Variation in the iodine concentration of foods: considerations for dietary assessment

Authors: CARRIQUIRY, ALICIA - University Of Iowa; SPUNGEN, JUDITH - Food And Drug Administration(FDA); MURPHY, SUZANNE - University Of Hawaii; Pehrsson, Pamela; DWYER, JOHANNA - National Institutes Of Health (NIH); JUAN, WENYEN - Food And Drug Administration(FDA); WIRTZ, MARK - Food And Drug Administration(FDA); SWANSON, CHRISTINE - National Institutes Of Health (NIH)

Submitted to: The American Journal of Clinical Nutrition
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/16/2015
Publication Date: N/A
Citation: N/A

Interpretive Summary: Food composition tables are used to estimate the nutritional content of foods. Because the nutrient content of each food is given as a single summary value, it is likely that the actual food consumed by a survey participant will have a nutrient content that is not exactly equal to the table value. Discrepancies between the actual nutrient content of foods and the food composition table value can have a large impact on estimates such as the prevalence of inadequate or of excess consumption of a nutrient such as iodine. We investigated the variation in the iodine content of approximately 280 foods included in the Food and Drug Administration’s Total Diet Study. For each food, iodine was measured in four regions of the United States over an eight-year period, resulting in approximately 32 measurements per food. Second, using different statistical summaries over the 32 iodine measurements for each food (10th percentile, median, mean, and 90th percentile) and intake data from NHANES 2009-2010, we estimated the usual iodine intake distributions and the proportion of individuals with inadequate and potentially excessive iodine intakes. The iodine content of foods appears to vary noticeably across years and regions. As a result, estimates of iodine inadequacy can be significantly different depending on the summary statistic chosen (e.g., ranging from 1% to 60% for women aged 71 years and older). Even the choice of the mean vs. the median value is important in estimating inadequacy (e.g., 32% vs. 10% for older women). Thus, it is necessary to carefully choose the way that food composition values are summarized for nutrients such as iodine that have highly variable levels in foods.

Technical Abstract: Food composition tables are used to estimate the nutritional content of foods. Because the nutrient content of each food is given as a single summary value, it is likely that the actual food consumed by a survey participant will have a nutrient content that is not exactly equal to the table value. Discrepancies between the actual nutrient content of foods and the food composition table value can have a large impact on estimates such as the prevalence of inadequate or of excess consumption of a nutrient such as iodine. We investigated the variation in the iodine content of approximately 280 foods included in the Food and Drug Administration’s Total Diet Study. For each food, iodine was measured in four regions of the United States over an eight-year period, resulting in approximately 32 measurements per food. Second, using different statistical summaries over the 32 iodine measurements for each food (10th percentile, median, mean, and 90th percentile) and intake data from NHANES 2009-2010, we estimated the usual iodine intake distributions and the proportion of individuals with inadequate and potentially excessive iodine intakes. The iodine content of foods appears to vary noticeably across years and regions. As a result, estimates of iodine inadequacy can be significantly different depending on the summary statistic chosen (e.g., ranging from 1% to 60% for women aged 71 years and older). Even the choice of the mean vs. the median value is important in estimating inadequacy (e.g., 32% vs. 10% for older women). Thus, it is necessary to carefully choose the way that food composition values are summarized for nutrients such as iodine that have highly variable levels in foods.