Reducing the sample size of household nutrient monitoring based on the Fortification Assessment Coverage Tool (FACT) Surveys in four African countries Open Access

Abstract

Objective: Fortification quality is determined by analyzing the nutrient levels in household food samples and comparing them to international and national standards. This study has two main objectives: 1) Compare the fortification quality calculated from composite laboratory samples versus individual household samples to evaluate the feasibility of using composite samples as a replacement for individual samples, and 2) Identify the minimum number of samples required to provide an equivalent fortification quality estimation using a simulation study.

Method: Data analyses were based on 6665 household food (salt, maize flour, wheat flour and oil) samples collected from Nigeria, South Africa, Tanzania and Uganda using the Fortification Assessment Coverage Tool (FACT) survey instrument. For Objective 1, we utilized Fisher’s Exact Test to compare the fortification quality of individual household food samples versus composite laboratory samples. For Objective 2, we adopted the point of stability (POS) framework via bootstrap resampling. The critical POS determined by a pre-specified confidence level and a tolerable estimation error (i.e., width) represents the proportion of the reduced sample size over the entire sample size.

Results: The fortification quality estimated from composite laboratory samples were substantially different from that obtained from individual household samples. For example, in Nigeria, 100% of the composite laboratory salt samples were found to be overfortified; while 42.73% of the individual household salt samples were overfortified. In analyses of reduced sample sizes, we found various reduced sample sizes with different values of critical POS (confidence levels: 80%, 90% and 95%; widths: ±1.25% and ± 2.5%). Importantly, 45%, 70%, 50%, and 50% of household salt samples from Nigeria, South Africa, Tanzania, and Uganda respectively were needed to estimate the iodine fortification quality in salt.

Conclusion: This study found the fortification levels among composite laboratory samples were not comparable to the levels among individual household samples, suggesting the testing of individual household samples remains essential. A specific reduced sample size that is comparable with the entire individual household data set to estimate the fortification quality under the corresponding confidence levels and widths can be calculated by multiplying the POScrit by the entire sample size.

Table of Contents

Table of Contents

Introduction......................................................................................................................... 1

Methodology........................................................................................................................ 3

Study data and food specimen collection.......................................................................... 4

Ethical consideration........................................................................................................ 6

Data analyses.................................................................................................................... 6

Results................................................................................................................................ 14

Discussion.......................................................................................................................... 30

References.......................................................................................................................... 34

Appendices......................................................................................................................... 36

About this Master's Thesis

Rights statement

• Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Rollins School of Public Health
Department	Biostatistics
Subfield / Discipline	Biostatistics - MPH & MSPH
Degree	M.S.P.H.
Submission	Master's Thesis
Language	English
Research Field	Biology, Biostatistics Health Sciences, Nutrition
Keyword	bootstrap sample size reduction point of stability food fortification
Committee Chair / Thesis Advisor	Yi-An Ko, Emory University
Committee Members	O. Yaw Addo, Emory University Helena Pachón, Emory University

Last modified

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Reducing the sample size of household nutrient monitoring based on the Fortification Assessment Coverage Tool (FACT) Surveys in four African countries ()	2019-04-09 15:57:51 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions
	Reducing the sample size of household nutrient monitoring based on the Fortification Assessment Coverage Tool (FACT) Surveys in four African countries (the originial document - .doc)	2019-04-09 15:57:56 -0400	Press to Select an action Download