Omega-3 index = 8.3 (1909); fell to 3.8 (1999) (after introduction of seed oils)

Claude AI - Oct 2026

There's no measured 50-year time series of the Omega-3 Index itself. The test dates from 2004, and the first nationally representative US measurement was only made in 2021–2023. The one analysis that estimates the index over many decades is a model, the Blasbalg/Hibbeln study of 20th-century US food supply. Measured data from other populations support the direction of its result.

The main analysis

Blasbalg et al. (2011) used food-disappearance data and modeled tissue fatty acids. They estimated the US omega-3 index fell from

  • 8.28 (using 1909 traditional foods) or
  • 6.51 (using 1909 foods as produced today) to
  • 3.84 in 1999.

Over the same period, the n-3 share of tissue HUFA fell from about 37% or 31% to about 23%. [Blasbalg 2011] That works out to a decline of 54% using traditional foods, or 41% using current foods. [Blasbalg Table 6]

The main driver in the model was competition from omega-6, not falling fish intake. Per-capita soybean oil consumption rose more than 1000-fold, and linoleic acid availability rose from 2.79% to 7.21% of energy. The gap between the two 1909 scenarios also matters. Only the traditional-foods version showed large drops in dietary EPA, DHA and arachidonic acid. So part of the estimated decline comes from changes in how animal foods are produced, not just from what people chose to eat.

The model covers 90 years, but much of the change fits inside your 50-year window. Measured US adipose tissue shows that linoleic acid rose from 9.1% in 1959 to 21.5% in 2008, a 136% increase [Guyenet 2015].

Measured evidence

Evidence Period What changed
[Nunavik Inuit] 1992 → 2004 Plasma omega-3 down 25%, omega-6 up 9%; young adults lower than older ones
[Hisayama, Japan] 2002 → 2012 Serum EPA/AA ratio fell, with the largest drops in younger age groups
[US RBC cohorts] 1993 → 2019 5.17% (WHI, 1993) vs 5.30–5.95% in clinical labs 2011–2019
[NHANES] 2021 → 2023 Mean 4.12% overall, 4.25% in adults, about 3.6% in children and adolescents

Behind the Japanese data, fish consumption in Japan has fallen and meat consumption has risen sharply over the last 50 to 60 years [Japan diet].

How to read this

  • The model checks out against NHANES. Blasbalg's 1999 estimate of 3.84% is close to the 4.12% NHANES measured two decades later. That supports the model's endpoint. Its 1909 starting values cannot be checked against any measurement.
  • Recent US values look flat or slightly higher, not lower. The 2024 world map moved the US up a category, from "very low" to "low." The authors weren't sure whether that reflected real change, more supplement use, or different populations being sampled. [Omega-3 map 2024] The US cohorts in the table also differ in age and makeup, so they aren't a true trend line.
  • Declines are still under way where traditional diets are fading. The Inuit and Japanese data show this, and in both the young are lower than the old. That pattern suggests a future decline in national averages as older people are replaced.

I didn't find any study that re-tested archived blood from the 1970s or earlier for RBC EPA+DHA. That would be the direct way to check the Blasbalg estimates, if suitable samples exist.

Sources:

See related in VitaminDWiki