Measuring diet cost at the individual level: a comparison of three methods

Background/objectives: Household-level food spending data are not suitable for population-based studies of the economics of nutrition. This study compared three methods of deriving diet cost at the individual level. Subjects/methods: Adult men and women ( n =164) completed 4-day diet diaries and a f...

Full description

Saved in:
Bibliographic Details
Published inEuropean journal of clinical nutrition Vol. 67; no. 11; pp. 1220 - 1225
Main Authors Monsivais, P, Perrigue, M M, Adams, S L, Drewnowski, A
Format Journal Article
LanguageEnglish
Published London Nature Publishing Group UK 01.11.2013
Nature Publishing Group
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Background/objectives: Household-level food spending data are not suitable for population-based studies of the economics of nutrition. This study compared three methods of deriving diet cost at the individual level. Subjects/methods: Adult men and women ( n =164) completed 4-day diet diaries and a food frequency questionnaire (FFQ). Food expenditures over 4 weeks and supermarket prices for 384 foods were obtained. Diet costs (US$/day) were estimated using: (1) diet diaries and expenditures; (2) diet diaries and supermarket prices; and (3) FFQs and supermarket prices. Agreement between the three methods was assessed on the basis of Pearson correlations and limits of agreement. Income-related differences in diet costs were estimated using general linear models. Results: Diet diaries yielded mean (s.d.) diet costs of $10.04 (4.27) based on Method 1 and $8.28 (2.32) based on Method 2. FFQs yielded mean diet costs of $7.66 (2.72) based on Method 3. Correlations between energy intakes and costs were highest for Method 3 ( r 2 =0.66), lower for Method 2 ( r 2 =0.24) and lowest for Method 1 ( r 2 =0.06). Cost estimates were significantly associated with household incomes. Conclusion: The weak association between food expenditures and food intake using Method 1 makes it least suitable for diet and health research. However, merging supermarket food prices with standard dietary assessment tools can provide estimates of individual diet cost that are more closely associated with food consumed. The derivation of individual diet cost can provide insights into some of the economic determinants of food choice, diet quality and health.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
content type line 23
ISSN:0954-3007
1476-5640
1476-5640
DOI:10.1038/ejcn.2013.176