# Ordinary thiamine (vitamin B1) × HbA1c reduction in type 2 diabetes with CKD stage 3 Verdict 3084 · Evidence Check Stored-material review date: 2026-10-05 Original evidence date: 2026-09-15 Review-criteria version: 1.0 Public page/material preparation date: 2026-10-29 Clinical papers were not read directly; existing review summaries were checked. No new clinical-literature search was performed. Unverified does not mean that research is absent or that there is no effect. ## Assessment scope - Population: Adults with type 2 diabetes and CKD stage 3; not all adults with diabetes - Ingredient, formulation and route: Ordinary thiamine (vitamin B1); single-ingredient oral thiamine capsules; salt, release formulation and manufacturer unconfirmed - Dose: 300 mg/day, 150 mg/capsule; a study regimen, not a dosing recommendation - Comparator: Placebo; exact composition and arm-level background diabetes care unconfirmed - Measured outcome: Between-group difference in median HbA1c change at 24 weeks; a secondary trial endpoint - Duration: 24 weeks of treatment; subsequent persistence unconfirmed - Questions outside this scope: Benfotiamine, injections and other renal conditions; Substituting the primary (1st) renal outcome for HbA1c; Reusing S20's historical arithmetic for 3-month means or post-treatment results as the 24-week median effect; Confirmation of clinical-event prevention, an optimal dose, medication replacement or individual safety ## Checklist items ### N01 · What single question is defined by this table's population, thiamine, route, comparator, duration and HbA1c endpoint? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The original assessment question and claim conditions identify HbA1c with ordinary oral thiamine, 300 mg/day for 24 weeks versus placebo, in adults with T2D and CKD stage 3. - Why unverified: The question boundaries agree across the manuscript and structured fields. Verification of question identity is distinguished from confirmation of efficacy. - Reason for missing information: The question's scope is verified; gaps concerning salt, background treatment and analysis values remain in the following items. - Required materials and conditions: Records identifying the current question and the direct trial's population, route, formulation, duration and comparator. - Materials to obtain first: Records identifying the current question and the direct trial's population, route, formulation, duration and comparator. - Further research: When collecting new material, apply these eligibility criteria and separately record excluded molecules, renal conditions and endpoints. - Conditions for verification: Fix the question as HbA1c change with ordinary single-ingredient oral thiamine, 300 mg/day for 24 weeks versus placebo, in adults with T2D and CKD stage 3; distinguish other renal conditions, molecules and endpoints. - Why this condition is needed: Changing the population or clinical endpoint would produce a different effect table. - Application exceptions: Primary (1st) renal outcomes, benfotiamine, injections and other diabetes populations are not reused as direct results for this claim. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are internally designed; M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs - Related sources: S02 ### N02 · Can the direct trial's full CKD stage 3 and T2D inclusion and exclusion criteria be verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S02 is a trial in adults with T2D and CKD stage 3. The stored abstract summary does not contain detailed diagnostic criteria, eGFR ranges or the full inclusion and exclusion criteria. - Why unverified: The population label is verified, but detailed criteria and representativeness remain unverified. - Reason for missing information: The direct trial's detailed eligibility criteria, recruitment setting and distribution of renal conditions. - Required materials and conditions: The direct trial's diagnostic, eGFR, age, inclusion and exclusion criteria, and recruitment flow. - Materials to obtain first: The direct trial's diagnostic, eGFR, age, inclusion and exclusion criteria, and recruitment flow. - Further research: In comparative studies defining the same population, record criteria and recruitment flow in advance and separate other renal conditions. - Conditions for verification: Beyond identifying CKD3 and adult T2D, verify diagnostic criteria, renal-function ranges, age, exclusions and recruitment setting, or specify the unverified parts. - Why this condition is needed: Microalbuminuria and CKD3 cannot be assumed to define the same population. - Application exceptions: S01's microalbuminuria criteria are not used to fill gaps in S02's CKD3 selection criteria. - Remaining uncertainty: The direct trial's detailed eligibility criteria, recruitment setting and distribution of renal conditions. - Referenced methods: M03 §3.2.1; detailed population conditions are internally designed; M08 §8.1–8.2.3; internal application - Related sources: S02 ### N03 · Were baseline thiamine status and the method for identifying deficiency verified in the eligible population? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The baseline vitamin-status and deficiency-assessment values are blank, with an unconfirmed information status. - Why unverified: A CKD3/T2D population alone does not establish that this was a B1-deficiency treatment trial. - Reason for missing information: Direct information on baseline concentrations, selection by deficiency and stratum-specific effects. - Required materials and conditions: Baseline thiamine assay, concentrations, deficiency criteria and arm-level distributions. - Materials to obtain first: Baseline thiamine assay, concentrations, deficiency criteria and arm-level distributions. - Further research: In studies measuring deficiency status in advance, separately analyse stratum-specific effects and interactions. - Conditions for verification: Verify baseline B1 measurement, units, deficiency criteria and population distribution; distinguish deficiency screening from effect-modification analysis. - Why this condition is needed: Results in a general study population must not be conflated with treatment effects in deficiency. - Application exceptions: Deficiency-specific effects are not mandatory for the narrow HbA1c question; unknown status is not filled in as deficient or non-deficient. - Remaining uncertainty: Direct information on baseline concentrations, selection by deficiency and stratum-specific effects. - Referenced methods: M03 §3.2.1; detailed population conditions are internally designed; M03 §3.2.1–3.2.4; detailed dose, persistence and effect-modification conditions are internally designed - Related sources: S02 ### N04 · To what extent are ordinary thiamine's exact salt, release formulation, manufacturer and single-ingredient status verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The observed molecule is thiamine (salt unconfirmed), supplied as single-ingredient oral thiamine capsules. Salt, release formulation and manufacturer remain unconfirmed. - Why unverified: The molecule and capsule formulation are verified, but exact formulation details remain unresolved. - Reason for missing information: Exact salt, release formulation, manufacturing origin and batch specifications. - Required materials and conditions: Trial-product ingredients, salt, release formulation, manufacturer and batch specifications. - Materials to obtain first: Trial-product ingredients, salt, release formulation, manufacturer and batch specifications. - Further research: Preserve product specifications in single-ingredient comparisons with identified formulations; screen benfotiamine and combination products separately. - Conditions for verification: Distinguish the identity of ordinary single-ingredient thiamine capsules from the exact salt, release formulation and manufacturing origin; do not fill unconfirmed details. - Why this condition is needed: Trials of different molecules or formulations must not be combined as identical exposures. - Application exceptions: Benfotiamine, injections or other salts do not fill gaps in the unconfirmed product specifications. - Remaining uncertainty: Exact salt, release formulation, manufacturing origin and batch specifications. - Referenced methods: M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed - Related sources: S02 ### N05 · Was actual oral administration verified in the direct trial? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The direct study's actual administration route was verified as oral and attributed to S02. - Why unverified: The actual study-route field was checked, rather than reading only the planned route in the claim conditions. - Reason for missing information: Actual oral administration is verified; absorption and formulation differences remain separate gaps. - Required materials and conditions: Records of the trial intervention's actual administration route. - Materials to obtain first: Records of the trial intervention's actual administration route. - Further research: Manage studies using other routes in separate eligibility groups; do not combine them with oral results. - Conditions for verification: Compare the planned question route with the observed trial route and link the source location supporting actual oral administration. - Why this condition is needed: The planned route must be distinguished from the route actually used. - Application exceptions: Verification of an oral route does not verify effects of injections or oral bioavailability. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed - Related sources: S02 ### N06 · Are the nominal doses of 300 mg/day and 150 mg/capsule, and actual exposure, verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The recorded dose value is 300, described as 300 mg/day and 150 mg/capsule. Actual intake and detailed dosing frequency remain unconfirmed. - Why unverified: The nominal dose is verified, but it cannot be substituted for every participant's actual intake. - Reason for missing information: Dosing frequency, adherence and actual cumulative exposure. - Required materials and conditions: Trial dosing instructions, frequency, medication-taking records and actual intake. - Materials to obtain first: Trial dosing instructions, frequency, medication-taking records and actual intake. - Further research: Collect medication-taking records in same-dose comparisons; distinguish actual-exposure analysis from the effect of assignment. - Conditions for verification: Link the nominal daily and per-capsule doses; verify dosing frequency, actual intake and adherence separately. - Why this condition is needed: Differences between nominal dose and actual exposure may affect the analysis. - Application exceptions: Do not convert 300 to another unit or total dietary B1; do not turn the study dose into individual dosing instructions. - Remaining uncertainty: Dosing frequency, adherence and actual cumulative exposure. - Referenced methods: M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed; M03 §3.2.3, M08 §8.2.2–8.2.3; internal application - Related sources: S02 ### N07 · Are the treatment duration and measurement time point of the direct HbA1c result verified? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: Actual treatment duration is verified as 24 weeks; this table's endpoint is the between-group difference in median HbA1c change at 24 weeks. Subsequent persistence remains unconfirmed. - Why unverified: The time point for the current 24-week question is verified. Long-term and post-treatment effects are outside that verification. - Reason for missing information: The 24-week time point itself has no identified gap; subsequent persistence remains in N29. - Required materials and conditions: The 24-week treatment and measurement schedule, with time-point labels for each result. - Materials to obtain first: The 24-week treatment and measurement schedule, with time-point labels for each result. - Further research: Prespecify measurement times for each outcome and report post-treatment follow-up as a separate outcome. - Conditions for verification: Link 24 weeks of treatment to the corresponding HbA1c change time point; separate subsequent persistence. - Why this condition is needed: S20's on-treatment and post-treatment time points must not be combined with the current 24-week question. - Application exceptions: Verification of duration does not establish persistence of the effect afterward. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M03 §3.2.1–3.2.4; detailed dose, persistence and effect-modification conditions are internally designed - Related sources: S02 ### N08 · To what extent are the placebo comparator's identity, exact composition and masking conditions verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S02 is a randomized placebo-controlled trial. The claim conditions mark the placebo's exact composition as unconfirmed. - Why unverified: A placebo comparator is verified, but its composition and successful masking were not verified. - Reason for missing information: Placebo ingredients, matched appearance or taste, and masking implementation. - Required materials and conditions: Placebo ingredients, appearance, allocation codes and masking procedures. - Materials to obtain first: Placebo ingredients, appearance, allocation codes and masking procedures. - Further research: In comparisons using a matched placebo, record composition, masking procedures and their success. - Conditions for verification: Verify the placebo comparison and distinguish its composition, appearance, presence of active ingredients and details of masking implementation. - Why this condition is needed: The comparator's own exposure and measurement differences must not be misattributed to B1. - Application exceptions: The labels RCT or placebo alone do not establish successful double masking or allocation concealment. - Remaining uncertainty: Placebo ingredients, matched appearance or taste, and masking implementation. - Referenced methods: M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed; M08 §8.1–8.2.3; internal application; M03 §3.2.3, M08 §8.2.2–8.2.3; internal application - Related sources: S02 ### N09 · Can diabetes treatments, rescue medication and treatment changes in both arms be verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The comparator field records the exact composition and arm-level details of shared diabetes care as unconfirmed. - Why unverified: The current summary does not verify details showing that shared treatment was maintained identically. - Reason for missing information: Arm-level medications, doses, treatment changes, rescue treatment and analytical implications. - Required materials and conditions: Arm-level diabetes medications, renal treatments, and changes or rescue-treatment records. - Materials to obtain first: Arm-level diabetes medications, renal treatments, and changes or rescue-treatment records. - Further research: Prespecify background care and rescue criteria, report changes by arm and link them to analysis of the assignment effect. - Conditions for verification: Verify arm-level drugs, doses, changes, rescue treatment and analytical handling, or retain these as limitations on attributing the intervention's contribution. - Why this condition is needed: Differences in other treatments must not be attributed to the contribution of added thiamine. - Application exceptions: Do not require all medication changes to be identical; verify the actual effect of interest and how changes were handled. - Remaining uncertainty: Arm-level medications, doses, treatment changes, rescue treatment and analytical implications. - Referenced methods: M03 §3.2.3, M08 §8.2.2–8.2.3; internal application; M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed - Related sources: S02 ### N10 · Are total dietary B1 and concurrent-vitamin exposures verified by arm? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The total-dietary-intake and concurrent-vitamin values are blank and their information status is unconfirmed. Their linked-source lists are also empty. - Why unverified: The trial dose of 300 mg/day cannot be described as total B1 including diet or the entirety of the isolated added-exposure difference. - Reason for missing information: Arm-level details of total dietary intake and all concurrent vitamins. - Required materials and conditions: Arm-level dietary B1, supplements and concurrent-vitamin intake records. - Materials to obtain first: Arm-level dietary B1, supplements and concurrent-vitamin intake records. - Further research: In comparisons measuring concurrent exposures, record total intake and background vitamins to analyse the applicability of the main effect. - Conditions for verification: Distinguish the trial supplement dose from total diet, other supplements and concurrent vitamins. - Why this condition is needed: Exposure outside the trial medication must not be assumed equal. - Application exceptions: An empty linked-source list does not mean no concurrent vitamins were used; do not arbitrarily reclassify this as a combination-product trial. - Remaining uncertainty: Arm-level details of total dietary intake and all concurrent vitamins. - Referenced methods: M03 §3.2.2–3.2.3; direction labels and test adaptations are internally designed; M03 §3.2.3, M08 §8.2.2–8.2.3; internal application - Related sources: S02 ### N11 · Are randomization sequence, concealment and baseline balance verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S02 is recorded as a randomized placebo-controlled trial. Sequence generation, allocation concealment, separation of responsibilities and the full baseline table remain unverified in the current abstract summary. - Why unverified: The design label is verified, but the available material does not establish the adequacy of implementation. - Reason for missing information: Original information on the allocation sequence, concealment implementation and baseline balance. - Required materials and conditions: Trial methods describing sequence generation, concealment, separation of responsibilities and the baseline table. - Materials to obtain first: Trial methods describing sequence generation, concealment, separation of responsibilities and the baseline table. - Further research: Report implementation methods and arm-level baseline data in randomized comparisons, with planning before analysis. - Conditions for verification: Distinguish the randomized placebo-controlled design from actual sequence generation, concealment responsibilities and baseline balance. - Why this condition is needed: An RCT label does not replace review of bias in its implementation. - Application exceptions: Do not judge small samples or unread full texts as evidence of actual randomization failure. - Remaining uncertainty: Original information on the allocation sequence, concealment implementation and baseline balance. - Referenced methods: M08 §8.1–8.2.3; internal application - Related sources: S02 ### N12 · Can baseline HbA1c, renal function and balance of relevant prognostic factors be verified in both arms? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: Historical access was at abstract level; arm-level baseline HbA1c values and adjustment models are not verified in the currently stored material. - Why unverified: S20's baseline values of 9.2%/8.8% belong to another trial family and cannot fill S02's baseline gaps. - Reason for missing information: The direct trial's baseline HbA1c, renal function, prognostic factors and adjustment model. - Required materials and conditions: S02 arm-level baseline HbA1c, eGFR, medications, prognostic factors and adjustment plan. - Materials to obtain first: S02 arm-level baseline HbA1c, eGFR, medications, prognostic factors and adjustment plan. - Further research: Report outcome tables including baseline values and prespecified adjustment models to assess the implications of imbalance. - Conditions for verification: Verify the direct trial's arm-level baseline outcomes, renal status and analytical adjustment variables. - Why this condition is needed: Baseline differences and values from another trial family must not be conflated with the direct trial's effect. - Application exceptions: Do not assume a simple between-group median-change contrast necessarily uses a baseline-adjusted model. - Remaining uncertainty: The direct trial's baseline HbA1c, renal function, prognostic factors and adjustment model. - Referenced methods: M03 §3.2.1; detailed population conditions are internally designed; M08 §8.1–8.2.3; internal application; M08 §8.2.2, M15 §15.3.1; internal application - Related sources: S02, S20 ### N13 · Are medication adherence, intervention deviations, rescue treatment and their handling verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The nominal dose is verified, but the current summary lacks arm-level details on adherence, intervention changes and rescue treatment. - Why unverified: Nominal 24-week treatment cannot be read as complete medication-taking by every participant. - Reason for missing information: Actual adherence, arm changes, discontinuation, rescue treatment and handling details. - Required materials and conditions: Medication-taking records, arm changes, intervention discontinuation, rescue treatment and analytical handling. - Materials to obtain first: Medication-taking records, arm changes, intervention discontinuation, rescue treatment and analytical handling. - Further research: In comparisons collecting post-allocation deviation data, plan assignment-effect and adherence-effect analyses separately. - Conditions for verification: Verify actual medication-taking, arm changes, rescue drugs and how assignment effects and adherence effects were handled. - Why this condition is needed: The contribution of the chosen intervention must be distinguished from actual deviations. - Application exceptions: Treatment discontinuation and missing follow-up are separate; do not use the same denominator without support. - Remaining uncertainty: Actual adherence, arm changes, discontinuation, rescue treatment and handling details. - Referenced methods: M03 §3.2.3, M08 §8.2.2–8.2.3; internal application; M08 §8.2.2, M15 §15.3.1; internal application - Related sources: S02 ### N14 · Are the HbA1c assay, units, quality control and arm-level measurement conditions verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The HbA1c endpoint and 24-week time point are recorded, but assay and assessment details remain unconfirmed. No separate systematic search for assay interference was performed. - Why unverified: The endpoint is verified; equivalent measurement conditions and absence of interference were not verified. - Reason for missing information: Assay, measurement units, quality control, arm-level conditions and interference status. - Required materials and conditions: HbA1c assay, units, blood-sampling schedule, quality control and assay-interference information. - Materials to obtain first: HbA1c assay, units, blood-sampling schedule, quality control and assay-interference information. - Further research: Use matching assays and measurement schedules, assess interference and report criteria before analysis. - Conditions for verification: Specify the HbA1c assay, units, collection schedule, assessment conditions and scope of interference verification. - Why this condition is needed: Differences in measurement conditions must not be mistaken for differences in effects. - Application exceptions: An objective blood test alone does not establish absence of measurement bias or interference. - Remaining uncertainty: Assay, measurement units, quality control, arm-level conditions and interference status. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M08 Table 8.2.a, M03 §3.2.4; internal application - Related sources: S02 ### N15 · Is the reported hierarchy of the primary (1st) renal outcome and secondary (2nd) HbA1c outcome verified? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: Renal function is the trial's primary endpoint and HbA1c is secondary. The original primary-nonsignificant/secondary-positive flag is recorded as true. Its observation value remains blank while its information status remains verified. - Why unverified: The hierarchy reported in the original summary is verified. This does not mean a complete comparison against preregistration was performed. - Reason for missing information: The reported hierarchy is verified; registered prespecification and multiplicity plans remain gaps in N16. - Required materials and conditions: The trial's description of primary and secondary endpoints and their results. - Materials to obtain first: The trial's description of primary and secondary endpoints and their results. - Further research: Report analyses and multiplicity plans with primary and secondary (2nd) endpoints distinguished in advance. - Conditions for verification: Distinguish the paper's primary endpoint from this page's endpoint; preserve the original primary (1st) nonsignificant and secondary (2nd) positive findings. - Why this condition is needed: A favourable secondary (2nd) result must not be recast as a successful primary (1st) result. - Application exceptions: A nonsignificant primary renal outcome does not prove an HbA1c effect of 0 and is not used for new clinical regrading. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M08 Table 8.2.a, M04 §4.6.3; internal application; M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is internally designed - Related sources: S02 ### N16 · Can HbA1c timing, analysis and multiplicity be compared against registration, protocol and SAP? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The registered-primary-endpoint value is blank/unconfirmed, and S02's registration field is blank. The paper describes a primary (1st) renal endpoint, but independent comparison against the registry remains unverified. - Why unverified: No acquired registration or plan is recorded, so HbA1c analysis prespecification and agreement with the results cannot be verified. - Reason for missing information: Registration identity and timing, the HbA1c plan, SAP, changes and multiplicity methods. - Required materials and conditions: S02's original registration, time history, protocol, SAP and complete results. - Materials to obtain first: S02's original registration, time history, protocol, SAP and complete results. - Further research: Prespecify HbA1c analyses, time points and multiplicity handling; compare them with results and reasons for changes. - Conditions for verification: Distinguish the registrations or plans actually compared from material not acquired; verify HbA1c prespecification and analysis changes. - Why this condition is needed: Gaps in the material needed to assess selection of favourable findings must be made explicit. - Application exceptions: Do not turn unverified registration into confirmed nonregistration, selective reporting or absence of research. - Remaining uncertainty: Registration identity and timing, the HbA1c plan, SAP, changes and multiplicity methods. - Referenced methods: M08 Table 8.2.a, M04 §4.6.3; internal application; M04 §4.5, §4.6.3; the limits of this material conversion follow the user's scope and internal design - Related sources: S02 ### N17 · Are the allocated, completing and actually analysed populations and analysis-set definition verified for 24-week HbA1c? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The 35 participants were allocated as 18 to thiamine and 17 to placebo. The HbA1c analysis denominator is blank/unconfirmed, with an empty linked-source list. - Why unverified: Allocation denominators are verified, but the analysis denominator for this outcome was not verified. - Reason for missing information: 24-week HbA1c analysis n, ITT/PP definitions and the number providing outcome data. - Required materials and conditions: Outcome-specific analysis n for 24-week HbA1c, allocation/completion flow and analysis-set definitions. - Materials to obtain first: Outcome-specific analysis n for 24-week HbA1c, allocation/completion flow and analysis-set definitions. - Further research: Report outcome-specific flow and analysis denominators; specify assignment-effect and available-outcome analyses. - Conditions for verification: Distinguish allocation of 18/17 from actual HbA1c analysis n; verify ITT/PP and the outcome contributors. - Why this condition is needed: Allocated participants cannot fill gaps in the outcome denominator or precision. - Application exceptions: Do not automatically use 18/17 as the HbA1c analysis n, completion n or safety n. - Remaining uncertainty: 24-week HbA1c analysis n, ITT/PP definitions and the number providing outcome data. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are internally designed; M08 §8.2.2, M15 §15.3.1; internal application - Related sources: S02 ### N18 · Can the extent and handling of missing HbA1c data, losses and discontinuations be verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The outcome's analysis denominator and between-group interval remain unconfirmed; missing-data and dropout flow and handling are also unverified in the current abstract summary. - Why unverified: Enrollment of 35 participants does not establish complete 24-week outcome data or a complete-case analysis. - Reason for missing information: Outcome-specific losses, reasons, handling models and sensitivity analyses. - Required materials and conditions: HbA1c measurement losses, discontinuation and dropout reasons, missing-data handling and sensitivity analyses. - Materials to obtain first: HbA1c measurement losses, discontinuation and dropout reasons, missing-data handling and sensitivity analyses. - Further research: Report outcome-specific missingness reasons and prespecify suitable missing-data assumptions and sensitivity analyses. - Conditions for verification: Distinguish outcome-specific loss reasons and timing, intervention discontinuation and missing follow-up; verify handling and sensitivity analyses. - Why this condition is needed: The limitations that losses and analysis impose on interpretation must be identified. - Application exceptions: Do not newly calculate missing counts or dropout rates from allocation and completion counts belonging to another trial family. - Remaining uncertainty: Outcome-specific losses, reasons, handling models and sensitivity analyses. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are internally designed; M08 §8.2.2, M15 §15.3.1; internal application - Related sources: S02 ### N19 · Which between-group median-change comparison and analysis does the estimate −1.35 represent? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S02 records a between-group difference in median HbA1c change of −1.35. Adjustment, testing models and ITT/PP details remain unconfirmed in the direct trial. - Why unverified: The reported between-group measure is known, but it cannot be converted into an adjusted mean difference or an individual's probability of improvement. - Reason for missing information: Arm-level median changes, models, adjustment and analysis-population details. - Required materials and conditions: Arm-level HbA1c change distributions, the between-group estimand definition and analysis model. - Materials to obtain first: Arm-level HbA1c change distributions, the between-group estimand definition and analysis model. - Further research: Prespecify the HbA1c effect of interest and method for estimating the median-change contrast; report the model and denominator. - Conditions for verification: Distinguish the reported between-group median-change contrast from within-group changes, final values and adjusted mean differences; verify the analysis model. - Why this condition is needed: Medians must not be conflated with means, or between-group comparisons with within-group changes. - Application exceptions: S20's arithmetic on mean changes does not establish the model for this between-group median-change contrast. - Remaining uncertainty: Arm-level median changes, models, adjustment and analysis-population details. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M08 §8.2.2, M15 §15.3.1; internal application - Related sources: S02 ### N20 · To what extent are the HbA1c contrast of −1.35 and its units verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The reported value −1.35 is retained. The manuscript explicitly states that the value's units were not given in the abstract accessed historically. - Why unverified: The value is verified, but its units are not; units such as %p or mmol/mol were not filled in. - Reason for missing information: Explicit units and the results table for this contrast. - Required materials and conditions: An original results table, full paper or correction explicitly identifying the units of −1.35. - Materials to obtain first: An original results table, full paper or correction explicitly identifying the units of −1.35. - Further research: State HbA1c units and changes in outcome reports; provide the between-group estimate and dispersion together. - Conditions for verification: Both the reported value and units must be verified to interpret magnitude; do not assume unreported units such as percentage points (%p). - Why this condition is needed: Without units, clinical magnitude and direct comparison with other trials cannot be established. - Application exceptions: Do not retroactively assign customary HbA1c units or S20's % units to this value. - Remaining uncertainty: Explicit units and the results table for this contrast. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M15 §15.3.1–15.3.3; internal application - Related sources: S02 ### N21 · To what extent are the P value and model for this HbA1c test verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The HbA1c result is reported as P<0.01. The exact P value and testing details remain unverified, so no effect estimate was derived from P. - Why unverified: The significance threshold is verified; the exact P value, model and precision remain unconfirmed. - Reason for missing information: Exact P value, testing model and dispersion or interval for the effect. - Required materials and conditions: S02's test type, exact P value and analysis results table. - Materials to obtain first: S02's test type, exact P value and analysis results table. - Further research: Report estimates, intervals and testing models together; do not determine clinical importance from significance alone. - Conditions for verification: Distinguish verification of the reported P threshold, exact P value and testing model; do not convert these into effect dispersion, CI or importance. - Why this condition is needed: A record of statistical significance alone must not establish sufficiency or clinical importance. - Application exceptions: Do not reverse-convert a P threshold into SE/SD/CI or a user's probability of success. - Remaining uncertainty: Exact P value, testing model and dispersion or interval for the effect. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M15 §15.3.1–15.3.3; internal application - Related sources: S02 ### N22 · Can the CI, dispersion and denominator of the direct HbA1c contrast be verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The between-group interval and analysis denominator remain unconfirmed. Complete mean/SD/n or adjusted-contrast inputs were unavailable, so no new CI or SMD was created. - Why unverified: The original CX and EX descriptions concern verification limits in the accessed material; they do not mean the unread full paper has no data. - Reason for missing information: CI, dispersion, denominator and model for the same 24-week HbA1c contrast. - Required materials and conditions: S02's between-group HbA1c CI, dispersion, analysis n and model. - Materials to obtain first: S02's between-group HbA1c CI, dispersion, analysis n and model. - Further research: Report estimates and confidence intervals for the same effect of interest; specify missing-data and adjustment assumptions. - Conditions for verification: Link the CI, dispersion and actual analysis n for the same endpoint, time point and analysis; do not generalize material not acquired to nonreporting throughout the full paper. - Why this condition is needed: Significance or allocation n must not fill gaps in intervals and dispersion. - Application exceptions: Do not substitute renal-outcome CIs or another trial's means/SD for the current HbA1c contrast. - Remaining uncertainty: CI, dispersion, denominator and model for the same 24-week HbA1c contrast. - Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptations are internally designed; M15 §15.3.1–15.3.3; internal application - Related sources: S02 ### N23 · Can an applicable clinically important HbA1c difference and its relationship to the interval be verified for this population? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The manuscript limits its conclusion: clinically meaningful glycaemic improvement cannot be established. Units/CI for −1.35 and an important-difference criterion remain unverified. - Why unverified: The value and significance alone cannot establish clinical importance or failure to meet an important difference. - Reason for missing information: An applicable important-difference criterion and units/interval for the same measure. - Required materials and conditions: Evidence supporting an applicable important difference for CKD3/T2D HbA1c, and units/CI for the direct contrast. - Materials to obtain first: Evidence supporting an applicable important difference for CKD3/T2D HbA1c, and units/CI for the direct contrast. - Further research: Assess the interval against a previously justified important difference without inventing minimum sample sizes, durations or efficacy grades. - Conditions for verification: Units, an interval and a justified important difference are needed to interpret their relationship; retain the unverified criterion. - Why this condition is needed: A laboratory signal must be distinguished from improvement meaningful to patients. - Application exceptions: Do not substitute an unverified criterion with no clinical effect or failure to meet an MCID; do not require an event trial for the narrow laboratory-value question. - Remaining uncertainty: An applicable important-difference criterion and units/interval for the same measure. - Referenced methods: M15 §15.3.1–15.3.3; internal application; M03 §3.2.4; internally designed extension boundaries - Related sources: S02 ### N24 · How should S01's lack of glycaemic benefit be read alongside the direct CKD3 result? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The earlier S01 microalbuminuria-trial abstract states that glycaemic control was not improved. It was not pooled as identical to CKD3/24-week S02. - Why unverified: The manuscript's unfavourable context and question differences are retained. Differences in duration and renal status were not declared conflicting solely from significance. - Reason for missing information: The context and its differences are verified; S01's detailed glycaemic analysis n and model are outside the current material. - Required materials and conditions: S01's glycaemic results, population, duration and analysis details, compared with S02. - Materials to obtain first: S01's glycaemic results, population, duration and analysis details, compared with S02. - Further research: Directly compare only studies addressing the same question; analyse other renal populations as stratified context. - Conditions for verification: Retain unfavourable context; distinguish whether findings address the same question using actual differences in population, duration and endpoints. - Why this condition is needed: Earlier no-difference reports must not be deleted while retaining only the positive direct finding. - Application exceptions: Do not extend the abstract's no-effect wording to proof of an effect of 0 for all B1, all T2D or CKD3. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is internally designed; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application; M03 §3.2.1; detailed population conditions are internally designed - Related sources: S01, S02 ### N25 · Were S01 and S20's shared registration number distinguished from independent replication? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S01 and S20 both use study identifier CTRI-2008-091-000112. S20 is recorded as context from the same trial. - Why unverified: The retained shared registration number and original roles distinguish the re-reported trial family. No new reading of the original registry was performed here. - Reason for missing information: The stored family relationship is verified; a cross-tabulation of actual outcome-specific participants in both reports remains unconfirmed. - Required materials and conditions: S01/S20 registration numbers, cohort linkage and outcome-specific participant flow. - Materials to obtain first: S01/S20 registration numbers, cohort linkage and outcome-specific participant flow. - Further research: Maintain a family map linking reports to unique trials and identify overlapping participants and analyses separately. - Conditions for verification: Link report identities, study identifiers and same/different family relationships as recorded; do not sum unverified total participant counts. - Why this condition is needed: Report counts must not be converted into independent replications or unique participant counts. - Application exceptions: Retain different outcomes under the same registration number, distinguishing their time points and analyses. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M04 §4.6.1–4.6.2 on distinguishing studies and reports; cohort/outcome linkage is an internal application; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application - Related sources: S01, S20 ### N26 · Which time points do S20's on-treatment and post-treatment HbA1c values and historical arithmetic represent? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S20 records thiamine 9.2→9.0→7.8% and placebo 8.8→8.5→8.4% at baseline, 3-month treatment end and 2-month off-treatment follow-up. Original C001's +0.1%p and C002's −1.0%p are historical arithmetic at different times. - Why unverified: The values, time points and original context-only/off-treatment arithmetic labels are verified. No new arithmetic was performed here. - Reason for missing information: Preservation of original values and time points is verified; these do not replace S20 analysis n/precision or S02's direct effect. - Required materials and conditions: S20 Table2's time-specific values, analysis denominators and original arithmetic records. - Materials to obtain first: S20 Table2's time-specific values, analysis denominators and original arithmetic records. - Further research: Prespecify on-treatment and post-treatment results as separate effects of interest; provide values and intervals for the same analysis. - Conditions for verification: Separate baseline, 3-month treatment-end and 2-month off-treatment means and historical calculations; do not reuse them as the current 24-week median effect. - Why this condition is needed: Improvement after discontinuation must not be reclassified as an on-treatment effect. - Application exceptions: Do not reuse arithmetic on rounded means in the original source as the current trial's adjusted mean difference, median difference or a new calculation. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs; M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is internally designed; M08 §8.2.2, M15 §15.3.1; internal application; M03 §3.2.1–3.2.4; detailed dose, persistence and effect-modification conditions are internally designed - Related sources: S20 ### N27 · Are independent replication and study counts verified for the same CKD3/24-week question? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: In the stored material, S02 is the decisive direct family, while S01/S20 share a registration in another population. Original R1 records this restricted direct-evidence scope. - Why unverified: The direct family is verified, but independent replication for the same question and worldwide study totals were not verified. - Reason for missing information: Independent replication, funding and exact unique participant counts for the same CKD3/24-week question. - Required materials and conditions: Eligible-study identities, families, funding and results for the same question. - Materials to obtain first: Eligible-study identities, families, funding and results for the same question. - Further research: Screen independent comparisons addressing the same question; compare condition differences and unfavourable findings by trial family. - Conditions for verification: Distinguish direct and adjacent trial families; limit unique-study and participant counts to what the stored material verifies. - Why this condition is needed: Adjacent material or reports under the same registration must not be counted as independent replications. - Application exceptions: Do not discard a result merely because it belongs to a single direct family. Without an exhaustive search, do not claim that only 1 direct study exists worldwide. - Remaining uncertainty: Independent replication, funding and exact unique participant counts for the same CKD3/24-week question. - Referenced methods: M04 §4.6.1–4.6.2 on distinguishing studies and reports; cohort/outcome linkage is an internal application; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application; Chamgap's own independence/attribution conditions. The referenced official sections do not certify every detail of these conditions - Related sources: S02, S01, S20 ### N28 · Are S02's funding, product provision, author employment/patents and declared interests verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The decisive trial's funding source is unconfirmed. Original I1 follows rubric precedent29; it does not establish actual mixed or independent funding. - Why unverified: The full-text funding section and declarations were not acquired, so actual sources of funding and independence cannot be verified. - Reason for missing information: S02 funding, products, employment, patents, analytical involvement and declarations. - Required materials and conditions: The direct paper's funding, COI, product provision and author/patent information. - Materials to obtain first: The direct paper's funding, COI, product provision and author/patent information. - Further research: In independent comparisons disclosing funding and product provision, record each party's role and analytical involvement. - Conditions for verification: Identify declarations, resources, product provision and unknowns by location; do not treat I1 as proof of actual mixed funding. - Why this condition is needed: Unverified funding must not be described as independent research or manufacturer-only research. - Application exceptions: Unconfirmed funding or company product provision alone does not automatically exclude a result or establish the magnitude of bias. - Remaining uncertainty: S02 funding, products, employment, patents, analytical involvement and declarations. - Referenced methods: Chamgap's own independence/attribution conditions. The referenced official sections do not certify every detail of these conditions - Related sources: S02 ### N29 · Can persistence beyond 24 weeks, an optimal dose or effects in particular subgroups be directly verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: S02 addresses 300 mg/day for 24 weeks; subsequent persistence remains unconfirmed. S20's post-treatment values in another population cannot replace this information. - Why unverified: The current summary has no persistence, optimal-dose or stratum-specific evidence under the same direct conditions. - Reason for missing information: Long-term, post-treatment, dose and subgroup information under the direct CKD3 conditions. - Required materials and conditions: Post-treatment CKD3 follow-up, direct dose comparisons and prespecified subgroup/interaction results. - Materials to obtain first: Post-treatment CKD3 follow-up, direct dose comparisons and prespecified subgroup/interaction results. - Further research: Design post-treatment follow-up, dose comparisons and prespecified stratum analyses separately from the current question; report estimates for each. - Conditions for verification: Verify post-treatment/long-term, direct dose-comparison and prespecified stratum/interaction information separately; do not infer these from the main effect. - Why this condition is needed: The chosen main effect must not be extended to maintenance effects or an optimal dose. - Application exceptions: Do not make all extension studies mandatory for verifying the current narrow 24-week question. - Remaining uncertainty: Long-term, post-treatment, dose and subgroup information under the direct CKD3 conditions. - Referenced methods: M03 §3.2.1–3.2.4; detailed dose, persistence and effect-modification conditions are internally designed; M03 §3.2.1; detailed population conditions are internally designed - Related sources: S02, S20 ### N30 · Does HbA1c change also answer questions about preventing complications or death, or improving quality of life? Verification status: Not applicable Reading scope: Stored prior-review summary read / external materials not read - Current findings: The record distinguishing laboratory markers from clinical events has a blank value and verified information status. Its warning against extending laboratory surrogates to clinical-event prevention is retained. - Why unverified: The current claim concerns HbA1c change; fewer clinical events are not the chosen question's endpoint. - Reason for missing information: Event prevention is not applicable to the current laboratory-value question; separate direct evidence supporting event benefits was not verified. - Required materials and conditions: Direct comparative evidence on clinical events and quality of life under the same population and exposure conditions. - Materials to obtain first: Direct comparative evidence on clinical events and quality of life under the same population and exposure conditions. - Further research: Evaluate clinical-event questions in studies with separately prespecified outcome definitions, durations and denominators. - Conditions for verification: Separate the glycaemic laboratory surrogate from clinical-event questions; do not extend the claim without direct event outcomes. - Why this condition is needed: A laboratory signal must not be converted into event prevention or replacement of existing medication. - Application exceptions: Unverified event trials are not a reason to delete the reported HbA1c value or assert no effect. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.2.4; internally designed extension boundaries; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations are question-specific internal designs - Related sources: S02 ### N31 · Can event definitions, collection and denominators underlying the statement that no serious adverse events were reported be verified? Verification status: Partly verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The original abstract states that no serious adverse events were reported; specific event counts, denominators and long-term exposure remain unconfirmed. - Why unverified: The presence of a harm statement is verified; actual counts of 0 people or event rates were not verified. - Reason for missing information: Event definitions, systematic collection, arm-level person/event counts, duration and safety denominators. - Required materials and conditions: S02 adverse-event and serious-event definitions, arm-level counts, exposure denominators, collection and discontinuation tables. - Materials to obtain first: S02 adverse-event and serious-event definitions, arm-level counts, exposure denominators, collection and discontinuation tables. - Further research: For the same exposure and population, report systematic harm collection, exposure time, event/person counts and discontinuation reasons. - Conditions for verification: Distinguish a nonreporting statement from actual event counts, people, collection methods and duration; do not promote it to 0 events or a safety guarantee. - Why this condition is needed: Efficacy summaries and harm information have different verification levels. - Application exceptions: Do not delete the HbA1c value because of safety-data gaps or automatically use allocation of 18/17 as the safety-analysis denominator. - Remaining uncertainty: Event definitions, systematic collection, arm-level person/event counts, duration and safety denominators. - Referenced methods: M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; detailed distinctions are internally designed; M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4, M04 §4.6.3; internal application; Chamgap's own interpretive boundaries. M19 principles are referenced; no official causality tool was applied - Related sources: S02 ### N32 · Are interactions and assay effects of concurrent exposures, including diabetes medications, verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The drug-interaction information value is blank and recorded as not separately searched. No separate comprehensive search covering all drug interactions was performed. - Why unverified: The current summary and historical search scope do not establish absence of interactions or individual safety. - Reason for missing information: Actual concomitant exposures, interactions, assay effects and collection scope. - Required materials and conditions: Arm-level concomitant exposures and relevant interaction or assay-effect information. - Materials to obtain first: Arm-level concomitant exposures and relevant interaction or assay-effect information. - Further research: In safety studies collecting concomitant exposures, distinguish underlying disease and assay effects in the assessment. - Conditions for verification: Specify separate interaction searches, actual concomitant exposure and review scope; do not interpret an unperformed search as no risk. - Why this condition is needed: Effects of concurrent medication and underlying disease must not be attributed to a single exposure. - Application exceptions: Do not turn a search not performed into confirmed interaction or absence of risk. - Remaining uncertainty: Actual concomitant exposures, interactions, assay effects and collection scope. - Referenced methods: M19 §19.1.2–19.2.4; internal application; M19 §19.2.2–19.2.4 on design considerations; detailed confounding records are internally designed; M19 §19.1.2–19.2.4; internal application - Related sources: S02 ### N33 · Can long-term or rare harms beyond 24 weeks and safety in vulnerable groups be directly verified? Verification status: Unverified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The manuscript retains unconfirmed event denominators and long-term exposure; it does not instruct treatment changes or self-medication at the study dose. - Why unverified: A harm statement in a small, short-term trial abstract cannot establish safety across all durations and vulnerable groups. - Reason for missing information: Direct safety information on long-term or rare harms and separate vulnerable groups. - Required materials and conditions: Long-term and rare-event data, vulnerable-group exposure/collection data and suitable comparisons. - Materials to obtain first: Long-term and rare-event data, vulnerable-group exposure/collection data and suitable comparisons. - Further research: Collect rare events and vulnerable-group events through long-term follow-up or suitable observational designs; report denominators and concomitant exposures. - Conditions for verification: Distinguish harm duration and extensions for the direct exposure/population; do not rule out rare risks because the sample is limited. - Why this condition is needed: Limited observations must not be extended to overall safety. - Application exceptions: Do not require every rare-harm study as an additional prerequisite for reporting the HbA1c value. - Remaining uncertainty: Direct safety information on long-term or rare harms and separate vulnerable groups. - Referenced methods: M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; internal application - Related sources: S02 ### N34 · Was reading the stored summary in this work distinguished from each paper's historical access level? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The whole stored manuscript summary was read here. Historical records identify publisher abstracts for S01/S02 and HTML full text for S20; external readings/attempts in this work were 0. - Why unverified: Reading levels distinguish current actions from historical records. External links are not treated as current full-text reading. - Reason for missing information: The current reading scope is verified; separate historical access dates are absent from the manuscript and remain blank for each document. - Required materials and conditions: Stored manuscript summary, bibliography, historical access levels, source locations and individual access-date records. - Materials to obtain first: Stored manuscript summary, bibliography, historical access levels, source locations and individual access-date records. - Further research: Maintain collection records identifying attempts, acquisition and reading levels by material, linked to original locations and versions. - Conditions for verification: Distinguish the stored manuscript actually read here from historical abstract/HTML access by material and date; retain current external full-text unread status. - Why this condition is needed: Links, summaries and historical access records must not be recast as current full-text reading. - Application exceptions: Do not copy S20's historical HTML access to S01/S02 full papers or current successful HTML access. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M04 §4.5–4.6 on collecting/selecting material; access/reading separation and the hyperlink contract are internally designed; M04 §4.5, §4.6.3; the limits of this material conversion follow the user's scope and internal design - Related sources: S01, S02, S20 ### N35 · What are the stored evidence date, current review date and search/correction scope? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: The original evidence/search date is 2026-09-15 and the conversion review date is 2026-10-05. The historical search was not exhaustive across worldwide databases; no new search was performed here. - Why unverified: The current review scope, original uncertainty and update conditions are verified. This is not completion of a current comprehensive literature verification. - Reason for missing information: Search and date boundaries are verified; new eligible literature, registrations and corrections were not checked here. - Required materials and conditions: Historical search records, correction/retraction/registration-check scope, and dates and identities of new material. - Materials to obtain first: Historical search records, correction/retraction/registration-check scope, and dates and identities of new material. - Further research: For separately authorized updates, record eligible searches and registration/correction tracking; preserve previous values and reasons for changes. - Conditions for verification: Distinguish the 2026-09-15 evidence date from the 2026-10-05 conversion review; qualify non-exhaustive searches and update triggers by scope. - Why this condition is needed: Absence of studies outside the search scope must not be claimed. - Application exceptions: Historical search and numeric records and the original reused162 material were not opened here; only their reference strings are retained. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M04 §4.5, §4.6.3; the limits of this material conversion follow the user's scope and internal design; M04 §4.6.3, M15 §15.3.1–15.3.3; status separation is internally designed - Related sources: S01, S02, S20 ### N36 · Were original C46, axes, blank values/empty lists and corrections distinguished from new condition-status judgments? Verification status: Verified Reading scope: Stored prior-review summary read / external materials not read - Current findings: Original grade C/score46 and B-S-R1-I1-EX-B1-CX, the primary-endpoint failure flag, blank values, empty linked-source lists and historical arithmetic are retained unchanged. No new efficacy grade, score or calculation is created. - Why unverified: Verified refers to preservation of the original assessment and the conversion scope, not to a newly clinically confirmed grade. - Reason for missing information: There is no gap in the preservation contract; original clinical and material gaps remain in the corresponding rows. - Required materials and conditions: The current manuscript's grade/score/axes, original observed-evidence/nutrition fields, corrections and existing calculation records. - Materials to obtain first: The current manuscript's grade/score/axes, original observed-evidence/nutrition fields, corrections and existing calculation records. - Further research: When a change is needed, separately preserve the entire original text, version, reason and date; distinguish clinical regrading from display conversion. - Conditions for verification: Link original values, historical assessments and calculations unchanged; do not automatically inherit them as condition statuses, new scores or probabilities. - Why this condition is needed: The original grade and model statuses must not be merged into a single quality or efficacy score. - Application exceptions: A structural PASS and model review do not constitute official method execution, independent clinical certification or completed publication. - Remaining uncertainty: The scope verified in the current record is distinguished from the remaining uncertainty in the original clinical evidence. - Referenced methods: M03 §3.1, §3.2.1–3.2.4 question-clarification principles; detailed separation rules are internally designed; M04 §4.5, §4.6.3; the limits of this material conversion follow the user's scope and internal design - Related sources: S01, S02, S20 ## Evidence and sources ### S01 Rabbani et al. — thiamine and early diabetic nephropathy (2009) [Official source](https://link.springer.com/article/10.1007/s00125-008-1224-4) Materials actually read: Stored manuscript summaries and condition-application records were read. External abstracts, full papers, registrations and analysis plans were not attempted, acquired or read in this work. Findings recorded in the summary: The earlier microalbuminuria-trial abstract describes no effect on glycaemic control. It was not pooled as an equivalent question to the CKD stage 3/24-week trial. Record level: The original review's summary and attribution are retained. This is not new reading or regrading of external clinical originals. Independent replication: S01/S20 are an adjacent family under the same registration, not direct CKD3/24-week replications | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Paper abstract | Material URL identified in the stored bibliography; current public availability not rechecked | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | | Full paper | Material identity and separate direct URL unconfirmed | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | ### S02 Phrawong et al. — thiamine in T2D with CKD stage 3 (online 2025; issue 2026) [Official source](https://www.benthamscience.com/article/148808) Materials actually read: Stored manuscript summaries and condition-application records were read. External abstracts, full papers, registrations and analysis plans were not attempted, acquired or read in this work. Findings recorded in the summary: Between-group difference in median HbA1c change −1.35, P<0.01; units and interval not stated in the abstract Record level: The original review's summary and attribution are retained. This is not new reading or regrading of external clinical originals. Independent replication: Decisive report from a single direct family; independent replication remains unconfirmed | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Paper abstract | Material URL identified in the stored bibliography; current public availability not rechecked | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | | Full paper | Material identity and separate direct URL unconfirmed | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | | Original trial registration | Material identity and separate direct URL unconfirmed | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | | Full protocol/statistical analysis plan | Material identity and separate direct URL unconfirmed | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | ### S20 Alam et al. — high-dose thiamine and T2D risk factors (2012) [Official source](https://www.iomcworld.com/open-access/effect-of-high-dose-thiamine-therapy-on-risk-factors-in-type-2-diabetics-26626.html) Materials actually read: Stored manuscript summaries and condition-application records were read. External abstracts, full papers, registrations and analysis plans were not attempted, acquired or read in this work. Findings recorded in the summary: Table 2 reports mean HbA1c of 9.2→9.0→7.8% with thiamine and 8.8→8.5→8.4% with placebo, at baseline, 3-month treatment end and 2-month off-treatment follow-up. Record level: The original review's summary and attribution are retained. This is not new reading or regrading of external clinical originals. Independent replication: S01/S20 are an adjacent family under the same registration, not direct CKD3/24-week replications | Material | Public availability | Access status | Actually read | Notes | |---|---|---|---|---| | Web full text of the paper (HTML) | Material URL identified in the stored bibliography; current public availability not rechecked | Not attempted | Unread | What was actually read here is the report summary in the stored manuscript. The paper-page URL and historical access level do not establish current full-text reading or acquisition of this material. | ## Review criteria and methods These are Chamgap's own application conditions, informed by the relevant sections of official methodological guidance. They do not constitute official assessment-tool certification or a new efficacy score. - M03 [Cochrane Handbook 6.5 (2024), chapter updated 2023-08](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-03) · §3.1, §3.2.1–3.2.3, MECIR C5/C7/C13 · Clearly record the question, eligibility conditions and reasons for changes; distinguish intervention variants. · Stored method principles, application records and copy hashes were checked. No new retrieval of external methods documents or execution of an entire official assessment tool was performed. - M04 [Cochrane Handbook 6.5.1 (2025), chapter updated 2025-03](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-04) · §4.5, §4.6.1–4.6.3, MECIR C42 · Distinguish and link studies and reports; do not omit studies solely because outcomes are unreported. · Stored method principles, application records and copy hashes were checked. No new retrieval of external methods documents or execution of an entire official assessment tool was performed. - M08 [Cochrane Handbook 6.5 (2024), chapter updated 2019-10](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-08) · §8.1–8.2.3, Table 8.2.a · Fix the outcome and effect of interest, then review randomization, deviations, missing data, measurement and selective reporting. · Stored method principles, application records and copy hashes were checked. No new retrieval of external methods documents or execution of an entire official assessment tool was performed. - M15 [Cochrane Handbook 6.5 (2024), chapter updated 2023-08](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-15) · §15.2.4, §15.3.1–15.3.3 · Distinguish effects, intervals, importance and applicability. Sample size alone does not determine precision. · Stored method principles, application records and copy hashes were checked. No new retrieval of external methods documents or execution of an entire official assessment tool was performed. - M19 [Cochrane Handbook 6.5 (2024), chapter updated 2019-10](https://www.cochrane.org/authors/handbooks-and-manuals/handbook/current/chapter-19) · §19.1.2, §19.2.1–19.2.4 · Harm questions may require designs, populations and follow-up different from efficacy questions. Case reports do not establish incidence. · Stored method principles, application records and copy hashes were checked. No new retrieval of external methods documents or execution of an entire official assessment tool was performed. Sections 19.1.2 and 19.2.1–19.2.4 of the stored harms chapter were additionally read. ## What the statuses mean - Verified: The specified record or boundary is verified in the stored material; this differs from confirmed efficacy or a new grade - Partly verified: Some parts of the condition are known while specific gaps remain - Unverified: The actual data or methods needed for the condition could not be verified within the currently stored material - Unread: External material was not read in this work; this differs from no research or an effect of 0 - Not applicable: An extension condition outside this independent question; not a substitute for unread-material status ## Change history 2026-10-29 · Public page/materials 1.0. Public pages and materials were prepared from the existing 36 items. No new clinical search, condition-status change or scoring was performed.