Serum Omega-3 Polyunsaturated Fatty Acids and Risk of Incident Diabetes in Men


Whether long-chain omega-3 polyunsaturated fatty acids (PUFA's) eicosapentaenoic acid (EPA; 20:5 n-3) and docosahexaenoic acid (DHA; 22:6n-3) found in fish and seafood are protective in development of Type 2 Diabetes (T2D) or other cardiovascular risk factors remains inconclusive. Likewise, the potential role of alpha-linolenic acid (ALA; 18:n-3) has not been explored. Previously, in a subsample of the 895 men who participated in the Kuopio Ischemic Heart Disease Risk Factor (KIHD) study, serum linoleic acid (an omega-6 fatty acid) was associated with a lower risk of T2D and impaired fasting glycemia, and reduced adverse changes in insulin and glucose concentrations but not other PUFA's. The association between serum n-3 fatty acids and hair mercury concentration and risk of T2D was the focus of this study.


The KIHD study involved a total of 2,682 Finnish men who were between ages 42-60 years old and free of T2D at baseline in 1984-1989. Serum levels of PUFA and hair mercury served as biomarkers. Self-administered four-day food records were also collected. Repeat examinations occurred at years four, 11 and 20. With a total of 1,241 men, or 80% of the baseline sample, completing the visits. Serum fatty acids, dietary intakes and plasma glucose were measured.


After 19.3 years, 19.2% of men developed T2D. In the multivariate adjusted models, men in the highest quartile of serum long-chain n-3 fatty PUFA levels had a 33% lower risk (95% CI 13-49) for incident T2D compared to men in the lowest quartile. Serum DHA was inversely associated with risk of T2D while the association with EPA was non-significant. Serum ALA was associated with lower risk of T2D after adjustment for age and exam year, but further adjustments attenuated the association. A dietary intake of fish, EPA, DHA and ALA were not associated with risk of T2D, nor was hair mercury associated.


Serum n-3 PUFA, a biomarker of fish and dietary n-3 intake, was associated with reduced risk of incident T2D, but dietary intake measures of fish, EPA and DHA were not.


The study provides valuable prospective data documenting the benefits of measuring serum levels of n-3 PUFA as being inversely associated with incident T2D, despite the absence of evidence from dietary data derived from this same cohort, confirming this association. The authors further extrapolate these inconsistencies as the underlying contributors to the inconclusive findings reported in previous studies and meta-analyses that have relied upon dietary data alone. It is important to recognize that the quality, accuracy and validity of diet assessment data vary considerably as well. The cost-conscious reliance on non-quantified and undifferentiated dietary assessment methods, including certain very limited food frequency questionnaires, self-administered, unedited food records or a single 24-hour dietary recall, is all too common in epidemiologic research. This study provides further evidence supporting the benefits of including objective biomarker measurements in determining associated risk factors, but it does not diminish the need for validated, quality-controlled diet assessment methods for purposes of documenting the relevant food sources and eating patterns that are potentially modifiable and amenable to intervention.


  1. Hu, F, Stampfer, M, Rimm, E, Ascherio, A, Rosner, B, Spiegelman, D, Willett, WC, Dietary fat and coronary heart disease: a comparison of approaches for adjusting for total energy intake and modeling repeated dietary measurements. Am J Epidemiol 1999; 149:531-40.
  2. Kristal, A, Peters, U, Potter, JD. Is it time to abandon the food frequency questionnaire? Cancer Epidemiol Biomarkers Prev 2005; 14:2826-28.
  3. Prentice RL, Huang Y, Kuller LH, Tinker LF, Van Horn L, Stefanick ML, Sarto G, Ockene J, Johnson KJ. Biomarker-calibrated energy and protein consumption and cardiovascular disease risk among postmenopausal women. Epidemiology 2011;22:170-179.

Keywords: Cardiovascular Diseases, Diabetes Mellitus, Type 2, Fatty Acids, Fatty Acids, Omega-3, Fatty Acids, Omega-6, Fatty Acids, Unsaturated, Myocardial Ischemia, Prediabetic State, alpha-Linolenic Acid

< Back to Listings