantly in either group.

⚠What it shows: Systematic review and meta-analysis of 16 RCTs of remotely delivered pelvic floor training in women with urinary incontinence; moderate-to-high heterogeneity between trials, and benefits differed sharply by life stage.Meta-analysis · Int J Nurs Pract

Across 25 randomised trials in 1,107 women with polycystic ovary syndrome, exercise training significantly reduced fasting insulin (SMD -0.28), HOMA-IR (SMD -0.50), triglycerides (-2.70 mg/dl), total cholesterol (-3.74 mg/dl) and CRP (SMD -0.61) versus non-exercise controls, while fasting glucose, LDL and HDL did not shift overall. In the network comparison, aerobic training and HIIT looked best for glucose handling, aerobic training for inflammation, and combined training for lipids.

⚠What it shows: Systematic review with pairwise and network meta-analyses of 25 RCTs totalling 1,107 women with PCOS aged on average 18 to 34; several effect sizes were small, absolute lipid changes were modest, and network rankings rest on indirect comparisons.Meta-analysis · BMC Womens Health

Pooling 15 randomised trials in 954 adults, fasting and 20%-30% calorie restriction had largely neutral effects on reproductive hormones, with no significant change in testosterone (fasting WMD −0.88 ng/mL, p = 0.24; calorie restriction −0.02 ng/mL, p = 0.65) and no significant effects on SHBG, DHEA-S, FSH, LH, oestradiol, AMH, prolactin or progesterone. The one exception was free testosterone, which fell by 4.9 pg/mL, a result driven by some of the smaller trials.

⚠What it shows: Systematic review and meta-analysis of 15 randomised trials in 954 adults with GRADE assessment, but there was significant heterogeneity for testosterone, the free-testosterone signal rested on smaller trials, and the authors did not assess symptoms, so a neutral hormone panel is not the same as clinical neutrality.Review · Food Sci Nutr

In a meta-analysis of cohort studies, every 10 µg/m3 increase in long-term ozone exposure was associated with modestly higher cardiovascular mortality (RR 1.0105) and cardiovascular morbidity (RR 1.0578). The mortality association was clearer for warm-season ozone (RR 1.0186) than for annual average exposure (RR 0.9971), and estimates were higher in females (RR 1.1135) than males (RR 1.0707).

⚠What it shows: Meta-analysis of observational cohort studies, so it shows association and not causation; the authors describe the long-term ozone evidence as sparse, the male and Asia-region estimates had confidence intervals crossing no effect, and most data came from Europe and the USA.Review · China CDC Wkly

In a network meta-analysis of 23 randomised trials in knee osteoarthritis, celecoxib ranked highest for pain on the visual analogue scale (mean difference -0.48, 95% CI -0.78 to -0.17) and for WOMAC total score, followed by glucosamine/chondroitin; krill oil ranked top on the WOMAC pain subscale, while probiotics and metformin had the most favourable safety rankings. Certainty for most of these adjuvant comparisons was low or very low.

⚠What it shows: Network meta-analysis of 23 RCTs with no direct head-to-head trials between the adjuvant options; GRADE certainty for most supplement comparisons was low or very low, mainly from severe imprecision, so the rankings are provisional rather than a league table.Systematic review · Front Med (Lausanne)

GPS-enabled physical activity programmes produced small but significant reductions in body weight and body fat percentage across nine randomised trials in adults, and adding dietary components did not consistently beat activity-only interventions. Short-term programmes of three months or less showed the greater weight effects.

⚠What it shows: Meta-analysis of nine RCTs (1363 adults, mostly female); effects were small, the body-fat result was highly heterogeneous across studies, and robustness was only supported for short-term outcomes.Meta-analysis · Nutrients