Geo-spatial patterns and associated risks of iron deficiency and infection among young Ghanaian children: implications for the safety of iron supplementation in malaria endemic areas

Abstract

Background: The safety and effectiveness of iron supplementation in malaria endemic areas may partly depend on host iron status; however, current methods for assessing iron deficiency risk tend to be confounded by infection and are infeasible to implement at a population level. Determining the geographical patterns of iron status and infection may provide a practical alternative means of identifying high risk populations for whom integrated anaemia and infection control programs are needed. Objective: Determine the geo-spatial factors associated with iron status and infection risk among 1943 Ghanaian children (6-35 months of age) before and after participating in a randomized iron home-fortification trial. Methods: Secondary spatial analyses of iron status and infection outcomes were conducted. Iron status was defined as serum ferritin concentration corrected for inflammation (C-reactive protein, CRP) using a regression-based method. Malaria and non-malaria infection outcomes iii iii were defined using four combinations of inflammation (CRP >5 mg/L) and malaria parasitaemia (with and without reported history of fever or concurrent axillary temperature >37.50 C). Analyses were performed using a geographical information system (GIS) and generalized linear geostatistical modelling with a Matern spatial correlation function. Results: After adjusting for demographic characteristics such as age, sex, and maternal education, none of the geo-spatial factors included in the iron status models (including elevation, and distance to a health facility) demonstrated associations at baseline or endline; however, there was significant residual spatial variation across the study area. Conversely, malaria parasitaemia at baseline was associated with greater distance to a health facility and lower elevation. These relationships did not remain at endline, nor when infection was defined using CRP only. Mapping the model outputs showed defined low-risk areas that tended to cluster around villages, particularly near the District centre. Conclusions: In a malaria endemic area, geographical location may play a role in the risk of iron deficiency and infection among children. Iron home-fortification likely alters the spatial risk profile of malaria and non-malaria infection in this setting, though additional research is needed to confirm the direction of these relationships