# Vitamin C × HbA1c in type 2 diabetes

Verdict 1585 · Evidence Check

Stored-material review date: 2026-10-04  
Original evidence cut-off: 2026-07-24  
Review-criteria version: 1.1  
Public page/material preparation date: 2026-10-07

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: People with type 2 diabetes; trial-specific age ranges, diagnostic criteria and baseline characteristics are unverified
- Ingredient, formulation and route: Assessment scope: oral vitamin C tablets or capsules. The chemical form and whether each trial used a single ingredient or a combination remain unverified
- Dose: The overall review covered 200–3,000 mg/day. Doses in the individual HbA1c trials are unverified
- Comparator: Required comparison: vitamin C added to shared background treatment versus placebo or no added vitamin C. This defines Chamgap's question; the actual comparators in the original trials are unverified
- Measured outcome: Difference in HbA1c versus the comparator, in percentage points. The exact group order, final values versus changes from baseline, and adjustment are unverified
- Duration: The overall review covered 2 weeks–1 year, mostly less than 6 months. Trial-specific HbA1c durations and an appropriate minimum duration are unverified or undetermined
- Questions outside this scope: Fasting or post-meal glucose, complications or mortality, replacing medication, optimal doses and efficacy in special populations are separate questions

## Checklist items

### N01 · Which population with type 2 diabetes does the effect apply to?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary identifies the population as people with type 2 diabetes. Trial-specific diagnoses, ages, baseline HbA1c and settings could not be verified from that summary.
- Why unverified: The named target population is verified, but trial-specific conditions needed to establish population alignment are unverified.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Obtain trial-specific diagnoses, ages, baseline HbA1c, disease duration, care settings and inclusion/exclusion criteria
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Prespecified stratified comparative trials including populations with different representativeness; first obtain tables of the included trials' characteristics
- Conditions for verification: Identify each trial's population, diagnoses and baseline characteristics, and explain matches or differences with the selected claim using locations in the original paper
- Why this condition is needed: The broad label 'type 2 diabetes' does not establish that the target populations are equivalent.
- Application exceptions: For mixed populations, separate results or documentation of the population mix and indirectness are acceptable. Other types of diabetes require a separate question.
- Remaining uncertainty: Who the findings apply to, and differences between deficient and non-deficient participants
- Referenced methods: M03 §3.2.1
- Related sources: S01, S02, S03

### N02 · What was actually administered?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary names oral tablets or capsules and states that superiority of a specific formulation has not been established. Each trial's chemical form, combination ingredients and product remain unverified.
- Why unverified: The route specified in the claim is verified, but original materials needed to match each trial's formulation and ingredients are insufficient.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Verify the chemical form of vitamin C, single ingredient versus combination, tablet versus capsule, oral route, product and content
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Obtain product analyses and manufacturing/content records; superiority of a particular formulation requires a direct comparative trial
- Conditions for verification: Verify each trial's actual ingredients, route, formulation and control ingredients, while retaining unverified details as unverified
- Why this condition is needed: Needed to avoid attributing effects from other formulations, routes or added ingredients to oral vitamin C.
- Application exceptions: Combination products may contribute if shared ingredients and between-group differences can be separated. Intravenous preparations and foods require separate comparisons.
- Remaining uncertainty: The contribution of vitamin C alone and equivalence between products
- Referenced methods: M03 §3.2.2–3.2.3
- Related sources: S01, S03

### N03 · Which dose is the efficacy question about?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The overall review's dose range of 200–3,000 mg/day is preserved. Dose, adherence, diet and baseline vitamin C status in each HbA1c trial remain unverified.
- Why unverified: Do not assume that the overall review's range represents identical conditions in the trials contributing to HbA1c.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Distinguish daily prescribed dose, dosing frequency, actual intake, adherence, baseline diet and vitamin C status
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Comparative trials with prespecified exposure measurement and adherence tracking; effect-modification analyses by baseline status
- Conditions for verification: Link dose, actual exposure, verification methods and duration for each trial contributing to the HbA1c analysis
- Why this condition is needed: If nominal dose differs from actual exposure, applicability and dose interpretation change.
- Application exceptions: When baseline deficiency information is absent, retain studies of the overall effect and leave only deficiency-specific applicability unverified.
- Remaining uncertainty: The exposure at which the results were observed
- Referenced methods: M03 §3.2.2; M15 §15.2.4
- Related sources: S01, S03

### N04 · Lower than what comparator?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored review summary does not provide each trial's comparator details or the group order used to calculate effects. The existing explanation describes a between-group mean difference, but original trial and synthesis tables were not checked.
- Why unverified: The actual comparison is not established because original descriptions of the comparison were not obtained.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Specify the actual placebo, no-treatment, usual-care or active comparator and fix the group order used to calculate the effect
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials explicitly describing placebo/no added vitamin C and shared treatment, or original descriptions of existing comparators
- Conditions for verification: Verify each trial's control conditions, group order and rationale for inclusion in the synthesis
- Why this condition is needed: A different comparator means a different question, even when the numerical result is the same.
- Application exceptions: Retain active-comparator results, but present them separately from the main question comparing against no added vitamin C.
- Remaining uncertainty: Mixing placebo, usual-care and active comparisons
- Referenced methods: M03 §3.2.3
- Related sources: S01, S02, S03

### N05 · Can the contribution of added vitamin C be separated?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: Each group's concomitant diabetes treatment, medication adjustments, diet/exercise and other supplements could not be verified from the stored review summary.
- Why unverified: General advice against replacing medication is not an experimental record establishing identical treatment in both groups.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Verify shared diabetes treatment, medication adjustments, diet, exercise, other supplements and changes in both groups
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Add-on treatment trials with prespecified records of shared treatment, adjustment rules and rescue treatment
- Conditions for verification: Link baseline and follow-up concomitant treatments/changes, between-group differences and how they were handled
- Why this condition is needed: Check whether differences in other treatments contribute to the observed difference.
- Application exceptions: Do not require completely identical treatment. If treatment changes, review balance, analysis and alignment with the effect of interest.
- Remaining uncertainty: The contribution of treatment differences other than vitamin C
- Referenced methods: M03 §3.2.3; M08 §8.2.2–8.2.3
- Related sources: S01, S03

### N06 · Can intervention effects be distinguished from chance and confounding?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The reports are labelled systematic reviews/meta-analyses of RCTs. Randomization, concealment and baseline balance in the original trials are unverified.
- Why unverified: Distinguish the existence of a study-design label from a result-specific assessment of bias.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Randomization methods, allocation concealment, baseline balance and result-specific risk-of-bias information from comparative trials
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Preregistered comparative trials with appropriate randomization, concealment and result analyses; first obtain the original methods
- Conditions for verification: Obtain original evidence of the actual design, randomization/concealment and relevant baseline differences, and discuss their effect on results
- Why this condition is needed: The label 'randomized' alone does not resolve risk-of-bias concerns.
- Application exceptions: Also record and use non-randomized studies as supporting evidence. Judgments must fit parallel, crossover or cluster designs. A formal RoB 2 assessment is outside this review's scope.
- Remaining uncertainty: Differences arising during allocation
- Referenced methods: M08 §8.1–8.2.3
- Related sources: S01, S03

### N07 · Is this the effect of assignment or the effect of adhering to treatment?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary records a mean reduction and a WMD. Mixing final values with changes, adjustment, ITT/PP, and parallel/crossover/cluster designs remain unverified.
- Why unverified: Do not reinterpret −0.54 as a within-group change.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Distinguish ITT/PP, analysis populations, adjustment variables, final values versus changes, and handling of crossover/cluster designs
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Comparative trials or reanalysis materials providing a prespecified analysis plan and necessary correlation, period and cluster information
- Conditions for verification: Verify agreement or differences between the effect of interest, the actual analysis and the effect used in the synthesis
- Why this condition is needed: Needed to avoid combining different effects and analyses as if they were the same result.
- Application exceptions: Crossover/cluster requirements are marked not applicable or handled separately only after the design is verified.
- Remaining uncertainty: Alignment between the effect of interest and the actual analysis
- Referenced methods: M08 §8.2.2; M15 §15.3.1
- Related sources: S01, S03

### N08 · Who was lost to follow-up?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: 1,133/1,247 are recorded as outcome-specific synthesis samples for HbA1c. Group-specific allocation/completion/analysis numbers and dropout reasons are not recorded at this location.
- Why unverified: Not reported in the stored summary. This does not establish that the full paper failed to report it.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Group-specific numbers allocated, completing follow-up, measured and analysed; reasons for dropout, handling of missing data and sensitivity analyses
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials tracking dropout causes and follow-up with sensitivity analyses under different assumptions; obtain existing flow diagrams and analysis materials
- Conditions for verification: Link outcome- and time-specific denominators, missingness, handling methods and results under different assumptions
- Why this condition is needed: Review the possibility that missing data changed the observed result.
- Application exceptions: Do not use an arbitrary universal dropout-rate threshold. Examine causes of missingness and their relationship to the outcome.
- Remaining uncertainty: How much missing data and dropout could alter the result
- Referenced methods: M08 Table 8.2.a
- Related sources: S01, S03

### N09 · Are the HbA1c measurements comparable?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: HbA1c units in percentage points and cautions about glucose-meter/some assay interference are recorded. Trial-specific HbA1c assays, blinding and interference remain unverified.
- Why unverified: A glucose-meter caution is not evidence that the HbA1c results are erroneous.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Verify HbA1c assay methods, units, blood-sampling times, assay interference/red-cell limitations and assessor blinding
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials reporting assay methods, interference validation and blood-sampling rules; analytical method comparisons where needed
- Conditions for verification: Verify each trial's HbA1c methods, units, visits and measurement limitations. Distinguish percentage points from relative percentages
- Why this condition is needed: Needed to distinguish measurement differences from actual changes.
- Application exceptions: Do not generalize glucose-meter interference to all HbA1c assays. Review each assay method separately.
- Remaining uncertainty: Distinguishing differences in measurement methods from actual changes
- Referenced methods: M08 Table 8.2.a; M03 §3.2.4
- Related sources: S01, S03

### N10 · Does the observation period fit the question?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary describes an overall review range of 2 weeks–1 year, mostly less than 6 months. Durations of the individual trials contributing to HbA1c are unverified.
- Why unverified: Do not transfer the overall review's duration range to each individual HbA1c trial.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Record treatment, measurement and follow-up periods separately, with evidence supporting the time frame used to interpret HbA1c changes
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials scheduling visits based on the outcome's time characteristics; official original assay references supporting duration criteria require follow-up checking
- Conditions for verification: Specify the timeline of each contributing trial, the questions that timeline can answer and its limitations
- Why this condition is needed: Short-term observation and sustained effects are different questions.
- Application exceptions: Do not adopt an arbitrary minimum number of weeks. An appropriate duration is undetermined from these materials alone.
- Remaining uncertainty: Outcome timing and criteria for an appropriate follow-up period
- Referenced methods: M03 §3.2.2–3.2.4
- Related sources: S01, S03

### N11 · Exactly what difference was observed?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: S01 −0.54 percentage points (95% CI −0.90 to −0.17) and S03 WMD −0.51 percentage points (95% CI −0.81 to −0.20) are preserved. S02 restates −0.54.
- Why unverified: The stored locations of values and units are verified, but group order, analysis/time points and attribution to original paper tables are unverified.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Link effect estimates, measures, units, group order, confidence intervals, time points, denominators and adjustment to original materials
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Obtain original trial result tables/supplements and recheck the original numbers; define the unit of analysis before a new synthesis
- Conditions for verification: Make analysis attribution traceable so that distinct results are not mistaken for the same effect
- Why this condition is needed: Within-group changes, between-group differences and pooled estimates are different results.
- Application exceptions: Retain synthesis summaries as results, while distinguishing them from direct results of original trials.
- Remaining uncertainty: Definition of between-group results, synthesis models and included values
- Referenced methods: M15 §15.3.1–15.3.3
- Related sources: S01, S02, S03

### N12 · What conclusions does the interval allow?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: Outcome-specific HbA1c CIs and samples are available. The stored summary mentions some clinical-relevance thresholds, but contains no verified numerical threshold or source supporting its application.
- Why unverified: Precision sufficiency is not judged because no new importance threshold has been invented.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Link outcome-specific samples, variance and effect intervals to a justified important difference
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials with sample sizes based on important-difference, variance and dropout assumptions; first agree on the threshold and obtain existing variance information
- Conditions for verification: Compare the effects the CI excludes or allows—important effects, small effects or no difference—with an evidence-based threshold
- Why this condition is needed: Study or participant counts alone cannot establish sufficiency, and statistical significance does not establish efficacy.
- Application exceptions: If a clinical-importance threshold is absent, leave it undetermined. Do not impose an arbitrary minimum sample or trial count.
- Remaining uncertainty: Whether an important difference is excluded or allowed, and assumptions used to design the sample
- Referenced methods: M15 §15.3.1–15.3.3
- Related sources: S01, S03

### N13 · Can all prespecified results be examined?

Verification status: Review incomplete  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: Bibliographic records in the stored materials mainly concern reviews. Original trial registrations, protocols, SAPs and comparisons of prespecified versus reported HbA1c analyses were not verified from these materials.
- Why unverified: Registrations/plans were neither obtained nor searched in this review. Lack of registration or selective reporting is not established.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Compare registration, protocol and analysis plans with reported HbA1c time points, analyses and reasons for changes
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Trials publishing preregistration, protocols/SAPs and all results, or retrieval of existing registrations/supplements
- Conditions for verification: Compare prespecified and reported outcomes, analyses and time points, explaining changes or why comparison is impossible
- Why this condition is needed: Examine whether an analysis was selected because its results were favourable.
- Application exceptions: Retain the study's existence and results even if registration is absent or private. Do not turn an unverified status into confirmed selective reporting.
- Remaining uncertainty: Whether outcomes, analyses and time points were prespecified
- Referenced methods: M08 Table 8.2.a; M04 §4.6.3
- Related sources: S01, S03

### N14 · Do three review papers represent independent replication?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: Bibliographic identities exist for three review reports. The acquired summaries contain no original-trial/related-report family table or review-overlap table.
- Why unverified: Do not add original trial counts as 16+18 or participants as 1,133+1,247.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Link original trials, registrations, related reports and review inclusion tables into study families, recording overlap
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Obtain included-trial/registration lists and compare duplicates; then conduct independent replication in a different recruited population
- Conditions for verification: Trace report → original trial → inclusion in each review, and identify matching or unresolved identities
- Why this condition is needed: The same participants or trial must not be counted more than once.
- Application exceptions: Retain unresolved study families under provisional identities. Unique trial and participant counts remain undetermined.
- Remaining uncertainty: Unique trials/participants and independent replication
- Referenced methods: M04 §4.6.1–4.6.2
- Related sources: S01, S02, S03

### N15 · Are results consistent under different conditions?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: S03 I²=90.4% is recorded in the stored summary. Trial-specific opposing/no-difference/positive HbA1c values are unverified. Non-significant results and limitations for other outcomes are also preserved in the original record.
- Why unverified: Similar directions of two pooled averages do not verify independent replication or consistent effects in individual trials.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Record trial-specific opposing, no-difference and positive results, condition differences, heterogeneity and prespecified subgroups
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Independent multicentre replication under standardized conditions; test prespecified hypotheses about baseline status and dose
- Conditions for verification: Preserve the distribution of unique trial results, explanations of condition differences/inconsistency and uncertainty
- Why this condition is needed: One pooled average does not guarantee results in every population and setting.
- Application exceptions: Between-study comparison may be not applicable for a single study family, but NA is not assigned here because study families remain unresolved.
- Remaining uncertainty: Causes of inconsistency, locations of opposing results and reproducibility
- Referenced methods: M03 §3.2.1; M15 §15.3.1
- Related sources: S01, S02, S03

### N16 · Is a higher dose better?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary states that a standard diabetes dose and superiority of particular formulations are undetermined. Direct dose-comparison results could not be verified from the stored summary.
- Why unverified: Further searching is needed to establish whether dose studies exist beyond these stored materials.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Direct dose comparisons, exposure/dose response, and joint observation of benefits and harms
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Randomized dose comparisons using the same formulation and shared treatment. Determine sample sizes and doses after reviewing the evidence
- Conditions for verification: Dose differences within the same comparison framework, or dose-response evidence explicitly describing confounding limitations
- Why this condition is needed: Dose differences across trials may be confounded by differences in populations and durations.
- Application exceptions: The main HbA1c-change question can be retained without claiming dose superiority.
- Remaining uncertainty: Dose selection considering benefits and harms together
- Referenced methods: M03 §3.2.2; Chamgap's own dose-study design proposal
- Related sources: S01, S03

### N17 · Does the effect persist long term or after stopping treatment?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The limitation that most studies are short term is preserved. Post-treatment persistence, long-term medication changes and fixed extension follow-up results are unverified in the acquired summaries.
- Why unverified: A duration range including 1 year does not by itself verify long-term persistence or persistence after stopping treatment.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Verify long-term exposure, post-treatment visits, medication changes and HbA1c trajectories
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Comparative trials/follow-up studies with prespecified long-term extensions and post-treatment follow-up. Duration is undetermined
- Conditions for verification: Long-term comparative results with timelines, treatment continuation/discontinuation and loss to follow-up
- Why this condition is needed: A change at the end of treatment differs from a sustained effect.
- Application exceptions: Do not remove short-term studies as ineligible. Leave persistence separately unverified.
- Remaining uncertainty: Long-term maintenance and the effects of treatment changes
- Referenced methods: M03 §3.2.2; Chamgap's own persistence-study design proposal
- Related sources: S01, S03

### N18 · What harms were observed in the same population at the same dose?

Verification status: Partly verified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The existing summary records limited safety data and gastrointestinal/oxalate cautions. Group-specific event counts, denominators and durations in this population at these doses are unverified.
- Why unverified: Separate general cautions from events observed in trials. Do not record non-reporting as zero events.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Verify group-specific harm definitions, collection methods, event/person counts, denominators, exposure periods and discontinuations
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Obtain comparative trials/safety reports documenting both prespecified harm collection and unanticipated harms
- Conditions for verification: Distinguish events in both groups, observation periods, denominators, collection methods and non-reporting
- Why this condition is needed: Non-reporting cannot be interpreted as zero events or established safety.
- Application exceptions: Retain efficacy results even when harms are not reported in an efficacy study. Present safety separately.
- Remaining uncertainty: Actual harm frequency and collection methods
- Referenced methods: M19 §19.1.2–19.2.4
- Related sources: S02, S04

### N19 · What about harms outside short trials and vulnerable populations?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary cautions about kidney disease/stone histories and iron overload. Long-term exposure and absolute risks in specific populations are unverified in these materials.
- Why unverified: The existence of a caution group is not a numerical risk estimate or evidence of no efficacy.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Exposure, comparisons, occurrence and long-term harm data in existing caution groups, including people with kidney disease/stone histories or iron overload
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Long-term cohorts/registries with denominators, exposure and appropriate comparators; special-population data require separate searching
- Conditions for verification: Verify design-specific event definitions, exposure, comparisons/confounding, denominators, durations and population-applicability limitations
- Why this condition is needed: The restricted populations and durations of efficacy trials may not capture rare harms.
- Application exceptions: Also preserve observational studies, registries and case signals. Without control of confounding and denominators, do not establish incidence rates or causality.
- Remaining uncertainty: Long-term and rare harms, and risks in vulnerable populations
- Referenced methods: M19 §19.1.2, §19.2.2–19.2.4
- Related sources: S04

### N20 · Is the effect greater in deficient participants or particular populations?

Verification status: Unverified  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The stored summary describes possible differences by low baseline vitamin C, dose and duration. Prespecified subgroup and interaction evidence is unverified.
- Why unverified: Do not upgrade a hypothesis to verified efficacy in a particular population.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Baseline vitamin C, HbA1c, medication/disease characteristics, prespecified effect-modification hypotheses and interactions
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Prespecified stratified trials or agreed individual-participant-data analyses. Label subgroups not defined in advance as exploratory
- Conditions for verification: Verify prespecified hypotheses and definitions, subgroup samples/estimates, interactions and precision
- Why this condition is needed: Needed to avoid extending an overall average to a particular population.
- Application exceptions: Pregnancy, children and other types of diabetes are not automatic extensions of this claim.
- Remaining uncertainty: Effect differences by deficiency, disease severity and special population
- Referenced methods: M03 §3.2.1; Chamgap's own effect-modification proposal
- Related sources: S01, S03

### N21 · What might the current record have missed?

Verification status: Review incomplete  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The existing evidence cut-off field is 2026-07-24. On 2026-10-04, only stored materials and methodology were checked; no new clinical search or correction/retraction checks were performed.
- Why unverified: This case did not review the entire current clinical literature or its latest status. The existence and accessibility of materials beyond papers are also unverified.
- Reason for missing information: Not recorded in the stored summary / review of original papers incomplete. This does not establish that the original papers failed to report the information or were inaccessible
- Required materials and conditions: Search scope/dates, inclusion/exclusion, registrations/unpublished studies, corrections/retractions, and material-specific access/reading histories
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: In an authorized follow-up, search and select new literature/registrations/publication-status information and record access. This was not performed here
- Conditions for verification: Record the scope in which evidence is absent or unverified, by date and material type
- Why this condition is needed: The existence of a stored summary does not mean the entire current literature has been checked.
- Application exceptions: No new clinical search was performed. Only the currently preserved scope applies.
- Remaining uncertainty: Missing/unpublished/new studies, publication status and access scope
- Referenced methods: M04 §4.5, §4.6.3
- Related sources: S01, S02, S03, S04

### N22 · Does a change in HbA1c directly establish benefits for complications?

Verification status: Not applicable (as a mandatory condition for the selected outcome)  
Reading scope: Stored existing review summaries were read on 2026-10-04. Full papers, supplements, registrations, plans and original data were not read.

- Current findings: The selected question concerns numerical HbA1c change. Existing separate subclaims describe benefits for complications/mortality as unestablished. Those findings were not reassessed here.
- Why unverified: Clinical-event trials are not required for the HbA1c question itself. Extending to benefits for complications requires separate conditions.
- Reason for missing information: Not applicable with an explicit application exception
- Required materials and conditions: Questions about complications, quality of life or mortality require separate comparative results, events and durations for those outcomes
- Materials to obtain first: First obtain and compare the necessary original materials, including existing papers, supplements and tables of included trials.
- Further research: Long-term comparative trials based on clinical events, with a separately defined question. This is not a new claim or search in this case
- Conditions for verification: Identify the separate clinical question/outcome and direct evidence. Do not automatically extend HbA1c findings
- Why this condition is needed: A boundary distinguishing questions about biomarkers from questions about clinical events.
- Application exceptions: Do not require clinical-event trials as a mandatory condition for the selected question about numerical HbA1c change itself.
- Remaining uncertainty: Questions about complications, mortality and quality of life are outside this scope and require further searching
- Referenced methods: M03 §3.2.4; Chamgap's own boundary for extended claims
- Related sources: S01, S03

## Evidence and sources

### S01

Mason SA, Keske MA, Wadley GD. Effects of Vitamin C Supplementation on Glycemic Control and Cardiovascular Risk Factors in People With Type 2 Diabetes: A GRADE-Assessed Systematic Review and Meta-analysis of Randomized Controlled Trials. Diabetes Care. 2021;44(2):618-630. PMID: 33472962. DOI: 10.2337/dc20-1893.

[Publisher](https://doi.org/10.2337/dc20-1893) · [PubMed](https://pubmed.ncbi.nlm.nih.gov/33472962/)

Materials actually read: Relevant sections and citation records in the existing review summary were read  
Findings recorded in the summary: Stored summary: HbA1c analysis of 16 RCTs and 1,133 participants, −0.54 percentage points (95% CI −0.90 to −0.17).  
Record level: Values reported in a stored summary, not directly verified against the original paper  
Independent replication: Unverified; counts of review papers and original trials do not represent the number of independent replications

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Abstract | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Full text | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Supplement | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Protocol | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Registration | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Original data | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |

### S02

Fong C, Alesi S, Mousa A, et al. Efficacy and Safety of Nutrient Supplements for Glycaemic Control and Insulin Resistance in Type 2 Diabetes: An Umbrella Review and Hierarchical Evidence Synthesis. Nutrients. 2022;14(11):2295. PMID: 35684094. PMCID: PMC9182772. DOI: 10.3390/nu14112295.

[Publisher](https://doi.org/10.3390/nu14112295) · [PMC](https://pmc.ncbi.nlm.nih.gov/articles/PMC9182772/) · [PubMed](https://pubmed.ncbi.nlm.nih.gov/35684094/)

Materials actually read: Relevant sections and citation records in the existing review summary were read  
Findings recorded in the summary: Stored summary: restates the HbA1c mean difference of −0.54 percentage points. Not counted as a new independent trial result.  
Record level: Values reported in a stored summary, not directly verified against the original paper  
Independent replication: Unverified; counts of review papers and original trials do not represent the number of independent replications

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Abstract | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Full text | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | PMCID PMC9182772 appears in S02's stored bibliographic record. Public full-text availability or successful access was not verified in this review. |
| Supplement | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Protocol | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Registration | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Original data | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |

### S03

Nosratabadi S, Ashtary-Larky D, Hosseini F, et al. The effects of vitamin C supplementation on glycemic control in patients with type 2 diabetes: A systematic review and meta-analysis. Diabetes Metab Syndr. 2023;17(8):102824. PMID: 37523928. DOI: 10.1016/j.dsx.2023.102824.

[Publisher](https://doi.org/10.1016/j.dsx.2023.102824) · [PubMed](https://pubmed.ncbi.nlm.nih.gov/37523928/)

Materials actually read: Relevant sections and citation records in the existing review summary were read  
Findings recorded in the summary: Stored summary: HbA1c analysis of 18 RCTs and 1,247 participants, WMD −0.51 percentage points (95% CI −0.81 to −0.20), I² 90.4%.  
Record level: Values reported in a stored summary, not directly verified against the original paper  
Independent replication: Unverified; counts of review papers and original trials do not represent the number of independent replications

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Abstract | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Full text | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Supplement | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Protocol | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Registration | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |
| Original data | Unverified (no web access/search attempted in this review) | Access not attempted | Not read | The existing review summary was read, rather than the paper itself |

### S04

National Institutes of Health, Office of Dietary Supplements. Vitamin C: Fact Sheet for Health Professionals. Updated July 31, 2025. PMID: none. DOI: none.

[Official source](https://ods.od.nih.gov/factsheets/VitaminC-HealthProfessional/)

Materials actually read: Only citations and safety guidance in the existing review summary were read  
Findings recorded in the summary: The existing stored summary records cautions about gastrointestinal effects, oxalate/stones, iron overload and meter interference. These are not event-count/incidence data for this population.  
Record level: Preserved general caution information; not a new safety assessment  
Independent replication: Not applicable: not an efficacy replication trial

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Full web fact sheet | Unverified (original source not accessed in this review) | Access not attempted | Preserved citations/summary only |  |

## 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, and distinguish variations of the intervention. · Relevant sections of the original web text reviewed directly; not an application of the chapter's entire methodology
- 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 a study solely because an outcome is unreported. · Relevant sections of the original web text reviewed directly; not an application of the chapter's entire methodology
- 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 · Define the result and effect of interest before examining randomization, deviations, missing data, measurement and selective reporting. · Relevant sections of the original web text reviewed directly; not an application of the chapter's entire methodology
- 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 effect estimates, intervals, importance and applicability. Sample size alone does not determine precision. · Relevant sections of the original web text reviewed directly; not an application of the chapter's entire methodology
- 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 different designs, populations and follow-up than efficacy questions. Case reports do not establish incidence rates. · Relevant sections of the original web text reviewed directly; not an application of the chapter's entire methodology

## What the statuses mean

- Partly verified: Some required conditions were verified in the stored summary. This does not mean all conditions were met
- Unverified: Required information/conditions could not be established within this review's scope
- Review incomplete: Original-material comparisons/searches within the permitted scope were not performed
- Not reported: A scope-limited status meaning that the item was not stated in the particular material/section that was read
- Inaccessible: Used only when access was actually attempted and failed. Not assigned to the clinical materials in this review
- Insufficient evidence for the conditions: Used only after relevant studies within a defined scope have been reviewed and evidence answering the conditions is insufficient. Not applied across the board in this review
- Not applicable: A specific exception based on applicability conditions. Not a replacement status for unread materials
- Conflicting data: Used when compared data for the same question, time point and analysis are difficult to reconcile. Not assigned merely because review estimates differ
- Met: All required conditions can be explained using locations in original sources. Not synonymous with established efficacy

## Change history

2026-10-07 · Public page/materials 1.0. Public-facing pages and materials were prepared from the existing 22 items. No new clinical search, verification-status changes or scoring were performed.