CHAMGAP
Verdict No. 3124 · Search date 2026-09-16 · Methodology v1.0

Riboflavin,
does it really help with 161.3-km race finishing time in experienced adult ultramarathon runners?

30-Second Summary
D
Evidence Grade D · 30 · Safety caution
ChatGPT source review and self-verification · Codex technical integration
The calculated descriptive mean contrast is riboflavin minus placebo +0.18 hours (+10.8 minutes), not causal evidence of harm. Original CI, exact P and adjusted estimate are null. Assumption-dependent selected-sample reconstructions are recorded separately. [S01; calculations.json]
Safety is Caution, separate from efficacy. The paper reports two riboflavin participants with emesis within an hour of the 90-km dose and urine-color observations (3 versus 1), but systematic arm-level adverse-event, serious-event and harm-withdrawal denominators were not verified. Causality of emesis is not established; unreported events are not zero. NIH’s lack of a tolerable upper intake level reflects limited harm data, not unlimited high-dose safety. Pregnancy, lactation, children, renal/hepatic disease, long-term high doses and laboratory interference were not adequately characterized by this study. Study doses are not personal dosing advice and do not replace assessment or treatment of symptoms during endurance exercise. [S01 Results p4; S03 excess-intake/interaction sections]
What the
research shows
In a small placebo-controlled study at a 161.3-km race in 2016, finish times did not differ significantly between riboflavin and placebo. The Table 1 study-completer means were 27.26 and 27.08 hours in 18 and 14 runners. Alternating allocation and selected-finisher analysis do not establish no effect, equivalence or exclusion of important benefit. [S01 Table 1 p4]
What the
ads claim
Specific advertising was not systematically investigated. This assessment does not substantiate endurance enhancement in all athletes, equivalence of all B-vitamin forms, or prolonged high-dose use.

Four separate assessment dimensions

Effect direction and sizeThe calculated descriptive mean contrast is riboflavin minus placebo +0.18 hours (+10.8 minutes), not causal evidence of harm. Original CI, exact P and adjusted estimate are null. Assumption-dependent selected-sample reconstructions are recorded separately. [S01; calculations.json]
Evidence certaintyConfidence is very limited even for the narrow race-time question. This is a narrative description of allocation, missingness and estimand limits, not a formal GRADE category.
Applicabilityexperienced adult ultramarathon runners, 2016 WSER; finish-time data restricted to study completers
SafetySafety is Caution, separate from efficacy. The paper reports two riboflavin participants with emesis within an hour of the 90-km dose and urine-color observations (3 versus 1), but systematic arm-level adverse-event, serious-event and harm-withdrawal denominators were not verified. Causality of emesis is not established; unreported events are not zero. NIH’s lack of a tolerable upper intake level reflects limited harm data, not unlimited high-dose safety. Pregnancy, lactation, children, renal/hepatic disease, long-term high doses and laboratory interference were not adequately characterized by this study. Study doses are not personal dosing advice and do not replace assessment or treatment of symptoms during endurance exercise. [S01 Results p4; S03 excess-intake/interaction sections]

Chamgap rule-based score, not success probability or official GRADE

*

Useful facts when choosing a product

  • Capsules were reported as a single active nutrient, riboflavin; exact salt, batch and complete excipient certificate remain unverified. [S01]
  • The population was not established as deficient or replete using baseline measurements. [S01]
  • 5 mg of 10% beta-carotene powder corresponds to a nominal calculated 0.5 mg active amount. [S01; calculations.json]
ID

Chamgap Semantic Classification Code

Permanent code issued

S.riboflavin.oral-capsule-100mg-prerace-100mg-90km-one-event.experienced-adult-ultramarathon-runners-161-3km-finish-time.shorten.matched-color-placebo-maltodextrin-beta-carotene

Substances > Riboflavin (vitamin B2) > Oral capsule 100 mg pre-race plus 100 mg at 90 km in one event > 161.3-km finish time in experienced adult ultramarathon runners > Shorten > Matched-color maltodextrin/beta-carotene placebo

The descriptive study-completer contrast of +0.18 hours (+10.8 minutes) from 27.26 versus 27.08 hours is retained; alternating allocation and selection preclude relabeling it as no effect, equivalence or causal harm. The code identifies the intervention and claim; it never contains the evidence grade, score, or safety result.

Download semantic index (JSON) · Codebook v1

Exact Claim Classification

These independent facets prevent evidence from different forms, routes, populations, effects, and comparators from being mixed.

Intervention classS · Supplement or nutraceutical
Canonical ingredient or interventionRiboflavin (vitamin B2), ordinary form as reported
Source or part usedIsolated nutrient; manufacture/source organism or tissue not reported; botanical part not applicable
Formulation or processingInvestigator-prepared capsule containing reported single active nutrient riboflavin; full batch/excipient certificate unverified
Routeoral
Dose100 mg before start plus 100 mg at 90 km; planned total 200 mg in one event
Durationone 161.3-km ultramarathon event; not a daily multiweek course
Populationexperienced adult ultramarathon runners, 2016 WSER; finish-time data restricted to study completers
Effect or conditionfaster endurance race completion
Primary endpoint161.3-km race finish time in hours at the race finish; Table 1 selected completers
Comparatormatched-color placebo with 95 mg maltodextrin and 5 mg 10% beta-carotene powder per capsule; race care/diet not standardized
Duplicate-detection keyS|riboflavin_as_reported|oral|100mg_pre+100mg_90km|WSER2016_161.3km|adult_experienced_finishers_complete_data|race_finish_hours|colored_placebo|supplementation
01

What the research actually shows

This page concerns endurance finishing time in one race. A standardized laboratory time trial, 400-m recovery run, pain, CK and vitamin-status measures are separate; favorable recovery results are not converted into a race-time benefit. [S01 Methods/Results] Assignment alternated in order of arrival at study registration; it was not a random sequence. Of 44 starters (22/22), 37 finished (20/17), but the Table 1 times cover only 32 participants with complete study data (18/14). Baseline nutrient status, diet, adjusted effect and a prospective primary-outcome hierarchy remain unverified. [S01 pp2–6] The original reports oral riboflavin 100 mg 0.5–1 hour before the start and another 100 mg at 90 km. The planned total was 200 mg in one race, not a long-term daily 200- or 400-mg regimen. Each placebo capsule contained 95 mg maltodextrin plus 5 mg of 10% beta-carotene powder. [S01 Methods p2]

02

Why this is classified as D (30)

Axes B/P/R1/I2/E0/CX/B2 → unchanged native calculator D → fixed 30 anchor for one strength (I2). R1 follows case 31 for one comparable family, not a claim of confirmatory randomization. E0 is the nonsignificant-result pathway, not proof of exact zero. I2 classifies disclosed foundation/public support.

Counterpoint. No independently replicated same-boundary race endpoint was verified. Pain, recovery, swimming, combinations and status markers are not counted as replication. Unverified studies are not treated as nonexistent.

Rejudgment record. One same-boundary family; different endpoints not combined — Supplied rubric and unchanged calculator; primary-source checking and self-review

Stored scoring profile
EndpointPSymptom or function itself is the target - including patient reports and performance tests
Case application: Race finishing performance is the functional target itself: P, not restricted to patient self-report and not a biochemical surrogate
ReplicationR1Single confirmatory trial
Case application: One directly comparable family; case 31 prescribes R1 when adjacent studies are not comparable. The native single-confirmatory-trial label is qualified: this is an alternating-allocation preliminary study, not a randomized or confirmatory trial
IndependenceI2Decisive evidence is publicly or non-profit funded
Case application: I2 from directly read foundation/public support and conflict declarations. Full donor provenance, funder roles and product provision remain unreported; this is not independent replication or external review
Effect sizeE0Null
Case application: E0 rubric pathway for the reported nonsignificant Table 1 comparison; not literal zero, equivalence, or exclusion of important benefit
PrecisionCXNo pooled confidence interval could be confirmed
Case application: CX because original CI and adjusted causal estimand are unverified. Reconstructed selected-sample t intervals are assumption-dependent descriptive quantities, not a C1 justification

Stored derived and displayed grades match; this is not a current recalculation or validity check (D).

Review performed and remaining limitations

Original article, all eight PDF pages, numbers, rubric and bilingual self-review; not external independent clinical peer review

Batch certificate, exact salt and full excipients Baseline riboflavin marker/deficiency threshold and total dietary intake Exact eligibility age bounds, renal/hepatic/pregnancy subgroup status Race clock/device and timing adjudication details Original CI/exact P, adjusted effect and baseline performance balance Registration/protocol/SAP and prospective finish-time hierarchy Outcomes of all 44 starters in a valid assigned-population estimand Diet and race-care balance, conflicting placebo medication numerator Funder roles/donor provenance and product provision Systematic arm-level adverse-event/serious-event/withdrawal denominators; long-term/special-population safety Inaccessible 1984 swimmer original methods/results Failed native database searches and no exhaustive unreported-trial assurance

Preliminary RoB 2-informed review — not a complete formal assessment
RandomizationAllocation alternated by registration arrival, hence predictable and not a random sequence.
Deviations from assigned interventionsSecond-dose omission/emesis and pain-medication use reported; full planned exposure not established in every participant.
Missing outcome dataOf 44 starters, Table 1 includes 32 (18/14). Seven nonfinishers and five finishers with incomplete study data are distinct. Selected-finisher comparison does not estimate the effect in all assigned participants.
Outcome measurementObjective fixed-distance race time, but timing-device details and baseline equivalent-course performance adjustment unreported.
Selection of the reported resultReport focuses on pain/recovery; finish time is a characteristics-table row. Registration and planned hierarchy unverified, so it is not relabeled the primary efficacy endpoint.
Reasons for the certainty judgment — not formal GRADE
Risk of biassubstantial limitations
Inconsistencynot estimable from one comparable family; noncomparable evidence is not contradictory replication
Indirectnessfield-race selected completers only; original task broader than verified endpoint
Imprecisionauthor interval unverified; reconstructed selected-sample intervals span faster and slower times
Publication biasnot assessable with one family and limited search; absence of unreported studies not established

Search scope and limitations. 63 general-web queries; 3 native database search attempts failed; not exhaustive-search assurance

03

Evidence Table

StudyDesignSampleFundingEndpointResultWeight
Hoffman et al. 2017 placebo-controlled preliminary riboflavin study at the 2016 WSER ultramarathonProspective parallel placebo-controlled preliminary trial; alternating allocation by registration arrival, not a random sequence; participant and onsite-researcher blinding reported44 alternately assigned and started (22/22), 37 finished (20/17), and Table 1 finish times cover 32 complete-data study completers (18/14); no ITT finish-time estimateSupport reported from the WSER Foundation, Rossi Family Foundation and US Department of Veterans Affairs resources/facilities; no industry support reported and authors declared no competing interests; funder roles, product provision and all donor provenance unverified161.3-km 2016 race finishing time in hours among Table 1 study completers; not verified as prospectively primary and not a standardized laboratory time trial or 400-m recovery runRiboflavin 27.26±2.36 hours (n=18) versus placebo 27.08±2.88 (n=14), mean±SD; no significant Table 1 group difference. Descriptive contrast +0.18 hours (+10.8 minutes); author exact P and CI unverifiedOnly directly comparable family at the locked finish-time boundary; nonsignificance is not extended to no effect, equivalence, causal harm or endurance in all athletes
§

Receipt — 10 References

Evidence access cutoff: 2026-09-16. Each citation states its actual access level (full-text transcription, abstract, index, or other) and remaining limitations. Bibliographic checking does not certify the study results or treatment efficacy.

Reference 1
checked
Reference 2
checked
Reference 3
checked
Reference 4
checked
Reference 5
checked
Reference 6
checked
Reference 7
checked
Reference 8
checked
Reference 9
checked
Reference 10
checked
Restricted to study-completer finishing time in one 161.3-km race among experienced adult ultramarathon runners; not standardized time-trial, 400-m recovery, all-athlete or long-term-dosing evidence.
Technical integration by: Codex · Evidence date: 2026-09-16 · Corrections: 7

Correction log — 7

Corrections applied to this verdict, in chronological order. Changes are logged, not erased.

  • · —
  • · —
  • · —
  • · —
  • · —
  • · —
  • · —

Cite this verdict

Riboflavin and 161.3-km ultramarathon finishing time: current evidence D·30 Evidence Grade D card
[Chamgap] Riboflavin and 161.3-km ultramarathon finishing time: current evidence D·30 — Evidence Grade D·30. 10 cited sources checked. Source: https://chamgap.com/en/verdicts/sports/oral-riboflavin-ultramarathon-finishing-time-161km/ · CC BY 4.0

CC BY 4.0 — free to use with attribution; do not distort grades, numbers, or verdict meaning.

!

What this document does and does not do

Chamgap is an information source. It reports what research has and has not confirmed; it does not tell readers what to take or buy. That decision belongs to readers and, when needed, medical or legal professionals. This verdict reflects literature available up to the search date and may change as new research appears. Nothing here is medical advice.