Does moderate iron deficiency protect against childhood illness? A test of the optimal iron hypothesis in Tanzania.

Objectives: To test the hypothesis that moderate iron deficiency among children is associated with lower likelihood of infection.

Materials And Methods: We use data from a population representative cross sectional study of 1164 Tanzanian children aged 6-59 months from the Tanzania Demographic and Health Survey. Respondents' iron levels were assessed through serum transferrin receptor (sTfR) and anemia was assessed using hemoglobin. C-reactive protein (CRP) was used as a marker of infection.

Results: Nearly 25% of the children were categorized as normal (iron replete, non-anemic); 45% were IDE (low iron, non-anemic), 24% were classified as IDA (low iron, anemic), and 69 children (5.9%) were anemic but had no evidence of iron deficiency. IDE was not associated with a lower likelihood of elevated CRP compared to iron replete, non-anemic children; 45% of normal children had elevated CRP compared to 51% of IDE children (P = 0.10). IDA, by contrast, was associated with a higher likelihood of elevated CRP (68%, P < 0.001). These results were unchanged when child, maternal, and household controls were added to a logistic regression model.

Discussion: Our results do not support the optimal iron hypothesis as conventionally formulated. The fact that we did not find an effect where some other studies have may be due to differences in study design, sample (e.g., age), or the baseline pathogenic ecology. Alternatively, it may be more fruitful to investigate iron regulation as an allostatic system that responds to infections adaptively, rather than to expect an optimal pre-infection value.