#596: Why Do Omega-3 Trials Show Mixed Results?

#596: Why Do Omega-3 Trials Show Mixed Results?

From Sigma Nutrition Radio by Danny Lennon

March 3, 2026 · 1h 6m

About this episode

This episode explores the mixed results of omega-3 trials and the factors influencing these outcomes.

Omega-3 fatty acids (particularly EPA and DHA) have a long history in nutrition and cardiovascular medicine, yet the clinical trial literature is often perceived as inconsistent. This episode examines why some randomized trials show clear benefit while others show null or mixed findings, and how differences in trial design, dose, population risk, and outcome selection can materially change what we observe. A key theme is separating (1) the persistent cultural narratives around omega-3s (including origin stories that do not hold up well to modern evidence) from (2) the more precise, mechanistic and clinical questions about where supplemental EPA/DHA may reduce cardiovascular risk. The discussion focuses heavily on understanding heterogeneity: why "omega-3 supplementation" is not a single, uniform exposure, and why subgroup patterns (e.g., secondary prevention, higher baseline triglycerides, and higher doses) may explain much of the apparent conflict in the evidence. Note: This discussion is taken from a previous episode of the podcast. The audio has been remastered and improved, and now study notes and full transcript are available. Timestamps [04:10] Omega-3 historical context…

People in this episode

Host: Danny Lennon

Topics covered

  • omega-3 fatty acids
  • clinical trials
  • cardiovascular risk
  • trial design
  • subgroup analysis
  • nutritional science

Keywords

  • omega-3
  • EPA
  • DHA
  • clinical trials
  • cardiovascular health
  • triglycerides
  • nutritional evidence

Mentioned in this episode

Organizations: Cochrane, Hu et al., VITAL, ASCEND, GISSI-Prevenzione, REDUCE-IT

More episodes of Sigma Nutrition Radio

Explore listener stats, chart rankings, contacts and more on the Sigma Nutrition Radio podcast page.