25 corrections on 10 pages. Most arrive together: one paper read again against every page that cites it, and every page fixed the same day.
Radaelli 2025: adults with sarcopenia said to have been excluded
October 4, 2026 · We got it wrong
What it said
The page said adults diagnosed with sarcopenia were "the population Radaelli excluded".
What it says now
The page says adults diagnosed with sarcopenia were a small share of Radaelli's trials.
Why
Radaelli et al. 2025 list sarcopenic participants among those eligible, and report participants with sarcopenia in 9 of their studies. The pages were written from our notes on the papers, and our checks compared them with those notes. The fix was checked against the paper and by an independent reviewer before it was applied.
Radaelli R, Rech A, Molinari T, et al. Effects of Resistance Training Volume on Physical Function, Lean Body Mass and Lower-Body Muscle Hypertrophy and Strength in Older Adults: A Systematic Review and Network Meta-analysis of 151 Randomised Trials. Sports Med. 2025;55:167-192Read the paper
Found by
Our own reading of the whole page against its paper.
Yan et al. 202517 corrections · 7 pages
October 4, 2026 · We got it wrong
Yan 2025: a dose-response model presented as observed results
What it said
The pages said meaningful improvements "appeared above about 120 minutes" of training a week, or "appeared above approximately 600 METs-min/week", and called the walking-speed relationship "linear".
What it says now
The pages say that in the authors' dose model clinically meaningful gains in walking speed may come above about 600 MET-minutes a week, that grip gains peak and then decline, and that walking speed rose without a plateau.
Why
The thresholds come from the authors' Bayesian dose-response model, which says clinically meaningful walking-speed gains may be achieved above approximately 600 MET-minutes a week; the relationship had no plateau, and the paper does not call it linear. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: a grip-strength comparison between trials presented as a tested frequency prescription
What it said
The pages said the 24 trials "produced a specific prescription" whose extra weekly session "more than doubled the benefit" or "the improvement" in leg force and power, that frequency "mattered more than any other" variable or was "the dominant variable", and that "the frequency answer changed the prescription". The review's own study page and the prescription claim said "The third session more than doubled the result" with the "Same exercises. Same effort per session", that one variable "stood above the rest", that the type of exercise "matters less than consistency", and called the authors' suggested start "The tested prescription" or "an evidence-based prescription".
What it says now
The pages say trials with three sessions a week gained more than twice as much grip strength as trials with two, that resistance type and training frequency changed the results, and that the starting dose is the authors' suggestion.
Why
The doubling (3.18 kg against 1.42 kg) is for grip strength, and it compares groups of different trials, not the same people training more often. The review names resistance type and training frequency together as key effect modifiers. Its three sessions a week of about 40 minutes is the authors' suggested starting point, not a programme the trials tested. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: the grip test called the least responsive measure and a failed test, and a proposal for future diagnostic criteria described as already in use
What it said
The creatine study page said "the grip test paradox changed which measurements doctors should trust". The review's own study page and the prescription claim called grip strength the "least responsive measure" of whether training works, said "The test failed" and called it "the worst tool", called the knee-strength result "The strongest response to resistance training in the entire analysis", and listed the timed up-and-go among the tests the authors favour.
What it says now
The pages say grip strength, a test used to diagnose the condition, showed only a limited response to training, and that the authors propose leg-strength and walking tests as better markers of it; the review's study page and the prescription claim add that most programmes did not train grip directly, and name the tests: gait speed, the sit-to-stand test and knee extension strength.
Why
The authors found that grip strength responded little to training, likely because most trials did not train grip directly, and that gait speed, the sit-to-stand test and knee extension strength showed larger and more consistent improvements; they suggest these may warrant consideration as endpoints in future revisions of the diagnostic criteria. Nothing has changed yet. Grip strength still improved significantly, and knee extension was the only result measured on a standardised scale, so it cannot be ranked as the strongest. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: a review that found no effect of load described as an intensity finding
What it said
The volume claim sent readers on to this review as the answer to "how hard each set should be", with "its own counter-intuitive curve — where the relationship between effort and results peaks, then drops". The Radaelli study page said it had "a different intensity finding", and the creatine study page linked it as "what frequency and intensity actually improves function".
What it says now
The pages describe it as evidence on training once muscle loss has been diagnosed: heavier loads did no better than lighter ones, trials with three sessions a week gained more grip strength than trials with two, and the authors' dose model has grip gains peaking at one weekly dose, then falling.
Why
The review compared loads of 70% of one-repetition maximum or more against lighter loads and found no difference for grip or knee strength. The curve that peaks and then falls is grip strength against total weekly training dose in the authors' model, not effort per set. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: a caveat about short trials and protein credited to the researchers
What it said
The review's study page said the lack of muscle growth "may simply reflect how short these studies were", that longer programmes "especially combined with enough protein" might produce visible growth, and that "The researchers acknowledged this limitation"; the prescription claim said the result "may reflect insufficient intervention duration rather than a biological ceiling".
What it says now
The pages give the limitation the authors do name: few of the studies measured muscle mass. They note that most trials ran 26 weeks or less, and that the paper does not report protein intake (on the claim, what the participants ate).
Why
The authors' limitation for muscle mass is that few studies measured it. They tie the trials' short durations, 26 weeks or less, to how far the dose-response findings generalise, not to muscle growth, and the paper does not mention protein. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: a source quote the paper does not contain
What it said
The claim quoted the paper as saying: "The optimal recommendation is approximately 120 minutes per week, conducted over at least 3 sessions, with each session lasting about 40 minutes at moderate intensity."
What it says now
The claim quotes the paper's own sentence: "clinical practice may consider initiating resistance training at approximately 120 min per week of moderate intensity, delivered in three sessions, and progressively increasing training frequency or weekly exercise dose as tolerated".
Why
The quoted sentence is in neither the paper nor our notes on it; it was composed when the page was written. The paper offers about 120 minutes a week in three sessions as a possible starting point, not as an optimal recommendation, and does not say "at least 3 sessions". The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: a statistic about differences between trials read as frequency explaining 60% of who improved
What it said
The pages said training frequency "explained 60% of who improved and who did not" or accounted for "60% of the outcome — more than exercise type, intensity, or session length", and a chart on the claim gave the remaining 40% to exercise type, intensity and session length.
What it says now
The pages say that in a meta-regression across the trials, frequency accounted for 60% of the differences between trials' grip strength results; the chart marks the rest as not explained by frequency.
Why
The 60.25% is the share of the between-trial differences in grip strength results that training frequency explained in a meta-regression with one variable at a time. It does not describe who improved, it is for grip only, and the review attributes the rest to no other variable. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: combined resistance training described as weights plus balance or flexibility work
What it said
The pages said the "combined" training that improved walking speed most mixed resistance with "balance or flexibility work" (on the claim, "balance training"), and described constant resistance as "traditional weight machines and free weights".
What it says now
The pages say trials combining constant and variable resistance improved walking speed the most and trials using constant resistance built more knee extension strength; the claim adds that the paper does not say which equipment the trials used.
Why
In the review, combined resistance training is constant and variable resistance together, not resistance plus balance or flexibility work, and the paper does not name the trials' equipment. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: findings in adults diagnosed with sarcopenia extended to people only at risk of it
What it said
The study page said the findings apply to "Older adults at risk of sarcopenia but not yet formally diagnosed"; a question on the claim asked whether training twice a week is "enough to prevent sarcopenia"; and an evidence note on the claim described all of "approximately 1,400 total participants" in its two reviews as sarcopenic adults.
What it says now
The study page lists only adults with diagnosed sarcopenia; the question now asks whether, with sarcopenia, training twice a week is enough; and the note separates the review's 951 sarcopenic adults from the second review's healthy older adults.
Why
Every trial in the review enrolled adults aged 60 or over already diagnosed with sarcopenia, so it tested training as a treatment, not as prevention. The claim's second review is a separate study, of older adults the page describes as healthy. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: muscle mass said not to have grown in a single trial of adults with sarcopenia, and the strength gain put down to the nervous system
What it said
The creatine claim said that in 24 trials of older adults with sarcopenia "the muscles themselves didn't grow. Not in a single trial", and that the gains "weren't about size. They were about power." The Radaelli study page said muscle mass "did not measurably increase in studies under six months". The prescription claim said muscle mass "did not change. Not in any study, not in any subgroup", that "Every trial agreed", and that "The brain learned to use what was already there"; the review's own study page said "The explanation is neurological."
What it says now
The pages say strength rose clearly while muscle mass, overall, did not rise significantly, that few of the trials measured it, and that a significant gain showed in the trials that used constant resistance. The review's study page and the prescription claim add that the trials measured no nerve or motor-unit outcome, so they do not show why strength rose more than muscle mass.
Why
The review found no significant overall improvement in muscle mass against usual care, but it does not say mass changed in no trial: a measure of muscle mass rose significantly in the trials that used constant resistance, and the authors note that few of their studies measured mass. The review does not tie the result to studies under six months, and it analysed no power outcome. The review measured no nerve, motor-unit or coordination outcome; it discusses them only through other studies. Its zero heterogeneity was for one muscle-mass measure, in 10 studies and 405 people, not agreement among all 24 trials; the other measure, the muscle mass index, varied widely between its 7 trials. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: results from different sets of trials described as the same people on the same programme
What it said
The pages said the grip and knee-strength results came from the "Same people. Same months of training" ("Same study. Same 951 people. Same exercise program."), and the claim measured knee extension in "the same participants".
What it says now
The pages say the grip and knee-strength results were pooled from different sets of trials; knee extension, for example, from the 11 trials that measured it.
Why
Of the review's 24 trials, 17 (646 people) measured grip strength and 11 (436 people) measured knee extension, in different programmes. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: several details of the review were wrong
What it said
The study page listed CNKI among the databases searched, gave 24 as the number of records screened, said the paper was "Not indexed" in PubMed and left out its PubMed and PMC numbers, said the trials' duration was "Not explicitly stated", and said most ran "between 8 and 26 weeks" (the claim: "8 to 26 weeks"). The claim gave the muscle-mass interval as -0.07 to 0.12 and the walking-speed interval as 0.04 to 0.12 with I² 59.3%, and said "Some outcomes showed sensitivity to study exclusion".
What it says now
The pages give the paper's details: it searched PubMed, Embase, CENTRAL and Web of Science; 3,022 records were screened and 81 full texts read; PMID 41212331 and PMCID PMC12602684; most trials ran 26 weeks or less; the muscle-mass interval is -0.10 to 0.15 and the walking-speed interval 0.03 to 0.14 with I² 51.7%; leave-one-out checks did not change any result, and publication bias was found for knee extension, whose adjusted result stayed significant.
Why
Each detail was checked against the paper in a full read of both pages and was wrong as written. The pages were written from our notes on the paper, and our checks compared them with those notes.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: side effects reported by four trials presented as a safety record for all 24
What it said
The pages said there were "No serious adverse events" ("Zero serious injuries across 24 clinical trials"), a consistent answer from "951 people across 24 trials", that doing more of the training, more often, "is also safe", and that the only side effects, "mild soreness in the first week or two", went away on their own.
What it says now
The pages say four of the 24 trials reported side effects, all mild and reversible, mostly joint discomfort, muscle soreness or fatigue at the start of training that typically eased once intensity or technique was adjusted, and that none led anyone to withdraw.
Why
Only four of the review's 24 trials reported adverse events. All were mild, reversible and led to no withdrawals, and most were transient symptoms that typically resolved after training intensity or technique was adjusted. The review describes no side effects for the other 20 trials, and it did not test whether more training, more often, is safe. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: strength and function gains told without the authors' finding that they fell short of clinically meaningful change
What it said
The Radaelli study page said this review found resistance training "significantly improved leg strength and functional movement". The review's own study page said "the functional evidence is strong", answered "emphatically yes", said physical performance improved too, and gave "High confidence for functional improvement"; the prescription claim called the knee gain "Functionally significant" and the improvement "real, functional, and meaningful for daily independence".
What it says now
The Radaelli study page says resistance training significantly improved leg strength, grip and walking speed, though the grip and walking gains fell short of what the authors count as clinically meaningful. The review's study page and the prescription claim say every gain with a clinical threshold fell short of it, while the leg-strength gain was large; the study page adds that the short physical performance battery did not improve.
Why
The gains in grip strength, walking speed, the timed up-and-go and the sit-to-stand test were statistically significant but below the minimal important differences the authors set, which they read as little to no clinical benefit; the short physical performance battery showed no effect. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: the review cited under titles it does not have
What it said
The claim cited the review as "Resistance Training for Sarcopenia in Older Adults: A Systematic Review and Meta-Analysis", and the study page's summaries for AI systems named it "Optimal resistance training prescriptions for sarcopenia".
What it says now
Both pages give the paper's title: "Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis".
Why
The titles were composed or shortened rather than copied from the paper. The pages were written from our notes on the paper, and our checks compared them with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: the trials said to have tested exercise alone, with no nutritional interventions
What it said
The claim said "These trials all studied exercise alone — no nutritional interventions."
What it says now
The claim says the paper does not report what the participants ate.
Why
The review does not say this. Its inclusion rules allowed resistance training as the main part of a programme alongside minor additional elements, and it does not report what participants ate. The page was written from our notes on the paper, and our checks compared it with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Yan 2025: where participants lived described as where they trained
What it said
The page said older adults who trained "in gym or clinic settings" gained more knee strength, and those who trained "in community centers" gained more everyday mobility.
What it says now
The page says adults living in long-term institutional care gained more knee extension strength, while those living in the community improved more on the timed up-and-go.
Why
The review's setting subgroups compare older adults receiving long-term institutional care with those living in the community: where they lived, not where they trained. The page was written from our notes on the paper, and our checks compared it with those notes. The fix was checked word for word against the paper and by an independent reviewer before it was applied.
Yan R, Chen Y, Zhang R, et al. Optimal resistance training prescriptions to improve muscle strength, physical function, and muscle mass in older adults diagnosed with sarcopenia: a systematic review and meta-analysis. Aging Clin Exp Res. 2025;37(1):320Read the paper
Found by
Our own reading of the whole page against its paper.
Bonham et al. 20257 corrections · 3 pages
October 4, 2026 · We got it wrong
Night-shift fasting trial: several details of the trial were wrong
What it said
The claim: "Same workers, same jobs, same rosters"; "a team in Melbourne"; meal replacements listed among the trial's plans; "Neither timing produced better weight loss than the other"; "The marker your doctor checks got better the same way"; a chart giving the calorie-restriction group 5-8 kg. The study page: one withdrawal for adverse events; "No stated conflicts of interest"; 149/250 and 170/250; dropout as 25/83 (33%) and 21/82 (27%).
What it says now
The groups were assigned at random; the trial ran in Melbourne and Adelaide; meal replacements were tested in a different trial; after adjusting for starting weight the day-shift/day-off group came out lighter than calorie restriction, which the authors put down to its finishers being heavier at the start; HbA1c did not change in any group; calorie-restriction finishers lost 5.3 kg; six people withdrew for adverse events (five unrelated to the study, one with mood changes); one author declares a consultancy with Vanda Pharmaceuticals, unrelated to this work; 149/248 and 170/248; dropout 24 (29%), 25 (33%) and 21 (27%), as the paper gives them.
Why
Each detail was checked against the paper in a full read of the page and was wrong as written.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting trial: the cholesterol result covered total and LDL cholesterol, and weight loss may have contributed
What it said
The claim said fasting and dieting "performed identically on every measure except one" (total cholesterol) and that the cholesterol gap was "larger than what the extra weight loss alone would explain"; the study page gave the cited meta-analysis as "~3.3 mg/dL LDL per extra kg".
What it says now
The group fasting on day shifts or days off ended with lower total and LDL cholesterol than the calorie-restriction group, an exploratory result. The authors note that group's larger weight loss could have contributed and that the finding needs validating. The meta-analysis they cite gives 1.28 mg/dL lower LDL per kg lost.
Why
Both total and LDL cholesterol differed; the paper names the larger weight loss as a possible contributor rather than ruling it out; 3.3 mg/dL was our total for 2.6 kg, given as a per-kg figure.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting trial: wrong DOI and a made-up title in the guide's source list
What it said
DOI 10.1016/j.ebiom.2025.105669 and the title "SWIFt: intermittent fasting vs continuous energy restriction in shift workers".
What it says now
DOI 10.1016/j.ebiom.2025.105803 and the paper's title, "Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial".
Why
The DOI was not the paper's (the paper gives 10.1016/j.ebiom.2025.105803), and the title was composed rather than copied from the paper.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting claim: a quote shown as the paper's was not in the paper
What it said
A quote presented as the paper's: "...These findings suggest that for night shift workers with overweight/obesity, CER may be preferred over modified intermittent fasting due to its simplicity and fewer reported adverse events."
What it says now
The paper's own words: "At 24 weeks, weight and HOMA-IR were not different between the modified IF groups compared with CER. Clinically significant improvements in weight and metabolic health were achieved for the majority of night shift workers who remained in the intervention at 24 weeks."
Why
The quoted sentences are not in the paper, and their advice to prefer calorie restriction reverses the authors' conclusion that intermittent fasting is a viable alternative to it. Our pipeline wrote them and presented them as a quote.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting trial: 61% is of those who finished six months; of everyone who started it was 40%
What it said
"most people who tried any steady approach made it work"; "When shift workers tried any structured eating plan, roughly six in ten lost real weight"; "61% lost real weight" in its summary line, and "On any plan" under its 61% figure.
What it says now
61% of those who stayed in the trial for six months lost at least 5% of their weight; counted against everyone who started, it was 40%.
Why
The paper reports 61% of those who provided 24-week data and 40% of everyone randomised. "Tried" counted the wrong group.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting trial: it had no do-nothing group, so it cannot show the diets "beat doing nothing"
What it said
"Both beat doing nothing" (the claim's audio, the guide's text and audio). The guide also described the trial as testing "whether any structured eating plan actually works for shift workers" and asked "does structured eating reverse the damage?".
What it says now
The trial compared two fasting schedules with daily calorie restriction. It had no group without a diet plan, so it cannot say how either compares with doing nothing.
Why
The paper states that a non-intervention control arm was not included. Our pages read the weight lost within each group as a win over doing nothing.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.
October 4, 2026 · We got it wrong
Night-shift fasting trial: the headaches were reported during weight loss and called mild, not "persistent" or "lasting"
What it said
The claim said about one in five fasters "developed persistent headaches", called them "documented side effects of intermittent fasting, not signs that the body is ‘adapting’" and "a cost with no payoff"; the guide said "lasting headaches"; the study page said the direction "is consistent across IF studies".
What it says now
In the trial, 19% and 22% of the two fasting groups, and none on daily calorie restriction, reported headaches during the weight-loss phase. The authors call the adverse effects mild and say they were previously associated with this kind of modified fasting. The pages now say that and no more.
Why
The paper gives no duration for the headaches, ties them to modified fasting regimens rather than intermittent fasting in general, and never tests whether they mean anything for weight loss. Our pages went beyond its adverse-event results.
Bonham MP, Leung GKW, Rogers M, et al. Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial. eBioMedicine. 2025;117:105803.Read the paper
Found by
Our own reading of the whole page against its paper.