Comparison of Blood Levels of Riboflavin and Folate With Dietary Correlates Estimated From a Semi-Quantitative Food-Frequency Questionnaire in Older Persons in Portugal

Abstract

Since information regarding biochemical parameters of riboflavin and folate status is limited in some populations of older adults, a food-frequency questionnaire is often used to estimate riboflavin and folate status. However, the performance of this type of questionnaire among this age group has not been comprehensively evaluated. Thus, we sought to assess riboflavin and folate status in older adults living in Portugal and to validate findings from a semiquantitative food-frequency questionnaire (FFQ), by comparison to these blood measures. We used a cross-sectional study to investigate riboflavin in red blood cells (as Glutathione Reductase Activity Coefficient; EGRAC) and folate in the serum of 88 older persons (66.7% female), aged between 60 and 94 years, recruited from seven adult day care community centers in Porto, Portugal. Forty-six subjects had low EGRAC levels (<1.2), with a group mean concentration of 1.17 and median of 1.10 (range 1.00–2.10). For daily riboflavin dietary intakes from FFQ, the mean was 3.34 mg, the median 3.37 mg, and range 0.66–4.81 mg. The Spearman correlation between these two measures was r = 0.073, (P = 0.497) and Pearson correlation, after adjustment for energy, was r = 0.263, P = 0.013. All participants were above the 7 nmol/L serum folate cut-off for adequacy. Spearman correlation coefficient between serum and FFQ measures was r = −0.10, (P = 0.359), and the Pearson correlation, after adjustment for energy and following loge transformation, was r = −0.58, (P = 0.593). Thus riboflavin and folate intakes estimated by FFQ correlated poorly with EGRAC and folate serum values.