Using household consumption and expenditure surveys to make inferences about food consumption, nutrient intakes and nutrition status

Abstract

Household consumption and expenditure surveys (HCES) are multipurpose surveys that are routinely conducted to collect data on household food consumption and availability in more than 120 countries. HCES are increasingly being used to calculate proxy estimates of food consumption, nutrient intakes, and nutrition status, often at the individual level. Rarely, however, do they collect information on meal participation, despite growing evidence that it is an increasingly important and variable component of the quantity of food consumed or available in a household. This paper explores the significance of adjusting for meal participation in making inferences about apparent food consumption and nutrient intakes. It focuses on two distinct sets of additional information requirements for enhancing the reliability and precision of measures of food consumption: (1) individual household members’ and household guests’ meal-eating behaviors, and (2) the number and apparent nutritional significance of meals. While the most comprehensive and precise accounting of intakes of individual food consumption and nutrients requires both types of information, the magnitude of the changes required in HCES questionnaires to capture them is likely to be prohibitive. Consequently, for many HCES, a “second best” approach may be the most effective method, at least in the short term. The paper empirically explores some of the relatively few HCES that currently attempt to capture some of these information requirements. In addition, it assesses their value-added to prioritize the global agenda for strengthening HCES measurement of food consumption in support of more evidence-based nutrition policy making.