Within-Person Variation in Nutrient Intakes across Populations and Settings: Implications for the Use of External Estimates in Modeling Usual Nutrient Intake Distributions

Abstract

Determining the proportion of a population at risk of inadequate or excessive nutrient intake is a crucial step in planning and managing nutrition intervention programs. Multiple days of 24-h dietary intake data per subject allow for adjustment of modeled usual nutrient intake distributions for the proportion of total variance in intake attributable to within-individual variation (WIV:total). When only single-day dietary data are available, an external adjustment factor can be used; however, WIV:total may vary by population, and use of incorrect WIV:total ratios may influence the accuracy of prevalence estimates and subsequent program impacts. WIV:total values were compiled from publications and from reanalyses of existing datasets to describe variation in WIV:total across populations and settings. The potential impact of variation in external WIV:total on estimates of prevalence of inadequacy was assessed through simulation analyses using the National Cancer Institute 1-d method. WIV:total values were extracted from 40 publications from 24 countries, and additional values were calculated from 15 datasets from 12 nations. Wide variation in WIV:total (from 0.02 to 1.00) was observed in publications and reanalyses. Few patterns by population characteristics were apparent, but WIV:total varied by age in children (< vs. >1 y) and between rural and urban settings. Simulation analyses indicated that estimates of the prevalence of inadequate intake are sensitive to the selected ratio in some cases. Selection of an external WIV:total estimate should consider comparability between the reference and primary studies with regard to population characteristics, study design, and statistical methods. Given wide variation in observed ratios with few discernible patterns, the collection of ≥2 days of intake data in at least a representative subsample in population dietary studies is strongly encouraged. In the case of single-day dietary studies, sensitivity analyses are recommended to determine the robustness of prevalence estimates to changes in the variance ratio.

Keywords: dietary assessment; measurement error; micronutrients; variance components; variance ratio; within-individual variation.

FIGURE 1

Ratios of within-individual to total variance in nutrient intakes from publications or reanalyzed datasets. The total number of ratio estimates (points) and studies, respectively, represented by each boxplot are shown in parentheses on the vertical axis. Data include analyses on overlapping populations when combined and disaggregated analyses were performed by season, age, sex, or physiological status, or including and excluding the contribution of iron intake from cooking pots. Ratios for analyses including intake from supplements are not shown. “Folic acid” refers to the synthetic form found in fortified foods, while “folate” refers to natural dietary forms or dietary folate equivalents. WIV:total variance, ratio of within-individual to total variance.

FIGURE 2

Ratios of within-individual to total variance in intakes of vitamin A (A), thiamin (B), iron (C), zinc (D), folate (E), and vitamin B-12 (F) from publications or reanalyzed datasets, by age and sex group. The total number of ratio estimates (points) and studies, respectively, represented by each boxplot are shown in parentheses on the vertical axis. Women included nonpregnant, nonlactating; pregnant; lactating; or mixed populations. Children >1 y included age ranges falling between 1 and 20 y of age. Children ≤1 y included ages up to 13 mo in some cases. Ratios for other age-sex groupings are not shown due to small numbers of available estimates. Data include analyses on overlapping populations when combined and disaggregated analyses were performed by season, age, or physiological status or including and excluding the contribution of iron intake from cooking pots. WIV:total variance, ratio of within-individual to total variance.

FIGURE 3

Simulated effect of varying the ratio of within-individual to total variance on the prevalence of folate (A), vitamin A (B), and zinc (C) inadequacy among women of reproductive age in Cameroon and Bangladesh using the 1-d NCI method. Prevalence using WIV:total calculated from 2-d data on the same population is marked by diamonds. The range of WIV:total observed in existing datasets among NPNL or mixed (NPNL and lactating/pregnant) populations is indicated by shaded boxes. NCI, National Cancer Institute; NPNL, nonpregnant and nonlactating; WIV:total variance, ratio of within-individual to total variance.

FIGURE 4

Modeled distributions of usual zinc intake in women in Cameroon (A) and Bangladesh (B) assuming a low within-individual to total variance ratio (WIV:total = 0.35) or high ratio (WIV:total = 0.80) using the 1-d NCI method. The EAR for zinc of 6 mg/d for nonpregnant, nonlactating women assuming a mixed or refined vegetarian diet [as determined by the International Zinc Nutrition Consultative Group (24)] was used as the cutoff for inadequacy. EAR, Estimated Average Requirement; NCI, National Cancer Institute; WIV:total, ratio of within-individual to total variance.