Development of a Screening Tool to Assess Folate Intake from Food

• Published in: Journal of the American Dietetic Association Vol. 99; no. 9; p. A89
• Main Authors: Phillips, M.N., Ilich, J.Z., Duffy, V.B.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.09.1999
• ISSN: 0002-8223; 1878-3570
• DOI: 10.1016/S0002-8223(99)00708-7

Adequate folate intake (≥400 μg/d) in women of childbeanng age may decrease the risk of neural tube defects. Assessing folate intake through dietary records is time intensive. Therefore, we designed and conducted initial validity testing on the Folate Intake Tool (FIT). The FIT is a food frequency survey comprised of top folate sources as assessed by frequency of consumption and amount per serving. The original FIT contained 72 items. Twenty-nine females (childbearing aged, mixed ethnic background) were recruited from a university community. Each completed the FIT and a 3-day food record (FR). The FIT was telephone interviewed and subjects reported usual intake over the previous year (average interview time: 12–13minutes). Total intake for FIT and FR was estimated with the 1998 folate fortification levels. For the FIT, total folate intake was calculated from frequency of intake and serving size; for the FR, with the Food Processor (version 7.2). Data were analyzed with nonparametric statistics (significance criterion: p<.05). Mean folate intake from the FIT (437.36±40.31 μg (SE)) did not differ significantly from that of the FR (381.02±30.98 μg). Grouping folate intake as adequate/inadequate (above/below 400 μg/d) identified 19 of 29 inadequate from the FR, 12 of 29 from the FIT. The FIT correlated significantly with the FR (rho=0.58, p=.001). Of the 72 FIT items, 24 were primary contributors to estimated folate intake. These items included the fortified cereals, breads, grains, and some fruits and vegetables. The calculated mean intake from these 24 items (302.53±19.89 μg) was less than that from the 72 items. However, the 24-item FIT continued to correlate significantly with the FR (rho=.67, p<.0001). The 24-item FIT classified more subjects with inadequate intakes (25 of 29). Thus the sensitivity (correctly identifying inadequate intake) reached 95% while specificity (correctly identifying adequate intake) was only 30%. These preliminary analyses suggest that the FIT can identify those with inadequate intakes of folate with as few as 24 items.