Objective: To identify the season with the highest prevalence of underweight among young children and to examine geographical variation in seasonality of underweight.
Design: This analysis is based on monthly data from a clinic-based growth monitoring programme that forms part of the National Health Information System. A regression-based technique is used to identify seasonal patterns in both underweight prevalence and attendance nationally and in 60 different districts.
Setting: The analysis covers the period 1988-1995 and is based in Zimbabwe.
Subjects: The analysis is based on weight-for-age measurements of Zimbabwean children less than 5 years old, who attended health centres as part of a growth monitoring programme.
Results: Nationally, a small but significant increase in levels of underweight takes place during January-March. Participation in growth monitoring also varies seasonally and could account for the increase observed. No evidence of seasonal variation in underweight prevalence is found in the majority of districts studied, although 11 of the districts showed a similar pattern to the national data set. This peak in the incidence of poor nutritional status also coincides with the period of food scarcity before harvest, which is also associated with higher prevalence of diarrhoea and malaria. No differences in seasonality of under-nutrition were found between districts with predominantly subsistence agriculture and those with more commercial forms of agriculture.
Conclusions: Seasonal variation in child weight-for-age exists in some parts of Zimbabwe, but its effects on cross-sectional prevalence studies are likely to be small. There are no readily discernible differences between areas that show evidence of seasonality in levels of underweight and those that do not.