Vitamin B12 and insomnia symptoms

What the research actually shows

Vitamin B12 and insomnia symptoms — current Chamgap assessment

TASK-1053 / R01-093 / existing site ID 3168 · operation=update · current review date 2026-09-21

1. Thirty-second answer

In the closest human record available, a cohort reported as 238 patients with vitamin B12 below 200 ng/L had a mean Pittsburgh Sleep Quality Index (PSQI) score of 9.77 before B12 treatment and 4.31 about two months later. Lower PSQI is the better direction, so sleep quality was reported to improve within this cohort.[S01]

The −5.46 points are only the arithmetic difference between two reported means. The accessible abstract does not report a concurrent comparator, and the actual route, dose, formulation, full active composition, adult-only scope, insomnia diagnostic entry criteria, complete paired denominator, controlled effect and confidence interval remain unverified. The change therefore is not treated as a comparator-adjusted B12 effect, a demonstrated insomnia treatment effect or an MCID response.[S01,S02]

Current assessment: grade=null, score=null, grading_status=not_applicable_or_policy_missing. This is not a ?/no-human-study decision: human PSQI data exist. Current faceted-classification rule 4.1 requires mandatory claim-identity fields such as intervention, administration route, population, efficacy claim, core endpoint and comparator for a normal efficacy verdict. The principal record does not establish the actual administration route and does not establish that the analyzed population is adults with insomnia. Rather than infer oral treatment, dose, formulation or an effect axis from other studies, this update remains a completed ungraded evidence report.

2. Separate the planned question from the observed evidence

The question is whether vitamin B12 improves insomnia symptoms in adults with insomnia symptoms. The planned boundary included sole oral B12 and an appropriate comparator. Planned values were not converted into study facts.

The principal 2024 record reports 238 patients with serum B12 below 200 ng/L. This count is not relabeled 238 adults with diagnosed insomnia or 238 complete paired PSQI cases. Its conclusion names cyanocobalamin, but that molecular name does not establish oral, intramuscular or intravenous administration, a single-ingredient product, dose, frequency or clock time.[S01,S02]

The descriptive endpoint is the mean PSQI score about two months after treatment. PSQI is a patient-reported sleep-quality measure in which lower scores are the better direction. It is not relabeled ISI, AIS, objective sleep monitoring or a verified registered primary endpoint.

3. What the principal record actually reports

The change is sizeable descriptively, but a same-cohort before/after record cannot by itself separate B12 from passage of time, cointerventions, lifestyle changes, regression to the mean or other changes during deficiency correction. These are possible sources of confounding, not verified mediators in this cohort.

4. Why other human records are not counted as matching replication

• Djokic 2019 [S03] studied 60 insomnia patients but used a multi-ingredient product containing melatonin, magnesium, B6, folate and B12. Its AIS result does not isolate B12.
• Bouloukaki 2023 [S04] was a cross-sectional study of 512 adults. Low B12 was associated with AIS insomnia symptoms, while the PSQI result in the same sample was not significant. This is not supplementation efficacy.
• Takahashi 1999 [S05] compared high- and low-dose methylcobalamin in delayed sleep-phase/non-24-hour sleep-wake disorders, not ordinary insomnia versus placebo.
• Mayer 1996 [S06] compared cobalamin forms in healthy volunteers; alerting/shorter-sleep signals are not insomnia-treatment evidence.
• Infant RCT 2022 [S07] measured actigraphic sleep in infants and is not adult-insomnia replication.

Human B12/sleep evidence therefore exists, but the populations, interventions, comparators, scales and purposes differ too much to count these as independent replications of the principal two-month PSQI claim.

5. Application of the current rules

? — not applicable

The scoring rule reserves ? for a claim endpoint with no human study at all, supported by reproducible search records. Human PSQI data do exist here, so no_human_study=false; ? is not used.

Mandatory claim identity — the operative upstream limitation

Current rules/faceted_classification.md section 4.1 requires a normal efficacy claim to identify the intervention, route, population, efficacy claim, key endpoint and comparator, among other mandatory fields. The principal record establishes B12 treatment and PSQI before/after values, but the actual route is unverified and directness to an adult-insomnia population is unverified. It is therefore not converted into the planned oral-B12/adult-insomnia efficacy input.

Observational-evidence workaround — reviewed but does not fill identity gaps

The current score sheet explicitly says observational evidence should not automatically be treated as worthless or ungradable; when observational evidence is decisive, design limitations should be reflected in the bias axis rather than by fabricating a weaker effect code. That rule was reviewed. It does not create an unreported route, formulation or direct population, so it does not resolve this upstream identity problem.

Unknown funding and the I1 rule

Funding, product supply and conflicts for the principal study were not verified. The score sheet says not to assign independent I2 when funding/conflict information has not actually been opened; the fallback is I1. Because this record is excluded from normal grading at the claim-identity gate, I1 is not submitted as a calculator input, but unknown funding is explicitly not promoted to I2.

6. Why the seven numerical axes remain null

This is a completed report excluded before normal efficacy calculation, not a malformed calculator input used to obtain a raw grade.

• claim_type: the planned question is clinical efficacy, but a directly identified oral-B12/adult-insomnia claim is not established.
• endpoint: two-month PSQI is retained descriptively and is not relabeled ISI, AIS, objective sleep or a registered primary.
• replication: combination, serum-level, circadian, healthy-volunteer and infant records are not counted as matching replication.
• independence: unknown funding is not I2; I1 would be the fallback if normal grading were reached, but no execution axis is created here.
• effect: −5.46 is a within-cohort difference of reported means, not a comparator effect; no E+/E0/E− is invented.
• bias: before/after observation and subjective outcome limit causal attribution, but inaccessible details are not fabricated into a defect count.
• precision: no controlled-effect CI/SE or paired covariance is available; precision or equivalence is not calculated.

7. Calculator and arithmetic records

The unchanged current rules/calculator.py was read and fingerprinted, but no clinical grading function or CLI was executed because mandatory claim-identity fields were not sufficient for a normal calculator input. Nonexecution is not recorded as PASS, zero errors or exit code 0.

The only arithmetic reproduced is:

4.31 − 9.77 = −5.46

This is not a new study statistic or a controlled treatment estimate. No paired test, confidence interval, standardized effect, MCID or NNT was reconstructed.

8. Evidence limitations

1. The principal record is a low-B12 patient before/after report; the accessible abstract does not report concurrent comparator information.
2. The 238 count is not confirmed as 238 adults with insomnia or 238 complete paired PSQI cases.
3. Actual B12 route, dose, formulation and full active composition are unverified.
4. Insomnia diagnosis, ISI/AIS entry criteria and chronicity are unverified.
5. Exact PSQI P value, controlled-effect CI, paired-change variance and applicable MCID are unavailable.
6. Concomitant sleep, behavioral, mood, light, lifestyle and other treatment changes are unverified, so improvement cannot be attributed solely to B12.

9. Evaluation-framework limitation

The framework does include an observational-study bias workaround and an I1 fallback for unknown funding, so this is not “ungraded simply because it is observational.” The upstream problem is the mandatory claim-identity boundary in section 4.1: route and direct population identity are needed to define the claim without silently inventing facts.

This ungraded status therefore means evidence exists, but the available record cannot be mapped to the planned normal efficacy claim without creating unreported study details. It is not a finding of no effect.

10. Work not completed in this update

The 2026-09-21 update reused the supplied prior research and access records. No new comprehensive literature search or new S01 full-text retrieval was performed in this cycle. Prior records show the publisher full text was inaccessible with HTTP 403.[S02]

Still unverified are the actual route, dose, formulation, age range, insomnia diagnosis, paired denominator, registration/original primary outcome, funding/product support, exact endpoint P/CI, detailed cointerventions and adverse-event denominators. These remain “unverified/not reported,” not “absent.”

11. Current conclusion

What can be said: a cohort reported as 238 patients with low B12 had a mean PSQI score of 9.77 before B12 treatment and 4.31 about two months later—a within-cohort change in the better direction.

What cannot be established: that B12 itself caused the change; that the same effect applies to adults with diagnosed insomnia; that adding B12 in nondeficient adults treats insomnia; or which route, dose or formulation would be effective. Lack of controlled comparison is neither proof of no effect nor proof of efficacy.

12. Safety and application

Safety is unknown for the principal sleep record because adverse-event ascertainment, denominators and withdrawals were not verified. Reused MHRA/NIH general B12 safety context is not converted into a study-specific safety rate.[R01,R02]

Correction of deficiency is different from adding B12 to a nondeficient adult for insomnia. The B12 bottle shown in explanatory images is a conceptual illustration, not verification of the study product, formulation or route. This report is not individual dosing, route, timing or treatment-change advice.

13. Revision triggers

The same ID 3168 can be reassessed if any of the following becomes available:

• full text or original data establishing route, dose, formulation, ages, insomnia diagnosis, paired denominators and comparator details;
• a controlled sole-B12 study directly matching adults with insomnia symptoms;
• a prespecified controlled effect with uncertainty estimates;
• an official change to the claim-identity rules that validly maps the present evidence into a normal efficacy input.

Current status is completed_with_uncertainty, editorial status ready_with_uncertainty, with null grade and score. Existing ID 3168, slug, category, URLs and first-publication timestamp are preserved. No server deployment was performed in this Chat.

Sources

[S01] Durmaz A, Topaloğlu M. Effects of Vitamin B12 on the Quality of Sleep and Life. 2024. DOI 10.22034/ircmj.2024.199361. Primary abstract accessed through the authors’ institution.

[S02] Publisher article/PDF access attempt for the same paper; full text/tables inaccessible with HTTP 403.

[S03] Djokic G, et al. The Effects of Magnesium - Melatonin - Vit B Complex Supplementation in Treatment of Insomnia. 2019. PMID 31850132.

[S04] Bouloukaki I, et al. Association of Vitamin B12 Levels with Sleep Quality, Insomnia, and Sleepiness in Adult Primary Healthcare Users in Greece. 2023. PMID 38063594.

[S05] Takahashi K, et al. Double-blind test on the efficacy of methylcobalamin on sleep-wake rhythm disorders. 1999. PMID 10459691.

[S06] Mayer G, et al. Effects of vitamin B12 on performance and circadian rhythm in normal subjects. 1996. PMID 8914118.

[S07] The effect of vitamin B12-supplementation on actigraphy measured sleep pattern; a randomized control trial. 2022. PMID 34999324. Infant population.

[R01] MHRA vitamin B12/cobalt-allergy safety update, 2023; reused prior safety extraction.

[R02] NIH Office of Dietary Supplements, Vitamin B12 Fact Sheet for Health Professionals; reused prior safety extraction.