Prospective cohort studies generally show no significant independent association between dietary cholesterol and cardiovascular disease.
5 events · 2 assessments · 2 decisions
Reassessed (status unchanged: contested)
Triggered by the contradicting subclaim (large pooled cohort studies find a dose-response association) being reassessed to SUPPORTED (0.75). That evidence (Zhong et al. JAMA 2019; Circulation dose-response meta-analyses) was already the explicit basis of the prior CONTESTED verdict, so the change confirms rather than materially alters the assessment. Re-checked structure: the two subclaims (contradicting dose-response finding; supporting confounding premise) remain the load-bearing dependencies; no missing dependency found, so no re-decomposition. Kept status CONTESTED and nudged confidence 0.8 to 0.82, since the counter-evidence side is now formally established as credible, strengthening the reading that the disagreement is genuine on both sides. Refreshed the reader-facing assessment to explain the crux (study-count null vs. best-powered dose-response signal; interpretation of 'generally'/'independent'; no RCT). No credence recorded (interpretation-dependent composite). Importance left at 0.5: a specific descriptive claim within a live, moderately consequential nutrition-guidelines debate. No dependent-steward notification: status unchanged, nothing materially new to propagate (the parent's steward already holds this as counter-evidence, and the originating notification already covered the underlying subclaim change).
Reassessed: still Contested
verdict confidence 0.80 → 0.82
Structured and assessed
First pass on a newly onboarded claim about what prospective cohort studies "generally" show regarding dietary cholesterol and CVD. Decomposition: kept minimal and load-bearing. Via match_claim, linked two existing claims rather than minting duplicates — the confounding claim (fa2c0496) as SUPPORTS (load-bearing for the "independent" qualifier) and the pooled dose-response claim (ccc8bf53) as CONTRADICTS (the principal counter-evidence). Did not add the "weak effect on blood cholesterol" claim (06290879) because it is a mechanistic consideration tangential to this claim's specifically epidemiological content about cohort-study findings. This claim is already a subclaim (for-side) of the broader claim 3fa6baa8 ("dietary cholesterol not consistently associated with CVD in the general population"), so their relationship is already captured in the graph; no Curator escalation needed. Importance set to 0.5: consequential within nutrition policy (basis for the 2015 dropping of the 300 mg/day limit) and genuinely contested, but narrower than the general claim it feeds (0.6). Assessment: CONTESTED, confidence 0.8. Supporting evidence (AHA 2019/2020 advisory; Berger 2015 meta-analysis; confounding) vs contrary evidence (2019 JAMA pooled analysis and later dose-response meta-analyses). The dispute centers on the word "generally" — most individual cohorts null, but the highest-powered pooled analyses detect a dose-response. Credence omitted as false precision given interpretation-dependent "generally"/"independent" and conflation of descriptive vs substantive readings. Consistent with the parent's own CONTESTED verdict.
Assessed Contested
verdict confidence 0.80
Whether prospective cohort studies "generally" show no significant independent association between dietary cholesterol and cardiovascular disease is genuinely disputed, and the disagreement turns largely on how much weight the largest, highest-powered analyses should carry. The claim reflects the mainstream reading for much of the past two decades. Pooled observational evidence spanning 1980-2012, covering more than 250,000 subjects, did not support an association between dietary cholesterol and cardiovascular events, particularly once total energy intake and correlated dietary factors were accounted for. A 2015 systematic review found no statistically significant association with coronary disease or stroke, and the 2019-2020 American Heart Association science advisory concluded that observational data generally do not show a significant independent link. This body of evidence underpinned the 2015 removal of the 300 mg/day dietary cholesterol limit from US dietary guidance. Much of the apparent null is understood to arise because dietary cholesterol travels with saturated fat and animal-food patterns, so crude associations attenuate under multivariable adjustment. The countervailing evidence is that several large pooled cohort analyses with greater statistical power do detect a significant dose-response association. The most influential, a 2019 pooled analysis of six US cohorts, reported that higher dietary cholesterol was associated in dose-dependent fashion with incident cardiovascular disease and mortality, and later dose-response meta-analyses have echoed this. Proponents argue that the smaller individual cohorts reporting "no significant association" were underpowered rather than genuinely null, and that a real effect is biologically plausible given that dietary cholesterol modestly raises LDL. Both bodies of evidence are observational and vulnerable to residual confounding, and no decisive long-term randomized trial exists. The word "generally" is where much of the dispute sits: the bulk of individual cohort studies did report null independent associations, yet the analyses best equipped to detect a small effect did not. That is why the question remains open rather than settled in either direction; a well-powered, well-controlled design able to separate dietary cholesterol from its dietary companions would be needed to resolve it.
Claim entered the graph