Alpha¹ by KaimenConfidential

Kaimen Rigor review

Effectiveness of kangaroo mother care before clinical stabilisation versus standard care among neonates at five hospitals in Uganda (OMWaNA): a parallel-group, individually randomised controlled trial and economic evaluation.

Tumukunde V, Medvedev MM, Tann CJ, Mambule I, Pitt C, Opondo C, Kakande A, Canter R, Haroon Y, Kirabo-Nagemi C, Abaasa A, Okot W, Katongole F, Ssenyonga R, Niombi N, Nanyunja C, Elbourne D, Greco G, Ekirapa-Kiracho E, Nyirenda M, Allen E, Waiswa P, Lawn JE, OMWaNA Collaborative Authorship Group

DOI: 10.1016/S0140-6736(24)00064-3
Record issued: 2026-08-10
Engine: 7.29.0
Exported: 2026-09-22

Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/10d39c70-617b-4252-80fb-cbf232e5fde6 is authoritative.

Kaimen Rigor review

Kaimen Rigor reviewed

4.25 / 5

Strong — minor issues to address

Interventional Study

Critical

High-priority

How this rating was calculated

Started at 5★ — deductions

IntegrityIntegrity concern−0.5★
ReportingEthical approvals partially met−0.25★

Model-reported confidence: highReviewer agreement: moderateEngine: 7.29.0Source: full text

Findings by category

01ArithmeticNone found 102OverreachOverstated conclusions · partly checked 103AuthenticityData authenticity concerns 104DisclosureReporting gaps 05AvailabilityNone found 06Copyedit3 minor suggestions

This Kaimen Rigor review uses Kaimen Rigor reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.

This is a well-designed and transparently reported randomised controlled trial: randomisation, concealment, power calculation, participant/maternal baseline data, ethics approvals, trial registration, and reporting guidelines are all adequately documented, and the verification components (6/6 statistics consistent, no retracted/unfound citations, all links live) found no substantive problems. The main weaknesses are reporting-level: threshold-only p-values in a few outcome tables, a Table 2 footnote that labels model-derived standard errors as 'mean (SD)', and the absence of an explicit Declaration of Helsinki/ICH-GCP compliance statement.

All three independent reviewer runs classified the paper as interventional and converged on passes for six of seven applicable dimensions; the two divergences (statistical analysis warn vs pass; data code availability warn vs pass) were resolved against the explicit scoring rules, with both positions preserved in the dimension details. key resources was excluded as not applicable (procedural trial with no investigational product). The statistics verification recomputed only 6 of the reported analyses (those with a test statistic + df or an effect estimate + CI); the remaining analyses are unverified, not confirmed.

Numerical inconsistencies

None found

Values that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.

Checked — nothing surfaced.

Recomputed 6 tests: 6 consistent, 0 inconsistent; 3 recomputed directly from the reported test statistics, 3 via agent-written checks.

Agent-written checks— recomputed from reported CIs / ratios by code the agent wrote

CONSISTENTreported p = .850 · recomputed p = .828
Recomputed RR 0.97 (95% CI 0.74–1.28), reported p=0.85
“RR 0.97 [95% CI 0.74–1.28]; p=0.85”
Find in source
Taken as given: 0.74–1.28 is a two-sided 95% confidence interval for the RR of 0.97, not a range, an IQR, or a different interval level; the RR is a RATIO measure, so the interval is symmetric on the log scale; p=0.85 is the p for THIS estimate, not for another comparison in the same sentence
Method: back the two-tailed p out of the log-scale CI width and compare it against the printed p
How we recomputed it: pCI(0.97, 0.74, 1.28, 1)
CONSISTENTreported p = .072 · recomputed p = .068
Recomputed RR 0.60 (95% CI 0.35–1.05), reported p=0.072
“RR 0.60 [95% CI 0.35–1.05]; p=0.072”
Find in source
Taken as given: 0.35–1.05 is a two-sided 95% confidence interval for the RR of 0.60, not a range, an IQR, or a different interval level; the RR is a RATIO measure, so the interval is symmetric on the log scale; p=0.072 is the p for THIS estimate, not for another comparison in the same sentence
Method: back the two-tailed p out of the log-scale CI width and compare it against the printed p
How we recomputed it: pCI(0.6, 0.35, 1.05, 1)
CONSISTENTreported p = .043 · recomputed p = .050
Recomputed RR 0.86 (95% CI 0.74–1.00), reported p=0.043
“RR 0.86 [95% CI 0.74–1.00]; p=0.043”
Find in source
Taken as given: 0.74–1.00 is a two-sided 95% confidence interval for the RR of 0.86, not a range, an IQR, or a different interval level; the RR is a RATIO measure, so the interval is symmetric on the log scale; p=0.043 is the p for THIS estimate, not for another comparison in the same sentence
Method: back the two-tailed p out of the log-scale CI width and compare it against the printed p
How we recomputed it: pCI(0.86, 0.74, 1, 1)
CONSISTENTreported p = .960 · recomputed p = .963Reviewers 1, 2
Crude 7-day mortality comparison
“81 (7·5%) of 1083 neonates in the intervention group and 83 (7·5%) of 1102 in the control group died (adjusted RR 0·97 [95% CI 0·74–1·28]; p=0·85).”
Find in source
Taken as given: The crude 2x2 table counts are 81 events/1002 non-events for intervention and 83 events/1019 non-events for control.; The crude p-value reported is 0.96, which is consistent with the chi-square test for independence.
Method: Pearson chi-square test from 2x2 cell counts
How we recomputed it: pChi2x2(81,1002,83,1019)
CONSISTENTreported p = .307 · recomputed p = .307Reviewers 1, 2
Crude 28-day mortality comparison
“119 (11·3%) of 1051 neonates in the intervention group and 134 (12·8%) of 1049 in the control group died (RR 0·88 [0·71–1·09]; p=0·229).”
Find in source
Taken as given: The crude 2x2 table counts are 119 events/932 non-events for intervention and 134 events/915 non-events for control.; The crude p-value reported is 0.307, which is consistent with the chi-square test for independence.
Method: Pearson chi-square test from 2x2 cell counts
How we recomputed it: pChi2x2(119,932,134,915)
CONSISTENTreported p < .001 · recomputed p = <.001Reviewer 2
Crude chi-square test for hypothermia at 24 h (secondary outcome).
“The proportion of neonates with hypothermia at 24 h was 40·9% (448 of 1096) in the intervention group and 53·1% (585 of 1101) in the control group (adjusted RR 0·76 [95% CI 0·70–0·83]; p<0·0001).”
Find in source
Taken as given: Crude chi-square test approximates the p-value; the adjusted p-value is <0.0001.; Cell counts are 448, 648, 585, 516.
Method: pChi2x2 from 2×2 table.
How we recomputed it: pChi2x2(448, 1096-448, 585, 1101-585)

Overstated conclusions

1 finding · worst low

Conclusions that reach past what the paper's own results support — including a significance claim that no longer holds when the statistic is recomputed, and efficacy resting on an unvalidated surrogate endpoint.

Conclusions only partially backed by the presented evidenceAssessed

10 major claims checked against the paper's own evidence: all adequately supported.

partialReviewer 3
KMC initiated before stabilisation decreased the economic cost of neonatal care to society and providers.
The mean cost differences were small and not statistically significant (societal –$7·2, p=0·33; provider –$9·4, p=0·058), so the claim that costs 'decreased' is stronger than the presented evidence.
Evidence: Mean economic cost to society: US$359·1 vs $365·9 (adjusted mean difference –$7·2 [–21·5 to 7·2]; p=0·33); provider costs –$9·4 [–19·0 to 0·3]; p=0·058.
“Our economic evaluation found that, compared with standard care, KMC initiated before stabilisation decreased the economic cost of neonatal care to society and providers.”
Discussion ¶1Find in source
supportedReviewers 1, 2
KMC initiated before stabilisation did not reduce early neonatal mortality (7 days).
The primary outcome shows no statistically significant difference between groups (adjusted RR 0.97, 95% CI 0.74–1.28, p=0.85).
Evidence: Primary outcome in Table 2: adjusted RR 0.97 (95% CI 0.74–1.28), p=0.85.
“From randomisation to 7 days of age, 81 (7·5%) of 1083 neonates in the intervention group and 83 (7·5%) of 1102 in the control group died (adjusted RR 0·97 [95% CI 0·74–1·28]; p=0·85).”
Table 2Find in source
supportedReviewers 1, 2
KMC was cost-effective from the societal and provider perspectives.
The economic evaluation shows that the intervention is cost-effective, with a high probability of being cost-saving even when no value is placed on averted deaths.
Evidence: Economic results: adjusted mean societal cost difference –$7.2 (95% CI –21.5 to 7.2), 97% probability of cost-effectiveness from provider perspective.
“Even if policy makers place no value on averting neonatal deaths, the intervention would have 97% probability from the provider perspective and 84% probability from the societal perspective of being more cost-effective than standard care.”
ResultsFind in source
supportedReviewers 1, 2
Secondary outcomes including hypothermia and daily weight gain were significantly improved in the intervention group.
Hypothermia at 24h was significantly reduced (RR 0.76, p<0.0001) and daily weight gain at 28 days was significantly higher (mean difference 0.75 g, p=0.047).
Evidence: Table 2: adjusted RR for hypothermia 0.76 (95% CI 0.70–0.83), p<0.0001; adjusted mean difference for daily weight gain 0.75 (95% CI 0.01–1.49), p=0.047.
The proportion of neonates with hypothermia at 24 h was 40·9% (448 of 1096) in the intervention group and 53·1% (585 of 1101) in the control group (adjusted RR 0·76 [95% CI 0·70–0·83]; p<0·0001). ... Mean weight gain at 28 days was 7·8 g per day (SD 0·3) in the intervention group and 7·1 g per day (0·3) in the control group (adjusted mean difference 0·75 [0·01–1·49]; p=0·047).
Table 2reviewer’s wording
supportedReviewers 1, 2
A pooled meta-analysis showed a significant relative reduction in 28-day mortality of 19% overall.
The meta-analysis of three trials (OMWaNA, iKMC, eKMC) shows a significant 19% reduction (95% CI 0.71–0.93, p=0.0019).
Evidence: Results, meta-analysis: 'A pooled meta-analysis... showed a relative reduction of 19% in 28-day mortality (95% CI 0·71–0·93; p=0·0019).'
A pooled meta-analysis of the OMWaNA trial, WHO’s iKMC trial, and the eKMC trial in The Gambia showed a relative reduction of 19% in 28-day mortality (95% CI 0·71–0·93; p=0·0019).
Resultsreviewer’s wording
supportedReviewer 2
KMC before stabilisation decreased the economic cost of neonatal care to society and providers.
The economic results show lower mean costs in the intervention group, though not statistically significant at conventional levels.
Evidence: Results: provider economic costs –$9.4 (–19.0 to 0.3, p=0.058); societal economic costs –$7.2 (–21.5 to 7.2, p=0.33).
Across all five hospitals in the OMWaNA trial, the mean economic cost to society per neonate (n=2221) was similar between the intervention group (US$359·1) and the control group ($365·9; adjusted mean difference –$7·2 [95% CI –21·5 to 7·2]; p=0·33).
Results ¶6reviewer’s wording
supportedReviewers 2, 3
KMC initiated before stabilisation did not reduce early neonatal mortality.
The primary outcome was null (RR 0·97, 95% CI 0·74–1·28, p=0·85), directly supporting the claim.
Evidence: Primary outcome: 81/1083 (7.5%) vs 83/1102 (7.5%) deaths at 7 days, adjusted RR 0·97 (0·74–1·28), p=0·85.
“KMC initiated before stabilisation did not reduce early neonatal mortality; however, it was cost-effective from the societal and provider perspectives compared with standard care.”
AbstractFind in source
supportedReviewers 2, 3
KMC initiated before stabilisation was cost-effective from the societal and provider perspectives compared with standard care.
The cost-effectiveness analysis reports high probabilities of cost-effectiveness (97% provider, 84% societal) even at zero value per death averted, supporting the claim.
Evidence: Incremental net benefit analysis: 97% probability (provider) and 84% probability (societal) of being more cost-effective than standard care.
“Even if policy makers place no value on averting neonatal deaths, the intervention would have 97% probability from the provider perspective and 84% probability from the societal perspective of being more cost-effective than standard care.”
ResultsFind in source
supportedReviewer 3
A pooled meta-analysis of the three trials showed a significant relative reduction in 28-day mortality of 19% overall and 14% across African sites.
The paper reports its own pooled analyses with CIs and p-values (overall RR 0·81 equivalent, 19% reduction; African-only 14% reduction, p=0·043), supporting the claim.
Evidence: Pooled meta-analysis: 19% relative reduction in 28-day mortality (95% CI 0·71–0·93; p=0·0019); African sites only 14% reduction (0·74–1·00; p=0·043).
A pooled meta-analysis of the OMWaNA trial, WHO's iKMC trial, and the eKMC trial in The Gambia showed a relative reduction of 19% in 28-day mortality (95% CI 0·71–0·93; p=0·0019)... A meta-analysis including African sites only across the three trials showed a relative reduction of 14% in 28-day mortality (0·74–1·00; p=0·043).
Results ¶4reviewer’s wording
supportedReviewer 3
Secondary outcomes, including hypothermia at 24 h and daily weight gain at 28 days, were significantly improved in the intervention group.
Hypothermia at 24 h (RR 0·76, p<0·0001) and daily weight gain (mean difference 0·75 g/day, p=0·047) are significantly improved, supporting the claim.
Evidence: Hypothermia at 24 h: 40·9% vs 53·1% (adjusted RR 0·76 [0·70–0·83]; p<0·0001); daily weight gain: 7·8 vs 7·1 g/day (adjusted mean difference 0·75 [0·01–1·49]; p=0·047).
“Secondary outcomes, including hypothermia at 24 h and daily weight gain at 28 days, were significantly improved among neonates in the intervention group.”
AbstractFind in source

Data authenticity concerns

1 finding · worst low

An adversarial read for patterns associated with data that may not be genuine: results that look too clean, implausibly large effects, duplicated data or images, and methods that do not match the results reported.

Methods and results do not matchAssessed

1 integrity concern flagged (0 high).

lowmethod result mismatch
Table 2 labels parenthetical values for model-derived continuous outcomes as 'mean (SD)', but values such as daily weight gain 7·8 (0·3) across 761 neonates and duration of hospital admission 7·3 (0·2) are implausibly small as standard deviations of raw data and are most likely standard errors from regression models.
Duration of hospital admission, days | 7·3 (0·2); 1083 (97·6%) | 6·1 (0·1); 1102 (99·2%) ... Daily weight gain at 28 days, g per day | 7·8 (0·3); 761 (68·6%) | 7·1 (0·3); 731 (65·8%) | 0·78 (0·02 to 1·55) | p=0·045 | 0·75 (0·01 to 1·49) | p=0·047
Table 2reviewer’s wording

Reporting gaps

1 finding · worst medium

Required detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.

Ethics/consent reporting incompleteAssessed

Eight Dimensions6 pass · 1 warn

The introduction cites the iKMC trial, eKMC trial, and WHO guidelines, discusses their limitations (e.g., underpowered in The Gambia, lack of economic data), and clearly states the rationale for the study in sub-Saharan Africa level 2 facilities. The limitations of prior work are addressed in terms of context and economic evaluation.

Sub-criteria

prior work citedADEQUATE

premise rationaleADEQUATE

limitations addressedADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 95% conf

The paper establishes a strong scientific premise by citing prior trials and guidelines, identifying evidence gaps, and providing a logical rationale for the study.

Evidence

paraphrase[Introduction, paragraphs 3-4]

“We aimed to compare the effectiveness and safety of KMC initiated before stabilisation versus standard care among neonates ≤2000 g in sub-Saharan Africa. Additionally, we aimed to assess the incremental costs and cost-effectiveness of this intervention from the societal perspective.”

Introduction ¶5Find in source

paraphrase[Discussion, paragraph 3]

“The iKMC trial reported a significant relative reduction of 25% in their secondary outcome of 28-day mortality... This observed reduction seemed to have been primarily driven by the trial site in India...”

PASSReviewer 2· 95% conf

The paper clearly establishes the scientific foundation by citing prior trials (iKMC, eKMC) and identifying evidence gaps regarding KMC before stabilisation in sub-Saharan Africa, cost-effectiveness, and resource requirements.

Evidence

paraphrase[Introduction, paragraph 3]

“One trial coordinated by WHO (the iKMC trial) … reported that immediate KMC led to a 25% reduction in their secondary outcome of mortality within 28 days … Another trial conducted in a level 2 neonatal unit in The Gambia found no reduction in mortality with KMC before stabilisation compared with standard care; however, this study was underpowered …”

“The findings of these two trials left an evidence gap regarding the effects of immediate KMC at facilities without neonatal intensive care, where most births in LMICs occur, and questions regarding possibly lower impact in sub-Saharan Africa.”

Introduction ¶4Find in source

“However, resource requirements were not well defined, including the incremental costs of initiating KMC before stabilisation and necessary facility infrastructure. Indeed, there are few economic evaluations of KMC overall.”

Introduction ¶4Find in source

PASSReviewer 3· 90% conf

The scientific foundation is well established: prior trials (iKMC, eKMC-Gambia) are cited with their strengths and weaknesses, and the rationale for the study (facility level 2 vs 3, regional uncertainty, economic gap) follows logically.

Evidence

paraphrase[Introduction, paragraph 3]

“One trial coordinated by WHO (the iKMC trial)... reported that immediate KMC led to a 25% reduction in their secondary outcome of mortality within 28 days... Another trial conducted in a level 2 neonatal unit in The Gambia found no reduction in mortality with KMC before stabilisation compared with standard care; however, this study was underpowered because of improvements in the quality of standard care that contributed to a 50% reduction in neonatal mortality compared with baseline.”

Introduction ¶3Find in source

paraphrase[Research in context, Evidence before this study]

“We searched PubMed... from Jan 1, 1990, to July 1, 2023... We found no relevant additional trials.”

Randomization was computer-generated with permuted blocks of varying sizes, stratified by birthweight and site. The independent statistician was masked. A power analysis was conducted assuming 25% baseline mortality and 80% power to detect a 5.6% absolute reduction. Inclusion/exclusion criteria are detailed. Missing data were handled by complete-case analysis with sensitivity analyses.

Sub-criteria

controlsN/A

power analysisADEQUATE

blinding levelsADEQUATE

outlier handlingADEQUATE

randomization unitADEQUATE

inclusion exclusionADEQUATE

randomization methodADEQUATE

replicate distinctionN/A

independent replicationN/A

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 95% conf

The trial design is rigorous with computer-generated randomization, allocation concealment, pre-specified sample size, and clear inclusion/exclusion criteria.

Evidence

“A random allocation sequence was computer-generated with permuted blocks of varying sizes, stratified by birthweight (700–1000 g, 1000–1499 g, or 1500–2000 g) and recruitment site.”

MethodsFind in source

paraphrase[Methods, Randomisation and masking]

“Parents, caregivers, and health-care workers were unmasked to treatment allocation; however, the independent statistician who conducted the analyses was masked to treatment allocation.”

“A sample size of 2188 neonates (1094 per group) was estimated to be required to detect an absolute reduction in 7-day mortality of 5·6% (22·4% relative reduction) at a 5% significance level (two-sided) and 80% power, allowing for 20% attrition.”

Statistical analysisFind in source

PASSReviewer 2· 95% conf

The study design is rigorous: a parallel-group, individually randomised trial with computer-generated permuted blocks, stratification, allocation concealment, a pre-specified power analysis, clear inclusion/exclusion criteria, and intention-to-treat analysis. Blinding was not possible for participants/staff but the independent statistician was masked.

Evidence

“A random allocation sequence was computer-generated with permuted blocks of varying sizes, stratified by birthweight (700–1000 g, 1000–1499 g, or 1500–2000 g) and recruitment site. Allocation concealment was done by programming the allocation sequence into the screening database and revealing treatment group only when screening of eligible neonates was complete.”

MethodsFind in source

Statistical analysisFind in source

“Masking of parents, caregivers, or health-care workers was not possible due to the nature of the KMC intervention; however, the independent statistician who conducted the analyses was masked to treatment allocation.”

MethodsFind in source

PASSReviewer 3· 92% conf

Core design elements are reported adequately: computer-generated random allocation with permuted blocks and stratification, clear unit of randomisation, masking of the independent statistician with rationale, a priori power calculation, and detailed eligibility criteria.

Evidence

“A random allocation sequence was computer-generated with permuted blocks of varying sizes, stratified by birthweight (700–1000 g, 1000–1499 g, or 1500–2000 g) and recruitment site.”

MethodsFind in source

Statistical analysisFind in source

Table 1 provides baseline characteristics by group, including sex (approximately 50% female), mean age at screening, gestational age, birthweight, mode of delivery, and maternal age, education, and income. Both sexes are enrolled, so sex justification is not required.

Sub-criteria

demographicsADEQUATE

sex reportedADEQUATE

sex justifiedN/A

age weight healthADEQUATE

housing conditionsN/A

species strain sourceN/A

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 95% conf

Demographic and clinical characteristics of neonates and mothers are well reported, including sex, age, weight, gestational age, and maternal education.

Evidence

paraphrase[Table 1]

“Sex: Female 558 (50·3%) in intervention group, 561 (50·5%) in control group.”

paraphrase[Table 1]

“Age at screening, h: 20·3 (11·6) in intervention group, 21·5 (11·5) in control group. Weight at screening, kg: 1·5 (0·3) in both groups.”

paraphrase[Table 1]

“Mothers: Age 25·6 (6·0) years in intervention group, 25·6 (6·2) in control group. Highest level of education: Primary school 35·3% vs 37·0%, secondary school 53·5% vs 53·7%, university degree 7·2% vs 6·4%.”

PASSReviewer 2· 95% conf

Key biological variables (sex, age, weight, gestational age, and maternal demographics) are well reported in Table 1 and baseline characteristics. Both sexes are enrolled, so sex_justified is not applicable. Health status is defined through inclusion criteria.

Evidence

paraphrase[Table 1, Baseline characteristics]

“Sex: Female 558 (50·3%) [Intervention], 561 (50·5%) [Control]; Male 552 (49·7%) [Intervention], 550 (49·5%) [Control]”

paraphrase[Table 1, Baseline characteristics]

“Age at screening, h: 20·3 (11·6) [Intervention], 21·5 (11·5) [Control]; Gestational age at screening, weeks: 32·3 (2·4) [Intervention], 32·3 (2·2) [Control]; Weight at screening, kg: 1·5 (0·3) [both groups]”

paraphrase[Table 1, Baseline characteristics]

“Age (actual or estimated), years: 25·6 (6·0) [Intervention], 25·6 (6·2) [Control]; Highest level of education: Primary school, Secondary school, University degree, None; Monthly income, US$: 9·4 (0·0–54·9) [Intervention], 5·5 (0·0–54·9) [Control]”

PASSReviewer 3· 85% conf

Sex, age, weight, and health status are reported quantitatively for both groups, and maternal demographics (age, education, income) are provided; the study enrols both sexes so a single-sex justification is not required.

Evidence

“Female | 558 (50·3%) | 561 (50·5%) | | Male | 552 (49·7%) | 550 (49·5%)”

Table 1Find in source

paraphrase[Results, paragraph 1]

“Mean age of neonates at screening was 20·3 h (SD 11·6) in the intervention group and 21·5 h (11·5) in the control group. In both groups, mean gestational age, measured by Ballard score, was 32·3 weeks and mean birthweight was 1·5 kg.”

paraphrase[Table 1]

“Mothers... Age (actual or estimated), years | 25·6 (6·0) | 25·6 (6·2)... Highest level of education | Primary school | 332 (35·3%) | 349 (37·0%)”

The paper reports approval by the Uganda Virus Research Institute, Uganda National Council of Science and Technology, and LSHTM ethics committees with protocol numbers. Written informed consent was obtained. However, no statement of adherence to the Declaration of Helsinki or ICH-GCP is provided, which is expected for a clinical trial.

Sub-criteria

iacuc statementN/A

informed consentADEQUATE

irb ethics statementADEQUATE

regulatory complianceNOT REPORTED

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 90% conf

The study has ethics approval from named committees and informed consent was obtained, but the paper does not explicitly state compliance with a recognized regulatory framework such as the Declaration of Helsinki.

Evidence

“The study was approved by the Research Ethics Committees of Uganda Virus Research Institute (GC/127/19/06/717), Uganda National Council of Science and Technology (HS 2645), and London School of Hygiene & Tropical Medicine (16972).”

MethodsFind in source

“Written informed parental consent was obtained for all participants.”

MethodsFind in source

absence[Throughout paper]

No statement of compliance with Declaration of Helsinki or ICH-GCP found

WARNReviewer 2· 80% conf

The study reports ethics approval from three named committees with protocol numbers and states written informed consent was obtained. However, it does not explicitly state adherence to the Declaration of Helsinki or another recognized regulatory framework, which is required for full compliance.

Evidence

MethodsFind in source

“Written informed parental consent was obtained for all participants.”

MethodsFind in source

absence[Methods, Study design]

No explicit statement of adherence to a regulatory framework such as the Declaration of Helsinki.

WARNReviewer 3· 60% conf

Named research ethics committees approved the study with protocol numbers and written informed parental consent is documented, but the paper does not state adherence to a named regulatory framework (e.g., Declaration of Helsinki or ICH-GCP), so regulatory_compliance is not reported.

Evidence

MethodsFind in source

“Written informed parental consent was obtained for all participants.”

MethodsFind in source

absence[Methods (ethics statements)]

No named regulatory framework (e.g., Declaration of Helsinki) is stated

The intervention is kangaroo mother care (skin-to-skin contact, wrap) and standard care (incubator/radiant heater). The paper does not identify specific brands or manufacturers of these devices. Since this is a behavioral/procedural trial, the dimension is not applicable per the scoring guidelines.

Sub-criteria

mycoplasma testingN/A

reagents identifiedN/A

organisms identifiedN/A

antibodies identifiedN/A

cell line authenticationN/A

software tools identifiedN/A

Per reviewer (3) — the votes behind the verdict

N/AReviewer 1· 95% conf

The trial is a procedural intervention (KMC) with no drug, biologic, device, or bench reagent as the investigational product. No key biological or chemical resources are used.

Evidence

“Neonates were naked, except for hat and diaper; placed prone and skin-to-skin on the caregiver’s chest; and secured using a KMC wrap.”

MethodsFind in source

paraphrase[Methods, Statistical analysis and Data collection]

“The paper identifies Stata version 18.1 and REDCap for data capture, but these are not considered key biological or chemical resources for this trial type.”

N/AReviewer 2· 95% conf

The study is a procedural trial (KMC versus standard care) with no investigational drug, biologic, device, or bench reagent. The dimension does not apply.

Evidence

paraphrase[Methods, Procedures]

“The intervention is kangaroo mother care (skin-to-skin contact) and standard care (incubator/radiant heater). No drugs, biologics, or devices are tested.”

N/AReviewer 3· 80% conf

The intervention is a care procedure (KMC) with no drug, biologic, device, or bench reagent, so no key biological or chemical resource is applicable; analysis software (Stata 18.1, Microsoft Excel) is named.

Evidence

“All statistical analyses were performed with Stata (version 18.1).”

Statistical analysisFind in source

paraphrase[Methods, Procedures]

“The intervention is skin-to-skin KMC positioning with a wrap; control is incubator/radiant heater care — no investigational product.”

All statistical tests are named (modified Poisson, Cox, linear regression). Effect sizes are reported with 95% CIs. Statistical software (Stata 18.1) is identified. Data presentation includes tables with n, percentages, means, SD, IQR, and a CONSORT flow diagram. Assumptions are handled by robust standard errors and standard methods. Some p-values are reported as '<0.0001' which is not exact, but this is a minor issue.

Sub-criteria

tests namedADEQUATE

exact p valuesINADEQUATE

effect sizes ciADEQUATE

data presentationADEQUATE

software identifiedADEQUATE

assumptions verifiedADEQUATE

mathematical plausibilityPASS

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

The statistical analysis is well-described with appropriate tests, effect sizes with confidence intervals, and software identification. Most p-values are exact, though some are reported as thresholds (e.g., p<0.0001).

Evidence

paraphrase[Methods, Statistical analysis]

“Risk ratios (RRs) with 95% CI were estimated for 7-day and 28-day mortality and were compared between the intervention group and the control group with modified Poisson regression models. ... Median time to stabilisation... hazard ratios (HRs) with 95% CIs comparing between the intervention and control groups were calculated with Cox proportional hazards regression models. Mean differences in admission duration... were estimated by use of linear regression models.”

paraphrase[Results, Table 2]

“Adjusted RR 0·97 (95% CI 0·74–1·28); p=0·85. ... Adjusted RR 0·88 (0·70–1·09); p=0·23.”

paraphrase[Results, Table 2]

“Adjusted RR 0·76 (0·70–0·83); p<0·0001. ... Adjusted HR 1·31 (1·13–1·53); p=0·0004.”

WARNReviewer 2· 85% conf

Statistical methods are appropriate and well-described, but several p-values are reported only as thresholds (e.g., p<0.0001) rather than exact values, which is imprecise reporting. The dimension is otherwise strong.

Evidence

paraphrase[Table 2, Secondary outcomes]

“Hypothermia at 24 h: adjusted RR 0·76 (0·70–0·83); p<0·0001”

paraphrase[Methods, Statistical analysis]

“Risk ratios (RRs) with 95% CI were estimated … with modified Poisson regression models … Median time to stabilisation … were estimated … with Cox proportional hazards regression models.”

paraphrase[Table 2, Primary and secondary outcomes]

“RR 0·97 (95% CI 0·74–1·28); p=0·85”

PASSReviewer 3· 85% conf

Statistical methods are named and appropriate (modified Poisson, Cox, linear regression), effect estimates are reported with 95% CIs, exact p-values are mostly given, software is identified, and per-group denominators are stated; the reported percentages are arithmetically plausible.

Evidence

“Risk ratios (RRs) with 95% CI were estimated for 7-day and 28-day mortality and were compared between the intervention group and the control group with modified Poisson regression models. Modified Poisson regression was preferred over binomial regression for obtaining RRs since binomial regression is prone to non-convergence.”

Statistical analysisFind in source

“From randomisation to 7 days of age, 81 (7·5%) of 1083 neonates in the intervention group and 83 (7·5%) of 1102 in the control group died (adjusted RR 0·97 [95% CI 0·74–1·28]; p=0·85).”

Results ¶2Find in source

“All analyses were performed with Microsoft Excel and Stata (version 18.1).”

MethodsFind in source

The paper states: 'Data from the study will be deposited online at LSHTM Data Compass (https://datacompass.lshtm.ac.uk/), with access subject to approval.' This is a concrete route with a named repository and conditions (managed access). For a clinical trial with patient-level data, repository deposit with a persistent identifier is not expected due to privacy; controlled access satisfies the field norm. Code sharing is not required because the analysis used standard Stata commands. Thus the single applicable criterion is adequate.

Sub-criteria

code sharingN/A

accession numbersN/A

repository depositN/A

data availability statementADEQUATE

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 80% conf

A data availability statement is present with a plan to deposit data in LSHTM Data Compass under managed access, but the data are not yet deposited and no code is shared.

Evidence

paraphrase[Data sharing]

“Data from the study will be deposited online at LSHTM Data Compass (https://datacompass.lshtm.ac.uk/), with access subject to approval.”

absence[Data sharing]

No repository deposit with persistent identifier provided

absence[Throughout paper]

No custom code sharing mentioned

PASSReviewer 2· 90% conf

The data availability statement names a concrete repository (LSHTM Data Compass) with managed access, which is adequate for patient-level data. Repository deposit and accession numbers are not applicable for identifiable trial data. Code sharing is not applicable as only standard software was used.

Evidence

paraphrase[Data sharing]

“Data from the study will be deposited online at LSHTM Data Compass (https://datacompass.lshtm.ac.uk/), with access subject to approval.”

PASSReviewer 3· 85% conf

A concrete data-availability statement is present naming a repository (LSHTM Data Compass) with managed access subject to approval; repository/accession and code-sharing criteria are not applicable to identifiable patient-level trial data.

Evidence

“Data from the study will be deposited online at LSHTM Data Compass ( https://datacompass.lshtm.ac.uk/ ), with access subject to approval.”

Data availabilityFind in source

Methods are complete for replication. Trial registration number is provided. CONSORT guideline is referenced. All pre-specified outcomes are reported in results. Limitations are discussed in detail. Conclusions are proportional to the evidence. Funding sources and conflicts of interest are declared.

Sub-criteria

funding coiADEQUATE

trial registrationADEQUATE

reporting guidelineADEQUATE

methods completenessADEQUATE

all outcomes reportedADEQUATE

limitations discussedADEQUATE

conclusions proportionalADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 95% conf

The manuscript is well-reported with detailed methods, trial registration, CONSORT adherence, transparent outcome reporting, limitations discussion, and funding/COI statements.

Evidence

paraphrase[Summary, Methods]

“This trial is registered with ClinicalTrials.gov (http://Clinicaltrials.gov), NCT02811432 (https://clinicaltrials.gov/ct2/show/NCT02811432).”

paraphrase[Methods, Study design]

“This trial is reported in accordance with CONSORT guidelines (p 4).”

paraphrase[Discussion, paragraph 5]

“This trial also had limitations. Our sample size assumed a baseline mortality rate of 25%... However, we observed lower reductions in mortality rate than were expected... Given the nature of KMC, it was impossible to mask participants and hospital staff...”

PASSReviewer 2· 95% conf

The manuscript is transparent: methods are detailed, the trial is registered, CONSORT is followed, all outcomes are reported (including non-significant), limitations are discussed, conclusions are proportional, and funding/COI are disclosed.

Evidence

paraphrase[Abstract and Methods]

“This trial is registered with ClinicalTrials.gov (http://Clinicaltrials.gov), NCT02811432 (https://clinicaltrials.gov/ct2/show/NCT02811432).”

paraphrase[Methods, Study design]

“This trial is reported in accordance with CONSORT guidelines (p 4).”

paraphrase[Discussion, paragraph 4]

“This trial also had limitations. Our sample size assumed a baseline mortality rate of 25% … we observed lower reductions in mortality rate than were expected … Given the nature of KMC, it was impossible to mask participants and hospital staff to treatment allocation, but the independent statistician … was masked.”

PASSReviewer 3· 90% conf

Reporting is complete: methods are replicable, the trial is registered (NCT02811432), CONSORT and CHEERS guidelines are referenced, all prespecified outcomes are reported, limitations are discussed at length, conclusions are proportional, and funding/COI are declared.

Evidence

paraphrase[Methods, Study design]

“This trial is reported in accordance with CONSORT guidelines ( p 4). The study was approved by the Research Ethics Committees of Uganda Virus Research Institute (GC/127/19/06/717)... This trial is registered with ClinicalTrials.gov (http://Clinicaltrials.gov) , NCT02811432”

“Our sample size would have been sufficient to detect an absolute reduction of 4·8% (relative reduction of 26·7%), even with an 18% mortality rate. However, we observed lower reductions in mortality rate than were expected from baseline data in both the control and intervention groups, which reduced our power to detect the prespecified relative reduction of 5·6%.”

Discussion ¶5Find in source

paraphrase[Acknowledgments / Declaration of interests]

“The OMWaNA trial was supported by a grant from the Joint Global Health Trials scheme... (MR/S004971/1)... Declaration of interests We declare no competing interests.”

Registered (1 ID: ClinicalTrials.gov). Reporting guideline cited: CONSORT.

ClinicalTrials.gov

NCT02811432

Reporting guidelines: CONSORT

Broken references and links

None found

References checked against Crossref, OpenAlex and Retraction Watch for retractions and resolvability, plus declared data and code links probed for whether they resolve to content matching the paper.

Checked — nothing surfaced.

Checked 44 references by DOI: 37 verified — 7 no DOI (shown, not verified).

NO DOI
Levels & trends in child mortality: report 2023
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
Born too soon: decade of action on preterm birth
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
WHO recommendations for care of the preterm or low-birth-weight infant
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
The cost-savings of implementing kangaroo mother care in Nicaragua
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
Kangaroo mother care: a practical guide
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
Official exchange rate (LCU per US$, period average)
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
NO DOI
Kangaroo mother care: implementation strategy for scale-up adaptable to different country contexts
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).

3 data/code links checked; 3 live.

datahttps://datacompass.lshtm.ac.uk/LIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttp://Clinicaltrials.govLIVEHTTP 200
Resolves, but the content could not be matched to the paper.
datahttps://clinicaltrials.gov/ct2/show/NCT02811432LIVEHTTP 200
Resolves, but the content could not be matched to the paper.

Copyediting

3 minor

Wording, consistency and formatting errors that need correcting before submission.

No major wording or formatting errors. 3 minor suggestions below.

3 copyedit issues flagged: mostly consistency, punctuation.

MINORpunctuationSummary, Findings
“adjusted relative risk [RR] 0·97 [95% CI 0·74–1·28]; p=0·85.”
→ Add closing parenthesis: 'p=0·85).'
Unbalanced parenthesis around the adjusted RR in the abstract.
MINORconsistencyTable 2 footnote and Results text
“Data are n/N (%), n (%), median (IQR), or mean (SD).”
→ Clarify whether the parenthetical values for model-derived outcomes (e.g., daily weight gain 7·8 (0·3); duration of admission 7·3 (0·2)) are SDs or standard errors.
Values such as 7·8 (0·3) and 7·3 (0·2) are implausibly small as raw-data SDs and are almost certainly standard errors from regression models; labelling them 'mean (SD)' is misleading.
MINORconsistencySummary vs Table 2
“(RR 0·88 [0·71–1·09]; p=0·229)”
→ Reconcile p=0·229 (abstract) with p=0·23 (Table 2) and the CI style ([...] vs (...)).
Same estimate is reported with slightly different rounding and bracket style in abstract versus table.

Kaimen Rigor Action ItemsFinal pass before journal submission

The published trial is methodologically robust and its headline conclusions (no early-mortality benefit; cost-effectiveness) are supported by the reported analyses and the verification checks performed. An informed reader should weigh three minor reporting gaps — the Table 2 'mean (SD)' labelling of model-derived standard errors (which warrants a correction/clarification), the threshold-only p-values, and the absent compliance statement — none of which overturn the findings or, on the evidence available, necessitate independent re-analysis.

1.
HIGHstatistics
Correct the Table 2 footnote so that parenthetical values for model-derived continuous outcomes (e.g., daily weight gain 7·8 [0·3]; duration of admission 7·3 [0·2]) are identified as standard errors from the regression models, not as 'mean (SD)' of raw data.
The integrity check flagged these values as implausibly small to be raw-data SDs — a method–result mismatch that could mislead readers and warrants a published correction.
2.
MEDIUMcopyedit
Reconcile the 28-day mortality estimate across the abstract (RR 0·88 [0·71–1·09]; p=0·229) and Table 2 (p=0·23) and standardise the bracket style ([...] vs (...)).
The same estimate is reported with inconsistent rounding and bracket formatting in the abstract versus the table.
3.
MEDIUMcopyedit
Fix the unbalanced parenthesis in the abstract findings by adding the closing ')' after 'p=0·85'.
The copyedit pass flagged an unbalanced parenthesis around the adjusted RR in the abstract.
4.
MEDIUMethics
Add an explicit statement of compliance with the Declaration of Helsinki (or ICH-GCP) to the Methods/ethics section.
All three reviewers flagged the absent regulatory-compliance statement as the only ethics reporting gap; the named approvals and consent documentation otherwise satisfy the criterion.
5.
MEDIUMstatistics
Report exact p-values for outcomes currently printed only as thresholds (e.g., replace 'p<0·0001' with the actual value in Table 2).
Threshold-only p-values constitute imprecise reporting of statistics the authors did compute, and exact values improve reproducibility.
6.
LOWdata code
Add a persistent identifier (DOI) for the LSHTM Data Compass deposit once available, and state when the dataset will be accessible.
The data-availability statement is concrete but prospective ('will be deposited'); a DOI strengthens the commitment to reproducibility.
7.
LOWdata code
Add a brief code-availability note for the Stata do-files used in the analysis.
Even where the code uses standard Stata commands, sharing or explicitly declining to share the analysis code enhances transparency.
8.
LOWstatistics
State whether the proportional-hazards assumption was tested for the Cox time-to-event analyses (e.g., test for non-proportionality).
Reviewer 2 noted the assumption is not explicitly verified for the Cox models reporting hazard ratios.
9.
LOWstatistics
Clarify the missing-data handling for the primary analysis (e.g., whether complete-case analysis was used for the primary outcome and multiple imputation only for costs).
Reviewer 1 asked for explicit specification of how missing data were handled for the primary endpoint versus the economic analysis.
10.
LOWother
Provide the rationale for the 5·6% absolute-reduction target chosen for the sample-size calculation as clinically meaningful.
Reviewer 1 suggested justifying why the specified effect size was clinically meaningful, which would aid readers interpreting the power analysis.

Full text (63,038 chars)Analyzed: Aug 10, 2026

RigorReviewer 1

RigorReviewer 2

RigorReviewer 3

RigorPre-Submission Reviewer

CopyeditAgent Scribe

CitationsCitation Integrity

StatisticsStatistics Consistency

ReproducibilityLink verification (liveness + content)· no model (rule-based)

PreregistrationDeterministic check· no model (rule-based)

The star rating is the report’s one-glance summary. Every paper starts at 5★ and loses stars for the concrete problems the review finds — so a rating is never a vague average, it’s a running total you can read line by line under “How this rating was calculated.”

How stars are deducted

Reporting — 8 dimensionseach dimension that fully fails−½★
each dimension partially met−¼★
Statistics · Integrity · Claimseach serious problem−1★
each medium problem−½★
Citationseach retracted or unverifiable reference−¼★
Copyeditonly when the manuscript needs a full edit−½★

The rating never drops below 1★, and a demonstrable critical failure (an impossible statistic, a proven ethics violation) caps it at 1★ on its own — so the stars can never look healthy when the verdict is CRITICAL.

The rating draws on a panel of agents. Three independent Kaimen Rigor reviewers grade the eight dimensions below across several independent passes (the shown verdict is their majority vote — steadier than any single run), isolate the paper’s major claims and check its own evidence backs them, and flag integrity concerns. Alongside them, a citation agent resolves every reference against Crossref, OpenAlex, and Retraction Watch; a statistics agent recomputes reported tests; and rule-based checks verify that declared data/code links actually resolve. Full text is required — an abstract-only submission is not analyzed.

The eight reporting dimensions

Scientific PremisePrior work, rationale, and how it addresses gaps.

Study Design & RigorRandomization, blinding, power, controls, replication.

Biological VariablesSex, age, species/strain, housing, demographics.

Ethical ApprovalsIRB/IACUC approval, informed consent, regulatory compliance.

Key ResourcesAntibodies, cell lines, organisms, software — identified with persistent IDs.

Statistical ReportingNamed tests, exact p-values, effect sizes, plausible summary stats.

Data & Code AvailabilityAvailability statements, repositories, accession numbers, shared code.

Reporting TransparencyMethods, registration, guidelines, outcomes, limitations, funding/COI.

Graded against NIH, MDAR, ARRIVE 2.0, CONSORT, EQUATOR, and RRID guidelines. A dimension that doesn’t apply to the study type is skipped, never penalized.

This Kaimen Rigor review is model-assisted and is not a substitute for formal expert review. It complements human evaluation by surfacing potential methodological concerns — verify each finding against the source.

Effectiveness of kangaroo mother care before clinical stabilisation versus standard care among neonates at five hospitals in Uganda (OMWaNA): a parallel-group, individually randomised controlled trial and economic evaluation.

DOI

10.1016/S0140-6736(24)00064-3

Record issued

2026-08-10

Engine

7.29.0

Exported

2026-09-22