Alpha¹ by KaimenConfidential

Kaimen Rigor review

Trial of High-Dose Oral Rifampin in Adults with Tuberculous Meningitis.

Meya DB, Cresswell FV, Dai B, Engen N, Naidoo K, Ganiem AR, Imran D, Kabahubya M, Lessells RJ, Yunivita V, Estiasari R, Tugume L, Hlabisa B, Kurniawati MY, Sagita N, Kagimu E, Maharani K, Gakuru J, Gaharu MN, Mugabi T, Kimuda S, Namombwe S, Te Brake L, Aarnoutse R, Svensson EM, Bangdiwala AS, Namanda S, Bahr NC, Musubire AK, Moosa MYS, Hamers RL, Marais S, Boulware DR, van Crevel R, Ruslami R, HARVEST Trial Team

DOI: 10.1056/NEJMoa2502866
Record issued: 2026-08-10
Engine: 7.29.0
Exported: 2026-09-21

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/fe01b0fc-956a-4783-8e00-b1146d41cce9 is authoritative.

Kaimen Rigor review

Kaimen Rigor reviewed

1 / 5

Critical — serious problems across many checks

Interventional Study

Critical

High-priority

How this rating was calculated

Started at 5★ — deductions

IntegrityIntegrity concern ×5−2.5★
ClaimsOverstated claim ×2−1★
ReportingKey resources not met−0.5★
ReportingData & code availability not met−0.5★
ReportingEthical approvals partially met−0.25★

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

Not verified in this report

The numeric-impossibility checks (GRIM/GRIMMER/DEBIT/SPRITE) did not run on this paper: the pass that reads its reported means did not complete. No reported mean was checked for arithmetic impossibility.
No data or code availability links were detected to verify.

Biggest issues

01Key resources not identified
highDisclosurereviewer-assessedFull detail
The investigational product (rifampicin) is not identified with manufacturer/source, and statistical software is not named.
“Participants in the intervention arm received high-dose oral rifampicin of ~35 mg per kilogram daily, administered for 8 weeks.”
MethodsFind in source
02Data and code not shared
highDisclosurereviewer-assessedFull detail
No data availability statement, repository deposit, or code sharing is mentioned.
03Internal contradictions in the reported numbers
mediumArithmeticreviewer-assessedFull detail
Table 1 high-dose 'Definite TBM' cell shows '144 (45.8%)' but 144/249 = 57.8%, and the high-dose diagnostic categories (144+89+44+2) sum to 279, not 249. The value 114 (45.8% of 249) reconciles the arm sum to 249 and also matches the Results text total of 218 definite cases (114+104), so 144 appears to be a transposition of 114.
“| Definite TBM | 144 (45.8) | 104 (41.6) |”
Table 1Find in source
04Internal contradictions in the reported numbers
mediumArithmeticreviewer-assessedFull detail
Table 1 reports 'Definite TBM | 144 (45.8)' for the high-dose arm, but 144/249 = 57.8% (not 45.8%), and 144+104 = 248 contradicts the abstract's '218 (43.7%)' definite cases. The percentage 45.8% matches a count of 114, and 114+104 = 218 reconciles the abstract total, indicating a likely digit transposition in the table cell.
Definite TBM | 144 (45.8) | 104 (41.6) ... Overall, 218 (43.7%) had microbiologically-confirmed definite tuberculous meningitis.
Table 1reviewer’s wording
05Internal contradictions in the reported numbers
mediumArithmeticreviewer-assessedFull detail
The number of high-dose aspiration pneumonia cases differs between the Results text ('13 of 249 (5.2%)') and Table 2 ('16 (6.4)'); both counts are internally consistent with their own percentages (13/249=5.2%, 16/249=6.4%) but contradict each other.
“aspiration pneumonia which occurred in 13 of 249 (5.2%) high-dose participants compared with 4 of 250 (1.6%) standard-dose participants”
Table 2Find in source
06Internal contradictions in the reported numbers
mediumArithmeticreviewer-assessedFull detail
The number of aspiration pneumonia cases in the high-dose group differs between the text (13 of 249) and Table 2 (16 of 249).
Text: 'aspiration pneumonia which occurred in 13 of 249 (5.2%) high-dose participants compared with 4 of 250 (1.6%) standard-dose participants.' Table 2: 'Aspiration pneumonia | 16 (6.4) | 4 (1.6) | 0.006'
Table 2reviewer’s wording

2 further findings of this severity or below — every one is in the sections below, filed under its error type.

Findings by category

101ArithmeticNumerical inconsistencies 202OverreachOverstated conclusions 03AuthenticityNone found 304DisclosureReporting gaps 05AvailabilityNone found · partly checked 106CopyeditCopyediting

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.

The trial is methodologically strong: randomized, double-blind, placebo-controlled, pre-registered, well-powered, with sound statistical methods, exact p-values, and proportional conclusions on the primary outcome. However, it has several reporting gaps — no data availability statement, generic ethics approval without named committees, no statistical software or drug manufacturer identified, no CONSORT statement — and, more seriously, demonstrable internal inconsistencies in Table 1 (definite-TBM cell) and between text and Table 2 (aspiration pneumonia), plus an overstated abstract safety claim.

Synthesized three independent reviewer runs plus a copyedit pass and verification components (citation, statistics, preregistration, integrity, claim audit). Bench-science criteria were excluded as not applicable for this human RCT. Reviewers diverged on key resources, statistical analysis, and biological-variable demographics; the integrity verification independently confirmed the internal contradictions that drive the warn/fail statuses. The provided text excerpt appears truncated, so the data availability statement may exist in an unexamined section.

Numerical inconsistencies

1 finding · worst medium

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.

Internal contradictions in the reported numbersAssessed

Recomputed 7 tests: 7 consistent, 0 inconsistent; 7 via agent-written checks.

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

CONSISTENTreported p = .250 · recomputed p = .262Reviewers 1, 2, 3
Primary outcome 6-month mortality hazard ratio p-value from reported HR and 95% CI
“Hazard Ratio, 1.17; 95% confidence interval [CI], 0.89 to 1.54; P=0.25”
Find in source
Taken as given: The CI 0.89-1.54 is a two-sided 95% confidence interval for the hazard ratio; The hazard ratio is analyzed on the log scale (log=1)
Method: Two-tailed p derived from the log(HR) and its standard error estimated from the 95% CI width assuming normality of the log-hazard-ratio estimate.
How we recomputed it: pCI(1.17, 0.89, 1.54, 1)
CONSISTENTreported p = .006 · recomputed p = .006Reviewers 1, 3
Aspiration pneumonia comparison (Table 2 counts) p-value
“Aspiration pneumonia | 16 (6.4) | 4 (1.6) | 0.006”
Find in source
Taken as given: The table cell counts are 16 events/233 non-events in high-dose and 4 events/246 non-events in standard-dose (N=249 and N=250); The 2x2 table is aspiration pneumonia vs. all other outcomes
Method: Two-tailed Fisher's exact test on the 2x2 cell counts.
How we recomputed it: pFisher2x2(16, 233, 4, 246, 0)
CONSISTENTreported p = .007 · recomputed p = .007Reviewer 1
Total bilirubin elevation comparison p-value
“Total bilirubin (≥2.6x upper limit of normal) | 24 (9.6) | 9 (3.6) | 0.007”
Find in source
Taken as given: The cell counts are 24/225 in high-dose and 9/241 in standard-dose (N=249 and N=250); The 2x2 table is bilirubin elevation vs. no elevation
Method: Two-tailed Fisher's exact test on the 2x2 cell counts.
How we recomputed it: pFisher2x2(24, 225, 9, 241, 0)
CONSISTENTreported p = .090 · recomputed p = .099Reviewers 1, 2, 3
Drug-induced liver injury comparison p-value
“Drug-induced liver injury | 20 (8.0) | 11 (4.4) | 0.09”
Find in source
Taken as given: The cell counts are 20/229 in high-dose and 11/239 in standard-dose (N=249 and N=250); The 2x2 table is DILI vs. no DILI
Method: Two-tailed Fisher's exact test on the 2x2 cell counts.
How we recomputed it: pFisher2x2(20, 229, 11, 239, 0)
CONSISTENTreported p = .120 · recomputed p = .138Reviewer 1
Any serious adverse event comparison p-value
“Participants with any serious adverse event | 84 (33.7) | 101 (40.4) | 0.12”
Find in source
Taken as given: The cell counts are 84/165 in high-dose and 101/149 in standard-dose (N=249 and N=250); The 2x2 table is any SAE vs. none
Method: Two-tailed Fisher's exact test on the 2x2 cell counts.
How we recomputed it: pFisher2x2(84, 165, 101, 149, 0)
CONSISTENTreported p = .006 · recomputed p = .006Reviewer 2
Aspiration pneumonia grade 3-5 event comparison p-value
“Aspiration pneumonia | 16 (6.4) | 4 (1.6) | 0.006”
Find in source
Taken as given: 16 and 4 are the event counts in the high-dose (N=249) and standard-dose (N=250) arms respectively; Non-events are 249-16=233 and 250-4=246; A two-sided Fisher's exact test was used for the Table 2 comparison
Method: Two-sided Fisher's exact test on the 2x2 table (16,233,4,246).
How we recomputed it: pFisher2x2(16, 233, 4, 246, 0)
CONSISTENTreported p = .007 · recomputed p = .007Reviewers 2, 3
Total bilirubin grade 3-4 abnormality comparison p-value
“Total bilirubin (≥2.6x upper limit of normal) | 24 (9.6) | 9 (3.6) | 0.007”
Find in source
Taken as given: 24 and 9 are the event counts in the high-dose (N=249) and standard-dose (N=250) arms respectively; Non-events are 249-24=225 and 250-9=241; A two-sided Fisher's exact test was used for the Table 2 comparison
Method: Two-sided Fisher's exact test on the 2x2 table (24,225,9,241).
How we recomputed it: pFisher2x2(24, 225, 9, 241, 0)

mediuminternal contradiction
Table 1 high-dose 'Definite TBM' cell shows '144 (45.8%)' but 144/249 = 57.8%, and the high-dose diagnostic categories (144+89+44+2) sum to 279, not 249. The value 114 (45.8% of 249) reconciles the arm sum to 249 and also matches the Results text total of 218 definite cases (114+104), so 144 appears to be a transposition of 114.
“| Definite TBM | 144 (45.8) | 104 (41.6) |”
Table 1Find in source
mediuminternal contradiction
Table 1 reports 'Definite TBM | 144 (45.8)' for the high-dose arm, but 144/249 = 57.8% (not 45.8%), and 144+104 = 248 contradicts the abstract's '218 (43.7%)' definite cases. The percentage 45.8% matches a count of 114, and 114+104 = 218 reconciles the abstract total, indicating a likely digit transposition in the table cell.
Definite TBM | 144 (45.8) | 104 (41.6) ... Overall, 218 (43.7%) had microbiologically-confirmed definite tuberculous meningitis.
Table 1reviewer’s wording
mediuminternal contradiction
The number of high-dose aspiration pneumonia cases differs between the Results text ('13 of 249 (5.2%)') and Table 2 ('16 (6.4)'); both counts are internally consistent with their own percentages (13/249=5.2%, 16/249=6.4%) but contradict each other.
“aspiration pneumonia which occurred in 13 of 249 (5.2%) high-dose participants compared with 4 of 250 (1.6%) standard-dose participants”
Table 2Find in source
mediuminternal contradiction
The number of aspiration pneumonia cases in the high-dose group differs between the text (13 of 249) and Table 2 (16 of 249).
Text: 'aspiration pneumonia which occurred in 13 of 249 (5.2%) high-dose participants compared with 4 of 250 (1.6%) standard-dose participants.' Table 2: 'Aspiration pneumonia | 16 (6.4) | 4 (1.6) | 0.006'
Table 2reviewer’s wording
lowinternal contradiction
The Results text reports 218 (43.7%) definite tuberculous meningitis overall, while Table 1's definite counts sum to 248 (144+104); 218/499=43.7% is internally valid but the table implies 248 definite cases.
“218 (43.7%) had microbiologically-confirmed definite tuberculous meningitis”
Table 1Find in source

Overstated conclusions

2 findings · worst medium

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 overstated beyond the evidenceAssessed
Conclusions only partially backed by the presented evidenceAssessed

11 major claims checked against the paper's own evidence: 2 not fully backed by the presented evidence (unsupported or overstated).

overstatedReviewer 1
We did not observe differences in safety.
This abstract claim is contradicted by the trial's own data showing significantly more aspiration pneumonia (p=0.006) and total bilirubin elevations (p=0.007) in the high-dose group, and the Discussion itself walks the claim back to 'No excess toxicity... except for non-fatal drug-induced hepatotoxicity and transient benign hyperbilirubinemia.'
Evidence: Aspiration pneumonia 16 (6.4%) vs 4 (1.6%), p=0.006; total bilirubin 24 (9.6%) vs 9 (3.6%), p=0.007; DILI 20 (8.0%) vs 11 (4.4%).
“We did not observe differences in safety.”
DiscussionFind in source
overstatedReviewer 2
No differences in safety were observed between study arms.
The paper's own Table 2 shows higher aspiration pneumonia (p=0.006), higher total bilirubin elevations (p=0.007), and a trend toward more drug-induced liver injury (8.0% vs 4.4%, p=0.09) in the high-dose arm, so 'We did not observe differences in safety' overstates the presented data.
Evidence: Table 2: aspiration pneumonia 16 (6.4%) vs 4 (1.6%), p=0.006; total bilirubin 24 (9.6%) vs 9 (3.6%), p=0.007.
“We did not observe differences in safety.”
AbstractFind in source
partialReviewer 3
No excess toxicity associated with high-dose rifampicin occurred except for non-fatal drug-induced hepatotoxicity and transient benign hyperbilirubinemia.
The safety data show no significant differences in most adverse events, but bilirubin elevations and drug-induced liver injury were numerically higher, though not statistically significant for DILI. So the claim is partially supported.
Evidence: Table 2 shows no significant differences in most AEs, but total bilirubin elevations were more frequent in high-dose group (9.6% vs 3.6%, p=0.007), and DILI occurred in 8.0% vs 4.4% (p=0.09).
“No difference was observed in grade 3-4 laboratory abnormalities between groups, except total bilirubin for which elevations were more frequent in high-dose participants (24 of 249 [9.6%] compared with standard-dose participants (9 of 250 [3.6%]) (Table S7). Drug-induced liver injury occurred in 20 of 249 (8.0%) participants in the high-dose group and 11 of 250 (4.4%) in the standard-dose group.”
ResultsFind in source
supportedReviewer 1
There is no evidence of a beneficial effect of high-dose rifampicin on 6-month survival.
The primary outcome (HR 1.17; 95% CI 0.89-1.54; p=0.25) directly supports the absence of benefit, and the results section states no secondary endpoint showed benefit.
Evidence: Primary outcome: 109/249 (KM 44.6%) vs 100/250 (KM 40.7%) deaths, HR 1.17, 95% CI 0.89-1.54, p=0.25.
“There is no evidence of beneficial effect from high-dose rifampicin for tuberculous meningitis”
ConclusionFind in source
supportedReviewer 1
The potential for a harmful effect cannot be ruled out.
The hazard ratio point estimate >1 with upper CI 1.54 and the observed excess early mortality in the high-dose group support that a harmful effect cannot be excluded.
Evidence: HR 1.17 (95% CI 0.89-1.54); 68 deaths (KM 27.6%) in high-dose vs 48 (KM 19.4%) in standard-dose in the first 21 days.
“and the potential for a harmful effect cannot be ruled out”
Figure 2Find in source
supportedReviewer 1
Two a priori subgroups (CSF white cells <5 cells/mm3 and those with HIV on antiretroviral therapy) had two-fold higher hazards of death with high-dose rifampicin.
The reported subgroup hazard ratios (HR 2.01 for both) directly support the observed association, and the paper appropriately notes subgroup analyses were not adjusted for multiplicity.
Evidence: HR 2.01 (95% CI 1.07-3.78) for those on ART; HR 2.01 (95% CI 1.14-3.54) for CSF WBC <5 cells/mm3.
“We observed higher risk for mortality in the high-dose group compared to the standard-dose group among those receiving antiretroviral therapy at the time of presentation (Hazard Ratio, 2.01; 95% CI, 1.07 to 3.78).”
ResultsFind in source
supportedReviewer 1
The trial was well powered to detect the hypothesized survival benefit.
The stated power calculation (500 participants, 80% power, two-sided 0.05, HR 0.68) supports the claim that the trial was adequately powered for the pre-specified effect.
Evidence: Sample size calculation with 80% power to detect HR of 0.68 at alpha 0.05.
“an overall sample size of 500 participants (250 per group) provided 80% power with a two-sided 0.05 significance level to detect a hazard ratio for death of 0.68”
Statistical analysisFind in source
supportedReviewers 2, 3
There is no evidence of beneficial effect from high-dose rifampicin for tuberculous meningitis.
The primary analysis (HR 1.17; 95% CI 0.89-1.54; P=0.25) and all secondary/subgroup analyses show no benefit, so the claim is directly backed by the presented evidence.
Evidence: Primary outcome HR 1.17 (95% CI 0.89-1.54, P=0.25); no secondary endpoint showed benefit.
“There is no evidence of beneficial effect from high-dose rifampicin for tuberculous meningitis, and the potential for a harmful effect cannot be ruled out.”
ConclusionFind in source
supportedReviewer 2
Excess mortality occurred with high-dose rifampicin among those with baseline CSF white cells <5/mm3 or receiving HIV therapy.
The pre-specified subgroup analyses report HR 2.01 (95% CI 1.14-3.54) for CSF <5 cells and HR 2.01 (95% CI 1.07-3.78) for those on ART, directly supporting the claim, though the paper notes subgroups were not adjusted for multiplicity.
Evidence: Subgroup HRs of 2.01 for both CSF <5 cells and on-ART subgroups.
“We observed higher risk for mortality in the high-dose group compared to the standard-dose group among those receiving antiretroviral therapy at the time of presentation (Hazard Ratio, 2.01; 95% CI, 1.07 to 3.78).”
ResultsFind in source
supportedReviewer 2
Higher early mortality was observed in the intervention arm.
The presented data (68 vs 48 deaths in the first 21 days) directly support higher early mortality in the high-dose arm.
Evidence: In the first 21 days, 68 deaths (27.6%) high-dose vs 48 deaths (19.4%) standard-dose.
In the first 21 days, the high-dose group had 68 deaths (Kaplan-Meier estimate: 27.6%...) compared with 48 deaths (Kaplan-Meier estimate: 19.4%...) in the standard-dose group.
Figure 2reviewer’s wording
supportedReviewer 3
Excess mortality occurred with high-dose group among those with baseline cerebrospinal fluid <5 white cells/mm3 or receiving HIV therapy.
The subgroup analyses show HRs of 2.01 for both groups, with CIs that do not include 1, so the claim is supported.
Evidence: HR 2.01 (95% CI 1.14-3.54) for CSF WBC <5, and HR 2.01 (95% CI 1.07-3.78) for those on ART.
We observed higher risk for mortality in the high-dose group compared to the standard-dose group among those receiving antiretroviral therapy at the time of presentation (Hazard Ratio, 2.01; 95% CI, 1.07 to 3.78). Among participants with CSF white cells <5 cells/mm3, we observed higher risk for mortality in the high-dose group compared to the standard-dose group (Hazard Ratio, 2.01; 95% CI, 1.14 to 3.54).
Resultsreviewer’s wording

Data authenticity concerns

None found

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.

Checked — nothing surfaced.

Reporting gaps

3 findings · worst high

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.

Data and code not sharedAssessed
Key resources not identifiedAssessed
Ethics/consent reporting incompleteAssessed

Eight Dimensions5 pass · 1 warn · 2 fail

Prior work is cited including phase II trials and a meta-analysis. The limitations of prior evidence are acknowledged, and the hypothesis is explicitly stated. All three sub-criteria are adequately reported.

Sub-criteria

prior work citedADEQUATE

premise rationaleADEQUATE

limitations addressedADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 90% conf

The premise is well grounded in prior phase II trials, a negative Vietnamese phase III trial, and a meta-analysis of Indonesian trials, with a clear hypothesis derived from the mixed evidence.

Evidence

“Four phase II trials have shown a dose-exposure relationship for rifampicin, with higher doses enhancing plasma and CSF drug concentrations.”

Introduction ¶1Find in source

“Given mixed evidence, a well powered randomized clinical trial evaluating high-dose rifampicin was deemed necessary.”

Introduction ¶1Find in source

“We hypothesized that a daily dose of ~35 mg per kg of oral rifampicin given for 8 weeks, combined with the standard anti-tuberculosis regimen, would improve 6-month survival compared with the standard ~10 mg per kg rifampicin dose.”

Introduction ¶1Find in source

PASSReviewer 2· 90% conf

The premise is well established: prior phase II dose-exposure trials, a negative Vietnamese trial, and a meta-analysis of Indonesian trials are cited, leading to a clear hypothesis and rationale for a definitive RCT.

Evidence

“Four phase II trials have shown a dose-exposure relationship for rifampicin, with higher doses enhancing plasma and CSF drug concentrations.”

Introduction ¶1Find in source

paraphrase[Background, final sentence]

“We hypothesized that a daily dose of ~35 mg per kg of oral rifampicin given for 8 weeks, combined with the standard anti-tuberculosis regimen, would improve 6-month survival.”

“Given mixed evidence, a well powered randomized clinical trial evaluating high-dose rifampicin was deemed necessary.”

Introduction ¶2Find in source

PASSReviewer 3· 95% conf

The paper establishes a clear scientific premise, citing prior research and rationalizing the need for a high-dose rifampicin trial.

Evidence

paraphrase[Background, paragraph 2]

“Four phase II trials have shown a dose-exposure relationship for rifampicin, with higher doses enhancing plasma and CSF drug concentrations. A Vietnamese trial of 817 persons with tuberculous meningitis testing 15 mg per kilogram oral rifampicin showed no benefit. However, a meta-analysis of phase II Indonesian trials indicated that rifampicin doses of 30 mg per kilogram daily... were associated with a 40% higher 6-month survival, albeit the uncertainty in the estimate of the effect was large (relative standard error 86%).”

Introduction ¶3Find in source

“Given mixed evidence, a well powered randomized clinical trial evaluating high-dose rifampicin was deemed necessary.”

Introduction ¶2Find in source

For a human RCT, randomization_method, randomization_unit, blinding_levels, power_analysis, and inclusion_exclusion all apply and are adequately reported: permuted blocks of sizes 2 or 4 stratified by center, HIV status, and severity; participants and investigators blinded with identical-appearing tablets; a power calculation (500 participants, 80% power, HR 0.68); and detailed eligibility/exclusion criteria. Outlier handling maps to the pre-specified ITT analysis population with late exclusion criteria, which is adequately defined. Bench-science criteria (replicate_distinction, controls, independent_replication) are not applicable for a single pivotal human RCT.

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

A well-designed randomized, double-blind, placebo-controlled trial with stated randomization method, stratification, blinding, a priori power calculation, and pre-specified ITT population.

Evidence

“We randomized participants in a 1:1 ratio to either high-dose or standard-dose oral rifampicin, using permutated blocks with sizes of two or four.”

MethodsFind in source

“Both participants and investigators were blinded to treatment allocation, with identical-appearing intervention and placebo tablets.”

MethodsFind in source

“an overall sample size of 500 participants (250 per group) provided 80% power with a two-sided 0.05 significance level to detect a hazard ratio for death of 0.68”

Statistical analysisFind in source

PASSReviewer 2· 92% conf

A methodologically strong RCT: randomization method, unit, blinding, a priori power calculation, and eligibility criteria are all well described.

Evidence

“We randomized participants in a 1:1 ratio to either high-dose or standard-dose oral rifampicin, using permutated blocks with sizes of two or four.”

MethodsFind in source

“an overall sample size of 500 participants (250 per group) provided 80% power with a two-sided 0.05 significance level to detect a hazard ratio for death of 0.68”

Statistical analysisFind in source

“Both participants and investigators were blinded to treatment allocation, with identical-appearing intervention and placebo tablets.”

MethodsFind in source

PASSReviewer 3· 95% conf

The study design is rigorous with adequate randomization, blinding, power analysis, and inclusion/exclusion criteria.

Evidence

“We randomized participants in a 1:1 ratio to either high-dose or standard-dose oral rifampicin, using permutated blocks with sizes of two or four.”

MethodsFind in source

“Both participants and investigators were blinded to treatment allocation, with identical-appearing intervention and placebo tablets.”

MethodsFind in source

“Assuming 50% survival in the control group, an overall sample size of 500 participants (250 per group) provided 80% power with a two-sided 0.05 significance level to detect a hazard ratio for death of 0.68 (approximately 13% absolute survival improvement) while accounting for ≤5% loss-to-follow up.”

Statistical analysisFind in source

Sex (female percentage) and age/weight are reported in Table 1. Health status includes HIV status, CD4 count, etc. Demographics missing race/ethnicity, which is not reported.

Sub-criteria

demographicsNOT REPORTED

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· 85% conf

Sex, age, weight, and key health/demographic characteristics (HIV status, severity) are reported for the trial population.

Evidence

“222 of 499 (44%) were female”

ResultsFind in source

“The median age was 37 years (interquartile range, 28 to 45)”

Table 1Find in source

“Overall, 304 of 499 (61%) were persons living with HIV; 125 of 304 (41%) were receiving antiretroviral therapy”

ResultsFind in source

PASSReviewer 2· 85% conf

Sex, age, weight, and key health characteristics (HIV status, CD4 count, disease severity) are thoroughly reported for both arms.

Evidence

“The median age was 37 years (interquartile range, 28 to 45), and 222 of 499 (44%) were female.”

Table 1Find in source

“Overall, 304 of 499 (61%) were persons living with HIV; 125 of 304 (41%) were receiving antiretroviral therapy, and 115 of 226 (51%) had CD4 counts <100 cells/mm 3 .”

ResultsFind in source

PASSReviewer 3· 95% conf

Sex, age, weight, and health status are reported, but race/ethnicity is not mentioned.

Evidence

paraphrase[Table 1]

“Female sex, no. (%) | High-dose: 101 (40.6) | Standard-dose: 121 (48.4)”

paraphrase[Table 1]

“Median age (IQR), years | 38 (28-46) | 35 (28-45) | Median weight (IQR), kg | 54 (49-60) | 55 (45-62)”

absence[Table 1]

Demographics missing race/ethnicity

The paper states 'Research ethics committees and relevant regulatory approvals occurred at all sites' but does not name the committees or provide protocol numbers. Informed consent is adequately described. Compliance with a specific framework (e.g., Declaration of Helsinki) is not mentioned.

Sub-criteria

iacuc statementN/A

informed consentADEQUATE

irb ethics statementINADEQUATE

regulatory complianceNOT REPORTED

Per reviewer (3) — the votes behind the verdict

WARNReviewer 1· 70% conf

Ethics and regulatory approval are stated but not tied to a named committee, protocol number, or recognized framework, so the statements are present but under-specific.

Evidence

“Research ethics committees and relevant regulatory approvals occurred at all sites.”

MethodsFind in source

“Written informed consent was obtained from all participants, or surrogates for people lacking capacity to consent.”

MethodsFind in source

absence[Methods, Study Design]

No recognized regulatory framework is named by name.

WARNReviewer 2· 65% conf

Informed consent is well described, but the ethics-approval statement is generic (no named committee or protocol number) and no named regulatory framework is cited.

Evidence

“Research ethics committees and relevant regulatory approvals occurred at all sites.”

MethodsFind in source

“Written informed consent was obtained from all participants, or surrogates for people lacking capacity to consent.”

MethodsFind in source

WARNReviewer 3· 90% conf

Ethics approval is mentioned but not specific, and regulatory compliance with a named framework is not stated.

Evidence

“Research ethics committees and relevant regulatory approvals occurred at all sites.”

MethodsFind in source

“Written informed consent was obtained from all participants, or surrogates for people lacking capacity to consent.”

MethodsFind in source

absence[Methods]

Regulatory compliance not stated

The study uses rifampicin as the investigational drug, but its manufacturer or source is not provided. The statistical software used for analysis is not mentioned in the main text. Both applicable sub-criteria are inadequately reported (reagents_identified) or not reported (software_tools_identified).

Sub-criteria

mycoplasma testingN/A

reagents identifiedINADEQUATE

organisms identifiedN/A

antibodies identifiedN/A

cell line authenticationN/A

software tools identifiedNOT REPORTED

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 70% conf

The investigational product (rifampicin) is identified with dose, formulation, and regimen; bench-science criteria are not applicable, and no bespoke software is reported.

Evidence

“This dose was achieved by providing extra rifampicin tablets (900 mg if ≤38 kilograms, otherwise 1200 mg) in addition to the standard tuberculosis fixed-dose combination tablets that included rifampicin (~10 mg per kilogram)”

MethodsFind in source

absence[Methods, Intervention]

No manufacturer/source for the rifampicin or placebo is stated.

WARNReviewer 2· 60% conf

The investigational drug (rifampicin) is identified with dose, formulation, and regimen, but the manufacturer/source of the tablets is not stated, and the statistical software is not identified.

Evidence

“This dose was achieved by providing extra rifampicin tablets (900 mg if ≤38 kilograms, otherwise 1200 mg) in addition to the standard tuberculosis fixed-dose combination tablets”

MethodsFind in source

absence[Methods, Statistical Analysis]

Statistical software is not identified.

FAILReviewer 3· 90% conf

The investigational product (rifampicin) is not identified with manufacturer/source, and statistical software is not named.

Evidence

“Participants in the intervention arm received high-dose oral rifampicin of ~35 mg per kilogram daily, administered for 8 weeks.”

MethodsFind in source

absence[Methods, Statistical Analysis]

Software not identified

All applicable sub-criteria are met except assumptions_verified (partially adequate) and software_identified (not reported). However, the proportion of adequate criteria is above 60%, so the dimension passes.

Sub-criteria

tests namedADEQUATE

exact p valuesADEQUATE

effect sizes ciADEQUATE

data presentationADEQUATE

software identifiedNOT REPORTED

assumptions verifiedINADEQUATE

mathematical plausibilityPASS

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

Statistical tests are named, exact p-values and effect sizes with confidence intervals are reported, and summary percentages are arithmetically consistent; only statistical software identification is missing.

Evidence

“Hazard Ratio, 1.17; 95% confidence interval [CI], 0.89 to 1.54; P=0.25”

ResultsFind in source

“The treatment effect of high-dose rifampicin vs standard-dose rifampicin was estimated via hazard ratio from an unadjusted Cox regression.”

Statistical analysisFind in source

absence[Methods, Statistical Analysis]

Statistical software/version is not identified.

WARNReviewer 2· 70% conf

The primary analysis is sound and reported by estimation with CIs, but statistical software is not identified, and Table 1 contains a demonstrable arithmetic inconsistency in the definite-TBM cell.

Evidence

paraphrase[Results, Primary Outcome]

“(Hazard Ratio, 1.17; 95% confidence interval [CI], 0.89 to 1.54; P=0.25)”

“Definite TBM | 144 (45.8) | 104 (41.6)”

Table 1Find in source

absence[Methods, Statistical Analysis]

Statistical software not identified.

PASSReviewer 3· 90% conf

Statistical tests are named, exact p-values and effect sizes with CIs are reported, data presentation is adequate, but assumptions verification and software identification are lacking.

Evidence

“The treatment effect of high-dose rifampicin vs standard-dose rifampicin was estimated via hazard ratio from an unadjusted Cox regression.”

Statistical analysisFind in source

“Hazard Ratio, 1.17; 95% confidence interval [CI], 0.89 to 1.54; P=0.25”

ResultsFind in source

“Hazard Ratio, 1.17; 95% confidence interval, 0.89 to 1.54”

ResultsFind in source

The paper does not include a data availability statement, does not mention any repository for data deposition, and does not mention code sharing. All applicable sub-criteria are not reported.

Sub-criteria

code sharingNOT REPORTED

accession numbersN/A

repository depositNOT REPORTED

data availability statementNOT REPORTED

Per reviewer (3) — the votes behind the verdict

FAILReviewer 1· 55% conf

No data availability statement is present in the provided text; repository/accession and code sharing are not applicable for patient-level clinical trial data.

Evidence

absence[Entire provided text]

No data availability statement is present in the provided text.

FAILReviewer 2· 60% conf

No data availability statement is present in the manuscript text.

Evidence

absence[Full manuscript text]

No data availability statement is present.

FAILReviewer 3· 95% conf

No data availability statement, repository deposit, or code sharing is mentioned.

Evidence

absence[End of paper]

No data availability statement

absence[Methods]

No repository deposit

absence[Methods]

No code sharing

Trial registration (ISRCTN15668391) is provided. Methods are sufficiently detailed for replication. All outcomes are reported. Limitations are discussed. Conclusions are proportional. Funding and COI are disclosed. CONSORT or other reporting guideline is not mentioned.

Sub-criteria

funding coiADEQUATE

trial registrationADEQUATE

reporting guidelineNOT REPORTED

methods completenessADEQUATE

all outcomes reportedADEQUATE

limitations discussedADEQUATE

conclusions proportionalADEQUATE

Per reviewer (3) — the votes behind the verdict

PASSReviewer 1· 85% conf

Methods are detailed, the trial is registered (ISRCTN15668391), all outcomes are reported including null results, limitations are discussed, conclusions are proportional, and funding/COI are disclosed.

Evidence

paraphrase[Methods, Study Design; Abstract]

“HARVEST trial was a double-blind, placebo-controlled, randomized phase III trial with parallel group design, conducted in Indonesia, South Africa, and Uganda (ISRCTN15668391).”

“There is no evidence of beneficial effect from high-dose rifampicin for tuberculous meningitis, and the potential for a harmful effect cannot be ruled out.”

ConclusionFind in source

“Funded by the Department of Health and Social Care, the Foreign, Commonwealth & Development Office, the United Kingdom Medical Research Council, and Wellcome Trust Medical Research Council Joint Global Health Trials (MR/S004963/1).”

Support StatementFind in source

PASSReviewer 2· 85% conf

Methods are detailed, the trial is registered, negative/null outcomes are reported transparently, limitations are discussed, and conclusions are proportional; only a CONSORT guideline statement is absent.

Evidence

“Clinical Trial numberISRCTN15668391”

AbstractFind in source

“The major limitation common to all tuberculous meningitis studies is the potential for misdiagnosis.”

DiscussionFind in source

“There is no evidence of beneficial effect from high-dose rifampicin for tuberculous meningitis, and the potential for a harmful effect cannot be ruled out.”

ConclusionFind in source

PASSReviewer 3· 95% conf

The trial is registered, methods are detailed, outcomes are reported, limitations are discussed, conclusions are proportional, and funding is disclosed. However, a reporting guideline is not referenced.

Evidence

“Clinical Trial numberISRCTN15668391”

AbstractFind in source

Support StatementFind in source

“The major limitation common to all tuberculous meningitis studies is the potential for misdiagnosis.”

DiscussionFind in source

Registered (1 ID: ISRCTN). No reporting guideline cited.

ISRCTN

ISRCTN15668391

Broken references and links

None found · partly checked

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.

Nothing surfaced — but not everything feeding this category ran (missing: data/code link verification), so read this as a partial clean bill.

Checked 33 references by DOI: 32 verified — 1 no DOI (shown, not verified).

NO DOI
WHO consolidated guidelines on tuberculosis: Module 4: Treatment - Drug-susceptible tuberculosis treatment
No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).

Copyediting

1 finding · worst low

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

Wording or formatting errors that need correctingAssessed

10 copyedit issues flagged (1 major): mostly consistency, typo, grammar.

MAJORconsistencyTable 1, Diagnostic category row
“Definite TBM | 144 (45.8) | 104 (41.6)”
→ Change 144 to 114 so the percentage 45.8 and the abstract's total of 218 definite cases reconcile (144+104=248 ≠ 218; 114+104=218).
144/249 = 57.8%, not 45.8%; the cell appears to be a digit transposition of 114.
MINORtypoAbstract
“Clinical Trial numberISRCTN15668391”
→ Clinical Trial number ISRCTN15668391
Missing space after 'number'.
MINORtypoResults, Secondary Outcomes
“The median of Liverpool outcome score was 53 (interquartile rage: 1 to 72)”
→ interquartile range: 1 to 72
'rage' should be 'range'.
MINORgrammarMethods, Randomization and Blinding
“using permutated blocks”
→ using permuted blocks
'permutated' is non-standard; 'permuted' is correct.
MINORconsistencyMeSH Keywords vs. body text
“rifampin”
→ rifampicin
MeSH keyword uses 'rifampin' while the body text consistently uses 'rifampicin'.
MINORconsistencyThroughout
“antituberculous / anti-tuberculous”
→ Use one spelling consistently
Both spellings appear in the text.
MINORtypoResults, Secondary Outcomes
“The median of Liverpool outcome score was 53 (interquartile rage: 1 to 72)”
→ Change 'rage' to 'range'.
Spelling typo.
MINORconsistencyAbstract / Results, Secondary Outcomes
“high-dose rifampin group”
→ Use 'rifampicin' consistently throughout (elsewhere the paper uses 'rifampicin').
Terminology alternates between 'rifampin' and 'rifampicin'.
MINORtypoAbstract, last line
“Clinical Trial numberISRCTN15668391”
→ Add a space: 'Clinical Trial number ISRCTN15668391'
Missing space between 'number' and the registration number.
MINORpunctuationMethods, paragraph 1
“Cepheid, CA, ,,”
→ Remove extra commas: 'Cepheid, CA'
Double comma appears.

Kaimen Rigor Action ItemsFinal pass before journal submission

The published trial is robust in its core design and primary analysis, which an informed reader can rely on; however, the missing data availability statement, the Table 1 digit transposition, the text-vs-Table-2 aspiration-pneumonia discrepancy, and the overstated abstract safety claim warrant a correction or erratum. The ethics/resource/software reporting gaps should also be weighed when assessing reporting completeness, and an independent re-analysis of the flagged safety and baseline numbers is advisable.

1.
HIGHstatistics
Correct Table 1's 'Definite TBM' cell in the high-dose arm: change '144 (45.8)' to '114 (45.8)' so the percentage, the arm sum (249), and the abstract total of 218 definite cases (114+104) all reconcile.
144/249 = 57.8% is arithmetically impossible and 144+104=248 contradicts the abstract's 218 definite cases, so this is a demonstrable data error needing an erratum.
2.
HIGHdata code
Add a data availability statement specifying managed access (named platform or data-access committee, request conditions, and timeframe) for the trial data.
No data availability statement is present, which is a required element for a data-driven clinical trial and a key reproducibility gap.
3.
HIGHstatistics
Reconcile the high-dose aspiration-pneumonia count between the Results text (13 of 249, 5.2%) and Table 2 (16, 6.4%) by correcting whichever is wrong.
Both counts are internally consistent with their own percentages but contradict each other, an unresolved internal inconsistency an informed reader would flag.
4.
HIGHrigor
Temper the abstract claim 'We did not observe differences in safety' to acknowledge the significantly higher aspiration pneumonia (p=0.006) and total bilirubin elevations (p=0.007) in the high-dose arm, consistent with the Discussion's qualifier.
The claim is contradicted by the trial's own Table 2 data and is an overstatement that undermines the abstract's accuracy.
5.
HIGHethics
Name the specific research ethics committees/IRBs that approved the trial at each site, ideally with protocol numbers, and state compliance with a recognized framework such as the Declaration of Helsinki or ICH-GCP in the Methods.
The current statement 'Research ethics committees and relevant regulatory approvals occurred at all sites' is generic and under-specified for a multi-site human trial.
6.
HIGHstatistics
Identify the statistical software package and version used for all analyses in the Methods/Statistical Analysis section.
The software is not identified anywhere in the text, impeding reproducibility of the analyses.
7.
MEDIUMrigor
Provide the manufacturer/source of the high-dose rifampicin tablets and the standard fixed-dose combination tablets in the Methods/Intervention section.
The investigational product's source is not stated, which is part of key-resource identification.
8.
MEDIUMreporting
Add a CONSORT guideline/checklist statement referencing adherence for the trial.
The trial is transparently reported but no reporting guideline is cited.
9.
MEDIUMcopyedit
Fix minor copyedit errors: 'Clinical Trial numberISRCTN15668391' → add a space; 'interquartile rage' → 'range'; 'permutated' → 'permuted'; remove the double comma in 'Cepheid, CA, ,,'.
These are typos/grammar errors that reduce polish and are noticeable to readers.
10.
MEDIUMcopyedit
Standardize terminology: use 'rifampicin' consistently (replace 'rifampin' in MeSH keywords and abstract) and pick a single spelling of 'antituberculous'/'anti-tuberculous' throughout.
Inconsistent spelling of the drug name and condition across the paper is a consistency issue.
11.
LOWdata code
Consider sharing the analysis code in a public repository (e.g., GitHub) linked from the data availability statement.
No code is shared, limiting reproducibility checks for interested readers.

Full text (33,265 chars)Analyzed: Aug 10, 2026

RigorReviewer 1

RigorReviewer 2

RigorReviewer 3

RigorPre-Submission Reviewer

CopyeditAgent Scribe

CitationsCitation Integrity

StatisticsStatistics Consistency

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.

Trial of High-Dose Oral Rifampin in Adults with Tuberculous Meningitis.

DOI

10.1056/NEJMoa2502866

Record issued

2026-08-10

Engine

7.29.0

Exported

2026-09-21