The Food Consumption Score (FCS) is the World Food Programme’s main food access indicator, a proxy for household diet quantity and quality. The score is based on the number of days in a week eight food groups were eaten, and thresholds classify households as having poor, borderline or acceptable food consumption. The ability of the FCS and its thresholds to classify household’s consumption consistently are vital as food assistance is directed towards areas and population groups where prevalence of inadequate food consumption is high. As there are indications that the current thresholds underestimate inadequate consumption, this paper asks whether FCS thresholds can be identified corresponding to inadequate energy consumption, the quantity dimension of the score. The analysis uses household survey data that include comprehensive modules on food consumption as well as the information necessary to calculate the FCS from six countries. The results show that the FCS is significantly but not highly correlated with calorie intake. Not counting foods eaten in small quantities, clearly improves the association between the FCS and caloric intake, but the analysis suggests that in practice it is difficult to exclude the small quantities. Established sensitivity and specificity criteria for suitable thresholds are not met and this paper concludes that it is not appropriate to identify FCS thresholds that adequately correspond to caloric thresholds. The analysis illustrates that the FCS depicts both quantitative and qualitative aspects of food consumption and concludes that a future strategy should be to anchor thresholds in an indicator comprising both these dimensions.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
It was assumed that staples are rarely eaten in small quantities whilst consumption of small amounts of oil and sugar can be considered normal, following the suggestions in de Haen et al. (2011).
Consumption of only one egg or a piece of meat the size of a box of matches for four or more persons were considered small amounts. Cheese spread over the whole family meal or one piece of fruit added as flavouring were also regarded as small amounts.
Firstly, probing about more items—as would usually be the case in the food consumption module—may help people to remember items they would otherwise forget in a standard FCS module. Secondly, in the food consumption module we do not know whether two items belonging to the same food group were eaten on different days or on the same day. In the FCS module only the number of days of eating the food group should be counted. In both these instances the consumption module would give a higher FCS than the standard FCS module.
These thresholds are sometimes adjusted by adding 7 to each threshold in populations where the consumption of sugar and/or oil is frequent (7 days) for the great majority of surveyed households. The reason for these higher cut-offs is that several analyses of FCS data encountered populations where the majority consumed oil and sugar, even when consumption of other food groups was rare.
The food composition table for Central American countries was used for the three countries (Instituto de Nutrición de Centroamérica y Panamá, INCAP), while for Malawi and Nepal we relied on a conversion table supplied by the national statistical office; Uganda is based on the FAO’s Food Composition Table for use in Africa.
To calculate the adult male equivalents (AME) we followed the procedure of Smith and Subandoro (2007). For the Central American countries, energy requirements by sex and age were taken from the adjusted energy and protein requirement tables prepared by INCAP. For the LSMS countries, the average energy requirements recommended by FAO (1985) were used. Light activity level was used as a normative standard. Adjustments for pregnancy and lactation were only done for the Central American countries, as information on lactation and pregnancy was not available in the LSMS surveys. AME factors for each individual were calculated by comparing his or her energy requirements with those of an adult man. Summing these individual factors gives the AME for the household.
The Receiver Operating Characteristic (ROC) analysis which graphically plots the true positive rates versus the false positive rates (sensitivity against 1-specificity) for the benchmark binary variables against all possible cut-offs point for the proxy variable. Each point on the curve represents a sensitivity/specificity pair corresponding to a particular threshold.
The other national stunting figures are from, respectively; the Malawi Demographic and Health Surveys 2010; the Guatemala Encuesta Nacional de Salud Materno Infantil 2008–09; the El Salvador Encuesta Nacional de Salud Familiar FESAL 2008; and the Honduras Encuesta Nacional de Demografía y Salud ENDESA 2005–2006. Although these surveys are all, except Malawi, older then the surveys analyzed we do not expect the stunting figures to have changed considerably and, in particular, not to have been reduced to the extent that they match the level of prevalence of poor and borderline FCS.
For the Central American countries two different modules were used for calculating the FCSs: The FCS excluding small quantities ex post was calculated based on the food consumption module and the standard FCS (including a question on small quantities) was calculated based on the WFP’s FCS module. Therefore, the correlation difference for the Central American countries between the standard FCSs and the FCS excluding small quantities ex post may be a result of using two different modules for the calculation rather than the effect of excluding small quantities. To check if this was an issue we calculated the standard FCS based on the food consumption module. The correlations in the second column in Table 2 changed only slightly. For El Salvador compare 0.33 with 0.37 for Guatemala compare 0.36 with 0.33 and for Honduras compare 0.20 with 0.24. Thus for illustrative purposes we used the FCS from the WFP’s FCS module as our base.
Except for Malawi where the FCS excludes small quantities from the score during the interview.
Arimond, M., Wiesmann, D., Becquey, E., Carriquiry, A., Daniels, M. C., Deitchler, M., et al. (2010). Simple food group diversity indicators predict micronutrient adequacy of women’s diets in 5 diverse, resource-poor settings. The Journal of Nutrition, 140, 2059S–2069S.
Baumann, S. M, Webb, P., & Zeller, M. (2013). Validity of food consumption indicators in the Lao context: moving toward cross-cultural standardization. Food and Nutrition Bulletin, 34(1). The United Nations University, 2013.
Beegle, K., De Weerdt, J., Friedman, J., & Gibson, J. (2012). Methods of household consumption measurement through surveys: experimental results from Tanzania. Journal of Development Economics, 98, 3–18.
Central Bureau of Statistics: “Nepal Living Standard Survey 2010/12.” (2011). Statistical report. National Planning Commission Secretariat. Government of Nepal.
Cleveland, W. S. (1979). Robust locally weighted regression and smoothing scatterplots. Journal of the American Statistical Association, 74(368), 829–836.
Coates, J., Rogers, B. L., Webb, P., Maxwell, D., Houser, R., & McDonald, C. (2007). Diet diversity study. World Food Programme report. Emergency Needs Assessment Service, Rome.
Deaton, A., & Drèze, J. (2009). Food and nutrition in India: facts and interpretations. Economic and Political Weekly, XLIV(7), 42–65.
de Haen, H., Klasen, S., & Qaim, M. (2011). What do we really know? Metrics for food insecurity and undernutrition. Food Policy, 36, 760–769.
FAO/WHO/UNU. (1985). Energy and protein requirements. Report of a Joint FAO/WHO/UNU Expert Consultation Rome: FAO.
Jensen, R. T., & Miller, N. H. (2010). A revealed approach to measuring hunger and undernutrition. NBER Working Paper No w16555, Cambridge, MA 2010.
Headey, D., & Ecker, O. (2013). Rethinking the measurement of food security: from first principles to best practice. Food Security, 5, 327–343.
Hoddinott, Y., & Yohannes, Y. (2002). Dietary diversity as a food security indicator. IFPRI Discussion Paper No. 136, Washington D.C.
National Statistical Office. (2012). Malawi Third Integrated Household Surveys (IHS3). 2010–11. Basic information document. NSO P.O BOX 333, Zomba, Malawi.
Ruel, M. T. (2003). Operationalizing dietary diversity; a review of measurement issues and research priorities. The Journal of Nutrition, 133(11), 3911S–3926S.
Smith, L., & Subandoro, A. (2007). Measuring food security. using household expenditure surveys. Food security in practice technical guide series. Washington, D.C., IFPRI.
Tiwari, S., Skoufias, E., & Sherpa, M. (2013). Shorter, cheaper, quicker, better : linking measures of household food security to nutritional outcomes in Bangladesh, Nepal, Pakistan, Uganda, and Tanzania. World Bank policy research working paper WPS6584. The World Bank.
Uganda Bureau of Statistics. (2009). Uganda National panel survey. 2009/10. Interviewer’s manual of instructions. UBoS, Kampala, Uganda.
WHO. (2003). Diet, nutrition and the prevention of chronic diseases. Geneva 28 2002, WHO Technical Report series, 916, World Health Organisation, Geneva.
Wiesmann, D., Hoddinott, J., Aberman, N. L., & Ruel, M. (2006). Review and validation of dietary diversity, food frequency and other proxy indicators for household food security. World Food Programme report. World Food Programme Rome.
Wiesmann, D, Bassett, L., Benson, T., & Hoddinott, J. (2009). Validating the World Food Programme’s food consumption score and alternative indicators of household food security. IFPRI Discussion Paper 00870, Washington D.C.
World Food Programme. (2008). Food consumption analysis. Calculation and use of the food consumption score in food security analysis. Technical guidance sheet. World Food Programme, Rome.
World Food Programme. (2009). Comprehensive food security & vulnerability analysis. Guidelines. World Food Programme Rome.
We are grateful to Daniel Maxwell and Anne Swindale as well as Doris Wiesmann and Cesar Revoredo-Giha for their valuable suggestions and advice for the Central American and the LSMS analyses, respectively. We are also in debt to Doris Wiesmann for sharing syntax files she created for a similar validation analysis. Thanks to Agnes Dhur for reviewing the survey design. We would also like to thank colleagues in WFP’s VAM service for fruitful discussion during the process of preparing this paper. In particular the analyses have benefitted from discussions and reviews from Margreet Barkhof, Rosella Bottone, Siemon Hollema, Joyce Luma, Kathryn Ogden and Peter Horjus. Finally, we would like to thank the Nepal Central Bureau, the Uganda Bureau of Statistics, and the Malawi National Statistical Office for providing data as well as the WFP country offices in Guatemala, El Salvador and Honduras for helping with the data collection for these countries.
Both authors are affiliated with the United Nations World Food Programme.
About this article
Cite this article
Lovon, M., Mathiassen, A. Are the World Food Programme’s food consumption groups a good proxy for energy deficiency?. Food Sec. 6, 461–470 (2014). https://doi.org/10.1007/s12571-014-0367-z