Associations of Dietary Cholesterol or Egg Consumption With Incident Cardiovascular Disease and Mortality
Zhong VW, Van Horn L, Cornelis MC, Wilkins JT, Ning H, Carnethon MR, Greenland P, Mentz RJ, Tucker KL, Zhao L, Norwood AF, Lloyd-Jones DM, Allen NB - JAMA, 2019
In this pooled analysis of six US cohorts, higher dietary cholesterol intake and higher egg consumption were each associated with higher risk of incident cardiovascular disease and all-cause mortality in a dose-response manner. Crucially, the egg associations became null after adjusting for total dietary cholesterol, suggesting that cholesterol content, not eggs per se, drives the signal. This is observational data and cannot establish causation.
Key findings
- Each additional 300 mg/day of dietary cholesterol was associated with higher incident CVD (adjusted HR 1.17; 95% CI 1.09-1.26; adjusted ARD 3.24%) and higher all-cause mortality (adjusted HR 1.18; 95% CI 1.10-1.26; adjusted ARD 4.43%).
- Each additional half an egg per day was associated with higher incident CVD (HR 1.06; 95% CI 1.03-1.10) and mortality (HR 1.08; 95% CI 1.04-1.11) before adjusting for dietary cholesterol.
- After adjusting for dietary cholesterol, the egg associations became null: incident CVD HR 0.99 (95% CI 0.93-1.05) and all-cause mortality HR 1.03 (95% CI 0.97-1.09).
- This is observational data and cannot establish causation; the pattern suggests cholesterol content, not eggs per se, drives the association.
Effect measures
- Hazard Ratio: 1.1795% CI 1.09-1.26
- Hazard Ratio: 1.1895% CI 1.10-1.26
- Hazard Ratio: 0.9995% CI 0.93-1.05
- Hazard Ratio: 1.0395% CI 0.97-1.09
Why this evidence tier (Moderate)
- Risk of bias:
- Pooled individual-participant analysis of six well-characterized prospective US cohorts with long follow-up, but observational: subject to confounding by overall diet and lifestyle and to exposure misclassification.
- Precision:
- Large pooled sample (29,615) with many events; the dietary-cholesterol intervals exclude the null, and the post-adjustment egg intervals are tight around 1.0.
- Directness:
- Directly relates dietary cholesterol and egg intake to clinical cardiovascular events and mortality.
- Consistency:
- The dose-response cholesterol signal and the null egg association after adjustment are internally coherent and consistent with the view that dietary cholesterol, not eggs as a food, carries the association.
- Funding / COI:
- Drew on six NIH/NHLBI-supported prospective cohorts; the analysis itself reported no industry funding. Author disclosures include Dr Wilkins (consulting fees from NGM Biopharmaceuticals) and Dr Mentz (research support and honoraria from multiple pharmaceutical companies); no other disclosures were reported. No funding conflict bearing on the dietary-cholesterol finding.
Moderate certainty. A large, long-follow-up cohort analysis linking dietary cholesterol to cardiovascular events and mortality, with the egg association explained by its cholesterol content; observational and non-causal.
- Population:
- 29,615 US adults pooled from 6 prospective US cohorts (mean age 51.6 years; 5,400 incident CVD events and 6,132 deaths during median 17.5-year follow-up); diet data collected March 1985 to August 2016.
- Conflicts of interest:
- No industry funding of the analysis. Author disclosures: Dr Wilkins reported consulting fees from NGM Biopharmaceuticals; Dr Mentz reported research support and honoraria from multiple pharmaceutical companies; no other disclosures were reported.
- Funding:
- Analysis based on six NIH/NHLBI-supported prospective US cohorts; no industry funding of the analysis reported.
Limitations
- Observational pooled cohort; associations are subject to confounding and cannot establish causation.
- Diet typically assessed once at baseline by self-report, with measurement error and no capture of dietary change.
- Residual confounding by overall diet quality and lifestyle cannot be excluded.
- No retraction, erratum, or expression of concern; the record carries only post-publication commentary letters.