Study archive · page 3

In a 26-week randomised trial of 786 adults with abdominal obesity, adding one avocado a day reduced total LDL particle concentration by 49.1 nmol/L versus a habitual diet, with no change in LDL particle size, other lipoprotein particles or apolipoproteins A or B.

⚠What it shows: Ancillary analysis of one RCT, n=786 adults with abdominal obesity; lipoprotein particles were exploratory outcomes, and ApoB and every other particle measure did not change — this is a shift in one marker, not a demonstrated reduction in heart disease.Study · J Clin Lipidol

Pooling 18 cohorts and 65 randomised trials in healthy people, the highest habitual anthocyanin intakes were associated with 26% lower cardiovascular disease incidence, 18% lower myocardial infarction and 11% lower type 2 diabetes risk, while trials using at least 50 mg/d improved flow-mediated dilation (net change 1.41% with chronic intake) and left lipids, blood pressure and glucose unchanged.

⚠What it shows: Systematic review and meta-analysis: the risk reductions come from prospective cohorts, which show association not causation and can be confounded by wider diet quality, while the trial evidence covers surrogate biomarkers such as vessel function rather than heart attacks or deaths. The review was part-funded by a blueberry industry council, and one author advises its grant committee, though the funder is stated to have had no role in study selection or analysis.Meta-analysis · Am J Clin Nutr

In a meta-analysis of randomised trials in people with overweight or obesity, hypocaloric low-carbohydrate diets were no better than low-fat diets for blood pressure, CRP, HDL, total cholesterol or body composition; they lowered triglycerides by 19.94 mg/dL but raised LDL by 20.00 mg/dL. Only very-low-carbohydrate versions significantly reduced flow-mediated dilation, a measure of blood-vessel function, by 2.12%.

⚠What it shows: Systematic review and meta-analysis of randomised trials in adults with overweight or obesity, with only 10 studies (n = 475) reporting the vascular outcome; diets were calorie-restricted, trial lengths and carbohydrate definitions varied, and the overall low-carb effect on vessel function was not statistically significant.Meta-analysis · Eur J Clin Nutr

Among 2009 cognitively unimpaired older adults aged 50 to 90, those with depression had a smaller CA23DG hippocampal subfield than those without, independent of brain amyloid and tau; within the depressed group, people using antidepressants had smaller CA1 and CA23DG volumes than those who did not.

⚠What it shows: Cross-sectional MRI comparison inside one large observational cohort (630 with depression, 1379 without), so it cannot show that depression causes tissue loss; the antidepressant difference is unrandomised and may reflect more severe or longer-standing depression rather than any effect of the drugs.Study · Transl Psychiatry

An environment quietly engineered to push people into an expansive rather than contractive posture reliably increased risk-taking on the balloon analogue risk task, an effect that held after accounting for discomfort with the posture and awareness of the manipulation. Effects on self-reported pride were partially mediated by participants' measured neck angle.

⚠What it shows: A single laboratory experiment using an incidental posture manipulation; the abstract reports no sample size or participant details, outcomes were self-report plus one computerised risk task, and the suggestion that standing desks or ergonomic setups matter in daily life is the authors' extrapolation rather than something they tested.Study · Br J Psychol

A Cochrane review of 11 randomised trials in people over 55 without cognitive impairment found that aerobic programmes raising cardiorespiratory fitness by roughly 14% coincided with better motor function (effect size 1.17), auditory attention (0.52), delayed memory (0.50) and cognitive speed (0.26) — though most comparisons showed no significant difference.

⚠What it shows: Cochrane systematic review of 11 RCTs in adults over 55 without known cognitive impairment, with a literature search dating to December 2005; the majority of comparisons yielded no significant result, the delayed memory effect rests on a single study, and the data cannot show that fitness gains caused the cognitive gains.Meta-analysis · Cochrane Database Syst Rev

Across 73 progressive resistance training trials in older adults (18 with blinded assessors, 55 unblinded), trials whose outcome assessors were blinded reported smaller lower-limb strength effects, a difference of -0.80 (95% CI -1.35 to -0.25) and -0.65 after adjusting for intention-to-treat analysis, suggesting unblinded trials exaggerated the gains.

⚠What it shows: Retrospective meta-regression of 73 resistance training trials in older adults published before 2007, with only 18 of them using blinded assessors; it compares trials rather than randomising blinding, so other differences between those trials could contribute, and the findings may not transfer to younger populations or newer trials.Meta-analysis · Am J Phys Med Rehabil

Pooling the only three eligible trials, totalling 37 untrained adults, blood flow restriction combined with neuromuscular electrical stimulation produced a medium but statistically non-significant effect on lower-limb muscle growth compared with control (d = 0.64, 95% CI -0.37 to 1.65, p = 0.176).

⚠What it shows: Systematic review and meta-analysis, but the entire evidence base is three studies in 37 untrained healthy adults aged 18-64, lower limb only, interventions of at least 14 days; the pooled effect was not statistically significant and the confidence interval crosses zero.Meta-analysis · Sci Rep

In a meta-analysis of nine randomised controlled trials in 1,166 older adults, tai chi reduced Pittsburgh Sleep Quality Index scores by an average of 1.53 points; weekly totals of 150-300 min produced a significant improvement, while less than 150 min did not.

⚠What it shows: Meta-analysis of nine RCTs (1,166 older participants) using self-reported sleep questionnaires, with high statistical heterogeneity (I2 = 78%) and no possibility of blinding participants to an exercise they are doing; the dose findings come from subgroup comparisons, not from trials designed to test dose.Meta-analysis · Eur Geriatr Med

A Cochrane review of 17 randomised trials in shift workers (556 relevant participants) could not determine whether bright light, napping or other non-drug measures reduce sleepiness at work or improve sleep between shifts. Pooled bright light at night cut sleepiness by 0.83 points on one scale, but that estimate rests on very low quality evidence.

⚠What it shows: Cochrane systematic review of 17 small RCTs totalling 556 participants; studies were too heterogeneous to pool in most cases and the evidence was rated low to very low quality, with only one comparison at moderate quality. Databases were searched to August 2015, so more recent trials are not included. Inconclusive is not the same as ineffective.Meta-analysis · Cochrane Database Syst Rev

Across 13 randomised trials of 561 adults aged 50 and over, adding HMB to resistance training produced no significant benefit for fat mass (SMD 0.24), muscle mass (SMD 0.05) or muscle strength (SMD 0.04). The authors concluded it cannot be recommended as a routine adjunct for people able to do structured exercise.

⚠What it shows: Meta-analysis of 13 randomised trials totalling 561 adults aged 50 and over; heterogeneity was high for the strength outcomes and trial durations and doses varied, and the findings say nothing about people unable to train or about younger lifters.Meta-analysis · Age Ageing

A critical review of the creatine literature concluded that the supplement's performance benefits have generally been shown in untrained people doing several exercise bouts under laboratory conditions, while the limited data available on highly trained athletes performing single competition-like tasks showed no benefit. The authors judged that widespread creatine use to improve competition performance was not justified on that evidence, though repeated-effort sports remained a plausible unconfirmed case.

⚠What it shows: A narrative critical review from 1997, not a modern pooled analysis: it rests on a small body of studies in highly trained athletes, reports no effect sizes, and says nothing about creatine's use during training blocks or in non-athletes.Meta-analysis · Int J Sports Med

In a 16-week randomised trial of adults with overweight and central obesity (105 per group), 75 minutes of weekly HIIT reduced total body fat mass versus control whether performed as one weekly session (-0.8 kg) or spread across three (-1.0 kg).

⚠What it shows: One single-centre RCT in Hong Kong adults with overweight and central obesity, 105 per group; fat-mass reductions versus control were modest (-0.8 to -1.0 kg over 16 weeks) and long-term durability is untested.RCT · Nat Commun

Pooling five European cohorts of 9922 men followed for an average of 25.6 years, men with low cardiorespiratory fitness who reported high occupational physical activity had higher cardiovascular (HR 1.27) and all-cause mortality (HR 1.22) than low-fitness men with low occupational activity. Among high-fitness men the cardiovascular association was no longer clear, though the all-cause association persisted.

⚠What it shows: Individual participant data meta-analysis of five observational European cohorts, male workers only (mean age 46.8), with occupational activity self-reported. Observational design means association, not proof of cause, and the results may not transfer to women or to other kinds of work.Meta-analysis · Scand J Work Environ Health

A meta-analysis of nine cohort studies found atopic dermatitis was associated with higher all-cause mortality (hazard ratio 1.222), along with raised risks of death from infectious, respiratory and gastrointestinal disease and cancer.

⚠What it shows: Meta-analysis of nine observational cohort studies; association, not causation. Results varied by study design and region, and the data cannot show whether treating eczema changes mortality risk.Meta-analysis · Expert Rev Clin Immunol

In a systematic review comparing dietary strategies for psoriasis, ketogenic and continuous calorie-restriction cohorts lost 10-25.6 kg and showed significant falls in Psoriasis Area and Severity Index scores, whereas intermittent fasting cohorts lost 0-2 kg and did not; no serious adverse effects were recorded with any approach.

⚠What it shows: A systematic review published in protocol form: it reports no number of included trials or participants, includes no head-to-head comparisons, and the authors flag heterogeneity and limited follow-up. The pattern links the amount of weight lost to skin severity rather than proving one diet beats another.Study · J Drugs Dermatol

Across 16 randomised trials with 1401 participants with type 2 diabetes, exercise significantly lowered HbA1c (SMD -0.46) and fasting blood glucose (SMD -0.52), with combined aerobic plus resistance training (SMD -0.71) beating aerobic exercise alone (SMD -0.32); the insulin resistance index did not improve significantly.

⚠What it shows: Meta-analysis of 16 randomised trials totalling 1401 people with type 2 diabetes; trial quality was assessed with the Cochrane risk-of-bias tool but individual protocols, durations and medication use varied, the BMI reduction only reached significance after removing an outlier study, and the insulin resistance index showed no significant change.Meta-analysis · Front Endocrinol (Lausanne)

Pooling 13 randomised trials in people with simple obesity, adding massage to a lifestyle programme was associated with larger falls in body weight (mean difference -4.85 kg) and BMI (-2.65) than lifestyle change alone, along with reductions in waist circumference, total cholesterol, triglycerides, LDL cholesterol, fasting insulin and HOMA-IR. HDL cholesterol also fell rather than rose.

⚠What it shows: Thirteen mostly small randomised trials drawn largely from Chinese-language databases, mixing adults with children and adolescents, and with extreme statistical heterogeneity for the blood-lipid outcomes (I² up to 98.3%); effects this large from adding massage are hard to explain and need independent replication before being taken at face value.Meta-analysis · Medicine (Baltimore)

A Cochrane review of four small laboratory trials (64 physically active, predominantly young male adults) found insufficient, very low quality evidence that whole-body cryotherapy reduces self-reported muscle soreness or improves subjective recovery compared with rest. Confidence intervals at 24, 48 and 72 hours included no difference, and no trial actively monitored adverse events.

⚠What it shows: Cochrane review of only four laboratory trials covering 64 physically active young adults, all but four of them male, mean age 23; every trial had design features carrying a high risk of bias and all outcomes were graded very low quality, with no data at all in women or elite athletes.Meta-analysis · Cochrane Database Syst Rev

Pooling 22 studies with 680 participants, resistance training was associated with higher serum insulin-like growth factor 1 in people older than 60, in both men and women, while participants younger than 60 showed lower IGF-1. The effect appeared at training durations both under and over 16 weeks.

⚠What it shows: Meta-analysis of 22 mostly small studies totalling 680 participants, searched to May 2020. It pooled a continuous blood marker using odds ratios, an unusual approach, and IGF-1 is a circulating marker rather than a measure of strength, muscle mass or health outcomes. The reversal in under-60s was unexpected and is not explained.Meta-analysis · Aging Male

A coordinated analysis of 14 longitudinal studies covering 52,039 older adults found that more physical activity was associated with a lower risk of moving from unimpaired to mildly impaired cognition, a lower risk of death, and substantially longer estimated life expectancy. Evidence of benefit after impairment had already begun was present but less consistent between studies.

⚠What it shows: Pooled analysis of 14 observational cohorts (52,039 adults, mean baseline ages 69.9-81.73) from North America and Europe; physical activity was self-reported and not randomised, so this shows association, not proof that activity causes the slower decline. Findings after impairment onset varied substantially between studies.Meta-analysis · J Gerontol A Biol Sci Med Sci

Living closer to a lead-releasing facility was associated with poorer episodic memory two years later in two cohorts of older adults (n = 1638 and n = 741): every 5 km closer corresponded to episodic memory scores 0.05 standard deviations lower.

⚠What it shows: Observational data from two cohorts of older adults; proximity to a facility is a stand-in for actual lead exposure, which was not directly measured. Association, not causation, and residual confounding is possible despite adjustment.Meta-analysis · Alzheimers Dement

In a meta-analysis of 14 randomised trials covering 1,233 adults with non-specific low back pain, strategies bolted onto exercise programmes to improve adherence showed no clear effect on adherence itself, yet pain and function improved slightly in the medium term on moderate-certainty evidence.

⚠What it shows: Fourteen randomised trials, 1,233 adults with non-specific low back pain; risk of bias was low in only six trials, adherence was measured inconsistently (device data in three trials, questionnaires in four), and certainty ranged from moderate to very low.Review · Musculoskelet Sci Pract

A network meta-analysis of 131 randomised trials in 11,457 perimenopausal women with anxiety, depression or panic symptoms found medication combined with psychotherapy ranked highest for depression scores, menopausal symptom scores and sleep quality, while no single strategy was best across all outcomes and the control condition ranked best for safety.

⚠What it shows: Network meta-analysis of 131 trials (11,457 women) that mostly ranks treatments indirectly using SUCRA probabilities rather than testing them head to head; rankings are sensitive to the quality and comparability of the included trials, symptom scales are subjective, and the finding that untreated controls ranked best for safety reflects side effects of active treatment.Systematic review · Front Psychiatry