10 The association between underweight and iron status in early childhood: cross-sectional and prospective study

Abstract Background The American Academy of Pediatrics (AAP) identifies poor growth as a risk factor for iron deficiency (ID). There is limited research to provide evidence of this association. Objectives To examine the association between underweight and ID in early childhood. Design/Methods We enrolled healthy, young children from primary care practices in Toronto, Canada, and used both a cross-sectional and prospective study design. Underweight was defined as a body mass index z-score (zBMI) <-2. In the cross-sectional cohort, children 12-29 months had concurrent measurement of zBMI and serum ferritin. In the prospective cohort, children 6-14 months had measurement of zBMI and at 15-29 months had measurement of serum ferritin. Multivariable regression models examined serum ferritin as a continuous variable (linear models) and categorical variable with ID defined as <12 µg/L (logistic model), adjusted for pre-specified covariates. Results For the cross-sectional cohort (n=1953), the mean age was 18.3 (SD 5.0) months, 51 (2.4%) were underweight, and 269 (13.8%) had ID. There was no association between underweight and serum ferritin (change in median serum ferritin 2.01 µg/L, 95% CI -2.22, 7.01, P=0.37) or underweight and ID (OR 0.98, 95% CI 0.43, 2.21, P= 0.95). In contrast, there was a strong negative association between overweight and serum ferritin (change in median serum ferritin -4.19 µg/L, 95% CI -6.93, -1.04, P=0.01), and a positive association between obesity and ID (OR 3.35, 95% CI 1.10, 10.25, P=0.03). For the prospective cohort (n=672), the mean age at outcome was 21.1 (SD 3.5) months, 34 (5.1%) were underweight, and 104 (15.5%) had ID. There was no association between underweight and serum ferritin or underweight and ID. Conclusion Using both a cross-sectional and prospective study design, we found no association between underweight and ID in young children 1 to 2 years of age. In contrast, we found a strong association between overweight/obesity and ID. For risk stratification and targeted screening, overweight/obesity may be more important than underweight as a risk factor for ID in young children.