Cover Image
close this bookFood Composition Data: A User's Perspective (UNU, 1987, 223 pages)
close this folderOther considerations
close this folderConsideration of food composition variability: What is the variance of the estimate of one-day intakes? Implications for setting priorities
View the document(introductory text...)
View the documentIntroduction
View the documentMagnitude of the reported variability of composition
View the documentImpact of composition variation on a one-day food intake
View the documentAdditional impact of a random error in intake estimation
View the documentSome implications for data analyses
View the documentValidation of food intake data: implications of food composition variation
View the documentSystematic errors in food composition data
View the documentRelevance to priorities for food composition data
View the documentConclusions
View the documentReferences

Some implications for data analyses

In assessment of the distribution of observed intakes, as in estimating the prevalence of inadequate or excessive intakes, the focus of attention should be on the estimation of the distribution of usual intakes. This is achieved by obtaining replicate one-day intakes. In the past these have been averaged for the individual to improve the estimate of his or her intake (eliminate the effect of day-to-day variation). A better approach involves application of an analysis of variance to estimate the distribution of usual intakes (the inter-individual variation component) [6]. Either way, a part of the random variation in food composition or in intake estimation will be factored out along with the removal of day-to-day variation. Thus, it can be demonstrated that the presence of variance affects the estimation of prevalence of inadequate intakes or of excessive intakes. However, the effect is surprisingly small! Improvement of the food composition data base by increasing replications will improve the confidence of the prevalence estimate, but the cost-effectiveness for this purpose needs to be examined very carefully.

Another application of computed nutrient intakes is in connection with epidemiologic studiesas in regression or correlation analyses of the relationship between observed intake and some biological outcome. It has been demonstrated by several authors that an error term in the estimation of usual intake can attenuate correlations and will bias regression slopes toward 0 if intake is used as the independent variable.

Table 4. Potential error associated with estimated one-day intakes

Nutrient

Diet HW1

Diet HW2

  Mean SD CV (%) Mean SD CV (%)
Protein 104.6 6.20 5.93 97.5 2.21 2.27
Calcium 1,540.2 80.77 5.24 1,135.2 61.31 5.40
Iron 8.03 1.19 14.85 10.4 1.66 16.00
Magnesium 250.1 15.70 6.28 222.4 13.04 5.86
Sodium 4,129.5 157.36 3.81 2,589.8 121.73 4.70
Zinc 11.6 0.909 7.85 13.3 1.64 12.33
Thiamine 2.10 0.375 17.92 0.715 0.076 10.59
Riboflavin 2.60 0.205 7.90 2.13 0.154 7.22
Niacin 15.9 0.908 5.72 13.5 0.879 6.53
Vitamin C 153.1 11.91 7.77 11.8 1.54 13.00
Vitamin B6 1.45 0.136 9.37 1.43 0210 14.62
Folacin 184.3 19.80 10.74 97.1 12 02 12.38
Vitamin A 3,798.4 281.24 7.40 5,142.0 603 61 11.74

Table 5. Impact of the number of food items in a record on the error term of computed nutrient intake for an individuala

Number of
foods in
record

CV of nutrient content of individual food serving

  10 15 20 25 30 35 40 45 50 55
2 7.07 10.61 14.14 17.68 21.21 24.75 28.28 31.82 35.36 38.89
3 5.77 8.66 11.55 14.43 17.32 20.21 23.09 25.98 28.87 31.75
4 5.00 7.50 10.00 12.50 15.00 17.50 20.00 22.50 25.00 27.50
5 4.47 6.71 8.94 11.18 13.42 15.65 17.89 20.12 22.36 24.60
10 3.16 4.74 6.32 7.91 9.49 11.07 12.65 14.23 15.81 17.39
15 2.58 3.78 5.16 6.45 7.75 9.04 10.33 11.62 12.91 14.20
20 2.24 3.35 4.47 5.59 6.71 7.83 8.94 10.06 11.18 12.30
25 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00 11.00
30 1.83 2.74 3.65 4.56 5.48 6.39 7.30 8.22 9.13 10.04

a. Table assumes that all foods make an equal contribution to nutrient intake.

A major concern in this regard is the day-to-day variation in intake [2, 3, 4, 5, 8], which can have a CV in the order of 25 to 35 per cent or even higher for some nutrients [2,3]. It is clear that food composition variation also contributes to the true total variability (it would be in addition to estimated intra-individual or within-person variation based on published food composition data). Clearly, improvement of the food composition data would improve correlation or regression analyses; it might not be very cost-effective. If we must continue to estimate food intakes on several days, and then average these to estimate "used intake," we again find that the random component of food composition variation will be reduced. The phenomenon is illustrated in table 9.

Table 6. Estimate of error term in one-day intakes associated with combined variability of food composition and error of the intake estimate (assumed measurement error CV = 10 per cent of reported intake, normally distributed)

Nutrient

Diet HW1a

Diet HW2a

  Mean SD CV (%) Mean SD CV ( %)
Protein 109.6 7.56 7.23 97.5 5.81 5.96
Calcium 1,540.2 103.7 6.74 1,135.2 82.52 7.26
Iron 8.03 1.23 15.35 10.40 1.73 16.62
Magnesium 250.0 17.72 7.08 222.4 15.51 6.97
Sodium 4,129.5 239.3 5.80 2,589.8 180.3 6.95
Zinc 11.58 1.00 8.67 13.32 1.76 13.22
Thiamine 2.10 0.395 18.85 0.716 0.080 11.13
Riboflavin 2.60 0.226 8.71 2.13 0.175 8.21
Niacin 15.89 1.18 7.43 13.46 1.28 9.49
Vitamin C 153.1 14.78 9.65 11.85 1.61 13.56
Vitamin B6 1.45 0.149 10.26 1.43 0.227 15.83
Folacin 184.3 21.12 11.46 97.07 12.72 13.10
Vitamin A 3,798.4 313.2 8.25 5,142.0 683.0 13.28

a For composition variability, see table 5

Table 7. Impact of random error in intake and food composition data on the error of calculated nutrient content of an individual serving of fooda

CV1

CV1

  0 5 10 15 20 25 30 35 40 45
0 0 5 10 15 20 25 30 35 40 45
5 5 7.1 11.2 15.8 21.6 25.5 30.5 35.4 40.4 45.3
10 10 11.2 14.2 18.1 22.4 27.0 31.8 36.6 41.4 46.3
15 15 15.8 18.1 21.3 25.2 29.4 33.8 38.4 43.1 47.9
20 20 20.6 22.4 25.2 28.6 32.4 36.6 40.9 45.4 50.1
25 25 25.5 27.0 29.4 32.4 35.9 39.8 43.9 48.2 52.7
30 30 30.5 31.8 33.8 36.6 39.8 43.4 47.3 51.4 55.7
35 35 35.4 36.6 38.4 40.9 43.9 47.3 51.0 55.0 59.1
40 40 40.4 41.4 43.1 45.4 48.2 51.4 55.0 58.8 62.8
45 45 45.3 46.3 47.9 50.1 52.7 55.7 59.1 62.8 66.8

a. All values expressed as CV = 100 x SD/Mean It is not important which variable is I or 2.