Date of this Version
Although the health benefits of omega-3 (n–3) fatty acids are widely appreciated, those of the omega-6 (n–6) fatty acids have become somewhat controversial in recent years, with some viewing them as harmful at current intakes (1, 2) and others (3) [including the American Heart Association (4)] supporting the status quo. Understanding the relations between intakes of these essential fatty acids and disease prevalence over time could inform the debate. To that end, this issue of the Journal contains a report by Blasbalg et al (5) that attempts to reconstruct those intakes and, from this information, to estimate fatty acid tissue content (eg, the Omega-3 Index [red blood cell eicosapentaenoic plus docosahexaenoic acid (EPA+DHA) (6)]). Low concentrations of the latter have been linked to increased risk of total mortality (7), cellular aging (8), depression (9), and acute coronary syndromes (10).
In reconstructing historical intakes, Blasbalg et al (5) used US Department of Agriculture (USDA) commodity disappearance data and made multiple assumptions and estimations regarding, for example, the completeness of USDA food and nutrient databases, the extent of wastage, the considerable differences in the amount and types of foods shipped in from outside of the United States (which would be particularly true for seafood), the proportions consumed by people (as opposed to animals or used for industrial purposes), and so forth. The USDA/Center for Nutrition Policy and Promotion approach used by these authors assumes that the fatty acid content of various foods did not change over the second half of the 20th century. In fact, many meat products, fats and oils (which were subjected to varying degrees of hydrogenation), and processed foods containing everchanging blends of commodity oils entered the food supply during this time. Consequently, Blasbalg et al’s reported intakes of both short-chain [α-linolenic acid (ALA) and linoleic acid (LA)] and long-chain, highly unsaturated fatty acids (HUFAs) of the n-6 and n-3 series cannot easily be independently verified and must remain as rough estimates only.