Objectives
To assess the association between different dietary diversity recall periods and haemoglobin concentration among non-pregnant women in rural Niger.
Methods
The REFANI-N study, a cluster-randomised trial, in Tahoua, Niger, is comparing the nutritional impact of two emergency unconditional cash transfers for âvery poor’ households. The study collected baseline and end-line information on 2,077 non-pregnant women, aged 15-49 years from 2,293 households. Food security indicators and haemoglobin data were collected to study the mechanisms by which cash transfers might affect nutrition. Household dietary diversity (HDD) was measured using 7-day and 24hr recall periods and women’s diet diversity (WDD) using a 7-day recall. HHD and WDD scores were calculated using standard methods. Haemoglobin was measured in peripheral blood samples using a portable photometer (HemoCue 301). We tested for mean haemoglobin differences between the top and bottom quartiles of HDD using t-test statistic.
Results
We collected haemoglobin on 2,025 women. The average values for HDD (24hrs and 7-day), WDD and haemoglobin for women were 4.07 (95%CI:3.98,4.16), 23.5 (95%CI:23.1,23.9), and 11.6 (95%CI:11.4,11.9) respectively. Hb concentration (mean 11.9 g/dL; 95%CI:11.9,12.0) was higher in women in the top quartiles of diet diversity when measured using 7-day recalls. However, this association disappeared when HHD was calculated using a 24hr recall period.
Conclusions
Dietary diversity indicators from a 7-day recall showed better associations with haemoglobin than those from a 24hr recall. Whether this pattern is affected by a cash transfer intervention or among those experiencing changes in food security will be tested.
