Ordinary thiamine (vitamin B1); the question targets an oral single ingredient, with actual salt/product unverified × Reduction of 12-week HDRS total when added at SSRI initiation in adult MDD inpatients

Required evidence, verification status and materials actually read
Item	AVerification status	BReading scope	CReview item	DCurrent findings	EWhy unverified	FRequired materials and conditions	GFurther research
1	Met	Full text not readStored summary read	N01 · Claim and independent-question boundariesWhich adults, adjunctive treatment and 12-week outcome does this table address?Required	The original claim concerns treatment added at SSRI initiation in adult MDD inpatients and reduction of HDRS total at 12 weeks. The question is fixed to single-ingredient oral ordinary thiamine versus placebo, with subclaims=[]. Diagnosis, response/remission, relapse and suicide prevention are separate questions. S01 S03	The identity and exclusion boundaries of the independent question were linked to the retained original text. Meeting this condition does not mean that candidate-trial eligibility or treatment efficacy has been verified.	Original claim, facets and subclaims, and clinical questions excluded from the same page	Future research should prespecify HDRS total at 12 weeks and response, remission and suicide-related events as separate outcomes.
2	Met	Full text not readStored summary read	N02 · Current source and identityDo the current published source and this summary have the same identity?Required	The current ID is 3120 and the canonical path is /verdicts/mood/oral-thiamine-mdd-hdrs-12-weeks/. The current base original and inventory snapshot have the same 99,833 bytes and SHA. The needs_id_assignment/server-not-deployed wording in the KO summary_note is preserved as historical content and is not used as a deployment record for this summary conversion.	The current record's identity was distinguished from historical status wording. The original site's current response or public availability was not checked in this review.	Current loader-selected path, original SHA, ID/canonical URL, version and actual publication receipt	Record identities, dates and publication events separately by version, without creating new research results.
3	Met	Full text not readStored summary read	N03 · Study and report familiesWere the 21 reports kept from being counted repeatedly as unique trials?Required	S01/S02/S10–S14 form one family, TF-GHALEIHA-2016. S03 is a different controlled study, TF-WATTANATHAMRAK-2019, but replication of the same comparison is unverified. S04 is TF-MOTI-2024 in PCOS/IVF and provides indirect material. The retained family table's 3 controlled candidates and 0 families eligible for efficacy grading are eligibility records from that time. S01 S02 S03 S04 S10 S11 S12 S13 S14	Original report counts were distinguished from family roles, and no new count of directly eligible trials was established.	Original reports, figures, registrations, repositories, a family linkage table and unique-participant identity information	Future syntheses should first create study-level outcome tables after comparing registration numbers, authors, recruitment institutions, periods and participant overlap.
4	Met	Full text not readStored summary read	N04 · Actual scope of readingWhat was read in this review, and what was the historical level of access?Required	The current stored summary for 3120 was read in this review, including its Korean text, structured evidence, 21 bibliographic records and historical access records. The records dated 2026-09-16 of partial access to S01, visual inspection of the S02 figure and access to the S04 HTML full text are history recorded in the original. External papers, figures, registrations and official materials were not accessed, acquired or read in this review; access was not attempted. S01 S02 S04 S07 S09 S13 S15	Current actual actions were separated from historical records. Past PDF failures were not labeled as failures in this review, and historical full-text access to S04 was not labeled as current full-text reading.	Retained source/citation, access_level/date/locator, and actual files and reading events	When materials are acquired in the future, record the sections and tables actually read, file SHA and access events separately for each material type.
5	Partly verified	Full text not readStored summary read	N05 · Adult inpatient populationWhat are the candidate trial's population and baseline depression severity?Required	S01 reports a total of 51 participants, all MDD inpatients, baseline HDRS >24, mean age 35.2 years and 53% women. Age bounds, sex distribution by arm and baseline scale means/variances are unverified. Do not turn means and percentages into identical characteristics of every adult. S01	The population label, setting and baseline threshold are preserved, but exact eligibility and balance between arms remain unknown.	Original eligibility criteria, age range, baseline tables by arm, inpatient setting and recruitment flow	Specify baseline severity and representativeness in the same inpatient MDD population and present balance between arms.
6	Unverified	Full text not readStored summary read	N06 · Diagnosis and special populationsHave the MDD diagnostic framework, episode, treatment resistance and major exclusions been verified?Required	The retained manuscript confirms that MDD was reported, but the DSM/ICD version, diagnostic interview, episode status and treatment resistance are unverified. Exclusions and population composition regarding pregnancy/lactation, bipolar disorder, dementia/delirium, renal/liver status and comorbidities are also unverified. S01	The retained location read explicitly describes a gap in access to the original methods, so the exact diagnostic and exclusion conditions cannot be said to have been met.	Original diagnostic interview/DSM/ICD, episode/resistance definitions, exclusion criteria and baseline tables by population	Prespecify diagnostic procedures and episode/resistance criteria, and analyze applicability to special populations separately.
7	Conflicting information	Full text not readStored summary read	N07 · Deficiency, nutrition and alcohol useWas the candidate population thiamine-deficient or thiamine-replete?Required	S08's secondary (2nd) narrative calls the S01 population thiamine-deficient. Neither deficiency nor repletion was verified in the accessed S01 original, and baseline markers, thresholds, nutrition, diet and alcohol use are unverified. This conflict concerns the level of verification of deficiency status, not conflicting numerical clinical effects. S01 S08	The secondary (2nd) assertion and the original-report verification gap were preserved, without arbitrarily adopting either status.	Original baseline thiamine tests/deficiency criteria, dietary/nutritional/alcohol-use records and the review's original citation location	Future research should measure baseline status and assess interaction by prespecified deficiency status.
8	Unverified	Full text not readStored summary read	N08 · Molecule and single-ingredient compositionCan ordinary thiamine be distinguished from the actual salt and single-ingredient product?Required	The question targets single-ingredient oral ordinary thiamine. S01 reports the ingredient name thiamine (vitamin B1), but the exact salt, product, source and single-ingredient composition could not be verified. Benfotiamine, fursultiamine/TTFD, sulbutiamine, allithiamine, phosphorylated derivatives, combinations and injections are not direct evidence for the same question. S01	The general ingredient name alone is not treated as meeting eligibility requirements for the exact molecule/salt and single-ingredient composition.	Original trial product name, salt, active/added ingredients, manufacturing/formulation documents and administration records	Future direct comparisons should predefine and disclose the salt, product and composition of ordinary thiamine and the control product.
9	Unverified	Full text not readStored summary read	N09 · Actual route of administrationWas oral administration verified in the candidate MDD original report?Required	The question concerns oral administration, but the actual oral route was not verified in retained S01. Oral-administration information in the S07 review is not a route verified in the original report, and oral-administration information in the PCOS study S04 does not establish the route in S01 either. S01 S04 S07	The current eligibility gap in the original report was maintained. An unverified oral route does not mean that a parenteral route has been verified.	Actual route of administration, product and prescription records in the candidate original report's Methods	Future comparisons should document oral administration and actual intake, and separate these from injection and derivative data.
10	Unverified	Full text not readStored summary read	N10 · Dose, intake and adherenceHave the actual prescribed dose and administration regimen in S01 been verified?Required	Daily dose, per-administration dose, frequency, meal timing, loading/maintenance and adherence are unverified from the S01 original. The S07 information on oral 300 mg/day and fluoxetine 20 mg/day is not verified in the original report. For S03 and S04, 300 mg/day is also retained only as the nominal dose in each separate study. S01 S03 S04 S07	Although the 12-week endpoint is preserved, the actual daily prescription and exposure cannot be established.	S01 original prescription/product, daily and per-administration doses, frequency, meal timing, adherence records and handling of deviations	Future research should record actual intake, adherence and changes, and assess dose optimization in separate direct comparisons.
11	Partly verified	Full text not readStored summary read	N11 · Incremental effect at SSRI initiationIs this an adjunctive-treatment comparison at initiation of a new SSRI?Required	S01 reports a comparison of added thiamine or placebo while a standardized SSRI was newly initiated and maintained at therapeutic levels. It is not verified as a trial of thiamine monotherapy or augmentation of a stable pre-existing antidepressant dose. The SSRI name, dose and actual treatment by arm are unverified. S01	The adjunctive comparison's structure is preserved, but the specific identity of the shared treatment and its implementation by arm remain unknown.	Records of SSRI initiation timing, drug/dose, maintenance at therapeutic levels and added ingredient/placebo in both arms	Compare only the added ingredient on top of the same SSRI initiation, and design augmentation of a stable dose as a separate stratum.
12	Partly verified	Full text not readStored summary read	N12 · Placebo and other treatmentsHow were placebo, psychotherapy, other drugs and rescue treatment compared?Required	The control treatment in S01 is placebo, and the shared treatment is a newly initiated standardized SSRI. Placebo composition/appearance and actual treatment by arm, including psychotherapy, other psychotropics and rescue treatment, are unverified because of the gap in access to the original methods. S01	A placebo comparison was reported, but specific verification of what was the same or different apart from the added ingredient remains outstanding.	Placebo composition/appearance, psychotherapy/other psychotropics by arm, rescue treatment and records of changes	Adjunctive comparisons should disclose concomitant treatment and rescue use by arm and handle these in a predefined analysis.
13	Partly verified	Full text not readStored summary read	N13 · Allocation and designHave the implementation of randomization, allocation concealment and baseline balance been verified?Required	S01 is reported as a randomized thiamine-versus-placebo trial. The full methods section needed to verify sequence generation, concealment, unit of allocation and baseline balance by arm was not acquired within the current retained scope. S01	The comparative design is named, but evidence verifying actual allocation is insufficient.	Sequence generation, concealment, unit of allocation, baseline tables by arm and participant flow in the original Methods	Future trials should predefine allocation and concealment and disclose participant flow and baseline data by arm.
14	Unverified	Full text not readStored summary read	N14 · Blinding and assessment implementationHave assessor blinding and the actual HDRS measurement procedure been verified?Required	The HDRS in S01 is retained as a clinician/expert assessment. The bibliographic URL for S13 contains double_blind, but actual implementation of assessor blinding, scale training, agreement and any unblinding are unverified. Do not conclude that measurement bias is absent from the title or URL alone. S01 S13	The assessor type is recorded, but implementation of blinding and measurement consistency for the selected outcome remain unknown.	Original blinding, unblinding/verification, assessor training/agreement and measurement procedures by arm	Define independent assessment procedures using the same scale and time point, and record measurement differences between arms.
15	Partly verified	Full text not readStored summary read	N15 · 51 participants and analysis denominatorsWere the total of 51 participants and arm-level randomized, analyzed and completed counts distinguished?Required	In S01, the 51 participants are the total reported in the abstract. Arm-level randomized, analyzed and completed n and the number analyzed for HDRS at 12 weeks are unverified. Do not divide the 51 participants into two arms or use them as the safety denominator at 12 weeks. S01	The reported total is preserved, but the arm-level analysis denominators for the outcome of interest cannot be verified.	Original CONSORT and allocation/completion/withdrawal tables, the HDRS analysis population at 12 weeks and n by arm	Present the full flow by arm and the analysis n for each outcome, and examine the link between outcomes and attrition.
16	Unverified	Full text not readStored summary read	N16 · Missing data and analysis populationsHave attrition, missing data, ITT/PP and the effect analyzed been verified?Required	The analysis n by arm, reasons for attrition/missingness, missing-data handling and ITT/PP analysis populations in S01 are unverified. Whether the analysis was adjusted or unadjusted, and whether the effect of interest concerns final values or changes, are also unresolved. S01	Even the same between-group difference can be interpreted differently depending on the analysis population and missing-data handling, so no analysis label was invented to fill the gap.	Original missingness/attrition by arm at 12 weeks, analysis n, SAP, ITT/PP and sensitivity results	Prespecify the distinction between the effect of assignment and the effect of adherence, and present missing-data assumptions and sensitivity analyses.
17	Partly verified	Full text not readStored summary read	N17 · HDRS version and total scoreWhat are the actual HDRS version, items, range and direction?Required	The selected outcome is the continuous depression total on the clinician-rated Hamilton Depression Rating Scale, with lower scores indicating improvement. The actual version, item count and score range are unverified. The GHQ-28 depression subscale with 7 items and individual sleep/appetite items are not the same total score. S01	The scale name, construct and direction are preserved, but the version and range needed to interpret the actual score units remain unknown.	Original scale version, items/range, scoring instructions and assessment result tables by arm	Future trials should prespecify the exact scale, version, total score and assessor procedures.
18	Met	Full text not readStored summary read	N18 · 12 weeks and intermediate visitsWere the endpoint result at 12 weeks and signals at 3 and 6 weeks distinguished?Required	The visits in retained S01 are at 0, 3, 6 and 12 weeks, and this table's principal time point is 12 weeks. The choice of principal time point is recorded as a review choice in the current manuscript, not as verified trial preregistration. The favorable signal at 6 weeks is preserved but does not replace the result at 12 weeks. S01 S02	The roles of the time points were distinguished. Meeting this condition is separate from verification of the effect/prescribed dose at 12 weeks.	Original visit schedule, principal time point in the registration/protocol, result tables by time and actual prescribed exposure	Future comparisons should prespecify the principal time point and repeated-measurement plan and disclose all intermediate results.
19	Unverified	Full text not readStored summary read	N19 · Between-group effects, within-group changes and the original effectHas the exact estimate of the between-group effect at 12 weeks been verified?Required	S01/S02 report a favorable signal at 6 weeks, a nonsignificant between-group mean difference at 12 weeks and improvement in both groups. The exact between-group difference at 12 weeks, arm means, changes and adjustment status are unverified. Within-group improvement does not establish a causal effect of thiamine. S01 S02	The numerical estimate of interest and its analysis definition are missing, so the magnitude of the incremental effect cannot be established.	Original arm-level total scores/changes at 12 weeks, adjusted/unadjusted between-group estimates and analysis definitions	Directly estimate the predefined incremental effect at 12 weeks between groups, and present within-group changes separately.
20	Unverified	Full text not readStored summary read	N20 · Variability, CI and graphCan the SD/SE and confidence interval at 12 weeks be interpreted?Required	The retained record explicitly states that Figure 1 was inspected only qualitatively and that graph digitization, arm means, SD/SE and effect/CI were not reconstructed. The confidence interval at 12 weeks and type of variability measure are unverified. The figure's S/S/L/M labels are not a verified MCID either. S01 S02	Without exact variability and intervals, it cannot be determined whether important benefit or harm has been excluded.	Original numerical tables at 12 weeks, SD/SE definitions, analysis n by arm, CI and model output	Report uncertainty intervals together with the variability/denominators used, and disclose the source data for the figure.
21	Unverified	Full text not readStored summary read	N21 · Exact P value and importanceHave the P value at 12 weeks and the criterion for a clinically important difference been verified?Required	The exact P value at 12 weeks and MCID are retained as unverified (null). Only a statement of nonsignificance is available; without an interval, clinically important benefit cannot be excluded. The response/remission P values in S07 are not P values for total-score severity. S01 S02 S07	The effect, interval and scale criterion needed to judge importance are all missing.	Original exact P value, CI and effect at 12 weeks, and evidence for an important difference appropriate to the HDRS version/population	Future research should justify the important difference with evidence and disclose the range of effects allowed by the CI.
22	Met	Full text not readStored summary read	N22 · Preserving nonsignificant and positive findingsWere favorable findings and statements of no difference both retained without exaggeration?Required	The favorable signal at 6 weeks, the nonsignificant between-group result at 12 weeks and within-group improvement in both arms were all preserved. The different controlled comparison in S03 and the indirect positive result in S04 are also retained separately. Nonsignificance does not mean an effect of 0, equivalence, noninferiority or exclusion of important benefit. S01 S02 S03 S04	The direction and applicability of the current reports were explained, without arbitrarily selecting only favorable findings.	All results by time, corrections, analysis definitions and exact original numerical tables	Future research should report all favorable/nonsignificant/unfavorable results according to prespecified time points and analyses.
23	Partly verified	Full text not readStored summary read	N23 · Funding, product provision and COITo what level of disclosure has independence been verified?Supplementary; required for claims of independence	S01 declares no external funding and no author conflicts, but internal support and product provision are unverified. Funding, product provision and COI in S03 are unverified. For S04, the IUMS grant 94-04-28-26840, support from the Sarem research center and the declaration of no COI are preserved. S01 S03 S04	Declarations were distinguished from actual verification gaps, and funding from another family was not assigned to the candidate trial.	Original funding, internal support, product provision, employee/patent relationships, declared COI and declarations in each report	Future research should disclose public, institutional and product support and author relationships item by item, and distinguish them from independent replication.
24	Unverified	Full text not readStored summary read	N24 · Registration, selective reporting and integrityWere the original registration, outcome hierarchy, corrections and scope of unverified information compared?Required	The registration identifier, timing and prespecified outcome hierarchy for S01 were not verified. A record states that no correction, retraction or expression of concern was found on the publisher page opened or in targeted searches at that time, but this is not an exhaustive integrity guarantee. There is no new registration or publication-status lookup in this review. S01	Without the original registration/protocol needed to compare plans and results, selective reporting is not determined.	Original registration ID, prospective timing, protocol/SAP, outcome changes, correction/retraction notices and actual access events	Future research should compare plans and results by version and record reasons for changes and unreported outcomes.
25	Partly verified	Full text not readStored summary read	N25 · Secondary (2nd) formulation and dose leadsWhat is the evidential level of the oral-administration, dose and SSRI information in the S07 review?Required	S07 provides leads of oral 300 mg/day and fluoxetine 20 mg/day while recording that the formulation was not specified. Only selected portions of PDF pages 4–6 were inspected at that time; this is not a record of reading all other pages. Do not promote this information to verified route, product, dose or version values from the S01 original. S01 S07	Tracing leads exist, but the original Methods needed for direct verification remain outstanding.	Relevant S07 tables/sections and original citations, and the S01 product, route, daily dose and scale Methods	Future material reviews should link each secondary (2nd) claim to a precise location in the original report and compare agreement.
26	Partly verified	Full text not readStored summary read	N26 · A separate controlled MDD signalIn S03, do the 50 participants, 300 mg/day and P values represent the same 12-week effect?Required	S03 studied 50 participants, adults with MDD aged 18–60 years in inpatient/outpatient care, adding thiamine 300 mg/day to standard treatment. It reports allocation by drawing lots and improvement in HRSD; the P-value sequence is .004, .031, .035, but the exact visits and effects of interest for the later values are unverified. Oral route, salt, single-ingredient composition, duration, scale version, n by arm and the placebo-controlled effect at 12 weeks were not verified. S03	The other controlled signal is preserved, but the dimensions needed to verify direct comparability with the selected question are insufficient.	Full S03 Methods/results, n by arm, route/salt, principal time point, analysis attribution of each P value and SSRI/placebo information	Future syntheses should compare studies after verifying exact PICOT agreement and visit-specific estimates, without automatically creating an average.
27	Conflicting information	Full text not readStored summary read	N27 · S03 publication-identity conflictAre the date, issue and citation year for S03 consistent?Supplementary; required for bibliographic identity	The publication-date field for S03 is 2019-09-24, the volume/issue is Vol.36 No.1 (2020), and the citation year is 2019. The original record preserves all these fields without inventing a resolution. The two PDF links could not be accessed at that time, and the original PDF was not read in this review either. S03	Because publication details differ within the same report, Conflicting information is retained only for the bibliographic scope.	The journal's canonical metadata, corrections to date/issue/citation information and publication events by version	Record changes in bibliographic identity together with the original values; this is not an efficacy experiment.
28	Met	Full text not readStored summary read	N28 · Indirect PCOS/IVF populationIs S04 a direct result for MDD inpatients at SSRI initiation?Required	S04 is a trial in women with PCOS, infertility and IVF, aged 18–40 years, using oral thiamine for 4 weeks at 300 mg/day. It excluded diagnosed psychological disorders and did not require a depression entry threshold. The GHQ-28 depression subscale with 7 items and a 0–21 range differs from HDRS total at 12 weeks in adult MDD inpatients. S04	The indirect role and reasons for exclusion were clearly linked. Oral-administration information and the product manufacturer in S04 do not establish the identity of S01.	Actual S04 eligibility, scale, duration, product, depression-subgroup data and original plan	To answer the MDD question, separately design a direct comparison in the same MDD/SSRI-initiation setting with HDRS at 12 weeks.
29	Partly verified	Full text not readStored summary read	N29 · PCOS denominators, numerical values and SDWere S04 denominators and positive results correctly attributed?Supplementary; required for describing indirect results	S04 randomized 35/35 participants, with 32/32 participants analyzed/completed. GHQ depression-subscale means (SDs) were 16.46 (SD 4.98) versus 15.84 (SD 5.53) at baseline and, at 4 weeks, 10.96 (SD 4.93) versus 16.25 (SD 4.87), with reported between-group P < .001. Baseline depression-domain scores ≥6 are retained as 30/28 participants, with a total of 6 participants below the threshold; the eligible depression-subgroup effect is unverified (null). S04	The original numerical values, SDs and denominators are linked, but they are not the selected MDD effect or a direct comparison in the relevant subgroup.	S04 original Tables 2–3, CONSORT, analysis n, actual depression-subgroup results and missing-data handling	Indirect studies should separate overall/subgroup populations and randomized/completed counts, and present scale-specific estimates and intervals.
30	Conflicting information	Full text not readStored summary read	N30 · S04 reporting and plan gapsHow were S04 thresholds, percentages, outcome hierarchy and registration retained?Supplementary; required for accuracy of indirect material	In S04, the abnormal threshold >6 in the text differs from Table 2's ≥6 notation. The retained note records that the printed percentages for 6/32 and 26/32 appear arithmetically inconsistent, but the original values are not silently corrected. Differences in wording about the general-health/infertility outcome hierarchy between the abstract and Methods also remain, and the original IRCT201510266917N3 registration and prospective timing are unverified. S04	The current conflicts are limited to the thresholds, reporting and hierarchy recorded in the retained manuscript. The table was not recalculated and the original registration was not checked in this review.	S04 table/text corrections, protocol, original IRCT registration/versions, CONSORT and raw data	Future material comparisons should disclose the original definitions of thresholds, denominators and hierarchy and reasons for changes, and check sensitivity.
31	Met	Full text not readStored summary read	N31 · Healthy participants, observational studies and deficiency case reportsDo the results in S16–18 directly answer the selected adjunctive-efficacy question?Supplementary; required to prevent extensions	S16 is a mood/cognition study in healthy young women who were not selected for depression. S17 concerns an observational association between thiamine status and depressive symptoms in older Chinese adults, not a controlled supplementation effect. S18 is a single parenteral→oral case of malnutrition and deficiency neuropathy due to depression; no result is borrowed as controlled antidepressant efficacy. S16 S17 S18	Each material's exclusion/context role was linked to its original retained location. This is not a declaration that research is absent worldwide or that no direct effect exists.	Population, design, exposure/route, endpoint and comparator in each original report, and directness to the selected MDD question	Design questions about healthy people, correction of deficiency and observational associations separately, and distinguish them from the exact MDD adjunctive-treatment trial.
32	Met	Full text not readStored summary read	N32 · Combinations, fatigue and other endpointsAre S19–21 HDRS results for ordinary thiamine alone?Supplementary; required to prevent extensions	S19 studies a B1+B2 combination and anxiety, stress and sleep. S20 NUTRIMUM concerns pregnancy and multiple micronutrients; S21 concerns fatigue-focused material in primary biliary cholangitis. The retained record did not grade these as the selected MDD HDRS result or extract numerical values. S19 S20 S21	Differences in ingredients, populations and endpoints were preserved, without pooling them as a direct single-ingredient B1 effect.	Ingredients, populations, screening scales, endpoints/times and ingredient-specific direct comparisons in each original report	If the contribution of B1 alone is needed, design a separate direct comparison with the other ingredients/treatments matched.
33	Unverified	Full text not readStored summary read	N33 · Youth registration candidateWere controlled adult results verified in NCT07793071?Supplementary; required for interpreting registration	S15 preserves only the indexed title of NCT07793071 concerning depressive symptoms in youth. At that time, the official registration page was an app shell, the API failed and controlled adult results were not verified. Details from 3rd-party mirrors were not used as verified original data. Registration access was not attempted in this review. S15	The existence of a candidate identifier/title is not verification of the original registration or adult results.	The official registration's identity, population, ingredient, route, design, time points, results, versions and actual acquisition events	After actually obtaining the registration content, compare population/design eligibility and separately track unpublished results as well.
34	Unverified	Full text not readStored summary read	N34 · Harm statements in the candidate trialDoes S01's statement of no adverse effects mean 0 events per arm?Supplementary; required for safety statements	The S01 abstract states that there were no adverse effects in either group. Arm-level event/exposure counts, serious events, AE discontinuations and ascertainment methods are null or unverified. Do not turn this sentence into 0 events per arm or an incidence of 0. S01	The reported wording is preserved, but there are no data to verify frequency or complete ascertainment.	S01 event/person counts by arm, exposure/follow-up n, duration, AE definitions/collection and reasons for discontinuation	Future trials should report systematic harm ascertainment and arm-level people/events, exposure and discontinuations separately from efficacy.
35	Unverified	Full text not readStored summary read	N35 · Psychiatric events and long-term harmsWere suicide-related events, manic switch, worsening and long-term safety assessed?Supplementary; required for those safety extensions	Suicide-related events, mania/hypomania and psychiatric worsening in S01 are unverified (null). Separate assessment, temporal relationships after exposure, discontinuation/re-exposure and long-term harms are unverified. Do not extend the candidate result at 12 weeks to long-term safety, rare events or safety in all populations, including pregnancy/lactation. S01	Symptom total scores and a general AE abstract alone cannot establish ascertainment, frequency or causality for separate events.	Psychiatric-event definitions, systematic ascertainment, arm-level n/duration, worsening/mania/suicide and long-term follow-up data	Collect psychiatric events using separate prespecified definitions and appropriate follow-up/surveillance, and separate them from efficacy total scores.
36	Partly verified	Full text not readStored summary read	N36 · Harms in other candidatesHow are the nausea/adverse-effect statements in S03 and S04 attributed?Supplementary; required for safety interpretation of other materials	S03 preserves abstract statements of nonsevere nausea and no serious adverse effects, but arm-level events/exposure n are unverified. S04 records a statement of no important adverse effects in the B1 arm and an exclusion criterion for B1 allergy. Arm-level events/exposure and independent re-extraction of CONSORT discontinuation reasons are unverified. S03 S04	The statements are attributed, but safety frequencies in other populations/products cannot be directly transferred to the candidate.	S03/S04 original events/people by arm, ascertainment, exposure/duration, discontinuations, products and CONSORT data	Safety syntheses should compare identities, populations, products and ascertainment, then separate unique events from frequencies.
37	Partly verified	Full text not readStored summary read	N37 · Separate oral-product SmPCDoes the Flamingo product information establish the candidate product and its harms?Supplementary; required for interpreting product safety	S06 is the SmPC for Flamingo's oral Thiamine Hydrochloride 100 mg Tablets. The frequencies of gastrointestinal adverse effects and hypersensitivity reactions are unknown, and hypersensitivity reports are retained as predominantly parenteral. The emc update of 2025-09-08 and text revision date of 2025-02-18 are historical version information, not identification of either MDD trial product. S06	The separate product's general harm context was linked, but the candidate product and incidence cannot be established.	Documentation linking the candidate trial product to this product, and the actual SmPC/version and original harm data for the oral product	Product-specific harm reviews should directly compare matched routes, products, exposed populations and event-occurrence/reporting systems.
38	Met	Full text not readStored summary read	N38 · Nutritional upper limit and general contextWhat do an unset UL and general nutritional/medication context mean?Supplementary; required for general safety extensions	The S05 ODS summary records that no UL has been established, but the FNB cautions that excessive intake may cause adverse effects. The existence of sections on deficiency, at-risk groups and medication interactions provides general context, not a source of depression efficacy evidence. An unset UL does not mean unlimited safety. S05	The safety/efficacy interpretation boundaries of the general material were checked. This does not mean that a specific concomitant-treatment risk in MDD or an individual dose has been verified.	Original official nutrition/excess-intake, vulnerable-population and interaction documents, and direct harm data matching the population and exposure	Assess concomitant-medication/excess-intake risks in the selected MDD population using separate actual exposure data.
39	Unverified	Full text not readStored summary read	N39 · Extensions to persistence, dose and remissionDoes the total score at 12 weeks also answer questions about optimal dose, maintenance, response/remission and prevention?Supplementary; required for those extension claims	The current candidate material does not establish optimal dose, long-term maintenance, persistence after discontinuation, effects in a deficient subgroup or augmentation of a stable SSRI. Response/remission, relapse and suicide prevention are also not the total-score outcome at 12 weeks in this table. Direct estimates for these extensions are unverified within the retained scope. S01 S07	Direct material needed for the extensions is unavailable, but this is not treated as failure of a mandatory event trial for the narrow total-score question.	Direct dose comparisons, follow-up after discontinuation, prespecified subgroups/interactions and response/remission/relapse/suicide-event data	Address each extension question in a separate study with a prespecified direct comparison, definition, time point and denominator.
40	Met	Full text not readStored summary read	N40 · Additional material and new researchWere original materials needed to fill current gaps distinguished from methods for future trials?Required	The first need is to obtain S01 original Methods, product/route, scale version, arm-level numerical values/CI at 12 weeks, analysis/flow, registration and harm data. Separately, methods are proposed for future research directly comparing single-ingredient oral ordinary thiamine in the same MDD/SSRI-initiation setting. No fixed sample size, dose, duration extension or efficacy grade is created. S01	The additional materials in each row are gaps in the record, while future research is distinguished as a methodological proposal that has not yet been carried out.	S01 original Methods/product, HDRS version, arm-level tables/CI at 12 weeks, flow, registration/SAP, safety and corrections	In a direct randomized placebo comparison that clearly defines the product/route, MDD criteria and SSRI initiation, prespecify HDRS at 12 weeks, arm-level n/missing data, CI, an important difference and harms, and disclose all results.
41	Met	Full text not readStored summary read	N41 · Reference dates, original verdict and exceptionWere the existing ?/null and calculator exception kept from being inherited as the new table's status?Required	The original evidence cutoff is 2026-09-16, and this summary's review date is 2026-10-05. Preserve the original grade ?, uncalculated score (null), claim type B, six unscored downstream axes (null) and the no-human-study flag false. Preserve the original explanation that the calculator returned validation errors and a pre-validation B at that time, but B was not adopted. Original document quality A and single-assistant self-review are not independent clinical certification.	Original values and the historical record of the exception were preserved, without newly inheriting statuses/scores, calculating or reassessing.	Original calculation inputs/errors, current-value exception, verification records and authoritative records of new evidence/condition-change events	If eligibility or the representation of rules is updated, separate the event, conditions and adjudication procedure, and preserve the entire previous original text.
42	Unverified	Full text not readStored summary read	N42 · Repeated visits and statistical modelHow were analyses across multiple visits/scales and the adjusted model prespecified?Required	The adjusted/unadjusted effect of interest and multiplicity-handling method in S01 are unverified. Visits at 0, 3, 6 and 12 weeks are preserved, but the repeated-measures model, covariates, primary/secondary endpoints and plans for multiple comparisons across time points were not verified. Do not reconstruct a statistical model from the difference between nonsignificance at 12 weeks and a favorable result at 6 weeks. S01 S02	Without material comparing the analysis plan and actual model, how the selected estimate was handled cannot be established.	Original protocol/SAP, analysis model, covariates, repeated-visit/multiplicity handling, effect of interest at 12 weeks and reasons for changes	Future direct comparisons should prespecify the effect of interest at 12 weeks, primary endpoint, visit-specific comparisons, covariates, multiplicity handling and sensitivity to missing data.

No items match your search or filter.

42 itemsWhat the statuses mean

Assessment scope

Stored-material review date: 2026-10-05
Original evidence cut-off: 2026-09-16
Scope actually checked: Stored Korean and English manuscripts and case-specific criteria and application records. Full papers, supplements, registrations, plans and original data were not read.
What the dates mean: This is the date the stored materials were checked, not the date of an up-to-date clinical literature search.
Population: Adult inpatients reported to have major depressive disorder (MDD) and baseline HDRS >24; diagnostic framework, age bounds and thiamine status unverified
Ingredient, formulation and route: Ordinary thiamine (vitamin B1); the question targets an oral single ingredient, with actual salt/product unverified · A single ingredient is the question target; the ordinary-thiamine salt and formulation are unverified in the original MDD report · The question targets oral administration; oral administration could not be verified in the accessed original MDD report
Dose: Unverified from the original MDD report; no thiamine-equivalent dose calculated
Comparator: Placebo + a newly initiated standardized SSRI in the candidate trial; exact drug, dose and other treatments unverified
Measured outcome: Clinician-rated HDRS total at 12 weeks; version, item count and score range unverified
Duration: 12 weeks (candidate trial endpoint; actual daily regimen unverified)
Questions outside this scope: This page does not assess diagnosis, response/remission, relapse or suicide-related events. The need to correct deficiency is also separate from the efficacy of additional supplementation in thiamine-replete adults with depression; the candidate population is not presumed deficient or replete.; Proceed as one independent table. Fix the question to adult MDD inpatients, addition at initiation of a new SSRI, and clinician-rated HDRS total at 12 weeks. Eligibility regarding actual oral administration, salt, single-ingredient composition and dose remains unverified in the original report. subclaims=[]; improvement at 6 weeks/within groups, remission/response, suicide prevention, deficiency correction, mood in healthy people, PCOS GHQ, observational studies/case reports, and combinations/fatigue are not pooled into the principal effect.

Evidence and sources

The links below lead to the sources. A link alone does not mean the original source was read or its latest publication status verified.

S01 · Ghaleiha et al. Adjuvant thiamine improved standard treatment in patients with major depressive disorder (2016)

Source

Material type: Candidate original report and partial publisher record
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Reports 51 participants (MDD inpatients), HDRS >24 and thiamine versus placebo at initiation of a new SSRI. The signal at 6 weeks is favorable, but the between-group mean difference at 12 weeks is nonsignificant, and the exact effect, CI, P value and n by arm are unverified. Statements of within-group improvement and no adverse effects are not, respectively, an incremental causal effect and an incidence of 0.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Paper abstract and partial publisher web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S02 · Ghaleiha 2016 original Figure 1

Source

Material type: Historical record of visual inspection of the original S01 figure
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A record of qualitative inspection at that time of HDRS trajectories at baseline and 3, 6 and 12 weeks. The favorable-at-6-weeks/nonsignificant-at-12-weeks statements are preserved, but the figure was not digitized and means, SD/SE or CIs were not reconstructed. The S/S/L/M labels are not a verified MCID.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Original figure Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S03 · Wattanathamrak. The Effectiveness of Adjunctive Thiamine (vitamin B1) with Standard Treatment in Patients with Major Depressive Disorder

Source

Material type: Abstract/bibliographic record of a separate original controlled MDD study
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: In 50 participants with MDD receiving inpatient/outpatient care, aged 18–60 years, 300 mg/day is added to standard treatment. The HRSD improvement P-value sequence .004/.031/.035 is preserved, but attribution to visits/estimands and the effect at 12 weeks, CI and n by arm are unverified. Allocation by drawing lots, the nausea statement and the publication-date/issue conflict are retained separately.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Paper abstract and journal bibliographic web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S04 · Moti et al. The Effects of Thiamine Supplementation on General Health and Infertility Treatment Outcomes in Women with PCOS (2024)

Source

Material type: Historical record of extraction from the HTML full text of the PCOS/IVF original report
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A PCOS/IVF study with 35/35 participants randomized and 32/32 participants analyzed, using oral treatment for 4 weeks at 300 mg/day. GHQ depression means (SDs) are 16.46(4.98)/15.84(5.53) at baseline and, at 4 weeks, 10.96(4.93)/16.25(4.87), with reported between-group P<.001. Eligibility, scale and time point differ, and an MDD subgroup effect is unverified. Notes on the original threshold/printed percentages and outcome hierarchy, and failure to obtain the original registration, are preserved.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Original protocol, analysis-plan and registration documents Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S05 · NIH Office of Dietary Supplements: Thiamin, Health Professional Fact Sheet

Source

Material type: Official nutrition/general harm context
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Preserve the unset UL and the FNB's caution about harm from excessive intake. These do not establish unlimited safety, a personal dose for the selected MDD product or efficacy. No specific medication interactions were newly extracted in this review.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full official web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S06 · Thiamine Hydrochloride 100 mg Tablets, Flamingo Pharma (UK), SmPC

Source

Material type: SmPC for a separate oral hydrochloride product
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Preserve the unknown frequency of gastrointestinal/hypersensitivity reactions and the route limitation that hypersensitivity reports are predominantly parenteral. The emc update of 2025-09-08 and text revision of 2025-02-18 are records from that time, not verification of the original MDD product or oral incidence.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full official product web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S07 · Sousa Santos and Braga Reis: Eficacia de la suplementación con tiamina para mejorar los síntomas depresivos. Revisión basada en la evidencia (2023)

Source

Material type: Historical record of selected PDF portions of a secondary (2nd) review
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: The review records oral 300 mg/day, fluoxetine 20 mg/day and unspecified formulation. Do not promote these to route, dose or product values verified in the S01 original. Preserve the limitations of access to selected portions and the distinction between response/remission P values and severity P values.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text review paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S08 · Dhir et al. Neurological, Psychiatric, and Biochemical Aspects of Thiamine Deficiency in Children and Adults (2019)

Source

Material type: Record of an indexed excerpt from a secondary (2nd) mechanistic/deficiency review
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: The review calls the S01 population deficient, but neither deficiency nor repletion was verified in the accessed original. Without original baseline tests and thresholds, the MDD population is not established as deficient or replete.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text review paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S09 · AIHTA Project Report 182, Combination Therapy with Antidepressants and Vitamin B (2026)

Source

Material type: Record of selected searches/bibliographic mapping in an HTA report
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A historical record stating that the entire 97-page report was not read. Bibliographic mapping at that time used only term searches and PDF page 68 (reference 86), without borrowing decisive efficacy facts.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full assessment report Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S10 · PubMed record 26984349

Source

Material type: Official PubMed index/abstract record
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A record of PMID 26984349, the same DOI and the indexed abstract. Preserve the original limitation that direct opening returned an empty result and this was not a completed search of the native PubMed database.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full official index web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S11 · Basel repository record for Ghaleiha 2016

Source

Material type: Abstract/metadata record in the Basel institutional repository
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A historical record that no exposed full-text file was obtained. Do not substitute the repository's default displayed date for the publisher's publication date or add separate participants/effects.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full institutional bibliographic web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S12 · Fribourg repository record for Ghaleiha 2016

Source

Material type: Abstract/metadata record in the Fribourg institutional repository
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Access is recorded as abstract/citation metadata, and notes and the extracted statement are empty strings. Do not fill these blanks with independent results, full-text acquisition or reported absence of an effect.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full institutional bibliographic web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S13 · ResearchGate record and publisher preview, Ghaleiha 2016

Source

Material type: ResearchGate aggregation and single-page original-report preview record
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A historical record of access only to the single-page preview at p.695 and the abstract; preview download failed. Citing-article content on the same page is not treated as the Ghaleiha original. This was not reading the entire full text.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S14 · SciSpace record, Ghaleiha 2016

Source

Material type: SciSpace aggregated metadata/abstract record
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A record stating that the full PDF was not acquired and AI-generated summaries were not used. Displayed abstract/bibliographic information is not independent replication or verification of the original full text.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S15 · ClinicalTrials.gov NCT07793071: Evaluating Thiamine (Vitamin B1) as Adjunctive Therapy for Youth With Depressive Symptoms

Source

Material type: Indexed-title record of the ClinicalTrials.gov youth candidate
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Only the youth depressive symptoms title for NCT07793071 was preserved. The original page is recorded as an app shell/API failure, and controlled adult results were not verified. Current status, design and age details were not checked in this review.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full official registration web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Original protocol, analysis-plan and results documents Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S16 · Benton et al. thiamine supplementation and mood/cognition in young women

Source

Material type: Indexed abstract of the original study in healthy young women
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Recorded as a mood/cognition study in healthy young women, not an objectively depression-selected population. PMID 9122365 is preserved, and a DOI was not adopted. No numerical efficacy results for the selected MDD question were extracted.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full index web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S17 · Zhang et al. Thiamine nutritional status and depressive symptoms are inversely associated among older Chinese adults

Source

Material type: Index/reference record of an original observational study in older Chinese adults
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Design screening of an observational study on an inverse association between thiamine nutritional status and depressive symptoms. Original numerical results were not extracted and are not transferred to adjunctive efficacy with an SSRI in MDD.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full official index web document Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.
Full-text paper Material URL unverified	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S18 · Melki et al. Thiamine Deficiency Neuropathy in a Patient with Malnutrition due to Melancholic Depression (2024)

Source

Material type: Indexed excerpt of a malnutrition/deficiency-neuropathy case report
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A case of deficiency neuropathy with malnutrition due to melancholic depression. A single case treated first parenterally and then orally is not controlled antidepressant efficacy evidence for ordinary oral single-ingredient B1, and no results were borrowed.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S19 · Tao et al. Impact of Vitamin B1 and Vitamin B2 Supplementation on Anxiety, Stress, and Sleep Quality (2025)

Source

Material type: Indexed excerpt of an original B1+B2 combination study
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A study of anxiety, stress and sleep with a B1+B2 combination, differing from the fixed depression-scale/single-ingredient B1 question. No original efficacy grading or decomposition of the B1 contribution was performed.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S20 · NUTRIMUM micronutrient trial (2024)

Source

Material type: Indexed excerpt of the original NUTRIMUM study
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: Pregnancy and a multiple-micronutrient formulation differ from ordinary thiamine alone. For this question, only the screening record of exclusion without efficacy extraction is preserved.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

S21 · Bager et al. high-dose thiamine in primary biliary cholangitis (2024)

Source

Material type: Indexed excerpt of an original high-dose thiamine study in PBC
Materials actually read: Summary read: the corresponding bibliography, extracted facts/structure, Korean content and historical access in the current stored manuscript were read. External original clinical/official materials were not read in this review.
Findings recorded in the stored summary: A disease population focused on fatigue in primary biliary cholangitis. This is not an HDRS efficacy result in a depression-selected population, and the previously completed record was not researched anew or regraded.
Record level: Report-specific results/context, bibliography, exclusion and access roles in the retained record dated 2026-09-16. This is not direct verification of original clinical data in this review.
Study overlap and independent replication: None

Public availability, access and reading status by material
Material	Public availability	Access status	Actually read	Notes
Full-text paper Open material ↗	Currently unverified (no external lookup)	Access not attempted	Not read	Distinguish reading a summary in the currently retained stored material from original clinical materials/official documents and historical levels of access.

Report counts versus independent study counts

These are retained family links for comparison candidates and indirect materials. TF-GHALEIHA-2016 (S01, S02, S10, S11, S12, S13, S14): strongest candidate anchor; no fully verified directly eligible family selected · unresolved_directness; not eligible for grading yet / TF-WATTANATHAMRAK-2019 (S03): additional controlled MDD signal; no pooling or replication grade · unresolved_route_form_scale_version_background_and_time / TF-MOTI-2024 (S04): controlled adjacent evidence, excluded from the locked MDD HDRS comparison · indirect_population_and_endpoint Same DOI/repositories/figure are one cohort. Do not sum S01/S02/S10–S14 participants. No meta-analysis. The comparison-candidate family count and grading-eligible count in the record from that time are not the total number of unique trials. The confirmed number of independent efficacy studies and the total participant count are unverified. Multiple reports from the same family are not added as independent replications.

Review criteria

Version v3120-THIAMINE-MDD-HDRS12-1.1 · Criteria record date 2026-10-05. These are Chamgap's own application conditions, informed by the referenced methods. They do not constitute official assessment-tool certification or a new efficacy score.

N01 · Which adults, adjunctive treatment and 12-week outcome does this table address?

Required materials and conditions: Original claim, facets and subclaims, and clinical questions excluded from the same page
Conditions for verification: Link the original /ko/effect to the population, addition at SSRI initiation and HDRS at 12 weeks, and explain overlap between parent and subordinate questions.
Requirement and applicability: Required
Application exceptions: Response, remission, relapse and suicide-related events remain separate questions rather than mandatory outcomes for this total-score question.
Why this condition is needed: Needed to prevent the question about total depressive symptoms from shifting to general mood, remission or suicide prevention.
Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M03 §3.2.4; custom boundaries on extensions

N02 · Do the current published source and this summary have the same identity?

Required materials and conditions: Current loader-selected path, original SHA, ID/canonical URL, version and actual publication receipt
Conditions for verification: Compare the current original and snapshot bytes, ID/canonical URL and reference date.
Requirement and applicability: Required
Application exceptions: Historical publication-readiness wording retained in KO is not deleted or reinterpreted as a site-publication event in this review.
Why this condition is needed: Needed to avoid confusing historical wording about unassigned identifiers and nondeployment with the current ID/URL.
Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design

N03 · Were the 21 reports kept from being counted repeatedly as unique trials?

Required materials and conditions: Original reports, figures, registrations, repositories, a family linkage table and unique-participant identity information
Conditions for verification: Explain the roles of the S01 series as one family, S03 as a separate candidate and S04 as an indirect family.
Requirement and applicability: Required
Application exceptions: Preserve the bibliography count, but do not estimate unresolved unique-study or participant counts or turn the three candidates into directly eligible trials.
Why this condition is needed: Needed to avoid counting figures, repositories, indexes and reviews for the same DOI as independent replications.
Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

N04 · What was read in this review, and what was the historical level of access?

Required materials and conditions: Retained source/citation, access_level/date/locator, and actual files and reading events
Conditions for verification: Separate existence, access, acquisition, reading and current/historical dates for each material type.
Requirement and applicability: Required
Application exceptions: Past failures, partial reading and reports of full-text access remain historical records; they are not reused as inaccessibility or full-text reading in this review.
Why this condition is needed: Needed to avoid promoting the summary of the current stored summary to reading the original paper, figure or full text.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design

N05 · What are the candidate trial's population and baseline depression severity?

Required materials and conditions: Original eligibility criteria, age range, baseline tables by arm, inpatient setting and recruitment flow
Conditions for verification: Distinguish the reporting level and unverified details of MDD inpatient status, HDRS >24 and means/percentages.
Requirement and applicability: Required
Application exceptions: Do not infer age ranges, arm composition or the status of all participants from mean age or the percentage of women.
Why this condition is needed: Needed to avoid applying mood studies in healthy people or other diseases to adult MDD inpatients.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed

N06 · Have the MDD diagnostic framework, episode, treatment resistance and major exclusions been verified?

Required materials and conditions: Original diagnostic interview/DSM/ICD, episode/resistance definitions, exclusion criteria and baseline tables by population
Conditions for verification: Link diagnosis, episode, treatment resistance, comorbidities and the scope of exclusions to the original methods.
Requirement and applicability: Required
Application exceptions: Do not fill in unverified exclusions, such as pregnancy, bipolar disorder or renal/liver disease, as if they had already been applied.
Why this condition is needed: Needed to avoid interpreting the reported MDD label as verification of a specific diagnostic procedure, treatment resistance or exclusion of bipolar disorder.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

N07 · Was the candidate population thiamine-deficient or thiamine-replete?

Required materials and conditions: Original baseline thiamine tests/deficiency criteria, dietary/nutritional/alcohol-use records and the review's original citation location
Conditions for verification: Separate baseline thiamine markers/thresholds, meals, nutrition and alcohol use from the verification level of secondary (2nd) statements.
Requirement and applicability: Required
Application exceptions: Do not use the secondary (2nd) review's word deficient to establish the original report's testing, deficiency threshold or replete status.
Why this condition is needed: Needed to avoid combining correction of deficiency with the effect of additional supplementation in a replete state.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

N08 · Can ordinary thiamine be distinguished from the actual salt and single-ingredient product?

Required materials and conditions: Original trial product name, salt, active/added ingredients, manufacturing/formulation documents and administration records
Conditions for verification: Show separately the question target and the level of verification of the actual label, salt, source and composition in the original report.
Requirement and applicability: Required
Application exceptions: Derivatives such as benfotiamine and TTFD, combinations and injections are excluded from the question, but the candidate product is not presumed to have been a single-ingredient preparation.
Why this condition is needed: Needed to avoid assigning the identity of derivatives, combinations or other products to the candidate trial.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed

N09 · Was oral administration verified in the candidate MDD original report?

Required materials and conditions: Actual route of administration, product and prescription records in the candidate original report's Methods
Conditions for verification: Distinguish the actual route in the original report from the question's route, and link the route-verification gap to its location.
Requirement and applicability: Required
Application exceptions: The review's oral-administration information is a lead for tracing the original report, not verification of the route in that report.
Why this condition is needed: Needed to avoid turning an oral-administration question into a fact of actual oral administration.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed

N10 · Have the actual prescribed dose and administration regimen in S01 been verified?

Required materials and conditions: S01 original prescription/product, daily and per-administration doses, frequency, meal timing, adherence records and handling of deviations
Conditions for verification: Distinguish daily/per-administration doses, frequency, meal timing, loading/maintenance and adherence from study duration.
Requirement and applicability: Required
Application exceptions: Do not arbitrarily convert hydrochloride mg or thiamine-equivalent amounts, or fill in unverified frequency, meal timing or adherence.
Why this condition is needed: Needed to avoid transferring 300 mg/day from another study or review into a verified dose for the candidate trial.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

N11 · Is this an adjunctive-treatment comparison at initiation of a new SSRI?

Required materials and conditions: Records of SSRI initiation timing, drug/dose, maintenance at therapeutic levels and added ingredient/placebo in both arms
Conditions for verification: Preserve actual background-treatment initiation and maintenance and the structure of the placebo add-on comparison.
Requirement and applicability: Required
Application exceptions: Newly initiated standard treatment and stable pre-existing treatment are not automatically pooled as the same incremental effect.
Why this condition is needed: Needed to avoid turning improvement from the SSRI itself into thiamine monotherapy efficacy or augmentation of a stable dose.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M08 §8.2.2, M15 §15.3.1; custom application

N12 · How were placebo, psychotherapy, other drugs and rescue treatment compared?

Required materials and conditions: Placebo composition/appearance, psychotherapy/other psychotropics by arm, rescue treatment and records of changes
Conditions for verification: Distinguish the placebo label from actual matching, shared treatment and rescue treatment/psychotherapy by arm.
Requirement and applicability: Required
Application exceptions: Do not assume that all changes must be identical; examine the actual effect of interest and how changes by arm were handled.
Why this condition is needed: Needed to avoid mixing differences in concomitant treatment with the selected ingredient's contribution.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application

N13 · Have the implementation of randomization, allocation concealment and baseline balance been verified?

Required materials and conditions: Sequence generation, concealment, unit of allocation, baseline tables by arm and participant flow in the original Methods
Conditions for verification: Link the reported randomized placebo comparison to gaps concerning actual sequence generation, concealment and allocation/baseline details.
Requirement and applicability: Required
Application exceptions: Do not turn unverified design details into an unsupported Not applicable designation for crossover/cluster considerations.
Why this condition is needed: Needed to avoid stating that allocation issues were resolved merely from the randomized label.
Referenced methods: M08 §8.1–8.2.3; custom application; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed

N14 · Have assessor blinding and the actual HDRS measurement procedure been verified?

Required materials and conditions: Original blinding, unblinding/verification, assessor training/agreement and measurement procedures by arm
Conditions for verification: Link assessor and participant/treatment blinding, timing, training and assessment agreement to the actual methods.
Requirement and applicability: Required
Application exceptions: A bibliographic title stating double blind does not establish successful blinding of every party or successful implementation of all assessment procedures.
Why this condition is needed: Needed to distinguish differences in the implementation of clinician assessment from actual differences in symptoms.
Referenced methods: M08 Table 8.2.a, M03 §3.2.4; custom application; M08 §8.1–8.2.3; custom application

N15 · Were the total of 51 participants and arm-level randomized, analyzed and completed counts distinguished?

Required materials and conditions: Original CONSORT and allocation/completion/withdrawal tables, the HDRS analysis population at 12 weeks and n by arm
Conditions for verification: Show the label for the reported total count and the unverified arm-level allocation/completion/analysis and principal-time-point denominators.
Requirement and applicability: Required
Application exceptions: Do not split the odd total participant count between arms or add it to S03 and S04.
Why this condition is needed: Needed to avoid turning the abstract's total participant count into the actual analysis denominator at 12 weeks.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application

N16 · Have attrition, missing data, ITT/PP and the effect analyzed been verified?

Required materials and conditions: Original missingness/attrition by arm at 12 weeks, analysis n, SAP, ITT/PP and sensitivity results
Conditions for verification: Distinguish reasons for attrition, missing data by arm, their handling, analysis populations and the effect of interest.
Requirement and applicability: Required
Application exceptions: If the original material does not specify the analysis, do not arbitrarily fill in ITT, PP, LOCF or another method.
Why this condition is needed: Needed to identify the material required to examine whether selection of completers/analyzed participants changes the between-group result.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application

N17 · What are the actual HDRS version, items, range and direction?

Required materials and conditions: Original scale version, items/range, scoring instructions and assessment result tables by arm
Conditions for verification: Separate the recorded clinician-rated total and lower-is-better direction from the unverified version/items/range.
Requirement and applicability: Required
Application exceptions: Do not infer 17/21 items or a score range from the HDRS label, or substitute a GHQ subscale.
Why this condition is needed: Needed to avoid treating other depression scales or individual items as the same effect as HDRS total.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M03 §3.2.4; custom application; M03 §3.2.4; custom boundaries on extensions

N18 · Were the endpoint result at 12 weeks and signals at 3 and 6 weeks distinguished?

Required materials and conditions: Original visit schedule, principal time point in the registration/protocol, result tables by time and actual prescribed exposure
Conditions for verification: Link the reported 0, 3, 6 and 12 weeks to the manuscript's chosen principal time point and its prespecification status.
Requirement and applicability: Required
Application exceptions: Do not describe the current manuscript's choice of 12 weeks as trial preregistration, or infer the actual daily prescription.
Why this condition is needed: Needed to avoid substituting a favorable intermediate time point for the selected 12-week question.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M08 Table 8.2.a, M04 §4.6.3; custom application

N19 · Has the exact estimate of the between-group effect at 12 weeks been verified?

Required materials and conditions: Original arm-level total scores/changes at 12 weeks, adjusted/unadjusted between-group estimates and analysis definitions
Conditions for verification: Distinguish the between-group mean difference at 12 weeks from final values/changes and within-group improvement.
Requirement and applicability: Required
Application exceptions: Do not reconstruct numerical values from the figure or use the 6-week result or response/remission P values to fill in the total-score result at 12 weeks.
Why this condition is needed: Needed to avoid turning within-group improvement in both arms into thiamine's incremental contribution.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N20 · Can the SD/SE and confidence interval at 12 weeks be interpreted?

Required materials and conditions: Original numerical tables at 12 weeks, SD/SE definitions, analysis n by arm, CI and model output
Conditions for verification: Show the actual meaning of the reported variability and intervals and the scope of missing values.
Requirement and applicability: Required
Application exceptions: Do not use qualitative inspection of the original Fig.1 as digitization or reconstruction of SD/SE or CI.
Why this condition is needed: Needed to avoid arbitrarily deriving precision from graph bars or sample size.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M15 §15.3.1–15.3.3; custom application

N21 · Have the P value at 12 weeks and the criterion for a clinically important difference been verified?

Required materials and conditions: Original exact P value, CI and effect at 12 weeks, and evidence for an important difference appropriate to the HDRS version/population
Conditions for verification: Explain whether the exact P value and CI at 12 weeks can be compared with an evidence-based important difference.
Requirement and applicability: Required
Application exceptions: Do not fill in unverified P values or MCIDs using conventional thresholds or P values for other outcomes.
Why this condition is needed: Needed to avoid declaring sufficient effect or precision from significance or participant counts alone.
Referenced methods: M15 §15.3.1–15.3.3; custom application; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed

N22 · Were favorable findings and statements of no difference both retained without exaggeration?

Required materials and conditions: All results by time, corrections, analysis definitions and exact original numerical tables
Conditions for verification: Preserve the signal at 6 weeks, nonsignificance at 12 weeks, within-group improvement and positive findings in other studies with their original attribution.
Requirement and applicability: Required
Application exceptions: Positive findings at other time points, in other populations or on other scales are not automatically treated as conflicts or replications.
Why this condition is needed: Needed to avoid selecting only improvement at 6 weeks to erase the 12-week gap, or turning nonsignificance into an effect of 0.
Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M15 §15.3.1–15.3.3; custom application

N23 · To what level of disclosure has independence been verified?

Required materials and conditions: Original funding, internal support, product provision, employee/patent relationships, declared COI and declarations in each report
Conditions for verification: Attribute separately the declarations and internal-support/product gaps in S01, unverified details in S03 and funding in S04.
Requirement and applicability: Supplementary; required for claims of independence
Application exceptions: Do not automatically exclude results because funding/COI exists, or conclude that bias is absent because it does not.
Why this condition is needed: Needed to avoid turning unverified funding or a declaration of no external funding into complete independence.
Referenced methods: Chamgap's own independence and attribution conditions. The official methodological sections used in this pilot do not validate all details of this item; M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application

N24 · Were the original registration, outcome hierarchy, corrections and scope of unverified information compared?

Required materials and conditions: Original registration ID, prospective timing, protocol/SAP, outcome changes, correction/retraction notices and actual access events
Conditions for verification: Distinguish the registration/plan gap for S01 from the scope of the correction search at that time.
Requirement and applicability: Required
Application exceptions: Do not retrospectively relabel the current manuscript's principal-time-point selection or historical targeted searches as preregistration or a complete, up-to-date integrity check.
Why this condition is needed: Needed to avoid turning failure to obtain a registration into confirmed nonregistration or selective reporting, or a guarantee of no retraction.
Referenced methods: M08 Table 8.2.a, M04 §4.6.3; custom application; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N25 · What is the evidential level of the oral-administration, dose and SSRI information in the S07 review?

Required materials and conditions: Relevant S07 tables/sections and original citations, and the S01 product, route, daily dose and scale Methods
Conditions for verification: Keep the secondary (2nd) information on oral 300 mg/day, fluoxetine 20 mg/day and unspecified formulation distinct as recorded.
Requirement and applicability: Required
Application exceptions: The review's tables and descriptions are leads for tracing the original report, not direct verification of the S01 product/route/dose.
Why this condition is needed: Needed to prevent secondary (2nd) information from automatically resolving original-report eligibility gaps.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N26 · In S03, do the 50 participants, 300 mg/day and P values represent the same 12-week effect?

Required materials and conditions: Full S03 Methods/results, n by arm, route/salt, principal time point, analysis attribution of each P value and SSRI/placebo information
Conditions for verification: Link the actual S03 population, standard-treatment control, HRSD and denominator/time-point gaps.
Requirement and applicability: Required
Application exceptions: Do not arbitrarily assign P=.004/.031/.035 to particular visits, analyses or 12 weeks.
Why this condition is needed: Needed to avoid combining another controlled MDD study as an exact replication of the placebo comparison at SSRI initiation and 12 weeks.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

N27 · Are the date, issue and citation year for S03 consistent?

Required materials and conditions: The journal's canonical metadata, corrections to date/issue/citation information and publication events by version
Conditions for verification: Preserve all of the published date 2019-09-24, the 2020 issue and the 2019 citation year.
Requirement and applicability: Supplementary; required for bibliographic identity
Application exceptions: Do not extend a bibliographic date conflict to a conflict in the clinical results or the entire treatment duration.
Why this condition is needed: Needed to avoid changing the original record by overwriting different bibliographic years with a single value.
Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N28 · Is S04 a direct result for MDD inpatients at SSRI initiation?

Required materials and conditions: Actual S04 eligibility, scale, duration, product, depression-subgroup data and original plan
Conditions for verification: Show the difference between PCOS/IVF at ages 18–40 years with GHQ at 4 weeks and MDD with HDRS at 12 weeks.
Requirement and applicability: Required
Application exceptions: Retain the indirect study's original data, but do not turn exclusion of psychological disorders and the absence of a depression entry threshold into an MDD sample.
Why this condition is needed: Needed to avoid transferring results for other populations, scales or time points to the selected effect merely because the trial used oral B1.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

N29 · Were S04 denominators and positive results correctly attributed?

Required materials and conditions: S04 original Tables 2–3, CONSORT, analysis n, actual depression-subgroup results and missing-data handling
Conditions for verification: Separately preserve the 35/35 randomized and 32/32 analyzed counts and the original GHQ means/SDs and P value.
Requirement and applicability: Supplementary; required for describing indirect results
Application exceptions: Do not convert means on another scale into an HDRS effect or calculate new between-group differences/changes.
Why this condition is needed: Needed to avoid confusing the 70 participants randomized with the 64 participants analyzed, or SD with total-score/subgroup results.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N30 · How were S04 thresholds, percentages, outcome hierarchy and registration retained?

Required materials and conditions: S04 table/text corrections, protocol, original IRCT registration/versions, CONSORT and raw data
Conditions for verification: Preserve historical quality notes, gaps in the original registration record and differences in scale thresholds/hierarchy.
Requirement and applicability: Supplementary; required for accuracy of indirect material
Application exceptions: Do not silently correct original percentages or infer prospective timing from the digits in the IRCT identifier.
Why this condition is needed: Needed to avoid resolving original reporting inconsistencies and failure to obtain registration in a favorable direction.
Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M08 Table 8.2.a, M04 §4.6.3; custom application; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed

N31 · Do the results in S16–18 directly answer the selected adjunctive-efficacy question?

Required materials and conditions: Population, design, exposure/route, endpoint and comparator in each original report, and directness to the selected MDD question
Conditions for verification: Show separately the differences among healthy young women, observations in older Chinese adults and a deficiency case treated by parenteral→oral administration.
Requirement and applicability: Supplementary; required to prevent extensions
Application exceptions: Do not delete indirect material, but do not invent numerical values for results that were not extracted.
Why this condition is needed: Needed to avoid combining mood in healthy people, nutritional-status associations and deficiency case reports into a placebo-controlled adjunctive effect in MDD.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application; M03 §3.2.4; custom boundaries on extensions

N32 · Are S19–21 HDRS results for ordinary thiamine alone?

Required materials and conditions: Ingredients, populations, screening scales, endpoints/times and ingredient-specific direct comparisons in each original report
Conditions for verification: Explain the actual screening roles of combination/pregnancy studies, anxiety/stress/sleep and PBC fatigue.
Requirement and applicability: Supplementary; required to prevent extensions
Application exceptions: Do not decompose the contribution of B1+B2 or multiple micronutrients into B1 alone.
Why this condition is needed: Needed to avoid attributing results for combinations, pregnancy or fatigue to the selected ingredient/total score.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M03 §3.2.1; detailed population conditions are custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application; M03 §3.2.4; custom boundaries on extensions

N33 · Were controlled adult results verified in NCT07793071?

Required materials and conditions: The official registration's identity, population, ingredient, route, design, time points, results, versions and actual acquisition events
Conditions for verification: Separate discovery of a candidate registration title from unverified original content/results.
Requirement and applicability: Supplementary; required for interpreting registration
Application exceptions: Do not use details from a 3rd-party mirror as the original registration, or equate youth with an adult population.
Why this condition is needed: Needed to avoid promoting an indexed title, app shell or API failure into actual adult results.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M08 Table 8.2.a, M04 §4.6.3; custom application; M03 §3.2.1; detailed population conditions are custom-designed

N34 · Does S01's statement of no adverse effects mean 0 events per arm?

Required materials and conditions: S01 event/person counts by arm, exposure/follow-up n, duration, AE definitions/collection and reasons for discontinuation
Conditions for verification: Separate the original wording from unverified event/person counts, exposure n, discontinuations and ascertainment.
Requirement and applicability: Supplementary; required for safety statements
Application exceptions: Do not arbitrarily use the total of 51 participants or 12 weeks as an exact exposure denominator or ascertainment scope.
Why this condition is needed: Needed to avoid turning the abstract's absence statement into complete ascertainment or an incidence of 0.
Referenced methods: M19 §19.1.2–19.2.4; custom application; M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M19 §19.1.2–19.2.4, M04 §4.6.3; custom application

N35 · Were suicide-related events, manic switch, worsening and long-term safety assessed?

Required materials and conditions: Psychiatric-event definitions, systematic ascertainment, arm-level n/duration, worsening/mania/suicide and long-term follow-up data
Conditions for verification: Check the scope of separate psychiatric ascertainment, exposure time and long-term/special-population information.
Requirement and applicability: Supplementary; required for those safety extensions
Application exceptions: Do not turn unverified suicide/mania/worsening into absence of events or a causal finding of worsening.
Why this condition is needed: Needed to avoid extending a general AE statement in a short-term abstract to psychiatric events or overall safety.
Referenced methods: M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M19 §19.2 design-specific harm questions; details of temporal relationships are custom-designed; M19 §19.1.2–19.2.4; custom application; Chamgap's own interpretive boundaries. Informed by M19 principles, not an official causality-tool implementation

N36 · How are the nausea/adverse-effect statements in S03 and S04 attributed?

Required materials and conditions: S03/S04 original events/people by arm, ascertainment, exposure/duration, discontinuations, products and CONSORT data
Conditions for verification: Distinguish populations, products and durations from the corresponding reports' harm statements and denominator gaps.
Requirement and applicability: Supplementary; required for safety interpretation of other materials
Application exceptions: Do not create event counts, exposure denominators or 0 events from descriptions of nausea or no important AEs.
Why this condition is needed: Needed to avoid adding events from other studies to harms in S01 or counting them as replications for the same product.
Referenced methods: M19 §19.1.2–19.2.4; custom application; M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M04 §4.6.1–4.6.2; adaptation for harm data is custom-designed

N37 · Does the Flamingo product information establish the candidate product and its harms?

Required materials and conditions: Documentation linking the candidate trial product to this product, and the actual SmPC/version and original harm data for the oral product
Conditions for verification: Separate the 100 mg oral product, historical version and unknown gastrointestinal/hypersensitivity frequencies from the unverified candidate product.
Requirement and applicability: Supplementary; required for interpreting product safety
Application exceptions: Do not transfer the frequency of hypersensitivity reported mainly with parenteral use into incidence in an oral trial.
Why this condition is needed: Needed to avoid assigning a separate hydrochloride product's identity to the candidate trial's actual product/route.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M19 §19.2.2–19.2.4, M03 §3.2.3; custom application; M19 §19.1.2–19.2.4; custom application

N38 · What do an unset UL and general nutritional/medication context mean?

Required materials and conditions: Original official nutrition/excess-intake, vulnerable-population and interaction documents, and direct harm data matching the population and exposure
Conditions for verification: Link the unset UL and excessive-intake warning in S05 to its role as context rather than efficacy evidence.
Requirement and applicability: Supplementary; required for general safety extensions
Application exceptions: Do not derive personal recommendations or MDD treatment effects from general information about intake/medication relationships.
Why this condition is needed: Needed to avoid using the absence of a UL or deficiency context as evidence of unlimited safety or MDD efficacy.
Referenced methods: M19 §19.1.2–19.2.4; custom application; Chamgap's own interpretive boundaries. Informed by M19 principles, not an official causality-tool implementation; M03 §3.2.1; detailed population conditions are custom-designed

N39 · Does the total score at 12 weeks also answer questions about optimal dose, maintenance, response/remission and prevention?

Required materials and conditions: Direct dose comparisons, follow-up after discontinuation, prespecified subgroups/interactions and response/remission/relapse/suicide-event data
Conditions for verification: Clearly distinguish the selected total score from unverified extension questions.
Requirement and applicability: Supplementary; required for those extension claims
Application exceptions: Do not require every extension trial as a mandatory condition for constructing this table's direct 12-week question.
Why this condition is needed: Needed to avoid extending a narrow symptom comparison to dose selection, persistence or clinical events.
Referenced methods: M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed; M03 §3.2.4; custom boundaries on extensions; M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed

N40 · Were original materials needed to fill current gaps distinguished from methods for future trials?

Required materials and conditions: S01 original Methods/product, HDRS version, arm-level tables/CI at 12 weeks, flow, registration/SAP, safety and corrections
Conditions for verification: Distinguish obtaining current original reports/plans from proposals for eligible direct comparisons and analysis/reporting.
Requirement and applicability: Required
Application exceptions: Do not adopt a new minimum sample size, number of trials, dose or MCID without justification.
Why this condition is needed: Needed to avoid presenting acquisition of a full text and conduct of new research as the same action.
Referenced methods: M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M08 §8.1–8.2.3; custom application; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M08 Table 8.2.a, M03 §3.2.4; custom application; M08 §8.2.2, M15 §15.3.1; custom application; M15 §15.3.1–15.3.3; custom application; M08 Table 8.2.a, M04 §4.6.3; custom application

N41 · Were the existing ?/null and calculator exception kept from being inherited as the new table's status?

Required materials and conditions: Original calculation inputs/errors, current-value exception, verification records and authoritative records of new evidence/condition-change events
Conditions for verification: Preserve the original cutoff and review dates, original grade/score, axes, exception and historical self-review.
Requirement and applicability: Required
Application exceptions: Do not adopt the historical calculator's validation error/pre-validation B as a new valid grade, or rerun the calculator.
Why this condition is needed: Needed to avoid turning an uncalculated value into no effect, no human research or a score of 0 points.
Referenced methods: M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

N42 · How were analyses across multiple visits/scales and the adjusted model prespecified?

Required materials and conditions: Original protocol/SAP, analysis model, covariates, repeated-visit/multiplicity handling, effect of interest at 12 weeks and reasons for changes
Conditions for verification: Explain the actual adjusted/unadjusted effect of interest, handling of multiple time points/endpoints and unverified plans.
Requirement and applicability: Required
Application exceptions: Without the original SAP, do not infer repeated-measures methods, multiplicity adjustments or covariate models.
Why this condition is needed: Needed to avoid mixing comparisons at 3, 6 and 12 weeks and selection of favorable analyses into the same 12-week estimate.
Referenced methods: M08 §8.2.2, M15 §15.3.1; custom application; M08 Table 8.2.a, M04 §4.6.3; custom application; M15 §15.3.1–15.3.3; custom application; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed

Referenced methods and the sections actually reviewed

M03 · Cochrane Handbook 6.5 (2024), chapter updated 2023-08

Official methodology · §3.1, §3.2.1–3.2.3, MECIR C5/C7/C13

Referenced principle: Clearly record the question, eligibility criteria and reasons for changes, and distinguish variations in the intervention.
Extent of review: Historical reuse of the actual retained HTML copies and SHAs for 1585 and the methodological principles directly reviewed at that time. This is not new reading of full methodological texts, official implementation or clinical certification in this review.

M04 · Cochrane Handbook 6.5.1 (2025), chapter updated 2025-03

Official methodology · §4.5, §4.6.1–4.6.3, MECIR C42

Referenced principle: Distinguish and link studies and reports, and do not omit a study solely because its outcomes are not reported.
Extent of review: Historical reuse of the actual retained HTML copies and SHAs for 1585 and the methodological principles directly reviewed at that time. This is not new reading of full methodological texts, official implementation or clinical certification in this review.

M08 · Cochrane Handbook 6.5 (2024), chapter updated 2019-10

Official methodology · §8.1–8.2.3, Table 8.2.a

Referenced principle: Hold the outcome and effect of interest fixed, then assess randomization, deviations, missing data, measurement and selective reporting.
Extent of review: Historical reuse of the actual retained HTML copies and SHAs for 1585 and the methodological principles directly reviewed at that time. This is not new reading of full methodological texts, official implementation or clinical certification in this review.

M15 · Cochrane Handbook 6.5 (2024), chapter updated 2023-08

Official methodology · §15.2.4, §15.3.1–15.3.3

Referenced principle: Distinguish effect estimates, intervals, importance and applicability. Sample size alone does not determine precision.
Extent of review: Historical reuse of the actual retained HTML copies and SHAs for 1585 and the methodological principles directly reviewed at that time. This is not new reading of full methodological texts, official implementation or clinical certification in this review.

M19 · Cochrane Handbook 6.5 (2024), chapter updated 2019-10

Official methodology · §19.1.2, §19.2.1–19.2.4

Referenced principle: Questions about harms may require designs, populations and follow-up different from efficacy questions. Case reports do not establish incidence.
Extent of review: Historical reuse of the actual retained HTML copies and SHAs for 1585 and the methodological principles directly reviewed at that time. This is not new reading of full methodological texts, official implementation or clinical certification in this review.

What the statuses mean

Met: The record's identity, boundaries and conditions for preservation/distinction can be explained. This differs from confirmation of treatment efficacy/safety or official certification.
Partly verified: Some information for the selected condition is linked, but specific gaps remain.
Unverified: Required values/conditions cannot be established within the material currently read. This does not mean no research, an effect of 0 or 0 events.
Conflicting information: A specific discrepancy in the level of verification of deficiency statements or in original bibliographic details/reporting/hierarchy. Differences in time points, denominators or designs alone are not automatically designated as conflicts.
Not applicable: Used only when there is an actual reason why the selected condition does not apply.
Not reported: Limited to a gap at a specific retained location actually read; not extended to nonreporting throughout the external original paper.
Inaccessible: Used only when an actual current access attempt has failed. This differs from external access not attempted in this review or historical failures.
Review incomplete: An actual pending state of individual table preparation/review, separate from unverified/unread clinical material.

Change history

2026-11-06 · Public page/material version 1.0

Public-facing display and data were prepared from the existing 42 items. No new clinical search, change in verification status or grading was performed.

The stored-material review date (2026-10-05) and original evidence cut-off (2026-09-16) were retained unchanged.

Researcher downloads

These materials provide the same evidence, scope and verification status as this page.

View all item details

N01 · Which adults, adjunctive treatment and 12-week outcome does this table address?

Current findings: The original claim concerns treatment added at SSRI initiation in adult MDD inpatients and reduction of HDRS total at 12 weeks. The question is fixed to single-ingredient oral ordinary thiamine versus placebo, with subclaims=[]. Diagnosis, response/remission, relapse and suicide prevention are separate questions.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The identity and exclusion boundaries of the independent question were linked to the retained original text. Meeting this condition does not mean that candidate-trial eligibility or treatment efficacy has been verified.
Reason for missing information: The question's identity is linked, but the actual product, route and effect remain separately unverified.
Required materials and conditions: Original claim, facets and subclaims, and clinical questions excluded from the same page
Materials to obtain first: Original claim, facets and subclaims, and clinical questions excluded from the same page
Further research: Future research should prespecify HDRS total at 12 weeks and response, remission and suicide-related events as separate outcomes.
Remaining uncertainty: The question's identity is linked, but the actual product, route and effect remain separately unverified.
Conditions for verification: Link the original /ko/effect to the population, addition at SSRI initiation and HDRS at 12 weeks, and explain overlap between parent and subordinate questions.
Why this condition is needed: Needed to prevent the question about total depressive symptoms from shifting to general mood, remission or suicide prevention.
Application exceptions: Response, remission, relapse and suicide-related events remain separate questions rather than mandatory outcomes for this total-score question.
Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M03 §3.2.4; custom boundaries on extensions

Related sources S01 S03

N02 · Do the current published source and this summary have the same identity?

Current findings: The current ID is 3120 and the canonical path is /verdicts/mood/oral-thiamine-mdd-hdrs-12-weeks/. The current base original and inventory snapshot have the same 99,833 bytes and SHA. The needs_id_assignment/server-not-deployed wording in the KO summary_note is preserved as historical content and is not used as a deployment record for this summary conversion.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The current record's identity was distinguished from historical status wording. The original site's current response or public availability was not checked in this review.
Reason for missing information: The production domain's response and actual deployment were not checked in this review.
Required materials and conditions: Current loader-selected path, original SHA, ID/canonical URL, version and actual publication receipt
Materials to obtain first: Current loader-selected path, original SHA, ID/canonical URL, version and actual publication receipt
Further research: Record identities, dates and publication events separately by version, without creating new research results.
Remaining uncertainty: The production domain's response and actual deployment were not checked in this review.
Conditions for verification: Compare the current original and snapshot bytes, ID/canonical URL and reference date.
Why this condition is needed: Needed to avoid confusing historical wording about unassigned identifiers and nondeployment with the current ID/URL.
Application exceptions: Historical publication-readiness wording retained in KO is not deleted or reinterpreted as a site-publication event in this review.
Referenced methods: M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design

Related sources

N03 · Were the 21 reports kept from being counted repeatedly as unique trials?

Current findings: S01/S02/S10–S14 form one family, TF-GHALEIHA-2016. S03 is a different controlled study, TF-WATTANATHAMRAK-2019, but replication of the same comparison is unverified. S04 is TF-MOTI-2024 in PCOS/IVF and provides indirect material. The retained family table's 3 controlled candidates and 0 families eligible for efficacy grading are eligibility records from that time.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Original report counts were distinguished from family roles, and no new count of directly eligible trials was established.
Reason for missing information: There is no new external verification of family identities, additional overlap or eligibility.
Required materials and conditions: Original reports, figures, registrations, repositories, a family linkage table and unique-participant identity information
Materials to obtain first: Original reports, figures, registrations, repositories, a family linkage table and unique-participant identity information
Further research: Future syntheses should first create study-level outcome tables after comparing registration numbers, authors, recruitment institutions, periods and participant overlap.
Remaining uncertainty: There is no new external verification of family identities, additional overlap or eligibility.
Conditions for verification: Explain the roles of the S01 series as one family, S03 as a separate candidate and S04 as an indirect family.
Why this condition is needed: Needed to avoid counting figures, repositories, indexes and reviews for the same DOI as independent replications.
Application exceptions: Preserve the bibliography count, but do not estimate unresolved unique-study or participant counts or turn the three candidates into directly eligible trials.
Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

Related sources S01 S02 S03 S04 S10 S11 S12 S13 S14

N04 · What was read in this review, and what was the historical level of access?

Current findings: The current stored summary for 3120 was read in this review, including its Korean text, structured evidence, 21 bibliographic records and historical access records. The records dated 2026-09-16 of partial access to S01, visual inspection of the S02 figure and access to the S04 HTML full text are history recorded in the original. External papers, figures, registrations and official materials were not accessed, acquired or read in this review; access was not attempted.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Current actual actions were separated from historical records. Past PDF failures were not labeled as failures in this review, and historical full-text access to S04 was not labeled as current full-text reading.
Reason for missing information: External full texts and old separate ZIP files/logs were not opened in this review.
Required materials and conditions: Retained source/citation, access_level/date/locator, and actual files and reading events
Materials to obtain first: Retained source/citation, access_level/date/locator, and actual files and reading events
Further research: When materials are acquired in the future, record the sections and tables actually read, file SHA and access events separately for each material type.
Remaining uncertainty: External full texts and old separate ZIP files/logs were not opened in this review.
Conditions for verification: Separate existence, access, acquisition, reading and current/historical dates for each material type.
Why this condition is needed: Needed to avoid promoting the summary of the current stored summary to reading the original paper, figure or full text.
Application exceptions: Past failures, partial reading and reports of full-text access remain historical records; they are not reused as inaccessibility or full-text reading in this review.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design

Related sources S01 S02 S04 S07 S09 S13 S15

N05 · What are the candidate trial's population and baseline depression severity?

Current findings: S01 reports a total of 51 participants, all MDD inpatients, baseline HDRS >24, mean age 35.2 years and 53% women. Age bounds, sex distribution by arm and baseline scale means/variances are unverified. Do not turn means and percentages into identical characteristics of every adult.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The population label, setting and baseline threshold are preserved, but exact eligibility and balance between arms remain unknown.
Reason for missing information: Age range, inclusion/exclusion criteria, baseline distributions by arm and baseline means/SDs are unverified.
Required materials and conditions: Original eligibility criteria, age range, baseline tables by arm, inpatient setting and recruitment flow
Materials to obtain first: Original eligibility criteria, age range, baseline tables by arm, inpatient setting and recruitment flow
Further research: Specify baseline severity and representativeness in the same inpatient MDD population and present balance between arms.
Remaining uncertainty: Age range, inclusion/exclusion criteria, baseline distributions by arm and baseline means/SDs are unverified.
Conditions for verification: Distinguish the reporting level and unverified details of MDD inpatient status, HDRS >24 and means/percentages.
Why this condition is needed: Needed to avoid applying mood studies in healthy people or other diseases to adult MDD inpatients.
Application exceptions: Do not infer age ranges, arm composition or the status of all participants from mean age or the percentage of women.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed

Related sources S01

N06 · Have the MDD diagnostic framework, episode, treatment resistance and major exclusions been verified?

Current findings: The retained manuscript confirms that MDD was reported, but the DSM/ICD version, diagnostic interview, episode status and treatment resistance are unverified. Exclusions and population composition regarding pregnancy/lactation, bipolar disorder, dementia/delirium, renal/liver status and comorbidities are also unverified.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The retained location read explicitly describes a gap in access to the original methods, so the exact diagnostic and exclusion conditions cannot be said to have been met.
Reason for missing information: Diagnosis and exclusions are gaps in the accessed portions of the original report. Do not conclude that they are unreported throughout the original paper.
Required materials and conditions: Original diagnostic interview/DSM/ICD, episode/resistance definitions, exclusion criteria and baseline tables by population
Materials to obtain first: Original diagnostic interview/DSM/ICD, episode/resistance definitions, exclusion criteria and baseline tables by population
Further research: Prespecify diagnostic procedures and episode/resistance criteria, and analyze applicability to special populations separately.
Remaining uncertainty: Diagnosis and exclusions are gaps in the accessed portions of the original report. Do not conclude that they are unreported throughout the original paper.
Conditions for verification: Link diagnosis, episode, treatment resistance, comorbidities and the scope of exclusions to the original methods.
Why this condition is needed: Needed to avoid interpreting the reported MDD label as verification of a specific diagnostic procedure, treatment resistance or exclusion of bipolar disorder.
Application exceptions: Do not fill in unverified exclusions, such as pregnancy, bipolar disorder or renal/liver disease, as if they had already been applied.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

Related sources S01

N07 · Was the candidate population thiamine-deficient or thiamine-replete?

Current findings: S08's secondary (2nd) narrative calls the S01 population thiamine-deficient. Neither deficiency nor repletion was verified in the accessed S01 original, and baseline markers, thresholds, nutrition, diet and alcohol use are unverified. This conflict concerns the level of verification of deficiency status, not conflicting numerical clinical effects.
Verification status: Conflicting information
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The secondary (2nd) assertion and the original-report verification gap were preserved, without arbitrarily adopting either status.
Reason for missing information: Original tests/criteria and arm-level data needed to determine baseline status are unverified.
Required materials and conditions: Original baseline thiamine tests/deficiency criteria, dietary/nutritional/alcohol-use records and the review's original citation location
Materials to obtain first: Original baseline thiamine tests/deficiency criteria, dietary/nutritional/alcohol-use records and the review's original citation location
Further research: Future research should measure baseline status and assess interaction by prespecified deficiency status.
Remaining uncertainty: Original tests/criteria and arm-level data needed to determine baseline status are unverified.
Conditions for verification: Separate baseline thiamine markers/thresholds, meals, nutrition and alcohol use from the verification level of secondary (2nd) statements.
Why this condition is needed: Needed to avoid combining correction of deficiency with the effect of additional supplementation in a replete state.
Application exceptions: Do not use the secondary (2nd) review's word deficient to establish the original report's testing, deficiency threshold or replete status.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

Related sources S01 S08

N08 · Can ordinary thiamine be distinguished from the actual salt and single-ingredient product?

Current findings: The question targets single-ingredient oral ordinary thiamine. S01 reports the ingredient name thiamine (vitamin B1), but the exact salt, product, source and single-ingredient composition could not be verified. Benfotiamine, fursultiamine/TTFD, sulbutiamine, allithiamine, phosphorylated derivatives, combinations and injections are not direct evidence for the same question.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The general ingredient name alone is not treated as meeting eligibility requirements for the exact molecule/salt and single-ingredient composition.
Reason for missing information: The candidate trial's product identity and actual composition are unverified.
Required materials and conditions: Original trial product name, salt, active/added ingredients, manufacturing/formulation documents and administration records
Materials to obtain first: Original trial product name, salt, active/added ingredients, manufacturing/formulation documents and administration records
Further research: Future direct comparisons should predefine and disclose the salt, product and composition of ordinary thiamine and the control product.
Remaining uncertainty: The candidate trial's product identity and actual composition are unverified.
Conditions for verification: Show separately the question target and the level of verification of the actual label, salt, source and composition in the original report.
Why this condition is needed: Needed to avoid assigning the identity of derivatives, combinations or other products to the candidate trial.
Application exceptions: Derivatives such as benfotiamine and TTFD, combinations and injections are excluded from the question, but the candidate product is not presumed to have been a single-ingredient preparation.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed

Related sources S01

N09 · Was oral administration verified in the candidate MDD original report?

Current findings: The question concerns oral administration, but the actual oral route was not verified in retained S01. Oral-administration information in the S07 review is not a route verified in the original report, and oral-administration information in the PCOS study S04 does not establish the route in S01 either.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The current eligibility gap in the original report was maintained. An unverified oral route does not mean that a parenteral route has been verified.
Reason for missing information: The original Methods needed to verify the actual route in S01 remain unverified.
Required materials and conditions: Actual route of administration, product and prescription records in the candidate original report's Methods
Materials to obtain first: Actual route of administration, product and prescription records in the candidate original report's Methods
Further research: Future comparisons should document oral administration and actual intake, and separate these from injection and derivative data.
Remaining uncertainty: The original Methods needed to verify the actual route in S01 remain unverified.
Conditions for verification: Distinguish the actual route in the original report from the question's route, and link the route-verification gap to its location.
Why this condition is needed: Needed to avoid turning an oral-administration question into a fact of actual oral administration.
Application exceptions: The review's oral-administration information is a lead for tracing the original report, not verification of the route in that report.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed

Related sources S01 S04 S07

N10 · Have the actual prescribed dose and administration regimen in S01 been verified?

Current findings: Daily dose, per-administration dose, frequency, meal timing, loading/maintenance and adherence are unverified from the S01 original. The S07 information on oral 300 mg/day and fluoxetine 20 mg/day is not verified in the original report. For S03 and S04, 300 mg/day is also retained only as the nominal dose in each separate study.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Although the 12-week endpoint is preserved, the actual daily prescription and exposure cannot be established.
Reason for missing information: S01 dose/conversion, actual exposure and adherence data are unverified.
Required materials and conditions: S01 original prescription/product, daily and per-administration doses, frequency, meal timing, adherence records and handling of deviations
Materials to obtain first: S01 original prescription/product, daily and per-administration doses, frequency, meal timing, adherence records and handling of deviations
Further research: Future research should record actual intake, adherence and changes, and assess dose optimization in separate direct comparisons.
Remaining uncertainty: S01 dose/conversion, actual exposure and adherence data are unverified.
Conditions for verification: Distinguish daily/per-administration doses, frequency, meal timing, loading/maintenance and adherence from study duration.
Why this condition is needed: Needed to avoid transferring 300 mg/day from another study or review into a verified dose for the candidate trial.
Application exceptions: Do not arbitrarily convert hydrochloride mg or thiamine-equivalent amounts, or fill in unverified frequency, meal timing or adherence.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed

Related sources S01 S03 S04 S07

N11 · Is this an adjunctive-treatment comparison at initiation of a new SSRI?

Current findings: S01 reports a comparison of added thiamine or placebo while a standardized SSRI was newly initiated and maintained at therapeutic levels. It is not verified as a trial of thiamine monotherapy or augmentation of a stable pre-existing antidepressant dose. The SSRI name, dose and actual treatment by arm are unverified.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The adjunctive comparison's structure is preserved, but the specific identity of the shared treatment and its implementation by arm remain unknown.
Reason for missing information: Drug, dose and arm-level treatment changes and maintenance data are unverified.
Required materials and conditions: Records of SSRI initiation timing, drug/dose, maintenance at therapeutic levels and added ingredient/placebo in both arms
Materials to obtain first: Records of SSRI initiation timing, drug/dose, maintenance at therapeutic levels and added ingredient/placebo in both arms
Further research: Compare only the added ingredient on top of the same SSRI initiation, and design augmentation of a stable dose as a separate stratum.
Remaining uncertainty: Drug, dose and arm-level treatment changes and maintenance data are unverified.
Conditions for verification: Preserve actual background-treatment initiation and maintenance and the structure of the placebo add-on comparison.
Why this condition is needed: Needed to avoid turning improvement from the SSRI itself into thiamine monotherapy efficacy or augmentation of a stable dose.
Application exceptions: Newly initiated standard treatment and stable pre-existing treatment are not automatically pooled as the same incremental effect.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M08 §8.2.2, M15 §15.3.1; custom application

Related sources S01

N12 · How were placebo, psychotherapy, other drugs and rescue treatment compared?

Current findings: The control treatment in S01 is placebo, and the shared treatment is a newly initiated standardized SSRI. Placebo composition/appearance and actual treatment by arm, including psychotherapy, other psychotropics and rescue treatment, are unverified because of the gap in access to the original methods.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: A placebo comparison was reported, but specific verification of what was the same or different apart from the added ingredient remains outstanding.
Reason for missing information: Placebo matching, concomitant treatment/changes by arm, rescue treatment and their analytical handling are unverified.
Required materials and conditions: Placebo composition/appearance, psychotherapy/other psychotropics by arm, rescue treatment and records of changes
Materials to obtain first: Placebo composition/appearance, psychotherapy/other psychotropics by arm, rescue treatment and records of changes
Further research: Adjunctive comparisons should disclose concomitant treatment and rescue use by arm and handle these in a predefined analysis.
Remaining uncertainty: Placebo matching, concomitant treatment/changes by arm, rescue treatment and their analytical handling are unverified.
Conditions for verification: Distinguish the placebo label from actual matching, shared treatment and rescue treatment/psychotherapy by arm.
Why this condition is needed: Needed to avoid mixing differences in concomitant treatment with the selected ingredient's contribution.
Application exceptions: Do not assume that all changes must be identical; examine the actual effect of interest and how changes by arm were handled.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application

Related sources S01

N13 · Have the implementation of randomization, allocation concealment and baseline balance been verified?

Current findings: S01 is reported as a randomized thiamine-versus-placebo trial. The full methods section needed to verify sequence generation, concealment, unit of allocation and baseline balance by arm was not acquired within the current retained scope.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The comparative design is named, but evidence verifying actual allocation is insufficient.
Reason for missing information: Sequence generation, concealment, unit of allocation and baseline balance are unverified.
Required materials and conditions: Sequence generation, concealment, unit of allocation, baseline tables by arm and participant flow in the original Methods
Materials to obtain first: Sequence generation, concealment, unit of allocation, baseline tables by arm and participant flow in the original Methods
Further research: Future trials should predefine allocation and concealment and disclose participant flow and baseline data by arm.
Remaining uncertainty: Sequence generation, concealment, unit of allocation and baseline balance are unverified.
Conditions for verification: Link the reported randomized placebo comparison to gaps concerning actual sequence generation, concealment and allocation/baseline details.
Why this condition is needed: Needed to avoid stating that allocation issues were resolved merely from the randomized label.
Application exceptions: Do not turn unverified design details into an unsupported Not applicable designation for crossover/cluster considerations.
Referenced methods: M08 §8.1–8.2.3; custom application; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed

Related sources S01

N14 · Have assessor blinding and the actual HDRS measurement procedure been verified?

Current findings: The HDRS in S01 is retained as a clinician/expert assessment. The bibliographic URL for S13 contains double_blind, but actual implementation of assessor blinding, scale training, agreement and any unblinding are unverified. Do not conclude that measurement bias is absent from the title or URL alone.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The assessor type is recorded, but implementation of blinding and measurement consistency for the selected outcome remain unknown.
Reason for missing information: Assessment blinding, training, agreement and verification of implementation are unverified.
Required materials and conditions: Original blinding, unblinding/verification, assessor training/agreement and measurement procedures by arm
Materials to obtain first: Original blinding, unblinding/verification, assessor training/agreement and measurement procedures by arm
Further research: Define independent assessment procedures using the same scale and time point, and record measurement differences between arms.
Remaining uncertainty: Assessment blinding, training, agreement and verification of implementation are unverified.
Conditions for verification: Link assessor and participant/treatment blinding, timing, training and assessment agreement to the actual methods.
Why this condition is needed: Needed to distinguish differences in the implementation of clinician assessment from actual differences in symptoms.
Application exceptions: A bibliographic title stating double blind does not establish successful blinding of every party or successful implementation of all assessment procedures.
Referenced methods: M08 Table 8.2.a, M03 §3.2.4; custom application; M08 §8.1–8.2.3; custom application

Related sources S01 S13

N15 · Were the total of 51 participants and arm-level randomized, analyzed and completed counts distinguished?

Current findings: In S01, the 51 participants are the total reported in the abstract. Arm-level randomized, analyzed and completed n and the number analyzed for HDRS at 12 weeks are unverified. Do not divide the 51 participants into two arms or use them as the safety denominator at 12 weeks.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The reported total is preserved, but the arm-level analysis denominators for the outcome of interest cannot be verified.
Reason for missing information: Arm-level randomized, analyzed and completed counts and measurement/exposure n at 12 weeks are unverified.
Required materials and conditions: Original CONSORT and allocation/completion/withdrawal tables, the HDRS analysis population at 12 weeks and n by arm
Materials to obtain first: Original CONSORT and allocation/completion/withdrawal tables, the HDRS analysis population at 12 weeks and n by arm
Further research: Present the full flow by arm and the analysis n for each outcome, and examine the link between outcomes and attrition.
Remaining uncertainty: Arm-level randomized, analyzed and completed counts and measurement/exposure n at 12 weeks are unverified.
Conditions for verification: Show the label for the reported total count and the unverified arm-level allocation/completion/analysis and principal-time-point denominators.
Why this condition is needed: Needed to avoid turning the abstract's total participant count into the actual analysis denominator at 12 weeks.
Application exceptions: Do not split the odd total participant count between arms or add it to S03 and S04.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application

Related sources S01

N16 · Have attrition, missing data, ITT/PP and the effect analyzed been verified?

Current findings: The analysis n by arm, reasons for attrition/missingness, missing-data handling and ITT/PP analysis populations in S01 are unverified. Whether the analysis was adjusted or unadjusted, and whether the effect of interest concerns final values or changes, are also unresolved.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Even the same between-group difference can be interpreted differently depending on the analysis population and missing-data handling, so no analysis label was invented to fill the gap.
Reason for missing information: Missing data/attrition, analysis methods, models and the estimand are unverified.
Required materials and conditions: Original missingness/attrition by arm at 12 weeks, analysis n, SAP, ITT/PP and sensitivity results
Materials to obtain first: Original missingness/attrition by arm at 12 weeks, analysis n, SAP, ITT/PP and sensitivity results
Further research: Prespecify the distinction between the effect of assignment and the effect of adherence, and present missing-data assumptions and sensitivity analyses.
Remaining uncertainty: Missing data/attrition, analysis methods, models and the estimand are unverified.
Conditions for verification: Distinguish reasons for attrition, missing data by arm, their handling, analysis populations and the effect of interest.
Why this condition is needed: Needed to identify the material required to examine whether selection of completers/analyzed participants changes the between-group result.
Application exceptions: If the original material does not specify the analysis, do not arbitrarily fill in ITT, PP, LOCF or another method.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application

Related sources S01

N17 · What are the actual HDRS version, items, range and direction?

Current findings: The selected outcome is the continuous depression total on the clinician-rated Hamilton Depression Rating Scale, with lower scores indicating improvement. The actual version, item count and score range are unverified. The GHQ-28 depression subscale with 7 items and individual sleep/appetite items are not the same total score.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The scale name, construct and direction are preserved, but the version and range needed to interpret the actual score units remain unknown.
Reason for missing information: Version, item count, score range and actual scoring procedures are unverified.
Required materials and conditions: Original scale version, items/range, scoring instructions and assessment result tables by arm
Materials to obtain first: Original scale version, items/range, scoring instructions and assessment result tables by arm
Further research: Future trials should prespecify the exact scale, version, total score and assessor procedures.
Remaining uncertainty: Version, item count, score range and actual scoring procedures are unverified.
Conditions for verification: Separate the recorded clinician-rated total and lower-is-better direction from the unverified version/items/range.
Why this condition is needed: Needed to avoid treating other depression scales or individual items as the same effect as HDRS total.
Application exceptions: Do not infer 17/21 items or a score range from the HDRS label, or substitute a GHQ subscale.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M03 §3.2.4; custom application; M03 §3.2.4; custom boundaries on extensions

Related sources S01

N18 · Were the endpoint result at 12 weeks and signals at 3 and 6 weeks distinguished?

Current findings: The visits in retained S01 are at 0, 3, 6 and 12 weeks, and this table's principal time point is 12 weeks. The choice of principal time point is recorded as a review choice in the current manuscript, not as verified trial preregistration. The favorable signal at 6 weeks is preserved but does not replace the result at 12 weeks.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The roles of the time points were distinguished. Meeting this condition is separate from verification of the effect/prescribed dose at 12 weeks.
Reason for missing information: The trial's prespecified outcome hierarchy and actual administration exposure are unverified.
Required materials and conditions: Original visit schedule, principal time point in the registration/protocol, result tables by time and actual prescribed exposure
Materials to obtain first: Original visit schedule, principal time point in the registration/protocol, result tables by time and actual prescribed exposure
Further research: Future comparisons should prespecify the principal time point and repeated-measurement plan and disclose all intermediate results.
Remaining uncertainty: The trial's prespecified outcome hierarchy and actual administration exposure are unverified.
Conditions for verification: Link the reported 0, 3, 6 and 12 weeks to the manuscript's chosen principal time point and its prespecification status.
Why this condition is needed: Needed to avoid substituting a favorable intermediate time point for the selected 12-week question.
Application exceptions: Do not describe the current manuscript's choice of 12 weeks as trial preregistration, or infer the actual daily prescription.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M08 Table 8.2.a, M04 §4.6.3; custom application

Related sources S01 S02

N19 · Has the exact estimate of the between-group effect at 12 weeks been verified?

Current findings: S01/S02 report a favorable signal at 6 weeks, a nonsignificant between-group mean difference at 12 weeks and improvement in both groups. The exact between-group difference at 12 weeks, arm means, changes and adjustment status are unverified. Within-group improvement does not establish a causal effect of thiamine.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The numerical estimate of interest and its analysis definition are missing, so the magnitude of the incremental effect cannot be established.
Reason for missing information: The exact effect at 12 weeks, final values/changes, model and arm-level numerical values are unverified.
Required materials and conditions: Original arm-level total scores/changes at 12 weeks, adjusted/unadjusted between-group estimates and analysis definitions
Materials to obtain first: Original arm-level total scores/changes at 12 weeks, adjusted/unadjusted between-group estimates and analysis definitions
Further research: Directly estimate the predefined incremental effect at 12 weeks between groups, and present within-group changes separately.
Remaining uncertainty: The exact effect at 12 weeks, final values/changes, model and arm-level numerical values are unverified.
Conditions for verification: Distinguish the between-group mean difference at 12 weeks from final values/changes and within-group improvement.
Why this condition is needed: Needed to avoid turning within-group improvement in both arms into thiamine's incremental contribution.
Application exceptions: Do not reconstruct numerical values from the figure or use the 6-week result or response/remission P values to fill in the total-score result at 12 weeks.
Referenced methods: M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 §8.2.2, M15 §15.3.1; custom application; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources S01 S02

N20 · Can the SD/SE and confidence interval at 12 weeks be interpreted?

Current findings: The retained record explicitly states that Figure 1 was inspected only qualitatively and that graph digitization, arm means, SD/SE and effect/CI were not reconstructed. The confidence interval at 12 weeks and type of variability measure are unverified. The figure's S/S/L/M labels are not a verified MCID either.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Without exact variability and intervals, it cannot be determined whether important benefit or harm has been excluded.
Reason for missing information: SD/SE, n by arm, CI and MCID are unverified, and no new calculation is performed.
Required materials and conditions: Original numerical tables at 12 weeks, SD/SE definitions, analysis n by arm, CI and model output
Materials to obtain first: Original numerical tables at 12 weeks, SD/SE definitions, analysis n by arm, CI and model output
Further research: Report uncertainty intervals together with the variability/denominators used, and disclose the source data for the figure.
Remaining uncertainty: SD/SE, n by arm, CI and MCID are unverified, and no new calculation is performed.
Conditions for verification: Show the actual meaning of the reported variability and intervals and the scope of missing values.
Why this condition is needed: Needed to avoid arbitrarily deriving precision from graph bars or sample size.
Application exceptions: Do not use qualitative inspection of the original Fig.1 as digitization or reconstruction of SD/SE or CI.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M15 §15.3.1–15.3.3; custom application

Related sources S01 S02

N21 · Have the P value at 12 weeks and the criterion for a clinically important difference been verified?

Current findings: The exact P value at 12 weeks and MCID are retained as unverified (null). Only a statement of nonsignificance is available; without an interval, clinically important benefit cannot be excluded. The response/remission P values in S07 are not P values for total-score severity.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The effect, interval and scale criterion needed to judge importance are all missing.
Reason for missing information: The exact P value, CI, MCID and evidence establishing clinical_importance are unverified.
Required materials and conditions: Original exact P value, CI and effect at 12 weeks, and evidence for an important difference appropriate to the HDRS version/population
Materials to obtain first: Original exact P value, CI and effect at 12 weeks, and evidence for an important difference appropriate to the HDRS version/population
Further research: Future research should justify the important difference with evidence and disclose the range of effects allowed by the CI.
Remaining uncertainty: The exact P value, CI, MCID and evidence establishing clinical_importance are unverified.
Conditions for verification: Explain whether the exact P value and CI at 12 weeks can be compared with an evidence-based important difference.
Why this condition is needed: Needed to avoid declaring sufficient effect or precision from significance or participant counts alone.
Application exceptions: Do not fill in unverified P values or MCIDs using conventional thresholds or P values for other outcomes.
Referenced methods: M15 §15.3.1–15.3.3; custom application; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed

Related sources S01 S02 S07

N22 · Were favorable findings and statements of no difference both retained without exaggeration?

Current findings: The favorable signal at 6 weeks, the nonsignificant between-group result at 12 weeks and within-group improvement in both arms were all preserved. The different controlled comparison in S03 and the indirect positive result in S04 are also retained separately. Nonsignificance does not mean an effect of 0, equivalence, noninferiority or exclusion of important benefit.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The direction and applicability of the current reports were explained, without arbitrarily selecting only favorable findings.
Reason for missing information: Exact numerical values, intervals and direct eligibility remain unverified.
Required materials and conditions: All results by time, corrections, analysis definitions and exact original numerical tables
Materials to obtain first: All results by time, corrections, analysis definitions and exact original numerical tables
Further research: Future research should report all favorable/nonsignificant/unfavorable results according to prespecified time points and analyses.
Remaining uncertainty: Exact numerical values, intervals and direct eligibility remain unverified.
Conditions for verification: Preserve the signal at 6 weeks, nonsignificance at 12 weeks, within-group improvement and positive findings in other studies with their original attribution.
Why this condition is needed: Needed to avoid selecting only improvement at 6 weeks to erase the 12-week gap, or turning nonsignificance into an effect of 0.
Application exceptions: Positive findings at other time points, in other populations or on other scales are not automatically treated as conflicts or replications.
Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M15 §15.3.1–15.3.3; custom application

Related sources S01 S02 S03 S04

N23 · To what level of disclosure has independence been verified?

Current findings: S01 declares no external funding and no author conflicts, but internal support and product provision are unverified. Funding, product provision and COI in S03 are unverified. For S04, the IUMS grant 94-04-28-26840, support from the Sarem research center and the declaration of no COI are preserved.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Declarations were distinguished from actual verification gaps, and funding from another family was not assigned to the candidate trial.
Reason for missing information: Internal support/product provision and employment/patents in S01 remain incompletely verified, and funding/COI information in S03 remains unverified.
Required materials and conditions: Original funding, internal support, product provision, employee/patent relationships, declared COI and declarations in each report
Materials to obtain first: Original funding, internal support, product provision, employee/patent relationships, declared COI and declarations in each report
Further research: Future research should disclose public, institutional and product support and author relationships item by item, and distinguish them from independent replication.
Remaining uncertainty: Internal support/product provision and employment/patents in S01 remain incompletely verified, and funding/COI information in S03 remains unverified.
Conditions for verification: Attribute separately the declarations and internal-support/product gaps in S01, unverified details in S03 and funding in S04.
Why this condition is needed: Needed to avoid turning unverified funding or a declaration of no external funding into complete independence.
Application exceptions: Do not automatically exclude results because funding/COI exists, or conclude that bias is absent because it does not.
Referenced methods: Chamgap's own independence and attribution conditions. The official methodological sections used in this pilot do not validate all details of this item; M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application

Related sources S01 S03 S04

N24 · Were the original registration, outcome hierarchy, corrections and scope of unverified information compared?

Current findings: The registration identifier, timing and prespecified outcome hierarchy for S01 were not verified. A record states that no correction, retraction or expression of concern was found on the publisher page opened or in targeted searches at that time, but this is not an exhaustive integrity guarantee. There is no new registration or publication-status lookup in this review.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Without the original registration/protocol needed to compare plans and results, selective reporting is not determined.
Reason for missing information: S01 registration/plans/timing and current publication status are unverified.
Required materials and conditions: Original registration ID, prospective timing, protocol/SAP, outcome changes, correction/retraction notices and actual access events
Materials to obtain first: Original registration ID, prospective timing, protocol/SAP, outcome changes, correction/retraction notices and actual access events
Further research: Future research should compare plans and results by version and record reasons for changes and unreported outcomes.
Remaining uncertainty: S01 registration/plans/timing and current publication status are unverified.
Conditions for verification: Distinguish the registration/plan gap for S01 from the scope of the correction search at that time.
Why this condition is needed: Needed to avoid turning failure to obtain a registration into confirmed nonregistration or selective reporting, or a guarantee of no retraction.
Application exceptions: Do not retrospectively relabel the current manuscript's principal-time-point selection or historical targeted searches as preregistration or a complete, up-to-date integrity check.
Referenced methods: M08 Table 8.2.a, M04 §4.6.3; custom application; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources S01

N25 · What is the evidential level of the oral-administration, dose and SSRI information in the S07 review?

Current findings: S07 provides leads of oral 300 mg/day and fluoxetine 20 mg/day while recording that the formulation was not specified. Only selected portions of PDF pages 4–6 were inspected at that time; this is not a record of reading all other pages. Do not promote this information to verified route, product, dose or version values from the S01 original.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Tracing leads exist, but the original Methods needed for direct verification remain outstanding.
Reason for missing information: Comparison of the review's leads with the original report and the actual formulation/product and dose in S01 remain unverified.
Required materials and conditions: Relevant S07 tables/sections and original citations, and the S01 product, route, daily dose and scale Methods
Materials to obtain first: Relevant S07 tables/sections and original citations, and the S01 product, route, daily dose and scale Methods
Further research: Future material reviews should link each secondary (2nd) claim to a precise location in the original report and compare agreement.
Remaining uncertainty: Comparison of the review's leads with the original report and the actual formulation/product and dose in S01 remain unverified.
Conditions for verification: Keep the secondary (2nd) information on oral 300 mg/day, fluoxetine 20 mg/day and unspecified formulation distinct as recorded.
Why this condition is needed: Needed to prevent secondary (2nd) information from automatically resolving original-report eligibility gaps.
Application exceptions: The review's tables and descriptions are leads for tracing the original report, not direct verification of the S01 product/route/dose.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources S01 S07

N26 · In S03, do the 50 participants, 300 mg/day and P values represent the same 12-week effect?

Current findings: S03 studied 50 participants, adults with MDD aged 18–60 years in inpatient/outpatient care, adding thiamine 300 mg/day to standard treatment. It reports allocation by drawing lots and improvement in HRSD; the P-value sequence is .004, .031, .035, but the exact visits and effects of interest for the later values are unverified. Oral route, salt, single-ingredient composition, duration, scale version, n by arm and the placebo-controlled effect at 12 weeks were not verified.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The other controlled signal is preserved, but the dimensions needed to verify direct comparability with the selected question are insufficient.
Reason for missing information: The visit/analysis for each P value, the original effect at 12 weeks, CI, exact route/composition and treatment by arm are unverified.
Required materials and conditions: Full S03 Methods/results, n by arm, route/salt, principal time point, analysis attribution of each P value and SSRI/placebo information
Materials to obtain first: Full S03 Methods/results, n by arm, route/salt, principal time point, analysis attribution of each P value and SSRI/placebo information
Further research: Future syntheses should compare studies after verifying exact PICOT agreement and visit-specific estimates, without automatically creating an average.
Remaining uncertainty: The visit/analysis for each P value, the original effect at 12 weeks, CI, exact route/composition and treatment by arm are unverified.
Conditions for verification: Link the actual S03 population, standard-treatment control, HRSD and denominator/time-point gaps.
Why this condition is needed: Needed to avoid combining another controlled MDD study as an exact replication of the placebo comparison at SSRI initiation and 12 weeks.
Application exceptions: Do not arbitrarily assign P=.004/.031/.035 to particular visits, analyses or 12 weeks.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

Related sources S03

N27 · Are the date, issue and citation year for S03 consistent?

Current findings: The publication-date field for S03 is 2019-09-24, the volume/issue is Vol.36 No.1 (2020), and the citation year is 2019. The original record preserves all these fields without inventing a resolution. The two PDF links could not be accessed at that time, and the original PDF was not read in this review either.
Verification status: Conflicting information
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Because publication details differ within the same report, Conflicting information is retained only for the bibliographic scope.
Reason for missing information: The meaning of the dates in the canonical metadata and the correspondence between corrections/versions are unverified.
Required materials and conditions: The journal's canonical metadata, corrections to date/issue/citation information and publication events by version
Materials to obtain first: The journal's canonical metadata, corrections to date/issue/citation information and publication events by version
Further research: Record changes in bibliographic identity together with the original values; this is not an efficacy experiment.
Remaining uncertainty: The meaning of the dates in the canonical metadata and the correspondence between corrections/versions are unverified.
Conditions for verification: Preserve all of the published date 2019-09-24, the 2020 issue and the 2019 citation year.
Why this condition is needed: Needed to avoid changing the original record by overwriting different bibliographic years with a single value.
Application exceptions: Do not extend a bibliographic date conflict to a conflict in the clinical results or the entire treatment duration.
Referenced methods: M04 §4.6.1–4.6.2 distinction between studies and reports; cohort/outcome linkage is a custom application; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources S03

N28 · Is S04 a direct result for MDD inpatients at SSRI initiation?

Current findings: S04 is a trial in women with PCOS, infertility and IVF, aged 18–40 years, using oral thiamine for 4 weeks at 300 mg/day. It excluded diagnosed psychological disorders and did not require a depression entry threshold. The GHQ-28 depression subscale with 7 items and a 0–21 range differs from HDRS total at 12 weeks in adult MDD inpatients.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The indirect role and reasons for exclusion were clearly linked. Oral-administration information and the product manufacturer in S04 do not establish the identity of S01.
Reason for missing information: Direct results matching the selected MDD question and an MDD subgroup effect are unverified.
Required materials and conditions: Actual S04 eligibility, scale, duration, product, depression-subgroup data and original plan
Materials to obtain first: Actual S04 eligibility, scale, duration, product, depression-subgroup data and original plan
Further research: To answer the MDD question, separately design a direct comparison in the same MDD/SSRI-initiation setting with HDRS at 12 weeks.
Remaining uncertainty: Direct results matching the selected MDD question and an MDD subgroup effect are unverified.
Conditions for verification: Show the difference between PCOS/IVF at ages 18–40 years with GHQ at 4 weeks and MDD with HDRS at 12 weeks.
Why this condition is needed: Needed to avoid transferring results for other populations, scales or time points to the selected effect merely because the trial used oral B1.
Application exceptions: Retain the indirect study's original data, but do not turn exclusion of psychological disorders and the absence of a depression entry threshold into an MDD sample.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application

Related sources S04

N29 · Were S04 denominators and positive results correctly attributed?

Current findings: S04 randomized 35/35 participants, with 32/32 participants analyzed/completed. GHQ depression-subscale means (SDs) were 16.46 (SD 4.98) versus 15.84 (SD 5.53) at baseline and, at 4 weeks, 10.96 (SD 4.93) versus 16.25 (SD 4.87), with reported between-group P < .001. Baseline depression-domain scores ≥6 are retained as 30/28 participants, with a total of 6 participants below the threshold; the eligible depression-subgroup effect is unverified (null).
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The original numerical values, SDs and denominators are linked, but they are not the selected MDD effect or a direct comparison in the relevant subgroup.
Reason for missing information: Subgroup effects, re-extraction of CONSORT events, intervals/missing data and direct comparison of the multiple-domain analysis remain outstanding.
Required materials and conditions: S04 original Tables 2–3, CONSORT, analysis n, actual depression-subgroup results and missing-data handling
Materials to obtain first: S04 original Tables 2–3, CONSORT, analysis n, actual depression-subgroup results and missing-data handling
Further research: Indirect studies should separate overall/subgroup populations and randomized/completed counts, and present scale-specific estimates and intervals.
Remaining uncertainty: Subgroup effects, re-extraction of CONSORT events, intervals/missing data and direct comparison of the multiple-domain analysis remain outstanding.
Conditions for verification: Separately preserve the 35/35 randomized and 32/32 analyzed counts and the original GHQ means/SDs and P value.
Why this condition is needed: Needed to avoid confusing the 70 participants randomized with the 64 participants analyzed, or SD with total-score/subgroup results.
Application exceptions: Do not convert means on another scale into an HDRS effect or calculate new between-group differences/changes.
Referenced methods: M08 Table 8.2.a, M15 §15.3, M19 §19.1.2–19.2.4; design-specific adaptation is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources S04

N30 · How were S04 thresholds, percentages, outcome hierarchy and registration retained?

Current findings: In S04, the abnormal threshold >6 in the text differs from Table 2's ≥6 notation. The retained note records that the printed percentages for 6/32 and 26/32 appear arithmetically inconsistent, but the original values are not silently corrected. Differences in wording about the general-health/infertility outcome hierarchy between the abstract and Methods also remain, and the original IRCT201510266917N3 registration and prospective timing are unverified.
Verification status: Conflicting information
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The current conflicts are limited to the thresholds, reporting and hierarchy recorded in the retained manuscript. The table was not recalculated and the original registration was not checked in this review.
Reason for missing information: Actual comparison of the original tables/plans, corrections and registration timing remains unverified.
Required materials and conditions: S04 table/text corrections, protocol, original IRCT registration/versions, CONSORT and raw data
Materials to obtain first: S04 table/text corrections, protocol, original IRCT registration/versions, CONSORT and raw data
Further research: Future material comparisons should disclose the original definitions of thresholds, denominators and hierarchy and reasons for changes, and check sensitivity.
Remaining uncertainty: Actual comparison of the original tables/plans, corrections and registration timing remains unverified.
Conditions for verification: Preserve historical quality notes, gaps in the original registration record and differences in scale thresholds/hierarchy.
Why this condition is needed: Needed to avoid resolving original reporting inconsistencies and failure to obtain registration in a favorable direction.
Application exceptions: Do not silently correct original percentages or infer prospective timing from the digits in the IRCT identifier.
Referenced methods: M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed; M08 Table 8.2.a, M04 §4.6.3; custom application; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed

Related sources S04

N31 · Do the results in S16–18 directly answer the selected adjunctive-efficacy question?

Current findings: S16 is a mood/cognition study in healthy young women who were not selected for depression. S17 concerns an observational association between thiamine status and depressive symptoms in older Chinese adults, not a controlled supplementation effect. S18 is a single parenteral→oral case of malnutrition and deficiency neuropathy due to depression; no result is borrowed as controlled antidepressant efficacy.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Each material's exclusion/context role was linked to its original retained location. This is not a declaration that research is absent worldwide or that no direct effect exists.
Reason for missing information: The selected incremental effect of single-ingredient oral thiamine at 12 weeks in MDD is not verified in these materials.
Required materials and conditions: Population, design, exposure/route, endpoint and comparator in each original report, and directness to the selected MDD question
Materials to obtain first: Population, design, exposure/route, endpoint and comparator in each original report, and directness to the selected MDD question
Further research: Design questions about healthy people, correction of deficiency and observational associations separately, and distinguish them from the exact MDD adjunctive-treatment trial.
Remaining uncertainty: The selected incremental effect of single-ingredient oral thiamine at 12 weeks in MDD is not verified in these materials.
Conditions for verification: Show separately the differences among healthy young women, observations in older Chinese adults and a deficiency case treated by parenteral→oral administration.
Why this condition is needed: Needed to avoid combining mood in healthy people, nutritional-status associations and deficiency case reports into a placebo-controlled adjunctive effect in MDD.
Application exceptions: Do not delete indirect material, but do not invent numerical values for results that were not extracted.
Referenced methods: M03 §3.2.1; detailed population conditions are custom-designed; M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application; M03 §3.2.4; custom boundaries on extensions

Related sources S16 S17 S18

N32 · Are S19–21 HDRS results for ordinary thiamine alone?

Current findings: S19 studies a B1+B2 combination and anxiety, stress and sleep. S20 NUTRIMUM concerns pregnancy and multiple micronutrients; S21 concerns fatigue-focused material in primary biliary cholangitis. The retained record did not grade these as the selected MDD HDRS result or extract numerical values.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Differences in ingredients, populations and endpoints were preserved, without pooling them as a direct single-ingredient B1 effect.
Reason for missing information: Direct results for the selected question and the individual contribution of B1 are unverified.
Required materials and conditions: Ingredients, populations, screening scales, endpoints/times and ingredient-specific direct comparisons in each original report
Materials to obtain first: Ingredients, populations, screening scales, endpoints/times and ingredient-specific direct comparisons in each original report
Further research: If the contribution of B1 alone is needed, design a separate direct comparison with the other ingredients/treatments matched.
Remaining uncertainty: Direct results for the selected question and the individual contribution of B1 are unverified.
Conditions for verification: Explain the actual screening roles of combination/pregnancy studies, anxiety/stress/sleep and PBC fatigue.
Why this condition is needed: Needed to avoid attributing results for combinations, pregnancy or fatigue to the selected ingredient/total score.
Application exceptions: Do not decompose the contribution of B1+B2 or multiple micronutrients into B1 alone.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed; M03 §3.2.1; detailed population conditions are custom-designed; M04 §4.6.1–4.6.2, M15 §15.3.1; custom application; M03 §3.2.4; custom boundaries on extensions

Related sources S19 S20 S21

N33 · Were controlled adult results verified in NCT07793071?

Current findings: S15 preserves only the indexed title of NCT07793071 concerning depressive symptoms in youth. At that time, the official registration page was an app shell, the API failed and controlled adult results were not verified. Details from 3rd-party mirrors were not used as verified original data. Registration access was not attempted in this review.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The existence of a candidate identifier/title is not verification of the original registration or adult results.
Reason for missing information: Original registration identity/content, population ages, results and current status are unverified.
Required materials and conditions: The official registration's identity, population, ingredient, route, design, time points, results, versions and actual acquisition events
Materials to obtain first: The official registration's identity, population, ingredient, route, design, time points, results, versions and actual acquisition events
Further research: After actually obtaining the registration content, compare population/design eligibility and separately track unpublished results as well.
Remaining uncertainty: Original registration identity/content, population ages, results and current status are unverified.
Conditions for verification: Separate discovery of a candidate registration title from unverified original content/results.
Why this condition is needed: Needed to avoid promoting an indexed title, app shell or API failure into actual adult results.
Application exceptions: Do not use details from a 3rd-party mirror as the original registration, or equate youth with an adult population.
Referenced methods: M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M08 Table 8.2.a, M04 §4.6.3; custom application; M03 §3.2.1; detailed population conditions are custom-designed

Related sources S15

N34 · Does S01's statement of no adverse effects mean 0 events per arm?

Current findings: The S01 abstract states that there were no adverse effects in either group. Arm-level event/exposure counts, serious events, AE discontinuations and ascertainment methods are null or unverified. Do not turn this sentence into 0 events per arm or an incidence of 0.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The reported wording is preserved, but there are no data to verify frequency or complete ascertainment.
Reason for missing information: Exact events/people by arm, exposure denominators, ascertainment and discontinuations are unverified.
Required materials and conditions: S01 event/person counts by arm, exposure/follow-up n, duration, AE definitions/collection and reasons for discontinuation
Materials to obtain first: S01 event/person counts by arm, exposure/follow-up n, duration, AE definitions/collection and reasons for discontinuation
Further research: Future trials should report systematic harm ascertainment and arm-level people/events, exposure and discontinuations separately from efficacy.
Remaining uncertainty: Exact events/people by arm, exposure denominators, ascertainment and discontinuations are unverified.
Conditions for verification: Separate the original wording from unverified event/person counts, exposure n, discontinuations and ascertainment.
Why this condition is needed: Needed to avoid turning the abstract's absence statement into complete ascertainment or an incidence of 0.
Application exceptions: Do not arbitrarily use the total of 51 participants or 12 weeks as an exact exposure denominator or ascertainment scope.
Referenced methods: M19 §19.1.2–19.2.4; custom application; M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M19 §19.1.2–19.2.4, M04 §4.6.3; custom application

Related sources S01

N35 · Were suicide-related events, manic switch, worsening and long-term safety assessed?

Current findings: Suicide-related events, mania/hypomania and psychiatric worsening in S01 are unverified (null). Separate assessment, temporal relationships after exposure, discontinuation/re-exposure and long-term harms are unverified. Do not extend the candidate result at 12 weeks to long-term safety, rare events or safety in all populations, including pregnancy/lactation.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Symptom total scores and a general AE abstract alone cannot establish ascertainment, frequency or causality for separate events.
Reason for missing information: Direct data on psychiatric events, long-term/special populations and timing/discontinuation are unverified.
Required materials and conditions: Psychiatric-event definitions, systematic ascertainment, arm-level n/duration, worsening/mania/suicide and long-term follow-up data
Materials to obtain first: Psychiatric-event definitions, systematic ascertainment, arm-level n/duration, worsening/mania/suicide and long-term follow-up data
Further research: Collect psychiatric events using separate prespecified definitions and appropriate follow-up/surveillance, and separate them from efficacy total scores.
Remaining uncertainty: Direct data on psychiatric events, long-term/special populations and timing/discontinuation are unverified.
Conditions for verification: Check the scope of separate psychiatric ascertainment, exposure time and long-term/special-population information.
Why this condition is needed: Needed to avoid extending a general AE statement in a short-term abstract to psychiatric events or overall safety.
Application exceptions: Do not turn unverified suicide/mania/worsening into absence of events or a causal finding of worsening.
Referenced methods: M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M19 §19.2 design-specific harm questions; details of temporal relationships are custom-designed; M19 §19.1.2–19.2.4; custom application; Chamgap's own interpretive boundaries. Informed by M19 principles, not an official causality-tool implementation

Related sources S01

N36 · How are the nausea/adverse-effect statements in S03 and S04 attributed?

Current findings: S03 preserves abstract statements of nonsevere nausea and no serious adverse effects, but arm-level events/exposure n are unverified. S04 records a statement of no important adverse effects in the B1 arm and an exclusion criterion for B1 allergy. Arm-level events/exposure and independent re-extraction of CONSORT discontinuation reasons are unverified.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The statements are attributed, but safety frequencies in other populations/products cannot be directly transferred to the candidate.
Reason for missing information: Harm denominators/ascertainment, events, discontinuations and product correspondence by arm are unverified.
Required materials and conditions: S03/S04 original events/people by arm, ascertainment, exposure/duration, discontinuations, products and CONSORT data
Materials to obtain first: S03/S04 original events/people by arm, ascertainment, exposure/duration, discontinuations, products and CONSORT data
Further research: Safety syntheses should compare identities, populations, products and ascertainment, then separate unique events from frequencies.
Remaining uncertainty: Harm denominators/ascertainment, events, discontinuations and product correspondence by arm are unverified.
Conditions for verification: Distinguish populations, products and durations from the corresponding reports' harm statements and denominator gaps.
Why this condition is needed: Needed to avoid adding events from other studies to harms in S01 or counting them as replications for the same product.
Application exceptions: Do not create event counts, exposure denominators or 0 events from descriptions of nausea or no important AEs.
Referenced methods: M19 §19.1.2–19.2.4; custom application; M19 §19.1.2–19.2.4; detailed distinctions are custom-designed; M04 §4.6.1–4.6.2; adaptation for harm data is custom-designed

Related sources S03 S04

N37 · Does the Flamingo product information establish the candidate product and its harms?

Current findings: S06 is the SmPC for Flamingo's oral Thiamine Hydrochloride 100 mg Tablets. The frequencies of gastrointestinal adverse effects and hypersensitivity reactions are unknown, and hypersensitivity reports are retained as predominantly parenteral. The emc update of 2025-09-08 and text revision date of 2025-02-18 are historical version information, not identification of either MDD trial product.
Verification status: Partly verified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The separate product's general harm context was linked, but the candidate product and incidence cannot be established.
Reason for missing information: Correspondence with the actual candidate product, frequency in the oral trial and the current SmPC version are unverified.
Required materials and conditions: Documentation linking the candidate trial product to this product, and the actual SmPC/version and original harm data for the oral product
Materials to obtain first: Documentation linking the candidate trial product to this product, and the actual SmPC/version and original harm data for the oral product
Further research: Product-specific harm reviews should directly compare matched routes, products, exposed populations and event-occurrence/reporting systems.
Remaining uncertainty: Correspondence with the actual candidate product, frequency in the oral trial and the current SmPC version are unverified.
Conditions for verification: Separate the 100 mg oral product, historical version and unknown gastrointestinal/hypersensitivity frequencies from the unverified candidate product.
Why this condition is needed: Needed to avoid assigning a separate hydrochloride product's identity to the candidate trial's actual product/route.
Application exceptions: Do not transfer the frequency of hypersensitivity reported mainly with parenteral use into incidence in an oral trial.
Referenced methods: M03 §3.2.2–3.2.3; adaptation for reverse-direction markers and tests is custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M19 §19.2.2–19.2.4, M03 §3.2.3; custom application; M19 §19.1.2–19.2.4; custom application

Related sources S06

N38 · What do an unset UL and general nutritional/medication context mean?

Current findings: The S05 ODS summary records that no UL has been established, but the FNB cautions that excessive intake may cause adverse effects. The existence of sections on deficiency, at-risk groups and medication interactions provides general context, not a source of depression efficacy evidence. An unset UL does not mean unlimited safety.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The safety/efficacy interpretation boundaries of the general material were checked. This does not mean that a specific concomitant-treatment risk in MDD or an individual dose has been verified.
Reason for missing information: Direct interactions and long-term/excess-intake safety for the selected population/actual product are unverified.
Required materials and conditions: Original official nutrition/excess-intake, vulnerable-population and interaction documents, and direct harm data matching the population and exposure
Materials to obtain first: Original official nutrition/excess-intake, vulnerable-population and interaction documents, and direct harm data matching the population and exposure
Further research: Assess concomitant-medication/excess-intake risks in the selected MDD population using separate actual exposure data.
Remaining uncertainty: Direct interactions and long-term/excess-intake safety for the selected population/actual product are unverified.
Conditions for verification: Link the unset UL and excessive-intake warning in S05 to its role as context rather than efficacy evidence.
Why this condition is needed: Needed to avoid using the absence of a UL or deficiency context as evidence of unlimited safety or MDD efficacy.
Application exceptions: Do not derive personal recommendations or MDD treatment effects from general information about intake/medication relationships.
Referenced methods: M19 §19.1.2–19.2.4; custom application; Chamgap's own interpretive boundaries. Informed by M19 principles, not an official causality-tool implementation; M03 §3.2.1; detailed population conditions are custom-designed

Related sources S05

N39 · Does the total score at 12 weeks also answer questions about optimal dose, maintenance, response/remission and prevention?

Current findings: The current candidate material does not establish optimal dose, long-term maintenance, persistence after discontinuation, effects in a deficient subgroup or augmentation of a stable SSRI. Response/remission, relapse and suicide prevention are also not the total-score outcome at 12 weeks in this table. Direct estimates for these extensions are unverified within the retained scope.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Direct material needed for the extensions is unavailable, but this is not treated as failure of a mandatory event trial for the narrow total-score question.
Reason for missing information: Direct comparisons/durations, definitions and subgroup interactions for each extension are unverified.
Required materials and conditions: Direct dose comparisons, follow-up after discontinuation, prespecified subgroups/interactions and response/remission/relapse/suicide-event data
Materials to obtain first: Direct dose comparisons, follow-up after discontinuation, prespecified subgroups/interactions and response/remission/relapse/suicide-event data
Further research: Address each extension question in a separate study with a prespecified direct comparison, definition, time point and denominator.
Remaining uncertainty: Direct comparisons/durations, definitions and subgroup interactions for each extension are unverified.
Conditions for verification: Clearly distinguish the selected total score from unverified extension questions.
Why this condition is needed: Needed to avoid extending a narrow symptom comparison to dose selection, persistence or clinical events.
Application exceptions: Do not require every extension trial as a mandatory condition for constructing this table's direct 12-week question.
Referenced methods: M03 §3.2.1–3.2.4; details of dose, persistence and effect modification are custom-designed; M03 §3.2.4; custom boundaries on extensions; M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed

Related sources S01 S07

N40 · Were original materials needed to fill current gaps distinguished from methods for future trials?

Current findings: The first need is to obtain S01 original Methods, product/route, scale version, arm-level numerical values/CI at 12 weeks, analysis/flow, registration and harm data. Separately, methods are proposed for future research directly comparing single-ingredient oral ordinary thiamine in the same MDD/SSRI-initiation setting. No fixed sample size, dose, duration extension or efficacy grade is created.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: The additional materials in each row are gaps in the record, while future research is distinguished as a methodological proposal that has not yet been carried out.
Reason for missing information: Neither material acquisition nor a new trial has been performed; the original eligibility/effect gaps remain.
Required materials and conditions: S01 original Methods/product, HDRS version, arm-level tables/CI at 12 weeks, flow, registration/SAP, safety and corrections
Materials to obtain first: S01 original Methods/product, HDRS version, arm-level tables/CI at 12 weeks, flow, registration/SAP, safety and corrections
Further research: In a direct randomized placebo comparison that clearly defines the product/route, MDD criteria and SSRI initiation, prespecify HDRS at 12 weeks, arm-level n/missing data, CI, an important difference and harms, and disclose all results.
Remaining uncertainty: Neither material acquisition nor a new trial has been performed; the original eligibility/effect gaps remain.
Conditions for verification: Distinguish obtaining current original reports/plans from proposals for eligible direct comparisons and analysis/reporting.
Why this condition is needed: Needed to avoid presenting acquisition of a full text and conduct of new research as the same action.
Application exceptions: Do not adopt a new minimum sample size, number of trials, dose or MCID without justification.
Referenced methods: M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M08 §8.1–8.2.3; custom application; M03 §3.2.3, M08 §8.2.2–8.2.3; custom application; M08 Table 8.2.a, M03 §3.2.4; custom application; M08 §8.2.2, M15 §15.3.1; custom application; M15 §15.3.1–15.3.3; custom application; M08 Table 8.2.a, M04 §4.6.3; custom application

Related sources S01

N41 · Were the existing ?/null and calculator exception kept from being inherited as the new table's status?

Current findings: The original evidence cutoff is 2026-09-16, and this summary's review date is 2026-10-05. Preserve the original grade ?, uncalculated score (null), claim type B, six unscored downstream axes (null) and the no-human-study flag false. Preserve the original explanation that the calculator returned validation errors and a pre-validation B at that time, but B was not adopted. Original document quality A and single-assistant self-review are not independent clinical certification.
Verification status: Met
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Original values and the historical record of the exception were preserved, without newly inheriting statuses/scores, calculating or reassessing.
Reason for missing information: The stage 0 native-calculator representation and original eligibility/effect at 12 weeks remain unverified.
Required materials and conditions: Original calculation inputs/errors, current-value exception, verification records and authoritative records of new evidence/condition-change events
Materials to obtain first: Original calculation inputs/errors, current-value exception, verification records and authoritative records of new evidence/condition-change events
Further research: If eligibility or the representation of rules is updated, separate the event, conditions and adjudication procedure, and preserve the entire previous original text.
Remaining uncertainty: The stage 0 native-calculator representation and original eligibility/effect at 12 weeks remain unverified.
Conditions for verification: Preserve the original cutoff and review dates, original grade/score, axes, exception and historical self-review.
Why this condition is needed: Needed to avoid turning an uncalculated value into no effect, no human research or a score of 0 points.
Application exceptions: Do not adopt the historical calculator's validation error/pre-validation B as a new valid grade, or rerun the calculator.
Referenced methods: M04 §4.5, §4.6.3; limitations of the current material conversion follow the user-defined scope and custom design; M03 §3.1, §3.2.1–3.2.4 principles for clarifying the question; detailed separation rules are custom-designed; M04 §4.5–4.6 principles for collecting/selecting materials; the separation of access from reading and the hyperlink contract are custom-designed; M04 §4.6.3, M15 §15.3.1–15.3.3; separation of statuses is custom-designed

Related sources

N42 · How were analyses across multiple visits/scales and the adjusted model prespecified?

Current findings: The adjusted/unadjusted effect of interest and multiplicity-handling method in S01 are unverified. Visits at 0, 3, 6 and 12 weeks are preserved, but the repeated-measures model, covariates, primary/secondary endpoints and plans for multiple comparisons across time points were not verified. Do not reconstruct a statistical model from the difference between nonsignificance at 12 weeks and a favorable result at 6 weeks.
Verification status: Unverified
Materials actually read: Summary read: the retained stored summary at the corresponding exact location was actually read. External clinical/official documents and old separate files were not read in this review.
Why unverified: Without material comparing the analysis plan and actual model, how the selected estimate was handled cannot be established.
Reason for missing information: The SAP, adjusted model, covariates, adjustments for multiple visits/endpoints and reasons for changes are unverified.
Required materials and conditions: Original protocol/SAP, analysis model, covariates, repeated-visit/multiplicity handling, effect of interest at 12 weeks and reasons for changes
Materials to obtain first: Original protocol/SAP, analysis model, covariates, repeated-visit/multiplicity handling, effect of interest at 12 weeks and reasons for changes
Further research: Future direct comparisons should prespecify the effect of interest at 12 weeks, primary endpoint, visit-specific comparisons, covariates, multiplicity handling and sensitivity to missing data.
Remaining uncertainty: The SAP, adjusted model, covariates, adjustments for multiple visits/endpoints and reasons for changes are unverified.
Conditions for verification: Explain the actual adjusted/unadjusted effect of interest, handling of multiple time points/endpoints and unverified plans.
Why this condition is needed: Needed to avoid mixing comparisons at 3, 6 and 12 weeks and selection of favorable analyses into the same 12-week estimate.
Application exceptions: Without the original SAP, do not infer repeated-measures methods, multiplicity adjustments or covariate models.
Referenced methods: M08 §8.2.2, M15 §15.3.1; custom application; M08 Table 8.2.a, M04 §4.6.3; custom application; M15 §15.3.1–15.3.3; custom application; M03 §3.2.4, M15 §15.3.1–15.3.3; question-specific outcome adaptation is custom-designed

Related sources S01 S02