Premium
Filling a dietary data gap? Estimating individual nutrient intakes from household‐level data in Ethiopia and Bangladesh
Author(s) -
Coates Jennifer,
Rogers Beatrice,
Blau Alexander,
Lauer Jacqueline
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.669.17
Subject(s) - nutrient , meal , environmental health , food consumption , consumption (sociology) , medicine , food science , biology , agricultural economics , economics , ecology , social science , sociology
Dietary intake data are needed for a wide variety of food policy purposes but are not consistently collected in many low‐income countries. Many countries collect household level data in Household Consumption and Expenditure Surveys (HCES), and efforts have been made to use these data to estimate individual food and nutrient intake, often based on the assumption that food is allocated to individual members according to their proportional energy requirements relative to an adult male (called an adult male equivalent, or AME). This study sought to validate the AME‐based method of estimating individual information from household‐level data using data sets from Bangladesh and Ethiopia that measured both household consumption and individual intakes. The study compared the allocation of nutrients (energy, protein, animal protein (ASP), and iron) predicted using the AME assumption with individual intakes across ten age‐sex groups. The study compared the accuracy of using AMEs based on energy with AME based on the relevant nutrient (protein, iron) and assessed the effect of further adjusting the data for meal partakers. Judged against the intakes reported by the food preparer, the energy‐weighted AME predictions resulted in more accurate intake estimates than the nutrient‐specific AME predictions. AME‐based predictions were accurate within ten percentage points of individually reported intakes in most cases, but were inaccurate among certain groups, notably infants 6–23 mo and (for some nutrients) children in Bangladesh. The AME was less accurate in predicting ASP than other nutrients. AME predictions were more accurate in Ethiopia than in Bangladesh. Adjusting for meal partakers did not consistently improve the AME‐based predictions. The authors conclude use of the energy AME can produce useful results for food and nutrition policy analysis but should be complemented with a subsample of individual intake data for further validation. Support or Funding Information This research work is made possible by the support of the American people through the United States Agency for International Development (USAID) under Agreement No. AID‐663‐A‐11‐00017. The contents of this document are the sole responsibility of the researcher & do not necessarily reflect the views of USAID or the United States Government.