Study archive · page 19

Pooling 39 randomised trials in 3,236 adults with cardiometabolic risk factors, an early time-restricted eating window reduced fasting blood glucose by 3.64 mg/dL and insulin resistance versus control, and early windows, especially combined with exercise, consistently ranked best across outcomes.

⚠What it shows: Promising but not settled: results pooled from 39 trials in 3,236 adults with cardiometabolic risk factors. The certainty of the evidence ranged from very low to moderate, and the rankings rest on a limited number of trials, so they could shift as more arrive.Systematic review · Front Nutr

In a Cochrane network meta-analysis of 18 studies with 15,282 adults with obesity and type 2 diabetes, Roux-en-Y gastric bypass was linked to 16.95% greater long-term weight loss than lifestyle and medical treatment, and it may also lower HbA1c and make partial diabetes remission more likely. For most other outcomes, including complete remission, serious adverse events and mortality, the evidence remained very uncertain.

⚠What it shows: Cochrane rigour applied to thin raw material: the weight-loss comparison rests on 8 trials with just 491 people, and the reviewers themselves rate most findings uncertain or very uncertain because the trials were small and serious events rare. The direction favours surgery; the size and safety picture are still genuinely open.Meta-analysis · Cochrane Database Syst Rev

A meta-analysis of 68 studies found cardiorespiratory fitness is markedly lower in people with type 1 and type 2 diabetes than in people without, with the larger gap in type 2. High physical activity levels partially closed the gap, while sedentary lifestyles and diabetes complications were linked to a wider one.

⚠What it shows: A robust pattern, but with a limit: the 68 pooled studies compared groups of people rather than testing an intervention, so this cannot prove diabetes causes the fitness gap or that activity alone would close it. The links to blood sugar control and body weight are associations, not demonstrated mechanisms.Meta-analysis · Med Sci Sports Exerc

Pooling 43 randomised trials with 3552 adults, aerobic training, resistance training and combined training all reduced subcutaneous abdominal fat, with combined training showing the largest reduction. Head to head, aerobic and resistance training did not differ significantly from each other.

⚠What it shows: A broad sweep of the evidence: results pooled from 43 trials in 3552 adults, most of them women, all point to exercise trimming the fat layer under the belly skin. The trials varied enough that outliers had to be removed, and the search only ran to late 2018, so newer trials are not counted.Meta-analysis · Adv Nutr

Pooled results from nine randomised trials in 2,476 postmenopausal women found hormone therapy did not improve muscle strength, and the result held regardless of treatment type, muscle group or treatment duration.

⚠What it shows: As solid as a null gets: results pooled from nine trials where 2,476 postmenopausal women were randomly assigned to hormone therapy, a placebo or nothing, and strength did not improve. The trials disagreed with each other a fair amount, so a very small effect in some settings cannot be completely ruled out.Meta-analysis · Menopause

In pooled results from 19 studies, pelvic floor muscle training, alone or combined with electrical stimulation, biofeedback or abdominal training, reduced the severity of postpartum urinary incontinence from before to after treatment, though training was not significantly better than control or education groups.

⚠What it shows: Solid but messier than the standard advice suggests: results pooled from 19 studies of postpartum women showed clear improvement over the course of training, yet the direct comparison against control or education groups did not reach significance. That leaves open how much of the improvement is natural recovery after childbirth.Meta-analysis · BJOG

Pooled data from 21 studies covering 5,589,862 pregnancies links rising daily maximum temperatures to adverse pregnancy and birth outcomes, with a 5% risk increase at 30.55 °C, 10% at 34.11 °C and 20% at 39.07 °C.

⚠What it shows: Strong signal, rough edges: risk clearly climbs with heat across results pooled from 21 studies of 5,589,862 pregnancies, but the underlying studies measured heat inconsistently and a single person handled the quality checks, so the exact temperature cutoffs are best read as estimates rather than hard lines.Meta-analysis · Nat Commun

Across 61 randomised trials in 4,100 postmenopausal women, aerobic exercise lowered systolic blood pressure by 4.5 mm Hg and diastolic by 2.4 mm Hg, raised HDL cholesterol by 2.4 mg/dL, and cut LDL by 3.6 mg/dL and triglycerides by 7.7 mg/dL. It also modestly reduced anxiety scores, though effects on depression were not statistically significant.

⚠What it shows: Strong evidence: results pooled from 61 trials in 4,100 postmenopausal women, which makes the blood pressure and cholesterol effects among the better-tested in this field. What it cannot yet say is whether those shifts translate into fewer heart attacks or strokes over the long run.Meta-analysis · Menopause

Aerobic training done with blood flow restriction cuffs produced greater muscle growth and strength gains than the same training without them, in pooled results from 7 trials of healthy adults aged 18 to 60.

⚠What it shows: Promising but not settled: the pooling covered only 7 trials in healthy adults aged 18 to 60, and most carried a high risk of bias. The benefit is plausible, but its true size could shrink as better studies arrive.Meta-analysis · J Strength Cond Res

In healthy older adults, eccentric resistance training produced slightly greater strength gains than traditional resistance training across 11 randomised trials, while muscle power, muscle size and functional capacity improved to a similar degree with both approaches.

⚠What it shows: A real but small edge: pooled results from 11 trials in healthy older adults favoured eccentric training for strength only, and both training styles produced large improvements in strength and everyday function on their own, so neither is a wrong choice.Meta-analysis · Ageing Res Rev

Pooled results from 12 randomised trials in 538 sarcopenic older adults show resistance training moderately improves grip strength and physical performance, yet did not significantly increase muscle mass. Twice-weekly moderate-intensity training, and programmes that included pulling exercises, predicted the biggest strength gains.

⚠What it shows: Fairly firm on strength: results pooled from 12 trials totalling 538 older adults with sarcopenia all point the same way. The trials were small, so the flat muscle-mass result may simply mean bigger or longer studies are needed to detect a change.Meta-analysis · BMC Geriatr

Pooling 36 trials in 1,552 people doing resistance training, whey protein added the most lean mass versus placebo, about 1.24 kg on average, with benefits plateauing around 30 to 35 g per day; in adults aged 65 and over, whey's edge narrowed while casein climbed the rankings.

⚠What it shows: Fairly solid for whey's edge: results pooled from 36 trials in 1,552 people lifting for at least six weeks. The dose plateau is an exploratory estimate, and the older-adult comparison rests on just nine trials, so those two details deserve the most doubt.Systematic review · Front Nutr

In adults under 50, adding creatine to resistance training produced a very small 1.19% reduction in body fat percentage but no change in absolute fat mass, across pooled results from 12 trials involving 266 people.

⚠What it shows: Solid as pooled evidence goes: 12 trials in 266 adults under 50 all point the same way, but the percentage change was tiny and several of the authors have financial ties to creatine companies, which is worth knowing when reading it.Meta-analysis · Nutrients

High-intensity exercise modestly reduced depressive symptoms across nine randomised trials in 514 people with depression, with high-intensity aerobic work the most effective modality and adults aged 60 or older benefiting most, though several depression rating scales showed no significant difference.

⚠What it shows: Treat this as a modest and slightly shaky signal: pooled results from nine trials in 514 people with depression pointed the right way overall, but several depression scales showed no clear benefit and the underlying studies disagreed with each other.Meta-analysis · Front Public Health

A simulation model calibrated to US survey data estimated that by age 40, about 1 in 7 male and 1 in 5 female individuals would have had an eating disorder, with annual prevalence peaking around age 21 and 95% of first cases beginning by age 25. The same model estimated that treating everyone with an eating disorder could avert 70.5 deaths per 100 000 people by age 40, compared with 41.7 deaths averted under current treatment coverage.

⚠What it shows: Treat these as modelled estimates rather than a head count: researchers simulated 100 000 people from birth to age 40 and tuned the model to US survey data from 2007 and 2011, and the uncertainty ranges around every figure are wide. It shows what the available data imply, not what was directly observed in a real population.Study · JAMA Netw Open

Across 29 randomised trials in 1,633 adults aged 60 and over, probiotic, prebiotic and synbiotic supplements raised beneficial gut bacteria such as Bifidobacterium, increased microbial diversity and lowered several inflammatory markers.

⚠What it shows: Encouraging but indirect: results pooled from 29 trials in 1,633 people aged 60 and over show shifts in bacteria and blood markers, not fewer illnesses or better strength or function. Different supplement types did different things, so no single product is proven here.Meta-analysis · Nutr J

In pooled trials of people with type 2 diabetes, microbiome therapies trimmed waist circumference by just 0.97 cm on average, and only probiotics reached significance at 0.62 cm, while blood sugar measures improved more consistently.

⚠What it shows: Believable but tiny: results pooled from 15 trials covering 443 treated and 387 control participants with type 2 diabetes, and the waist change is under a centimetre. The blood sugar improvements were more consistent than the waist ones, and prebiotics and synbiotics actually raised fasting insulin.Meta-analysis · J Acad Nutr Diet

A meta-analysis of 39 randomised trials in 2,237 people found probiotic fermented milk products lowered LDL cholesterol by 7.34 mg/dL and total cholesterol by 8.30 mg/dL on average, with no significant effect on HDL cholesterol or triglycerides.

⚠What it shows: Reasonably firm but modest: results pooled from 39 trials in 2,237 people, and the drop is small. The products, bacterial strains and doses varied widely between trials, so it is not clear which fermented milk, or which bacteria, actually does the work.Meta-analysis · Nutr Metab Cardiovasc Dis

Progressive muscle relaxation improved subjective sleep quality in adults across 14 trials involving 957 people, with clear benefits in adults under 55 but uncertain effects in those aged 55 and older, and doing more sessions was not linked to better sleep.

⚠What it shows: Encouraging but not airtight: pooled results from 14 trials in 957 people showed better sleep, but the studies disagreed with each other a lot and sleep was rated by questionnaire rather than measured by monitors, so the true size of the benefit is uncertain.Meta-analysis · Front Public Health

In a multi-cohort human exercise study comparing sprint-interval exercise with moderate-intensity exercise, the intensity of the session changed which proteins and metabolites appeared in the blood, in both untrained and trained participants. Muscle fibres and fat cells showed the broadest secretory and gene-expression changes, and some of the intensity-dependent proteins were linked to cardiometabolic health and disease in a large plasma database.

⚠What it shows: Early mechanism work rather than a health result: a human study in untrained and trained participants that mapped blood proteins and metabolites after sprint-interval and moderate-intensity sessions, with no disease or fitness outcomes measured. The cardiometabolic links come from matching those proteins against a large database, so nobody has yet shown that changing your intensity changes your health.Study · Cell Rep Med

Across 14 trials with 2797 older adults with osteoporosis or osteopenia, exercise training that mixed balance, resistance and aerobic work roughly halved the rate of falls and significantly reduced fear of falling.

⚠What it shows: Encouraging but shakier than it looks: results pooled from 14 trials in 2797 older adults with weakened bones, yet all but one trial had a high risk of bias and the studies disagreed with each other substantially. The direction of benefit is convincing; the exact size is not.Meta-analysis · J Geriatr Phys Ther

Pooled results from 13 randomised trials found that aerobic exercise programmes did not significantly increase long-term smoking quit rates, although quitters who exercised improved their aerobic fitness (VO2max or VO2peak). Not a single trial has tested exercise for vaping cessation.

⚠What it shows: A believable null: pooling 13 trials found no clear boost to long-term quit rates, even though two of the strongest trials, using three vigorous supervised sessions a week, did produce more quitters. Six of the 13 trials were low quality, and nobody has tested this for vaping at all.Meta-analysis · Int J Environ Res Public Health

Pooled data from 15 studies shows fasts of 24 hours or longer are associated with marked drops in the active thyroid hormones FT3 and total T3 and a moderate fall in TSH, while FT4 holds steady. That pattern points to reduced conversion of T4 into T3 rather than a shutdown of thyroid hormone production.

⚠What it shows: Solid on direction, fuzzy on size: results pooled from 15 studies of adults fasting at least 24 hours agreed that T3 falls while T4 holds, but the studies disagreed considerably on how big the FT3 drop was. None of this covers shorter intermittent fasting windows.Meta-analysis · Front Endocrinol (Lausanne)

In data from 353 locations across China, hot dry days were linked with a 10.18% excess rate of non-accidental deaths, compared with 3.21% for hot humid days, and high heat combined with low humidity raised risk by more than the sum of each alone.

⚠What it shows: Strong signal, not proof: researchers tracked death records and weather across 353 locations in China rather than testing anything, so other differences between places and days could shape the pattern. Whether the same holds in climates outside China is unknown.Meta-analysis · Environ Int