. "Appendix D: Assessing the Performance of the EAR Cut-Point Method for Estimating Prevalence." Dietary Reference Intakes: Applications in Dietary Assessment. Washington, DC: The National Academies Press, 2000.
The following HTML text is provided to enhance online
readability. Many aspects of typography translate only awkwardly to HTML.
Please use the page image
as the authoritative form to ensure accuracy.
DRI DIETARY REFERENCE INTAKES: Applications in Dietary Assessment
Box D-3Major findings—Distribution of requirements not symmetrical
The bias in the estimate of inadequacy that results from application of the EAR cut-point method when the distribution of requirements is skewed can be severe.
When skewness exceeds values around 2, the relative bias (estimated prevalence/true prevalence) is very large—over 100 percent.
Even though this simulation was limited in scope, results are striking enough for the Uses Subcommittee to recommend that the EAR cut-point method not be used to assess the prevalence of nutrient inadequacy for a nutrient with a skewed requirement distribution.
of 50,000, prevalence of inadequacy was estimated from the population itself. Therefore, the values shown in Table D-1 and in Figure D-19 represent the actual proportion of individuals with intakes below requirements (true prevalence) and the estimate obtained from application of the EAR cut-point method.
The only nutrient for which there is strong evidence indicating a skewed requirement distribution (at the time this report was published) is iron in menstruating women (FAO/WHO, 1988). In recent Institute of Medicine reports on Dietary Reference Intakes (DRIs)
TABLE D-1 True Prevalence of Inadequacy and Estimated Prevalence of Inadequacy of Iron Obtained Using the EAR Cut-point Method
Distribution of Requirements
Mean
Standard Deviation
Skewness
True Prevalence (%)
Estimated Prevalence (%)
Bias (%)
8.4
0.7
0.62
12
11
1
8.6
1.4
1.32
15
11
4
9.0
2.5
2.51
20
11
9
9.5
3.9
3.15
24
11
13
10.4
6.9
5.73
28
12
16
NOTE: The distribution of usual intakes is fixed to be normal with a mean of 12 mg and a standard deviation of 3 mg.
