Fatty acid intakes in healthy adults quantified using a food frequency questionnaire compared with red blood cell membrane fatty acid content: A validation study.

AIMS

Red blood cell membrane fatty acids can be used alongside self-reported dietary assessment methods, such as food frequency questionnaires, to measure the validity of self-reported intakes. This study aimed to validate fatty acid intakes estimated from the Australian Eating Survey food frequency questionnaire against red blood cell profiles of healthy Australian adults.

METHODS

Demographic data and dietary intakes of fatty acids were collected, and red blood cell membrane fatty acid composition was measured using gas chromatography. Bland-Altman plots and concordance correlation coefficients (ρ) examined levels of agreement between measures and were adjusted for confounders.

RESULTS

A total of 58 participants (67% female, mean age 39.6 years) yielded 105 observations. Total saturated, total monounsaturated, total polyunsaturated fats including Eicosapentaenoic acid, Docosahexaenoic acid, Docosapentaenoic acid, and linoleic acid were within limits of agreement with moderate associations (ρ = 0.26-0.59). All adjusted models weakened associations; except total saturated fat retained moderate association in all models (ρ = 0.24-0.58). Omega-3 supplement use weakened associations for all fatty acids, except saturated and total omega-3 polyunsaturated fats. Carbohydrate intake had the least impact on associations.

CONCLUSION

Self-reported intakes from the Australian Eating Survey food frequency questionnaire were in moderate agreement (ρ = 0.20-0.60) with red blood cell membrane fatty acids. This questionnaire may be used as an indicator of self-reported long-term dietary fatty acid intake in Australian adults, with caveats for individuals reporting extreme intakes and consideration for evaluating the effects of endogenous synthesis of fatty acids. Future studies are warranted to replicate findings and strengthen translation to other sub-populations.