Jakobsen · The American journal of clinical nutrition 2009 · pooled analysis of prospective cohort studies · n=344,696 participants (11 cohorts)

Major types of dietary fat and risk of coronary heart disease: a pooled analysis of 11 cohort studies.

Cited 1025 times in the scientific literature.

Level 3 - non-randomized controlled study

Pooled analysis of 11 prospective cohort studies

PubMed 19211817 · doi:10.3945/ajcn.2008.27124 · record verified 2026-08-29

What was done

Researchers pooled individual-level data from 11 American and European prospective cohort studies comprising 344,696 participants followed for 4 to 10 years. Statistical substitution models were used to examine how replacing 5% of energy intake from saturated fatty acids (SFAs) with equivalent energy from polyunsaturated fatty acids (PUFAs), monounsaturated fatty acids (MUFAs), or carbohydrates affected the risk of incident coronary heart disease (CHD) events and coronary deaths, including potential effect modification by age and sex.

What was found

Over follow-up, 5,249 coronary events and 2,155 coronary deaths occurred. Replacing 5% of energy intake from SFAs with PUFAs was associated with a 13% lower risk of coronary events (hazard ratio: 0.87; 95% CI: 0.77, 0.97) and a 26% lower risk of coronary deaths (hazard ratio: 0.74; 95% CI: 0.61, 0.89). Replacing 5% of energy from SFAs with carbohydrates was associated with a slight increase in coronary events (hazard ratio: 1.07; 95% CI: 1.01, 1.14) and no significant change in coronary deaths (hazard ratio: 0.96; 95% CI: 0.82, 1.13). MUFA substitution showed no significant association with CHD. Associations did not significantly differ by age or sex.

Why it matters

This pooled analysis clarifies that the cardiovascular effect of reducing saturated fat depends critically on what nutrient replaces it, supporting dietary guidelines that emphasize replacing saturated fats with polyunsaturated fats rather than carbohydrates or monounsaturated fats.

Limits

The findings derive entirely from observational cohort studies rather than randomized controlled trials, leaving potential for residual confounding and measurement error from self-reported dietary assessments. Specific carbohydrate quality (such as refined sugars versus whole grains) and subclasses of fatty acids were not distinguished in the abstract.

Cited by