Study archive · page 8

Digital cognitive behavioural therapy for insomnia produced a moderate-to-large reduction in insomnia severity in serving military personnel and veterans across five pooled randomised trials, with improvements maintained for up to 12 months where follow-up data existed.

⚠What it shows: Effective on the evidence so far, but the base is thin: results pooled from just five trials of serving personnel and veterans, and dropout plus patchy reporting left the risk of bias unclear or high. The certainty of the evidence was judged moderate rather than strong.Meta-analysis · Sleep Med

Pooled trial evidence in perimenopausal and postmenopausal women with chronic insomnia found non-drug approaches, led by cognitive behavioural therapy for insomnia along with acupuncture, yoga and massage, improved sleep quality on more measures than medication did. Neither approach increased total sleep time.

⚠What it shows: A careful pooling of 17 trials in menopausal women, mostly rated good to excellent in quality. Sleep was judged largely by questionnaires, and no treatment added actual sleep time, so the gains rest on how sleep felt rather than how long it lasted.Meta-analysis · Sleep Med

Pooled data from 47 studies of 9684 fathers across 17 countries put average sleep at 398.29 minutes a night in the first 24 months after birth, with pooled sleep quality scores in the poor range and just over one night awakening on average.

⚠What it shows: A solid snapshot of a neglected group: results pooled from 47 studies of 9684 fathers, though the studies disagreed with each other a great deal on every measure. Most of the data comes from North America, where paternity leave is scarce, so fathers elsewhere may fare differently.Meta-analysis · Sleep Health

In a systematic review of six randomised trials with 180 exercising participants, probiotic and synbiotic supplements improved nine of twelve primary sleep outcomes, most consistently how well people felt they slept and in some cases how quickly they fell asleep.

⚠What it shows: Early days: a sweep of all the studies turned up just six trials totalling 180 exercising people, and the results were compared side by side rather than pooled, so better perceived sleep is a promising hint, not a verdict.Systematic review · J Int Soc Sports Nutr

Supplementing with the olive phenolics oleuropein, hydroxytyrosol and tyrosol modestly lowered total cholesterol, triglycerides and insulin across 14 pooled trials involving 594 people, though the effects were small and the trials often disagreed.

⚠What it shows: Suggestive rather than settled: results pooled from 14 trials totalling 594 people found small drops in cholesterol, triglycerides and insulin, but the trials disagreed with each other on several outcomes. The authors themselves call for more studies before drawing firm conclusions.Meta-analysis · Crit Rev Food Sci Nutr

In older people with sarcopenia, strength training combined with whey protein or HMB supplementation improved fat-free mass, grip strength, chair-stand and walking performance across seven pooled randomised trials. Whey plus training significantly improved grip strength; HMB plus training did not.

⚠What it shows: Promising rather than proven, by the authors' own admission: results pooled from just seven trials of 472 older adults with sarcopenia, several carrying a moderate to high risk of bias, and the subgroup results did not all point the same way.Meta-analysis · Medicine (Baltimore)

Higher vitamin D levels are associated with a lower risk of colon cancer and with lower overall mortality, according to a review of the evidence. Brief daily UV exposure is usually enough for the body to make vitamin D, while extended sunbathing or sunbed use is inadvisable because it raises skin cancer risk.

⚠What it shows: Reasonably solid on the sun advice, softer on cancer: the vitamin D links come from studies that followed people rather than testing anything, so they show association, not proof that vitamin D prevents cancer. This sweep of the evidence dates from 2010, its authors call for more research, and the breast cancer picture remains unclear.Systematic review · Dtsch Arztebl Int

Pooling 17 randomised trials in 1,440 people with sarcopenia, Chinese herbal medicine, alone or added to conventional care, improved muscle mass, grip strength and 6-metre walking speed, though the authors' own checks pointed to publication bias.

⚠What it shows: Interesting but shaky: results pooled from 17 trials in 1,440 people did show gains, yet the reviewers' own checks suggested trials with poor results may never have been published, which would inflate the benefits. The formulas tested also varied widely, so this supports the broad approach rather than any single product.Meta-analysis · Phytother Res

Weight-bearing and strength exercise modestly improved bone density at the hip and spine in postmenopausal women across pooled randomised trials, gains the authors estimate would cut 20-year osteoporotic fracture risk by roughly 11% at the femoral neck and 10% at the spine.

⚠What it shows: A dependable direction with a modest size: results pooled from 28 trial comparisons of postmenopausal women who were not regular exercisers, training for at least 24 weeks. The bone changes were small and the fracture figures are projections from those changes, not counted broken bones.Meta-analysis · BMC Musculoskelet Disord

Pooled results from three randomised trials in 995 women with menopause-related vasomotor symptoms show elinzanetant reduced the frequency and intensity of symptoms versus placebo, and improved quality of life and sleep. Adverse events were similar between the drug and placebo.

⚠What it shows: Convincing as far as it goes: results pooled from three trials in 995 menopausal women, and the trials pointed the same way on the main outcomes. Three trials is still a modest base, and long-term safety beyond them is not answered here.Meta-analysis · Eur J Obstet Gynecol Reprod Biol

A meta-analysis of six randomised trials found no significant advantage of high-intensity interval training over moderate-intensity continuous training in women with PCOS, across weight, waist size, fitness, glucose and insulin measures, blood lipids and hormones.

⚠What it shows: Take it as permission rather than proof: only six trials were pooled and the authors rate the certainty of the evidence low to very low, so a genuine difference between the two styles could still have been missed.Meta-analysis · Front Endocrinol (Lausanne)

In a pooled analysis of 14 randomised trials covering 1,230 menopausal women for sleep and 482 for urinary incontinence, exercise favoured better sleep quality, while the incontinence benefit came specifically from pelvic floor strengthening. The trials disagreed strongly with one another.

⚠What it shows: Promising in direction, shaky on size: results pooled from 14 trials covering 1,230 women for sleep quality and 482 for urinary incontinence, most rated low risk of bias. The studies disagreed with each other enormously, so read the result as which way things moved rather than by how much.Review · J Menopausal Med

Results pooled from randomised trials suggest both structured cognitive training and physical exercise modestly improve global cognitive function in people with mild cognitive impairment; cognitive training also showed weaker benefits for executive function and delayed memory.

⚠What it shows: Encouraging but modest: results pooled from a set of trials in people diagnosed with mild cognitive impairment, and the authors themselves call the memory and executive function findings weak. The trials varied enough in design that better standardised studies are still needed.Meta-analysis · J Alzheimers Dis

Across 12 pooled randomised trials, exercise delivered through virtual reality significantly reduced depression compared with control conditions, though the certainty of the evidence was graded very low.

⚠What it shows: Promising but shaky: 12 pooled trials pointed the same way and the result held up when individual studies were removed, but the studies disagreed with each other in who they enrolled and how, and the authors rate their confidence in the finding as very low.Meta-analysis · BMC Psychiatry

In a systematic review of seven randomised trials covering 5663 people with impaired glucose tolerance, lifestyle programmes cut the risk of developing diabetes, with relative risk reductions of 25% to 65% across trials. The protection tended to fade once the programmes ended, and the one trial that tracked deaths found no difference.

⚠What it shows: About as solid as prevention evidence gets on the headline effect: seven trials in 5663 high-risk people followed for three to six years. The programmes varied widely, the benefit shrank once they ended, and the single trial that tracked deaths found no difference.Meta-analysis · Metabolism

A network meta-analysis of 47 randomised trials in 2,355 sedentary adults found combined aerobic plus strength training beat any single exercise type for lowering blood pressure, while HIIT ranked best for cutting total cholesterol and Tai Chi topped the rankings for LDL cholesterol and glycated haemoglobin.

⚠What it shows: Solid as a broad map, less so on the fine print: pooled results from 47 trials in 2,355 sedentary adults, most judged at low risk of bias. The 'best exercise' rankings lean partly on comparisons between trials that never faced each other head to head, so treat them as a guide rather than a final league table.Meta-analysis · Endocrine

A systematic review found only 13 randomised trials of vitamin D for treating or preventing infections in humans, with the strongest signals pointing to tuberculosis, influenza and viral colds. Serious side effects from supplementation were rare, but the reviewers concluded the evidence justified further research rather than changes to care.

⚠What it shows: A state of the field rather than a verdict: a sweep of all the trials on this up to 2009 turned up just 13, and they differed so much in doses and patients that no firm conclusion was possible. Serious side effects were rare across the lot.Systematic review · Endocr Pract

In a sweep of 12 recent randomised trials, probiotics showed a consistent direction of benefit for recurrent urinary tract infections, Helicobacter pylori eradication and respiratory infection symptoms, but the certainty of the evidence was rated low to moderate.

⚠What it shows: Promising direction, thin proof: a sweep of all the recent studies found only 12 small trials totalling 1200 people, with designs too varied to pool properly. The authors rate the certainty of the evidence low to moderate and say routine recommendations are not yet justified.Systematic review · Future Microbiol

Pooled across 10 studies of 1,005 adults with type 2 diabetes, VR-based physical activity and self-management programmes were followed by a 0.50 point drop in HbA1c, but they were no better than traditional approaches such as routine care or in-person education.

⚠What it shows: Decent pooled evidence but not the last word: results came from 10 studies covering 1,005 adults with type 2 diabetes, yet only five assigned people at random and four simply measured the same group before and after, which can flatter a result. Whether VR truly matches good in-person care still needs stronger trials.Meta-analysis · J Med Syst

Pooling 26 randomised trials in 4129 older adults, multicomponent digital health programmes raised daily steps by 822.8 per day and moderate-to-vigorous activity by 45.9 minutes a week. Sedentary time, light activity and total activity showed no statistically significant change.

⚠What it shows: Reasonably strong on brisk activity, weaker on steps: results pooled from 26 trials in 4129 older adults, with certainty rated moderate for moderate-to-vigorous activity and low for everything else. The trials disagreed a lot and many combined the technology with human coaching, so the app or tracker alone cannot take the credit.Meta-analysis · J Med Internet Res

Across 45 randomised trials with 7788 obese adults who had lost at least 5% of their body weight, behavioural programmes targeting both diet and physical activity reduced weight regain by an average of 1.56 kg over 12 months compared with controls.

⚠What it shows: About as sturdy as this field gets: results pooled from 45 trials in 7788 adults who had already lost at least 5% of their body weight, though the programmes varied a lot and the average benefit was small.Meta-analysis · BMJ

Pooled results from 34 randomised trials found that applying photobiomodulation, a light-based therapy, to muscles before exercise improved muscle endurance and speeded recovery of strength and muscle damage markers. The benefit appeared in athletes and physically inactive people but not in moderately active people.

⚠What it shows: Worth a raised eyebrow rather than a rush to buy a laser: results pooled from 34 trials show modest average benefits, but the strange split, helping athletes and the sedentary yet not regular exercisers, comes from comparisons between subsets of trials that are easy to over-read. The authors themselves note the field is still controversial.Meta-analysis · Lasers Med Sci

Across five randomised trials in 834 people living with both chronic low back pain and depressive symptoms, yoga combined with education showed the largest improvements in pain and mood versus usual care. The certainty of the evidence was rated low to very low.

⚠What it shows: A pointer, not a verdict: just five trials covering 834 people fed the comparison, and the authors rate their own certainty as low to very low. The yoga result rests on a thin slice of that evidence.Meta-analysis · Sci Rep

Pooled data from 31 cohort studies covering 414,477 people found that men consuming around 10 g of alcohol a day, roughly one small drink, had a 14% higher risk of developing hypertension, while women at the same intake showed no raised risk and no protective effect. Risk climbed in a straight line with intake in white, black and Asian groups, with black people showing the steepest rise per 10 g of daily alcohol.

⚠What it shows: About as solid as this question gets without a trial: results pooled from 31 studies that followed 414,477 people rather than testing drinking on them, so other habits that travel with alcohol could still explain part of the link. The sex and race comparisons rely on fewer studies each, which makes those specific numbers less certain than the overall pattern.Meta-analysis · Nutr Metab Cardiovasc Dis