Examining the causal effects of changes in iron status on brain and attentional functioning using structural equation modeling

Abstract

Iron deficiency (ID) in otherwise healthy adults can impair perceptual and cognitive functioning, along with brain dynamics (as measured by EEG). We used structural equation modeling (SEM) to test the hypothesis that changes in iron status lead to changes in brain functioning and then to changes in behavioral measures. The data used here were taken from 54 ID (sFt <15 µg/L) participants in a randomized double‐blind efficacy study. Blood, behavioral, and EEG measures were taken at baseline, endline, and at one of eight random time‐points in between. Five measures of iron status, behavioral data from three tasks measuring attentional capture and control, and peak amplitudes of a set of event related potentials (ERP) measures taken during the tasks at baseline were used as indicators of separate latent constructs in a set of contrasting SEMs. Thus far, a model with three latent factors provides the best description of the data, with the three latent factors being (a) systemic iron (sFt and sTfR) directly influencing (b) brain dynamics (ERP amplitudes), with brain dynamics then influencing (c) attentional function (reaction times from the three tasks). This model provides a better description of the data than does a model representing no causal link from blood to brain to behavior, or a model in which oxygen delivery (as indicated by hemoglobin, red blood cell distribution width, and mean cell volume) influences brain dynamics and then attentional functioning. These results suggest the utility of SEMs to relate changes in blood, brain, and behavior.