CHAMGAP
APPROVEDReviewed and approved by the Chamgap Editorial Team (2026-08-03). The draft was written by AI, the existence of all 3 cited sources was verified at the original page, and the verdict passed blind grading and adversarial audit. Methodology v0.6.
Verdict No. 2017 · Search date 2026-08-03 · Methodology v0.6

Tofu at least once weekly,
does it really help with Lower incidence of coronary heart disease?

30-Second Summary
C
Evidence Grade C · 48 · Safety acceptable
Frequent tofu consumers had fewer coronary events, but the evidence is observational
Tofu is generally safe in ordinary food amounts. People with soy allergy should avoid it, and tofu should not replace indicated cardiovascular treatment.
What the
research shows
The grade is C. People reporting tofu at least once weekly had lower coronary heart disease incidence than those reporting less than once monthly, HR 0.82, 95% CI 0.70 to 0.95. This was an observational cohort, not a trial assigning tofu, so causation is not established.
What the
ads claim
The accurate statement is that lower incidence occurred among people reporting frequent tofu intake. Saying that feeding tofu prevented coronary disease turns observation into experiment.
*

Useful facts when choosing a product

  • Tofu does contain isoflavones. In the USDA database, one category of firm silken tofu averaged 29.97 mg total isoflavones per 100 g, with substantial variation across products.
  • The 29.97 mg per 100 g value is a general measurement for USDA samples, not a direct assay of tofu eaten by cohort participants.
  • The presence of isoflavones does not establish that tofu acts identically to human estrogen.
  • The cohorts were supported by NIH grants, and no soy-association funding was identified.
  • Verdict 063 is C with 46 points for isolated soy isoflavones and menopausal symptoms, while verdict 513 is C with 58 points for soy protein and LDL reduction. Their interventions and outcomes differ from whole tofu and coronary incidence here.
  • Tofu has a long East Asian food history, but no exact classical date or quotation was verified.
Gap Measurement · Verdict 2017 · C 48
What advertising claims
What independent, higher-quality research supports
△ GAP
01

What the research actually shows

The 2020 Circulation paper, supplement, and HPFS dietary-validation report were cross-checked. Nonfatal myocardial infarction and coronary death were adjudicated from medical and death records. The 0.56 correlation with diet records applied to a combined tofu and soybean item, not tofu alone. Bias concerns include self-reported food-frequency exposure, limited item validity, residual confounding by health behavior and education, and multiple comparisons across tofu, soy milk, isoflavones, and female subgroups.

02

Why this is classified as C (48)

A publicly funded large prospective observational cohort found a binary clinical endpoint association whose confidence interval excluded 1. Self-reported food-frequency exposure, limited item validity, residual confounding by health behavior and education, and multiple comparisons raise the bias level, yielding C with 48 points.

Counterpoint. A long trial assigning tofu replacement against another protein source would be more direct but difficult to conduct for coronary events.

Rejudgment record. Cross-check applied — Assessed adjudicated events, confidence intervals excluding 1, self-reported food-frequency exposure, validity of the combined tofu and soybean item, residual confounding, and multiple comparisons

Scoring profile behind this grade
EndpointHHard endpoint - actual events such as death
ReplicationR1Single confirmatory trial
IndependenceI2Decisive evidence is publicly or non-profit funded
Effect sizeE+Meets the clinically important threshold

The scoring table and the verdict agree (C).

Sub-claim grades by effect

This ingredient is marketed for several effects. A single overall grade blends strong and weak claims together, so each effect is graded separately here. The overall grade reflects the strongest disconfirming or core claim.

Effect (sub-claim)GradeBasis
Eating tofu at least weekly prevents coronary heart diseaseCAn inverse association was observed, but causation is unconfirmed.
Tofu contains isoflavonesCThis is a verified composition fact, separate from coronary efficacy.

Cross-check — Codex and Claude

This verdict was drafted by Codex through literature review and source-existence checks, cross-checked through blind grading and adversarial audit, and settled by reapplying the methodology boundary rules. Cases with split grades were resolved through rejudgment.
03

Evidence Table

StudyDesignSampleFundingEndpointResultWeight
Study 1Pooled observational analysis of three prospective cohorts8,359US National Institutes of HealthNonfatal myocardial infarction or coronary heart disease deathAt least weekly versus less than monthly tofu HR 0.82 (0.70 to 0.95); confidence interval excluded 1Large clinical-endpoint analysis, but a single observational cohort
§

Receipt — 3 References

All 3 cited sources were verified for existence at the original page (as of 2026-08-03).

Ma L, et al. Isoflavone Intake and the Risk of Coronary Heart Disease in US Men and Women. Circulation. 2020;141:1127-1137.
checked
Feskanich D, et al. Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire. J Am Diet Assoc. 1993;93:790-796.
checked
USDA Database for the Isoflavone Content of Selected Foods, Release 2.0.
checked
Draft and rewrite: Codex (AI) · Verification: Codex blind grading and adversarial audit · Final adjudication: Claude
Reviewed and approved: Chamgap Editorial Team · Approval date: 2026-08-03 · Corrections: none

Cite this verdict

Tofu at least once weekly x lower coronary heart disease incidence Evidence Grade C card
[Chamgap] Tofu at least once weekly x lower coronary heart disease incidence — Evidence Grade C·48. 3 cited sources checked. Source: https://chamgap.com/en/verdicts/heart/tofu-weekly-coronary-heart-disease-incidence/ · 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.