Latent state trait modeling of children's cortisol at two points of the diurnal cycle

Abstract

One challenge in examining stable individual differences in basal activity of the HPA axis is controlling for internally or externally based situational factors that lead to day-to-day variation in ambulatory cortisol. Disturbed basal activity is of particular interest in studies with children, for whom a dysregulated HPA axis may play an etiologic role in emotional or health outcomes. The purpose of this study was to determine whether trait vs. situationally specific sources of variation can be identified at different points of the diurnal cycle in children and if so, whether state and trait components vary according to time of measurement. Early morning and late evening salivary cortisol was collected from 164 children aged 7 to 11 years. Samples were collected 30 min after wake-up and 30 min before bedtime on 3 weekdays. State, trait, and error components of cortisol levels were assessed using a latent state trait model. Possible influences of sampling day and outlier treatment on parameter estimates were examined. The results showed that a latent trait factor superimposed on state residuals and measurement error was identified for both early morning and late evening cortisol. Model fit was excellent and criteria for invariance tests were met. Trait factors accounted for 41% and 57% of the variance in morning and evening cortisol, respectively. These findings suggest cortisol attributed to trait factors can be identified and are of substantial magnitude at both the peak and nadir of the diurnal cycle. Latent state trait modeling is a potentially useful tool in understanding the role of stable individual differences in cortisol levels for development and health.