# Folic acid × Reduction of BDI-II depression severity as an antidepressant adjunct

Verdict 3109 · Evidence Check

Stored-material review date: 2026-10-05  
Original evidence date: 2026-09-15  
Review-criteria version: NGM-v3109-FOLATED-BDI-II-AUC-1.0  
Public page/material preparation date: 2026-11-05

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 clinician-diagnosed moderate-to-severe unipolar MDD; screened for folate/B12 deficiency; pregnancy/planned pregnancy excluded. Page applicability is restricted to nonlactating adults, but trial lactation status is unverified.
- Ingredient, formulation and route: 5 mg folic acid tablets overencapsulated in opaque hard gelatin capsules with lactose BP filler; salt, original tablet manufacturer, lot and purity unverified
- Dose: Folic acid 5 mg, 1 dose daily; common optimized antidepressant treatment
- Comparator: Matching lactose placebo under the same antidepressant optimization and usual-care policy; individual antidepressant drugs and doses were not identical across all participants
- Measured outcome: Time-weighted AUC average of the BDI-II total score over 0–25 weeks: between-group difference adjusted for baseline and stratification variables; lower scores indicate improvement
- Duration: 12 weeks of trial treatment; observation through 25 weeks from randomization
- Questions outside this scope: L-methylfolate/folinic acid/B-vitamin combinations, or improvement in blood folate alone; Extrapolating the initial single time point at 25 weeks or trials with other doses, durations, scales, subgroups or pediatric populations to the primary AUC outcome; Effect of 0/equivalence/ineffectiveness of all folic acid in all MDD; Pregnancy/lactation, long-term/rare harms, or individual dosing recommendations; Substituting biomarkers/remission/mortality for BDI-II AUC

## Checklist items

### N01 · Which folic acid adjunct regimen and BDI-II outcome are being asked about?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The selected question is whether oral folic acid 5 mg/day added for 12 weeks to common optimized antidepressant treatment in adults with unipolar MDD results in a lower 0–25-week time-weighted mean BDI-II than placebo. The subclaims list is empty.
- Why unverified: Confirms preservation of the boundaries; it is neither efficacy certification nor a verdict that all folic acid is ineffective.
- Reason for missing information: Confirms preservation of the boundaries; it is neither efficacy certification nor a verdict that all folic acid is ineffective.
- Required materials and conditions: Current original claim, actual population, 5 mg/day for 12 weeks, common antidepressant policy, placebo, and the 0–25-week time-weighted mean BDI-II
- Materials to obtain first: Current original claim, actual population, 5 mg/day for 12 weeks, common antidepressant policy, placebo, and the 0–25-week time-weighted mean BDI-II
- Further research: Review direct results within these boundaries separately from materials addressing other questions.
- Conditions for verification: Link the original claim to the boundaries of the actual regimen, comparison and time-weighted endpoint.
- Why this condition is needed: Needed to avoid extending the conclusion to all folic acid and all depression.
- Application exceptions: L-methylfolate, combination products, biomarkers and the single 25-week endpoint are not framed as the same question.
- Remaining uncertainty: Confirms preservation of the boundaries; it is neither efficacy certification nor a verdict that all folic acid is ineffective.
- Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are internally designed; M03 §3.2.1; detailed population criteria are internally designed; M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed
- Related sources: S02, S06, S07

### N02 · Are the actual folic acid product and unverified manufacturing details distinguished?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Folic acid tablets were overencapsulated in opaque hard gelatin capsules with lactose BP filler. Bilcare supply was reported, but the exact salt, manufacturing source, original tablet manufacturer, batch and purity remain unverified.
- Why unverified: Product form and supply are preserved, but manufacturing details and current comparison with the original source remain unverified.
- Reason for missing information: Product form and supply are preserved, but manufacturing details and current comparison with the original source remain unverified.
- Required materials and conditions: Active ingredient, 5 mg tablet, overencapsulation, filler, Bilcare supply, and documentation of salt/manufacturer/lot/purity
- Materials to obtain first: Active ingredient, 5 mg tablet, overencapsulation, filler, Bilcare supply, and documentation of salt/manufacturer/lot/purity
- Further research: Compare trial product, batch and ingredient records with the original medication section.
- Conditions for verification: Record the reported active ingredient and overencapsulation separately from unverified manufacturing details.
- Why this condition is needed: Needed to avoid equating the supplier with the manufacturing source or with other folate products.
- Application exceptions: The authorized Flamingo product does not establish the identity of the FolATED product.
- Remaining uncertainty: Product form and supply are preserved, but manufacturing details and current comparison with the original source remain unverified.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed
- Related sources: S02, S11

### N03 · Is 5 mg as 1 daily dose the same as actual total exposure?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The plan for oral folic acid 5 mg as 1 daily dose is verified. The current stored materials were not taken to verify every participant's actual total exposure and adherence, and no new cumulative dose was calculated.
- Why unverified: A nominal dose is available, but actual individual exposure has not been fully verified.
- Reason for missing information: A nominal dose is available, but actual individual exposure has not been fully verified.
- Required materials and conditions: Oral route, nominal amount for 1 dose and dosing frequency, dosing records and total exposure
- Materials to obtain first: Oral route, nominal amount for 1 dose and dosing frequency, dosing records and total exposure
- Further research: Link arm-specific dosing, missed doses and adherence records to the effect of interest.
- Conditions for verification: Distinguish the planned oral 5 mg as 1 daily dose from unverified actual exposure.
- Why this condition is needed: Needed to avoid turning the trial dose into an individual recommendation or a claim of actual complete adherence.
- Application exceptions: Do not calculate new cumulative doses, conversions or recommended doses.
- Remaining uncertainty: A nominal dose is available, but actual individual exposure has not been fully verified.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; 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

### N04 · Are 12 weeks of treatment distinguished from 25 weeks of observation?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Trial treatment lasted 12 weeks; observation extended through 25 weeks from randomization. The 0–25-week mean was not recast as 25 weeks of continuous treatment or a confirmed persistent effect after discontinuation.
- Why unverified: Confirms the time boundaries; it does not verify long-term efficacy or safety.
- Reason for missing information: Confirms the time boundaries; it does not verify long-term efficacy or safety.
- Required materials and conditions: End of trial treatment, observation measured from randomization, and visit times
- Materials to obtain first: End of trial treatment, observation measured from randomization, and visit times
- Further research: If persistence after treatment discontinuation is the question, review direct estimates and exposure at that time.
- Conditions for verification: Show 12 weeks of treatment separately from observation through 25 weeks.
- Why this condition is needed: Needed to avoid turning 25 weeks of observation into 25 weeks of treatment or a lasting long-term effect.
- Application exceptions: Do not impose all long-term follow-up as a requirement for this narrow question.
- Remaining uncertainty: Confirms the time boundaries; it does not verify long-term efficacy or safety.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed
- Related sources: S02, S07

### N05 · What were the matching placebo and common antidepressant policy?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The comparison used a visually matched lactose placebo, and both arms followed a common optimized-antidepressant and usual-care policy. This does not mean all participants received an identical drug and fixed dose.
- Why unverified: The actual placebo and policy are preserved, but individual concomitant-treatment histories have not been fully verified.
- Reason for missing information: The actual placebo and policy are preserved, but individual concomitant-treatment histories have not been fully verified.
- Required materials and conditions: Lactose placebo, visual matching for blinding, optimization/usual-care policy, and individual drug names/doses
- Materials to obtain first: Lactose placebo, visual matching for blinding, optimization/usual-care policy, and individual drug names/doses
- Further research: Link arm-specific antidepressants and medication changes to the selected BDI-II analysis.
- Conditions for verification: Distinguish the actual added-ingredient contrast from the common policy and unverified individual exposure.
- Why this condition is needed: Needed to avoid describing a shared policy as identical drugs and doses for all participants.
- Application exceptions: Do not automatically exclude comparative results because concomitant treatment was used.
- Remaining uncertainty: The actual placebo and policy are preserved, but individual concomitant-treatment histories have not been fully verified.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; 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

### N06 · Were medication changes, psychotherapy and dietary/supplement exposure verified?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The context allowing medication changes and psychotherapy is preserved. Complete individual antidepressant-change and rescue-treatment histories, and dietary, fortified-food and concurrent-supplement exposure remain unverified.
- Why unverified: The policy permitting changes is verified, but a complete comparison of actual concurrent-exposure balance and its analytical impact remains outstanding.
- Reason for missing information: The policy permitting changes is verified, but a complete comparison of actual concurrent-exposure balance and its analytical impact remains outstanding.
- Required materials and conditions: Arm-specific antidepressant changes, rescue treatment, psychotherapy, dietary/fortified-food/supplement exposure and their handling
- Materials to obtain first: Arm-specific antidepressant changes, rescue treatment, psychotherapy, dietary/fortified-food/supplement exposure and their handling
- Further research: Compare arm-specific changes and concurrent-exposure records with how the effect of assignment was handled.
- Conditions for verification: Separate the permitted clinical context from unverified actual treatment and exposure histories.
- Why this condition is needed: Needed to avoid describing permitted changes as absent or attributing the result solely to folic acid.
- Application exceptions: Do not impose a new condition that all changes must be identical.
- Remaining uncertainty: The policy permitting changes is verified, but a complete comparison of actual concurrent-exposure balance and its analytical impact remains outstanding.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; M03 §3.2.3, M08 §8.2.2–8.2.3; internal application; M08 §8.2.2, M15 §15.3.1; internal application; M19 §19.2.2–19.2.4 considerations for study design; detailed recording of confounding is internally designed
- Related sources: S02

### N07 · Which adult MDD population and screening criteria were used?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Participants were adults with clinician-diagnosed moderate-to-severe unipolar MDD under ICD10. BDI-II thresholds were ≥19 at screening and ≥17 at randomization, with a mixture of new and existing antidepressant users. Bipolar and psychotic depression were excluded; participants were not described as uniformly treatment-resistant.
- Why unverified: Confirms the retained diagnostic and eligibility boundaries; it does not establish an effect applicable to all adults or depression subtypes.
- Reason for missing information: Confirms the retained diagnostic and eligibility boundaries; it does not establish an effect applicable to all adults or depression subtypes.
- Required materials and conditions: Clinician diagnosis under ICD10, moderate/severe depression, BDI-II screening/randomization criteria, and new/existing antidepressant use
- Materials to obtain first: Clinician diagnosis under ICD10, moderate/severe depression, BDI-II screening/randomization criteria, and new/existing antidepressant use
- Further research: Link original inclusion/exclusion criteria, recruitment and baseline status to the actual denominators.
- Conditions for verification: Preserve BDI-II screening ≥19 and randomization ≥17, clinician diagnosis and the actual population boundaries.
- Why this condition is needed: Needed to avoid recasting the population as entirely treatment-resistant depression or as general low mood.
- Application exceptions: Do not generalize trial screening cutpoints into diagnostic criteria for all clinical care.
- Remaining uncertainty: Confirms the retained diagnostic and eligibility boundaries; it does not establish an effect applicable to all adults or depression subtypes.
- Referenced methods: M03 §3.2.1; detailed population criteria are internally designed; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed
- Related sources: S02, S04, S18

### N08 · Does deficiency screening establish normal nutritional status in everyone?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Folate and B12 deficiency were screened, but small subgroups with low values were also reported later. Normal nutritional status was not assumed for everyone, and some baseline biochemical units remain unverified.
- Why unverified: Screening and subgroups are preserved, but verification of every individual's nutritional status and comparison of units remain incomplete.
- Reason for missing information: Screening and subgroups are preserved, but verification of every individual's nutritional status and comparison of units remain incomplete.
- Required materials and conditions: Folate/B12 deficiency screening, baseline values/units and subgroups with low values
- Materials to obtain first: Folate/B12 deficiency screening, baseline values/units and subgroups with low values
- Further research: Compare baseline tests/units and eligibility with the prospective plan for stratified analyses.
- Conditions for verification: Preserve screening records, small low-value subgroups and unverified units separately.
- Why this condition is needed: Needed to avoid turning a screening policy into a claim that every participant had normal status.
- Application exceptions: Do not repurpose results from a small low-value subgroup as the overall primary effect or an effect in confirmed deficiency.
- Remaining uncertainty: Screening and subgroups are preserved, but verification of every individual's nutritional status and comparison of units remain incomplete.
- Referenced methods: M03 §3.2.1; detailed population criteria are internally designed; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed
- Related sources: S02, S03, S10

### N09 · Is restricting the page to nonlactating adults the same as verifying actual lactation status?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The trial excluded pregnancy and planned pregnancy. Page applicability is restricted to nonlactating adults, but actual trial lactation status could not be explicitly verified.
- Why unverified: Pregnancy exclusion is preserved, and lactation status remains unverified.
- Reason for missing information: Pregnancy exclusion is preserved, and lactation status remains unverified.
- Required materials and conditions: Exclusion of pregnancy/planned pregnancy, lactation inclusion/exclusion criteria and actual status
- Materials to obtain first: Exclusion of pregnancy/planned pregnancy, lactation inclusion/exclusion criteria and actual status
- Further research: Compare actual eligibility criteria, lactation status and evidence for special populations separately.
- Conditions for verification: Separate exclusion of pregnancy/planned pregnancy from unverified lactation status.
- Why this condition is needed: Needed to avoid turning a page restriction into a record that lactating participants were actually excluded from the trial.
- Application exceptions: Do not turn unverified lactation status into confirmed harm, efficacy or exclusion during lactation.
- Remaining uncertainty: Pregnancy exclusion is preserved, and lactation status remains unverified.
- Referenced methods: M03 §3.2.1; detailed population criteria are internally designed; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed; M19 §19.1.2–19.2.4; internal application
- Related sources: S02, S04, S18

### N10 · Are the subjective scale's direction and collection distinguished from blinding?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: BDI-II is a patient-reported total score for depressive symptoms; lower is better. Stored records describe a multicenter randomized, blinded design with a matching placebo, but original measurement and blinding implementation were not newly verified in this review.
- Why unverified: The scale, direction and design are preserved, but current comparison of implementation details with the original source remains outstanding.
- Reason for missing information: The scale, direction and design are preserved, but current comparison of implementation details with the original source remains outstanding.
- Required materials and conditions: BDI-II total score, lower-is-better direction, collection at visits, matching placebo/blinding, and limitations in implementation
- Materials to obtain first: BDI-II total score, lower-is-better direction, collection at visits, matching placebo/blinding, and limitations in implementation
- Further research: Link scale collection and blinding implementation at each visit to the selected analysis.
- Conditions for verification: Explain the patient-reported depression total score, reported blinding and the extent of implementation not cross-checked.
- Why this condition is needed: Needed to avoid asserting absence of all measurement bias solely from the scale name or double blinding.
- Application exceptions: Do not treat a subjective scale as an automatic failure or recast it as a biomarker.
- Remaining uncertainty: The scale, direction and design are preserved, but current comparison of implementation details with the original source remains outstanding.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 §8.1–8.2.3; internal application; M08 Table 8.2.a, M03 §3.2.4; internal application
- Related sources: S02, S07

### N11 · How should the time-weighted mean be interpreted?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: This is the time-weighted mean obtained by dividing the area between baseline and week 4, 12 and 25 scores by the actual observation period. Its unit is BDI-II points, not point-weeks, change from baseline or the week-25 endpoint.
- Why unverified: Confirms the retained definition; it is not a newly calculated AUC from individual-level data.
- Reason for missing information: Confirms the retained definition; it is not a newly calculated AUC from individual-level data.
- Required materials and conditions: Area across baseline and weeks 4/12/25, divided by the actual observation period; mean expressed in BDI-II points
- Materials to obtain first: Area across baseline and weeks 4/12/25, divided by the actual observation period; mean expressed in BDI-II points
- Further research: Compare definitions of the estimand and units in the original SAP and results tables.
- Conditions for verification: Link the final plan's time-weighted AUC average, units and adjusted between-group estimand.
- Why this condition is needed: Needed to avoid recasting it as point-weeks, within-group change or a single endpoint.
- Application exceptions: Do not perform new AUC calculations or conversions in this review.
- Remaining uncertainty: Confirms the retained definition; it is not a newly calculated AUC from individual-level data.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S02, S06, S07

### N12 · Were the initial 25-week endpoint and the final AUC the same from the outset?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The initial single time point at 25 weeks differs from the final AUC average. The final report describes a SAP before unblinding, and S07 identifies the passage about the unblinding meeting dated 2011-10-10; the exact amendment approval date and complete registration history remain unverified.
- Why unverified: The change and its reported location are preserved, but direct comparison of the full chronology remains incomplete.
- Reason for missing information: The change and its reported location are preserved, but direct comparison of the full chronology remains incomplete.
- Required materials and conditions: Original protocol, erratum, final SAP, unblinding/amendment approval dates and registration history
- Materials to obtain first: Original protocol, erratum, final SAP, unblinding/amendment approval dates and registration history
- Further research: Compare protocol, erratum and SAP versions, registration changes and approval timing.
- Conditions for verification: Show separately the initial single time point, final AUC, reportedly pre-unblinding SAP and unverified approval date.
- Why this condition is needed: Needed to avoid concealing the endpoint change and unverified details.
- Application exceptions: Do not turn an unavailable plan into confirmed selective reporting or a claim of an identical prospective plan from the outset.
- Remaining uncertainty: The change and its reported location are preserved, but direct comparison of the full chronology remains incomplete.
- Referenced methods: M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design; M08 Table 8.2.a, M04 §4.6.3; internal application
- Related sources: S04, S07, S08, S09, S18

### N13 · To what extent are the actual comparative design and allocation process verified?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: FolATED was a multicenter individually randomized, matched-placebo parallel-group trial under a common antidepressant optimization policy. During this summary conversion, the complete original text on allocation, concealment and baseline balance was not newly cross-checked.
- Why unverified: The design is preserved, but verification of detailed implementation does not extend beyond the current local records.
- Reason for missing information: The design is preserved, but verification of detailed implementation does not extend beyond the current local records.
- Required materials and conditions: Multicenter individually randomized parallel groups, matching placebo, allocation/concealment and baseline balance
- Materials to obtain first: Multicenter individually randomized parallel groups, matching placebo, allocation/concealment and baseline balance
- Further research: Link original randomization/concealment information and baseline tables to the selected outcome.
- Conditions for verification: Distinguish the recorded actual design from details not cross-checked in this review.
- Why this condition is needed: Needed to avoid extending the design label into complete verification of concealment and baseline balance.
- Application exceptions: Do not turn unverified details into a nonrandomized classification or automatic failure.
- Remaining uncertainty: The design is preserved, but verification of detailed implementation does not extend beyond the current local records.
- Referenced methods: M03 §3.2.1; detailed population criteria are internally designed; M08 §8.1–8.2.3; internal application
- Related sources: S02, S04

### N14 · What do the denominators of 475 and 440 represent?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: There were 475 valid randomized participants (237 folic acid, 238 placebo). The primary analysis included 440 (223 and 217) after excluding 35 with no postrandomization BDI-II at all. The 440 were not described as completers of every visit or as ITT including all 475.
- Why unverified: Confirms preservation of the denominator distinctions; it does not mean the impact of loss to follow-up has been resolved.
- Reason for missing information: Confirms preservation of the denominator distinctions; it does not mean the impact of loss to follow-up has been resolved.
- Required materials and conditions: Valid allocation 237/238, total 475; 35 with no follow-up BDI-II; mITT 223/217, total 440
- Materials to obtain first: Valid allocation 237/238, total 475; 35 with no follow-up BDI-II; mITT 223/217, total 440
- Further research: Compare the original flowchart, analysis definitions and loss to follow-up by arm.
- Conditions for verification: Preserve allocation, reasons for exclusion and primary-analysis denominators together with their arms.
- Why this condition is needed: Needed to avoid substituting the allocation denominator for analysis or completion denominators.
- Application exceptions: Do not create dropout rates, new denominators or new calculations in this review.
- Remaining uncertainty: Confirms preservation of the denominator distinctions; it does not mean the impact of loss to follow-up has been resolved.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S03, S06

### N15 · Does regression imputation include the 35 excluded participants?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The primary analysis retained assigned groups but was mITT, including only participants with at least 1 follow-up value. Regression imputation was used for missing data within the included 440. The 35 with no follow-up BDI-II were excluded from the primary analysis; this was not recast as ITT including all 475.
- Why unverified: The retained analysis and missing-data scope are available, but the impact of exclusions was neither fully resolved nor reanalyzed.
- Reason for missing information: The retained analysis and missing-data scope are available, but the impact of exclusions was neither fully resolved nor reanalyzed.
- Required materials and conditions: Assigned groups retained; inclusion requires at least 1 follow-up; imputation within the 440; sensitivity analysis of all randomized participants
- Materials to obtain first: Assigned groups retained; inclusion requires at least 1 follow-up; imputation within the 440; sensitivity analysis of all randomized participants
- Further research: Compare original missing-data reasons, imputation and sensitivity methods with the effect of interest.
- Conditions for verification: Separate the inclusion criteria for modified ITT from the scope of imputation and exclusions.
- Why this condition is needed: Needed to avoid calling imputation full ITT or removing the B1 defect.
- Application exceptions: Do not infer that imputation within the included group brought excluded participants into the primary analysis.
- Remaining uncertainty: The retained analysis and missing-data scope are available, but the impact of exclusions was neither fully resolved nor reanalyzed.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S02, S03, S07

### N16 · Are alternative analyses of the same trial counted as new replications?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Complete-case, repeated mixed-model and sensitivity analyses reweighting all randomized participants are analyses of the same FolATED trial. The record that completers of all visits and complete BDI cases have different denominators is preserved; exact counts absent from the current summary were not filled in.
- Why unverified: Sensitivity-analysis types and denominator differences are available, but comparison of all values and denominators in the original tables remains outstanding.
- Reason for missing information: Sensitivity-analysis types and denominator differences are available, but comparison of all values and denominators in the original tables remains outstanding.
- Required materials and conditions: Complete-case, repeated mixed-model and reweighted-allocation analyses; denominators for completing visits and for complete BDI data
- Materials to obtain first: Complete-case, repeated mixed-model and reweighted-allocation analyses; denominators for completing visits and for complete BDI data
- Further research: Compare original Tables 18–20 and participant flow with each analysis denominator and result.
- Conditions for verification: Preserve sensitivity analyses within the same family and the differences in denominators.
- Why this condition is needed: Needed to avoid turning analysis counts into trial counts or collapsing different completion denominators into one.
- Application exceptions: Do not infer exact completion denominators absent from the current summary.
- Remaining uncertainty: Sensitivity-analysis types and denominator differences are available, but comparison of all values and denominators in the original tables remains outstanding.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S02, S03, S07

### N17 · What exact comparison and interval does +1.09 points refer to?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The adjusted primary-analysis folic-acid-minus-placebo difference was +1.09 BDI-II points, 95% CI −0.48 to 2.66, P=0.173, with 223/217 participants. Additional improvement was not demonstrated on a lower-is-better scale. The result was not recast as within-group change or a difference at the 25-week endpoint.
- Why unverified: Confirms preservation of the reported effect and direction; the original table's SE conflict remains separately recorded.
- Reason for missing information: Confirms preservation of the reported effect and direction; the original table's SE conflict remains separately recorded.
- Required materials and conditions: Table 13 adjusted folic acid minus placebo, 0–25-week AUC average, CI/P, 223/217 and lower-is-better direction
- Materials to obtain first: Table 13 adjusted folic acid minus placebo, 0–25-week AUC average, CI/P, 223/217 and lower-is-better direction
- Further research: Compare original Table 13, narrative text, analysis definitions and printed values.
- Conditions for verification: Preserve +1.09, CI −0.48 to 2.66 and P=0.173 together with the original estimand.
- Why this condition is needed: Needed to avoid presenting a positive value as improvement or recasting it as a within-group or endpoint difference.
- Application exceptions: Do not calculate new effects, CIs, P values or standardized estimates.
- Remaining uncertainty: Confirms preservation of the reported effect and direction; the original table's SE conflict remains separately recorded.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M08 §8.2.2, M15 §15.3.1; internal application; M15 §15.3.1–15.3.3; internal application
- Related sources: S03, S06, S07

### N18 · Do the printed SE and CI/P in Table 13 agree?

Verification status: Conflicting printed values  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The original review records that the SE=0.86 printed in Table 13 does not agree with CI/P under the usual normal approximation. +1.09, CI −0.48 to 2.66, P=0.173 and SE 0.86 are all preserved; no new SE/CI/P was generated.
- Why unverified: The conflict between original printed values remains unresolved; no new calculation or correction was made in this review.
- Reason for missing information: The conflict between original printed values remains unresolved; no new calculation or correction was made in this review.
- Required materials and conditions: Printed SE 0.86, CI/P, and records of numerical conflicts in tables and narrative text
- Materials to obtain first: Printed SE 0.86, CI/P, and records of numerical conflicts in tables and narrative text
- Further research: During an authorized correction, compare the original table, analysis output and errata, and retain before-and-after values.
- Conditions for verification: Preserve both the printed values and the explanation of the conflict.
- Why this condition is needed: Needed to avoid deleting problematic printed values or silently replacing them with recalculated values.
- Application exceptions: Preserve the historical conflict record; do not conduct a new normal-approximation check or generate new values.
- Remaining uncertainty: The conflict between original printed values remains unresolved; no new calculation or correction was made in this review.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M15 §15.3.1–15.3.3; internal application
- Related sources: S03, S06

### N19 · Were +0.06 and −0.06 overwritten with the same sign?

Verification status: Conflicting printed values  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The original record that the unadjusted AUC difference differs between Table 12 (+0.06) and Table 18 (−0.06) is preserved. Neither value was used as an alternative or correction to the adjusted primary effect of +1.09.
- Why unverified: Resolution of the unadjusted sign conflict remains unverified; no correction or recalculation was made in this review.
- Reason for missing information: Resolution of the unadjusted sign conflict remains unverified; no correction or recalculation was made in this review.
- Required materials and conditions: Unadjusted AUC difference in Tables 12/18, the sign printed in each table, and distinction from the adjusted primary effect
- Materials to obtain first: Unadjusted AUC difference in Tables 12/18, the sign printed in each table, and distinction from the adjusted primary effect
- Further research: Compare original tables/directions, errata and analysis outputs as a separate correction event.
- Conditions for verification: Retain the original conflict between +0.06 in Table 12 and −0.06 in Table 18 unchanged.
- Why this condition is needed: Needed to avoid erasing the sign conflict or transferring it to the adjusted primary effect.
- Application exceptions: Differences between analyses are not automatically treated as conflicts, but the actual conflict recorded here is preserved.
- Remaining uncertainty: Resolution of the unadjusted sign conflict remains unverified; no correction or recalculation was made in this review.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S03, S06

### N20 · Does failure to demonstrate benefit mean an effect of 0, equivalence or ineffectiveness in all MDD?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The CI for +1.09 includes 0, and improvement was not demonstrated in this comparison. This does not mean an exact effect of 0, equivalence or ineffectiveness of every dose in all MDD. Positive adjacent results are also preserved.
- Why unverified: Confirms preservation of interpretive boundaries; it is not a new conclusion of equivalence or ineffectiveness.
- Reason for missing information: Confirms preservation of interpretive boundaries; it is not a new conclusion of equivalence or ineffectiveness.
- Required materials and conditions: Direction of the primary effect, CI, fixed question and original meaning of E0
- Materials to obtain first: Direction of the primary effect, CI, fixed question and original meaning of E0
- Further research: Review direct results for the same estimand and relevant clinical thresholds alongside additional evidence.
- Conditions for verification: Distinguish failure to demonstrate improvement in this narrow comparison from the range allowed by the interval.
- Why this condition is needed: Needed to avoid turning nonsignificance into established absence of effect or equivalence.
- Application exceptions: Do not remove positive results for other doses, populations or scales.
- Remaining uncertainty: Confirms preservation of interpretive boundaries; it is not a new conclusion of equivalence or ineffectiveness.
- Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M15 §15.3.1–15.3.3; internal application; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application
- Related sources: S06, S13, S15

### N21 · Is the initial 3-point power target the MCID for AUC?

Verification status: Unverified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: A validated between-group MCID applicable to the AUC average was not obtained. The initial 3-point power target for the single 25-week time point was not repurposed as an AUC MCID. The original C0 is preserved unchanged.
- Why unverified: Because an applicable importance threshold is unverified, exclusion of clinically important benefit is not established.
- Reason for missing information: Because an applicable importance threshold is unverified, exclusion of clinically important benefit is not established.
- Required materials and conditions: Reported CI, a validated between-group threshold for clinical importance of the AUC average, and the initial 3-point target at 25 weeks
- Materials to obtain first: Reported CI, a validated between-group threshold for clinical importance of the AUC average, and the initial 3-point target at 25 weeks
- Further research: Compare a justified threshold for the same endpoint and estimand with the CI.
- Conditions for verification: Separate the unverified applicable AUC MCID from the role of the initial design target.
- Why this condition is needed: Needed to avoid using a power target as an MCID or as evidence excluding clinically important benefit.
- Application exceptions: Do not create a universal minimum sample size, duration or study count, or a new MCID.
- Remaining uncertainty: Because an applicable importance threshold is unverified, exclusion of clinically important benefit is not established.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M15 §15.3.1–15.3.3; internal application
- Related sources: S02, S06

### N22 · Are endpoint means, adjusted endpoint differences and the primary AUC separated?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The adjusted 25-week difference was +1.27 points (CI −0.99 to 3.54; P=0.270); unadjusted means (SDs) were 22.61 (13.53) for folic acid and 22.33 (14.09) for placebo. These were not interchanged with the primary AUC difference of +1.09 or with within-group change.
- Why unverified: Confirms preservation of time-point, dispersion and analysis distinctions; the original tables were not newly verified in this review.
- Reason for missing information: Confirms preservation of time-point, dispersion and analysis distinctions; the original tables were not newly verified in this review.
- Required materials and conditions: Adjusted 25-week difference +1.27/CI/P, arm-specific unadjusted means and SDs, and distinct estimands
- Materials to obtain first: Adjusted 25-week difference +1.27/CI/P, arm-specific unadjusted means and SDs, and distinct estimands
- Further research: Compare Table 14 and Appendix 10/Table 42 with the original analysis definitions.
- Conditions for verification: Preserve the three types of values, the 25-week time point and units separately.
- Why this condition is needed: Needed to avoid combining different results into one primary effect.
- Application exceptions: Do not calculate new endpoint differences, standardized estimates or change scores.
- Remaining uncertainty: Confirms preservation of time-point, dispersion and analysis distinctions; the original tables were not newly verified in this review.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M08 §8.2.2, M15 §15.3.1; internal application
- Related sources: S10, S12

### N23 · Are FolATED reports and appendices counted as independent trials?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The current evidence links S01, S02, S03, S04, S06, S07, S08, S09, S10, S12 and S18 to one FOLATED trial family. Economic, genetic, biochemical and sensitivity analyses are also not new independent replications.
- Why unverified: Confirms preservation of family links; it does not mean 19 bibliographic records represent 19 trials.
- Reason for missing information: Confirms preservation of family links; it does not mean 19 bibliographic records represent 19 trials.
- Required materials and conditions: FOLATED links among the report, Methods/Results, protocol/erratum, SAP, registration and appendices
- Materials to obtain first: FOLATED links among the report, Methods/Results, protocol/erratum, SAP, registration and appendices
- Further research: Verify the unique recruitment/allocation identity and distinctions from other families at the original locations.
- Conditions for verification: Preserve the 11 linked reports S01/02/03/04/06/07/08/09/10/12/18 as one FOLATED family.
- Why this condition is needed: Needed to avoid inflating independent replication or participant counts from the number of documents.
- Application exceptions: Keep verified document identities separate from the unverified complete registration history.
- Remaining uncertainty: Confirms preservation of family links; it does not mean 19 bibliographic records represent 19 trials.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between 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, S02, S03, S04, S06, S07, S08, S09, S10, S12, S18

### N24 · Why is the positive female-subgroup HDRS-17 result not counted as direct replication?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The positive Coppen result is preserved: 127 participants in total; folic acid 500 µg/day added to fluoxetine 20 mg/day; 10-week female-subgroup HDRS17 endpoints of 6.8 (SD 4.1) versus 11.7 (SD 6.7), P<0.001. The exact female denominator and prospective interaction testing remain unverified; this is not replication of FolATED's overall AUC.
- Why unverified: A positive abstract result is available, but subgroup plans/denominators, the full text and direct applicability to the identical question remain unverified.
- Reason for missing information: A positive abstract result is available, but subgroup plans/denominators, the full text and direct applicability to the identical question remain unverified.
- Required materials and conditions: 127 participants, 500 µg/day, 10 weeks, fluoxetine 20 mg, female-subgroup HDRS17, and subgroup denominators/plans
- Materials to obtain first: 127 participants, 500 µg/day, 10 weeks, fluoxetine 20 mg, female-subgroup HDRS17, and subgroup denominators/plans
- Further research: Compare the abstract's scope with subgroup plans/denominators and the direct comparison in the original full text.
- Conditions for verification: Preserve 6.8 (SD 4.1) versus 11.7 (SD 6.7), P<0.001, and the different population stratum, dose, scale and duration.
- Why this condition is needed: Needed to avoid deleting favorable adjacent results or transferring them to FolATED's overall AUC.
- Application exceptions: Do not infer the exact female denominator or prospective interaction testing.
- Remaining uncertainty: A positive abstract result is available, but subgroup plans/denominators, the full text and direct applicability to the identical question remain unverified.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M03 §3.2.1; detailed population criteria are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed
- Related sources: S13

### N25 · What role does the uncorrected author proof for 2.5 mg over 8 weeks have?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The Sepehrmanesh uncorrected author proof reports 45 versus 45 participants, folic acid 2.5 mg/day for 8 weeks added to citalopram 20 mg/day, and endpoint BDI-II of 13.31 (SD 6.57) versus 19.11 (SD 8.59), P<0.001. The positive result is preserved without counting it as replication of the FolATED AUC; final-version and registration limitations are retained.
- Why unverified: Results are available from the accessed version, but the final version and registration, and direct replication of the selected question, remain unverified.
- Reason for missing information: Results are available from the accessed version, but the final version and registration, and direct replication of the selected question, remain unverified.
- Required materials and conditions: Sepehrmanesh 45/45, citalopram 20 mg, 2.5 mg/day for 8 weeks, BDI-II, and version/registration
- Materials to obtain first: Sepehrmanesh 45/45, citalopram 20 mg, 2.5 mg/day for 8 weeks, BDI-II, and version/registration
- Further research: Compare the final publication version, registration history, denominators and plans with the different exposures and estimands.
- Conditions for verification: Preserve the positive result of 13.31 (SD 6.57) versus 19.11 (SD 8.59), P<0.001, together with version limitations.
- Why this condition is needed: Needed to avoid calling it replication of the 5 mg, 25-week AUC solely because it uses the same scale name.
- Application exceptions: Do not upgrade an uncorrected author proof to complete verification of the final published version.
- Remaining uncertainty: Results are available from the accessed version, but the final version and registration, and direct replication of the selected question, remain unverified.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application
- Related sources: S15

### N26 · Was Resler metadata promoted to verified results?

Verification status: Unverified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Resler2008 is a candidate human trial of folic acid plus fluoxetine, but access to the full results failed. The exact dose, denominators and effect estimate remain unverified; it was retained only as a metadata candidate and excluded from replication counts and scoring.
- Why unverified: A bibliographic candidate exists, but no full-text verification is available to satisfy the result criteria.
- Reason for missing information: A bibliographic candidate exists, but no full-text verification is available to satisfy the result criteria.
- Required materials and conditions: PMID18716414/DOI, the scope of access failure, and actual dose/denominators/effect
- Materials to obtain first: PMID18716414/DOI, the scope of access failure, and actual dose/denominators/effect
- Further research: If separately obtaining the source is authorized, link the original results, design, exposure and denominators.
- Conditions for verification: Separate candidate identity from unverified full-text results.
- Why this condition is needed: Needed to avoid promoting a literature candidate to verified efficacy or a replication count.
- Application exceptions: Access failure does not mean absence of a study or an effect of 0.
- Remaining uncertainty: A bibliographic candidate exists, but no full-text verification is available to satisfy the result criteria.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application
- Related sources: S14

### N27 · Were reviews and trials in people younger than 18 combined with direct adult results?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Reviews S16/19 provide clues for literature discovery and mapping contrary results. S17 is a fluoxetine adjunct trial in people younger than 18, outside the adult MDD population boundary; it was not added to adult denominators or funding assessments for the direct trial.
- Why unverified: Confirms roles and exclusion boundaries; not all underlying trials in the reviews were verified in this review.
- Reason for missing information: Confirms roles and exclusion boundaries; not all underlying trials in the reviews were verified in this review.
- Required materials and conditions: Discovery role of reviews S16/19, population boundary for pediatric S17, and original families
- Materials to obtain first: Discovery role of reviews S16/19, population boundary for pediatric S17, and original families
- Further research: If needed, separately compare underlying trial identities, adult-population boundaries and actual results.
- Conditions for verification: Distinguish evidence mapping and exclusion from the adult population from unverified direct results.
- Why this condition is needed: Needed to avoid counting reviews and material outside the adult population as independent adult replications.
- Application exceptions: Preserve the roles of reviews and excluded material without deleting them.
- Remaining uncertainty: Confirms roles and exclusion boundaries; not all underlying trials in the reviews were verified in this review.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M03 §3.2.1; detailed population criteria are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M04 §4.6.1–4.6.2, M15 §15.3.1; internal application
- Related sources: S16, S17, S19

### N28 · Was the secondary SF-12 result merged into BDI-II worsening or confirmed harm?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The difference in the time-weighted SF-12 mental-health mean was −1.97 points for folic acid minus placebo (CI −3.49 to −0.44; P=0.012), a nominal secondary comparison favoring placebo. It was not merged into BDI-II worsening or established clinical harm.
- Why unverified: Confirms preservation of the unfavorable result and its role; it does not certify a causal harm relationship.
- Reason for missing information: Confirms preservation of the unfavorable result and its role; it does not certify a causal harm relationship.
- Required materials and conditions: SF12 mental-health AUC −1.97/CI/P, favoring placebo, and nominal secondary comparison
- Materials to obtain first: SF12 mental-health AUC −1.97/CI/P, favoring placebo, and nominal secondary comparison
- Further research: Compare original Table 13, the plan, multiplicity and clinical meaning for that scale.
- Conditions for verification: Preserve −1.97, CI −3.49 to −0.44, P=0.012 and the endpoint's role and multiplicity context.
- Why this condition is needed: Needed to avoid concealing an unfavorable result or repurposing it as a different endpoint.
- Application exceptions: Do not turn nominal secondary significance into established clinical harm.
- Remaining uncertainty: Confirms preservation of the unfavorable result and its role; it does not certify a causal harm relationship.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M08 Table 8.2.a, M04 §4.6.3; internal application; M03 §3.2.4; internally defined boundaries for extending the claim; M19 §19.1.2–19.2.4; internal application
- Related sources: S03, S06, S10

### N29 · Do small subgroups, response/remission or blood results replace the primary effect?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Small low-folate, genotype and sex subgroups, response, remission and biochemical results were reported, but they do not replace the overall primary outcome. Biochemical/response analyses in the appendices belong to the same FolATED family; effects absent from the current summary were not created.
- Why unverified: The roles of other analyses are available, but verification of complete plans, quantitative results and denominators is limited.
- Reason for missing information: The roles of other analyses are available, but verification of complete plans, quantitative results and denominators is limited.
- Required materials and conditions: Plans, denominators and individual results for low-folate/genotype/sex subgroups, response/remission and biochemistry
- Materials to obtain first: Plans, denominators and individual results for low-folate/genotype/sex subgroups, response/remission and biochemistry
- Further research: Separately compare prospective subgroup/interaction plans, denominators, multiplicity and original results.
- Conditions for verification: Preserve the existence and roles of other results and the boundaries of the primary effect.
- Why this condition is needed: Needed to avoid using favorable selected results as the overall BDI-II primary effect.
- Application exceptions: Do not fill in new quantitative effects or interactions absent from the current summary.
- Remaining uncertainty: The roles of other analyses are available, but verification of complete plans, quantitative results and denominators is limited.
- Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is an internal application; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M08 Table 8.2.a, M04 §4.6.3; internal application; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed; M03 §3.2.4; internally defined boundaries for extending the claim
- Related sources: S03, S10

### N30 · Does public funding mean there are no competing interests?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Funding for the decisive FolATED trial was NIHR HTA04/35/08; Andrew McCaddon's scientific-advisor/shareholder relationship with COBALZ Ltd and Bilcare supply are also preserved. Public funding does not mean absence of conflicts. Coppen funding was unverified in the accessed abstract.
- Why unverified: Public funding and disclosed relationships are available, but not all competing interests or funding for adjacent studies have been fully verified.
- Reason for missing information: Public funding and disclosed relationships are available, but not all competing interests or funding for adjacent studies have been fully verified.
- Required materials and conditions: NIHR HTA04/35/08, COBALZ advisory/shareholder roles, Bilcare supply and funding for adjacent studies
- Materials to obtain first: NIHR HTA04/35/08, COBALZ advisory/shareholder roles, Bilcare supply and funding for adjacent studies
- Further research: Compare the original funding/COI/supply information and the scope of adjacent-study disclosures.
- Conditions for verification: Preserve verified public funding together with disclosed author and supplier relationships.
- Why this condition is needed: Needed to avoid extending public funding into a guarantee of no conflicts or bias.
- Application exceptions: Do not automatically exclude results because a competing interest exists.
- Remaining uncertainty: Public funding and disclosed relationships are available, but not all competing interests or funding for adjacent studies have been fully verified.
- Referenced methods: Chamgap's own criteria for independence and attribution. The official methods sections used in this pilot do not endorse all of these detailed items
- Related sources: S01, S02, S13, S15, S18

### N31 · Are 33/45 and 6/14 counts of events or people?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Over 25 weeks, AE counts were 33 for folic acid and 45 for placebo; SAE counts were 6 and 14; possible drug reactions were 4 and 3. These are event counts and were not assumed to equal unique patient counts. Incidence, NNH and new risk differences were not generated.
- Why unverified: Reported event counts are available, but verification of unique-person denominators, complete collection, causality and comparative risk remains incomplete.
- Reason for missing information: Reported event counts are available, but verification of unique-person denominators, complete collection, causality and comparative risk remains incomplete.
- Required materials and conditions: 25-week AE 33/45, SAE 6/14, possible drug reactions 4/3, and unique-person denominators/collection
- Materials to obtain first: 25-week AE 33/45, SAE 6/14, possible drug reactions 4/3, and unique-person denominators/collection
- Further research: Compare event definitions and collection, exposure, unique people and missingness in original Tables 15–17.
- Conditions for verification: Distinguish reported event counts and observation duration from unverified person denominators.
- Why this condition is needed: Needed to avoid turning event counts into unique patient counts, incidence or NNH.
- Application exceptions: Do not calculate new harm rates, comparative risks or NNH.
- Remaining uncertainty: Reported event counts are available, but verification of unique-person denominators, complete collection, causality and comparative risk remains incomplete.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; detailed differentiation is internally designed; M19 §19.2 questions about harms by study design; details of temporal relationships are internally designed; M19 §19.2.2–19.2.4, M03 §3.2.3; internal application; M19 §19.2.2–19.2.4 considerations for study design; detailed recording of confounding is internally designed; 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 assessment tool was applied
- Related sources: S03

### N32 · Are 0 serious/unexpected reactions being read as 0 events overall?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Among the 4 and 3 possible drug reactions, 0 were serious or unexpected. This does not mean that all AE/SAE counts were 0, and it does not establish long-term absence of risk or equivalence.
- Why unverified: Confirms preservation of the scope of 0 events; it does not verify absence of all harms.
- Reason for missing information: Confirms preservation of the scope of 0 events; it does not verify absence of all harms.
- Required materials and conditions: Scope of 0 serious or unexpected reactions among the 4/3 possible drug reactions
- Materials to obtain first: Scope of 0 serious or unexpected reactions among the 4/3 possible drug reactions
- Further research: Compare original reaction classifications, seriousness/expectedness and collection scope.
- Conditions for verification: State both the category to which 0 refers and the overall event counts.
- Why this condition is needed: Needed to avoid extending 0 in one subcategory to all AE/SAE or to long-term absence of risk.
- Application exceptions: Do not turn failure to verify a significant difference into equivalence or an assurance of safety.
- Remaining uncertainty: Confirms preservation of the scope of 0 events; it does not verify absence of all harms.
- Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; detailed differentiation is internally designed; Chamgap's own interpretive boundaries. M19 principles are referenced; no official causality assessment tool was applied
- Related sources: S03

### N33 · Do 25/223 and 28/217 represent suicide attempts or deaths?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: In the 25-week imputed MINI analysis, the high-risk category included 25/223 participants receiving folic acid and 28/217 receiving placebo. These are not counts of actual suicide attempts or deaths by suicide. Separately counted arm-specific behavioral and death events remain unverified.
- Why unverified: Risk classification is preserved, but event data for comparisons of actual behavior and deaths remain unverified.
- Reason for missing information: Risk classification is preserved, but event data for comparisons of actual behavior and deaths remain unverified.
- Required materials and conditions: 25-week MINI high-risk category, imputed analysis with 223/217, and actual suicidal-behavior/death events
- Materials to obtain first: 25-week MINI high-risk category, imputed analysis with 223/217, and actual suicidal-behavior/death events
- Further research: Compare original MINI definitions, imputation, collection and actual event records with their respective denominators.
- Conditions for verification: Separate risk-category counts of 25/223 and 28/217 from unverified separately counted events.
- Why this condition is needed: Needed to avoid turning a scale classification into counts of actual behavior or deaths.
- Application exceptions: Do not fill unverified behavior or death counts with 0.
- Remaining uncertainty: Risk classification is preserved, but event data for comparisons of actual behavior and deaths remain unverified.
- Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; adaptations for each study design are internally designed; M03 §3.2.4; internally defined boundaries for extending the claim; M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; detailed differentiation is internally designed; 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 assessment tool was applied
- Related sources: S03

### N34 · What is the scope of documents on B12 masking and interactions?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The original safety explanation that folic acid can improve hematological signs of B12 deficiency while delaying recognition of neurological problems, and the antiepileptic/methotrexate context, are preserved. For NIH, access failed after partial search extracts; the Flamingo SmPC entry records historical access to an official label. These do not establish FolATED incidence, product identity or a depression indication.
- Why unverified: Safety context is available, but confirmation of the same events, causality and frequency in the direct trial, and current verification of original full texts, are separate matters.
- Reason for missing information: Safety context is available, but confirmation of the same events, causality and frequency in the direct trial, and current verification of original full texts, are separate matters.
- Required materials and conditions: Masking of hematological signs of B12 deficiency, neurological problems, antiepileptics/methotrexate, and partial NIH material/SmPC
- Materials to obtain first: Masking of hematological signs of B12 deficiency, neurological problems, antiepileptics/methotrexate, and partial NIH material/SmPC
- Further research: Separately compare original safety documents, case/study families, exposure, timing and alternative explanations.
- Conditions for verification: Record general safety context, the scope of material access, and boundaries between other products and trial data.
- Why this condition is needed: Needed to avoid using general context from official documents as FolATED incidence or independent events.
- Application exceptions: Citing an official document again does not constitute a new independent event or evidence of authorization for depression.
- Remaining uncertainty: Safety context is available, but confirmation of the same events, causality and frequency in the direct trial, and current verification of original full texts, are separate matters.
- Referenced methods: M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M19 §19.1.2–19.2.4; internal application; M19 §19.1.2–19.2.4; detailed differentiation is internally designed; M19 §19.2 questions about harms by study design; details of temporal relationships are internally designed; M19 §19.2.2–19.2.4 considerations for study design; detailed recording of confounding is internally designed; M04 §4.6.1–4.6.2; adaptations for harms data are internally designed; Chamgap's own interpretive boundaries. M19 principles are referenced; no official causality assessment tool was applied
- Related sources: S05, S11

### N35 · Does 25-week observation answer every long-term or special-population safety question?

Verification status: Unverified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The scope of verification for long-term and high-dose use, pregnancy, lactation, children, severe hepatic/renal disease and cancer history is separate. 12 weeks of treatment, 25 weeks of observation and official safety materials alone were not described as verifying all special populations or rare harms.
- Why unverified: Extended harm information remains unverified within the selected materials.
- Reason for missing information: Extended harm information remains unverified within the selected materials.
- Required materials and conditions: Direct exposure and events for high-dose/long-term use, pregnancy/lactation/children, hepatic/renal disease and cancer history
- Materials to obtain first: Direct exposure and events for high-dose/long-term use, pregnancy/lactation/children, hepatic/renal disease and cancer history
- Further research: Separately design or review material suited to the relevant exposure, population and long-term/rare events.
- Conditions for verification: Distinguish trial scope from each unverified extended harm question.
- Why this condition is needed: Needed to avoid expanding narrow observation into assurances for all vulnerable populations or rare harms.
- Application exceptions: Do not impose studies in all special populations as requirements for this narrow efficacy question.
- Remaining uncertainty: Extended harm information remains unverified within the selected materials.
- Referenced methods: M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed; M19 §19.1.2–19.2.4; internal application; Chamgap's own interpretive boundaries. M19 principles are referenced; no official causality assessment tool was applied
- Related sources: S02, S05, S11

### N36 · Is reading summaries distinguished from past reading of external sources?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The relevant locations and values in the current stored summaries were actually read in this review. The 19 bibliographic entries' full_text, abstract, metadata, partial-access/failure and uncorrected-author-proof levels are historical records. Actual report-specific access-date fields are absent and therefore blank; 2026-09-15 is kept separately as the evidence cutoff. Current external full texts have not been read.
- Why unverified: The current summaries are verified, but current reading of external full texts, their original bytes and report-specific access dates remain unverified.
- Reason for missing information: The current summaries are verified, but current reading of external full texts, their original bytes and report-specific access dates remain unverified.
- Required materials and conditions: Exact locations/values/SHA of current stored summaries, access levels of 19 bibliographic entries, report-specific access dates and evidence cutoff date
- Materials to obtain first: Exact locations/values/SHA of current stored summaries, access levels of 19 bibliographic entries, report-specific access dates and evidence cutoff date
- Further research: When original materials are actually obtained, separately preserve material types, dates, locations read and original bytes.
- Conditions for verification: Link the local summaries actually read in this review separately from past access levels and unread external sources.
- Why this condition is needed: Needed to avoid upgrading a summary link to current full-text verification or using the review date as an access date.
- Application exceptions: Do not label external access not attempted as access failure.
- Remaining uncertainty: The current summaries are verified, but current reading of external full texts, their original bytes and report-specific access dates remain unverified.
- Referenced methods: M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design
- Related sources: S01, S05, S09, S13, S14, S15

### N37 · Does the prior search scope represent a current exhaustive search or retraction check?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The evidence scope is ChatGPT_search_through_2026-09-15 and absence_proven=false. The ISRCTN identifier was cross-checked against the report and protocol, but access to the complete registration history failed. The 2009 protocol erratum is preserved; new clinical searches and external retraction/correction checks in this review total 0.
- Why unverified: Historical searches, partial registration information and a protocol correction are available, but there is no current exhaustive search or verification that no recent retractions exist.
- Reason for missing information: Historical searches, partial registration information and a protocol correction are available, but there is no current exhaustive search or verification that no recent retractions exist.
- Required materials and conditions: 2026-09-15 search scope, partial/failed registration access, protocol correction, and the scope of checking retractions/new evidence
- Materials to obtain first: 2026-09-15 search scope, partial/failed registration access, protocol correction, and the scope of checking retractions/new evidence
- Further research: During a separately authorized search or correction, preserve complete original texts, events, before-and-after values and versions.
- Conditions for verification: Link existing search/correction records to the scope of work currently not performed and information remaining unverified.
- Why this condition is needed: Needed to avoid turning 0 new searches in this review into 0 studies, proof of absence, or a claim that no recent corrections exist.
- Application exceptions: Do not merge a protocol erratum and a correction to original results into the same event.
- Remaining uncertainty: Historical searches, partial registration information and a protocol correction are available, but there is no current exhaustive search or verification that no recent retractions exist.
- Referenced methods: M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design; M08 Table 8.2.a, M04 §4.6.3; internal application
- Related sources: S04, S09, S18

### N38 · Are D30 and B/P/R1/I2/E0/B1/C0 being presented as a new verdict?

Verification status: Met  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: The original D/30 points/Caution and B/P/R1/I2/E0/B1/C0 are preserved. This is the original fixed anchor with I2 as 1 strength; no new score was calculated. E0 denotes a narrow failure to demonstrate benefit and was not changed to F/RX. Document quality A and single-assistant self-review are not independent clinical or journal certification.
- Why unverified: Confirms preservation of original values and review boundaries; no new clinical grade or certification was assigned.
- Reason for missing information: Confirms preservation of original values and review boundaries; no new clinical grade or certification was assigned.
- Required materials and conditions: Original grade/score/axes, 1 strength, document quality A, self-review and clinical certification=false
- Materials to obtain first: Original grade/score/axes, 1 strength, document quality A, self-review and clinical certification=false
- Further research: If a correction is separately authorized, record the original evidence, verdict, reasons and before-and-after values as an event.
- Conditions for verification: Preserve the original D30/Caution, 7 axes, meaning of E0, document quality and review level unchanged.
- Why this condition is needed: Needed to avoid turning table-criterion counts, colors or structural checks into a new efficacy score or clinical certification.
- Application exceptions: External content QA is inspection of the current table conversion; it is not retroactively treated as independent review of the original clinical assessment.
- Remaining uncertainty: Confirms preservation of original values and review boundaries; no new clinical grade or certification was assigned.
- Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design; Chamgap's own criteria for independence and attribution. The official methods sections used in this pilot do not endorse all of these detailed items; M15 §15.3.1–15.3.3; internal application
- Related sources: S01, S02, S06

### N39 · Are optimal dose, biomarkers, remission and mortality combined into the same answer?

Verification status: Not applicable  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Optimal dose, other folate chemical forms/B-vitamin combinations, blood-level increases, mortality and all clinical events were not automatically combined into the answer to the selected 5 mg/12-week BDI-II AUC question. Material suited to each extension is needed.
- Why unverified: Indicates that an extension outside the selected question is not applicable; it does not erase unverified information about the actual trial.
- Reason for missing information: Indicates that an extension outside the selected question is not applicable; it does not erase unverified information about the actual trial.
- Required materials and conditions: The selected AUC question and direct questions about other doses/durations, L-methylfolate, B-vitamin combinations and clinical events
- Materials to obtain first: The selected AUC question and direct questions about other doses/durations, L-methylfolate, B-vitamin combinations and clinical events
- Further research: If an extension is asked about, separately compare direct dose, follow-up and event data.
- Conditions for verification: Keep efficacy questions outside the selected question within separate boundaries.
- Why this condition is needed: Needed to avoid requiring all extension studies for the selected numerical outcome or extrapolating automatically.
- Application exceptions: Do not use inapplicability of an extension as a substitute for actual unverified product, plan or MCID information.
- Remaining uncertainty: Indicates that an extension outside the selected question is not applicable; it does not erase unverified information about the actual trial.
- Referenced methods: M03 §3.2.2–3.2.3; reverse-direction labeling and adaptations for testing are internally designed; M03 §3.2.4, M15 §15.3.1–15.3.3; outcome adaptations for each question are internally designed; M03 §3.2.1–3.2.4; details on dose, duration and effect modification are internally designed; M03 §3.2.4; internally defined boundaries for extending the claim
- Related sources: S02, S10, S13, S15

### N40 · Do historical deployment flags mean that this table has now been publicly published?

Verification status: Partly verified  
Reading scope: The exact locations and values in the preserved stored summaries were actually read; external sources and older separate original documents were not read in this review

- Current findings: Original ID3109, assigned code, original URL, server_deployed=true, server_write_performed=false and deployed_at 2026-09-16 are all preserved, along with the Korean statement that actual server implementation was not performed. This table is prepared locally, with 0 public deployments. Updates for new evidence, full texts, corrections or errors accumulate a history at the same ID/URL.
- Why unverified: Original metadata and update provisions are preserved, but authoritative correction of publication wording and actual site implementation were not performed in this review.
- Reason for missing information: Original metadata and update provisions are preserved, but authoritative correction of publication wording and actual site implementation were not performed in this review.
- Required materials and conditions: Original 3109/assigned code/URL, server_deployed true/write false, dates, update policy and before-and-after original text
- Materials to obtain first: Original 3109/assigned code/URL, server_deployed true/write false, dates, update policy and before-and-after original text
- Further research: During an authorized update, retain complete originals, original values/contrary results, reasons for changes and versions.
- Conditions for verification: Preserve original identity and wording, and distinguish local unpublished status and versions of criteria, observations and events.
- Why this condition is needed: Needed to avoid presenting historical identity/deployment wording as completed public publication of the current local sheet.
- Application exceptions: Do not arbitrarily correct differences in original metadata wording or extend deployment authorization.
- Remaining uncertainty: Original metadata and update provisions are preserved, but authoritative correction of publication wording and actual site implementation were not performed in this review.
- Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are internally designed; M04 §4.5–4.6 principles for collecting/selecting material; the separation of access from reading and the hyperlink contract are internally designed; M04 §4.6.3, M15 §15.3.1–15.3.3; the separation of statuses is internally designed; M04 §4.5, §4.6.3; limits on the present conversion of material reflect the user-defined scope and internal design
- Related sources: 

## Evidence and sources

### S01

FolATED: randomized trial and economic evaluation, HTA 18(48), 2014

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262306/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Report of the primary trial, public funding and COI; one family  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full text of the published report | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S02

FolATED Chapter 2: Methods

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262311/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Medication, eligibility and analysis policies; manufacturing and lactation unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full text of the report's Methods | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S03

FolATED Chapter 3: Results

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262313/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Denominators, effects, numerical conflicts, harms and sensitivity analyses  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full text of the report's Results | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S04

Roberts et al. 2007 FolATED protocol

[Source](https://link.springer.com/article/10.1186/1471-244X-7-65)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Initial plan for a single time point at 25 weeks; distinguished from the final AUC  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full protocol text | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S05

NIH ODS Folate health-professional fact sheet

[Source](https://ods.od.nih.gov/factsheets/Folate-HealthProfessional/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Access failed after partial search extracts; not a direct trial or incidence estimate  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Official NIH ODS web document | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S06

FolATED Table 13: adjusted primary AUC-average estimates

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262313/table/table13/?report=objectonly)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Primary effect +1.09/CI/P, SE conflict and secondary SF12  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Original report table | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S07

FolATED Appendix 5: final analysis plan

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262296/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: AUC, mITT/missing-data definitions and the source location for the record of the unblinding meeting; approval date unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full analysis plan text | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S08

FolATED protocol amendment history

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262312/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Protocol amendment chronology; not a substitute for the complete registration history  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full report appendix text | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S09

ISRCTN37558856 FolATED registration

[Source](https://www.isrctn.com/ISRCTN37558856)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Identifier cross-check and access failure after metadata; complete versions unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| ISRCTN metadata | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S10

FolATED Appendix 10: detailed clinical results

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262303/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: 25-week means/SDs, response/remission/biochemistry; not a new trial  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full text of the detailed results appendix | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S11

Flamingo Pharma folic acid 5 mg tablet SmPC, text revised 20 May 2026

[Source](https://www.medicines.org.uk/emc/product/14604/smpc)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Flamingo 5 mg label, revised 2026-05-20; not the FolATED product or authorization for MDD  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Official manufacturer label | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S12

FolATED Table 14: adjusted individual follow-up scores

[Source](https://www.ncbi.nlm.nih.gov/books/NBK262313/table/table14/?report=objectonly)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Adjusted single 25-week result +1.27; separate from the AUC  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Original report table | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S13

Coppen and Bailey 2000: folic acid added to fluoxetine

[Source](https://pubmed.ncbi.nlm.nih.gov/10967371/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Positive female-subgroup HDRS17 result with 500 µg over 10 weeks; female denominator/interaction unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Abstract/official index | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S14

Resler et al. 2008: folic acid plus fluoxetine

[Source](https://pubmed.ncbi.nlm.nih.gov/18716414/)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Metadata/partial-access failure; denominator/effect unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Metadata candidate | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S15

Sepehrmanesh et al.: Acid Folic Supplementation in MDD Treatment

[Source](https://www.researchgate.net/publication/306347545_Acid_Folic_Supplementation_in_Major_Depressive_Disorder_Treatment_A_Double-Blind_Randomized_Clinical_Trial)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Positive BDI-II result with 2.5 mg over 8 weeks; uncorrected proof, with final version/registration unverified  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Uncorrected author proof of the original article | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S16

Khalili et al. 2022/2023 folic acid depression systematic review

[Source](https://www.emerald.com/nfs/article/53/3/521/453862/The-effects-of-folic-acid-supplementation-on)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Discovery/contrary-evidence map; not itself direct replication of a primary trial  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Review abstract | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S17

Karbasi Amel and Haghshenas 2025: folic acid with fluoxetine below age18

[Source](https://brieflands.com/journals/jcp/articles/162489)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Fluoxetine adjunct in people younger than 18; excluded from the direct adult denominator  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full text of the pediatric original study | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S18

Roberts et al. 2009 FolATED protocol erratum

[Source](https://link.springer.com/article/10.1186/1471-244X-9-14)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: 2009-04-07 protocol erratum; separate from a correction of results  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Full protocol correction text | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

### S19

Altaf et al. 2021 folate adjunct SSRI/SNRI systematic review

[Source](https://www.sciencedirect.com/science/article/pii/S0965229921001114)

Materials actually read: Exact stored bibliographic information and linked summaries were actually read; external sources and older separate original documents were not read in this review  
Findings recorded in the summary: Discovery of literature on folate adjuncts to SSRI/SNRI; not counted as an independent trial  
Record level: Limited to the explicitly stated scope of historical stored bibliographic information/summaries; not new verification of clinical full texts  
Independent replication: 19 bibliographic records ≠ 19 trials; the 11 FolATED reports belong to one family, with other families, reviews and labels assigned distinct roles; no new trials or participants were summed

| Material | Public availability | Access status | Actually read | Notes |
|---|---|---|---|---|
| Preserved stored-summary bibliographic information, linked summaries and historical access | Preserved locally | Acquired | Summary read | Exact stored-summary pointers, values and SHA; not the original bytes of the original clinical full text |
| Review abstract/search metadata | Unverified (0 external lookups in this review) | Access not attempted | Not read | Distinguish historical access levels from access not attempted and sources not read in this review; the existence of a link does not mean the original source was verified in this review |

## 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 criteria and reasons for changes, and distinguish variations in the intervention. · Reuse of case 1585 methodological principles/versions and byte-for-byte copies of 5 HTML files; no new reading of complete methods texts or application of an official tool in this review
- 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, and do not omit a study solely because its outcomes are not reported. · Reuse of case 1585 methodological principles/versions and byte-for-byte copies of 5 HTML files; no new reading of complete methods texts or application of an official tool in this review
- 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 · Hold the outcome and effect of interest fixed, then assess randomization, deviations, missing data, measurement and selective reporting. · Reuse of case 1585 methodological principles/versions and byte-for-byte copies of 5 HTML files; no new reading of complete methods texts or application of an official tool in this review
- 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. · Reuse of case 1585 methodological principles/versions and byte-for-byte copies of 5 HTML files; no new reading of complete methods texts or application of an official tool in this review
- 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 · Questions about harms may require designs, populations and follow-up different from efficacy questions. Case reports do not establish incidence. · Reuse of case 1585 methodological principles/versions and byte-for-byte copies of 5 HTML files; no new reading of complete methods texts or application of an official tool in this review

## What the statuses mean

- Partly verified: Some preserved information is available, but gaps remain in manufacturing, exposure, plans, denominators, harms and original sources
- Unverified: Necessary information or conditions cannot be established within this review's scope
- Review incomplete: Original-source comparisons or searches were not performed within the permitted scope
- Not reported: A scope-limited status indicating that the item was not stated in the particular material or section read
- Inaccessible: Used only for material for which access was actually attempted and failed. Not assigned to clinical materials in this review.
- Insufficient eligible evidence: Used only when, after reviewing relevant studies within a defined scope, evidence is insufficient to answer the condition. Not applied universally in this review.
- Not applicable: An extension beyond the selected AUC comparison is not applicable; this does not substitute for actual unverified information
- Conflicting information: Used when comparison reveals information that is difficult to reconcile for the same question, time point and analysis. Not assigned solely because review estimates differ.
- Met: The stated preservation, boundary, distinction or linkage condition is verified; this does not mean new clinical efficacy verification, full-text reading or safety certification
- Conflicting printed values: The conflict between printed values in the original record is preserved; no new calculation or correction

## Change history

2026-11-05 · Public page/materials 1.0. A public-facing display and data were prepared from the existing 40 items. No new clinical search, change in verification status or scoring was performed.

## Material-specific source addresses

- S01-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S01-M2 · Full text of the published report](https://www.ncbi.nlm.nih.gov/books/NBK262306/)
- S02-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S02-M2 · Full text of the report's Methods](https://www.ncbi.nlm.nih.gov/books/NBK262311/)
- S03-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S03-M2 · Full text of the report's Results](https://www.ncbi.nlm.nih.gov/books/NBK262313/)
- S04-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S04-M2 · Full protocol text](https://link.springer.com/article/10.1186/1471-244X-7-65)
- S05-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S05-M2 · Official NIH ODS web document](https://ods.od.nih.gov/factsheets/Folate-HealthProfessional/)
- S06-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S06-M2 · Original report table](https://www.ncbi.nlm.nih.gov/books/NBK262313/table/table13/?report=objectonly)
- S07-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S07-M2 · Full analysis plan text](https://www.ncbi.nlm.nih.gov/books/NBK262296/)
- S08-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S08-M2 · Full report appendix text](https://www.ncbi.nlm.nih.gov/books/NBK262312/)
- S09-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S09-M2 · ISRCTN metadata](https://www.isrctn.com/ISRCTN37558856)
- S10-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S10-M2 · Full text of the detailed results appendix](https://www.ncbi.nlm.nih.gov/books/NBK262303/)
- S11-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S11-M2 · Official manufacturer label](https://www.medicines.org.uk/emc/product/14604/smpc)
- S12-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S12-M2 · Original report table](https://www.ncbi.nlm.nih.gov/books/NBK262313/table/table14/?report=objectonly)
- S13-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S13-M2 · Abstract/official index](https://pubmed.ncbi.nlm.nih.gov/10967371/)
- S14-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S14-M2 · Metadata candidate](https://pubmed.ncbi.nlm.nih.gov/18716414/)
- S15-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S15-M2 · Uncorrected author proof of the original article](https://www.researchgate.net/publication/306347545_Acid_Folic_Supplementation_in_Major_Depressive_Disorder_Treatment_A_Double-Blind_Randomized_Clinical_Trial)
- S16-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S16-M2 · Review abstract](https://www.emerald.com/nfs/article/53/3/521/453862/The-effects-of-folic-acid-supplementation-on)
- S17-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S17-M2 · Full text of the pediatric original study](https://brieflands.com/journals/jcp/articles/162489)
- S18-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S18-M2 · Full protocol correction text](https://link.springer.com/article/10.1186/1471-244X-9-14)
- S19-M1 · Preserved stored-summary bibliographic information, linked summaries and historical access: No external link · retained summary
- [S19-M2 · Review abstract/search metadata](https://www.sciencedirect.com/science/article/pii/S0965229921001114)