Skip to main content
ARS Home » Southeast Area » Little Rock, Arkansas » Arkansas Children's Nutrition Center » Research » Publications at this Location » Publication #230788
Title: The validity and reproducibility of food-frequency questionnaire–based total antioxidant capacity estimates in Swedish women

Author
item RAUTIAINEN, SUSANNE - KAROLINSKA INSTITUTE
item SERAFINI, MAURO - NATL INST FOOD & NUTR RES
item MORGENSTERN, RALF - KAROLINSKA INSTITUTE
item Prior, Ronald
item WOLK, ALICJA - KAROLINSKA INSTITUTE

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/11/2008
Publication Date: 5/15/2008
Citation: Rautiainen, S., Serafini, M., Morgenstern, R., Prior, R.L., Wolk, A. 2008. The validity and reproducibility of food-frequency questionnaire–based total antioxidant capacity estimates in Swedish women. American Journal of Clinical Nutrition. 87(5):1247-1253.

Interpretive Summary: Nutritional epidemiology studies often require methods for assessing the intake of various nutrients. A food frequency questionnaire (FFQ) is often used in which subjects indicate the frequency with which different foods have been consumed over a specified period of time. This report represents the first time in which antioxidant capacity intake has been assessed using food values obtained by three different antioxidant capacity assays: 1) oxygen radical absorbance capacity (ORAC), 2) total radical-trapping antioxidant parameters, and 3) ferric-reducing antioxidant power. The validity and reproducibility of the FFQ were evaluated in 2 random samples of human subjects from the Swedish Mammography Cohort. Validity was studied by comparing FFQ-based antioxidant capacity estimates with one measurement of plasma antioxidant capacity in 108 women. Reproducibility was studied in 300 women who completed 2 FFQs 1 year apart. Fruit and vegetables (mainly apples, pears, oranges, and berries) were the major contributors to FFQ-based antioxidant capacity measures. In the validity study, whole plasma ORAC was correlated with FFQ-based antioxidant capacity estimates from fruit and vegetables. Significant correlations between lipophilic plasma ORAC and FFQ-based antioxidant capacity estimates from fruit and vegetables also were observed. The study demonstrated that FFQ-based antioxidant capacity values represent valid and reproducible estimates that may be used in nutritional epidemiology to assess antioxidant intake from foods. Further studies in other populations are needed to confirm these results.

Technical Abstract: Total antioxidant capacity (TAC) provides an assessment of antioxidant activity and synergistic interactions of redox molecules in foods and plasma. We investigated the validity and reproducibility of food frequency questionnaire (FFQ)–based TAC estimates assessed by oxygen radical absorbance capacity (ORAC), total radical-trapping antioxidant parameters (TRAP), and ferric-reducing antioxidant power (FRAP) food values. Validity and reproducibility were evaluated in 2 random samples from the Swedish Mammography Cohort. Validity was studied by comparing FFQ-based TAC estimates with one measurement of plasma TAC in 108 women (54- to 73-y-old dietary supplement nonusers). Reproducibility was studied in 300 women (56–75 y old, 50.7% dietary supplement nonusers) who completed 2 FFQs 1 y apart. Fruit and vegetables (mainly apples, pears, oranges, and berries) were the major contributors to FFQ-based ORAC (56.5%), TRAP (41.7%), and FRAP (38.0%) estimates. In the validity study, whole plasma ORAC was correlated (Pearson) with FFQ-based ORAC(r_0.35), TRAP (r_0.31), and FRAP (r_0.28) estimates from fruit and vegetables. Correlations between lipophilic plasma ORAC and FFQ-based ORAC, TRAP, and FRAP estimates from fruit and vegetables were 0.41, 0.31, and 0.28, and correlations with plasma TRAP estimates were 0.31, 0.30, and 0.28, respectively. Hydrophilic plasma ORAC and plasma FRAP values did not correlate with FFQ-based TAC estimates. Reproducibility, assessed by intraclass correlations, was 0.60, 0.61, and 0.61 for FFQ-based ORAC, TRAP, and FRAP estimates, respectively, from fruit and vegetables. FFQ-based TAC values represent valid and reproducible estimates that may be used in nutritional epidemiology to assess antioxidant intake from foods. Further studies in other populations to confirm these results are needed.