A new study suggests that early trauma may shape how children’s cortisol responses to stress relate to anxiety and behavioural problems, with different trajectories emerging over the following two years. The research tracked 189 young people aged 8 to 15 from Charleston, South Carolina, of whom 107 had experienced at least one potentially traumatic event before age six and 82 had not. The study was led by Gretchen R. Perhamus of the Medical University of South Carolina.
Researchers assessed the stress response in a laboratory setting using a three‑minute social task in which each young person spoke to a camera as though auditioning for a show, while a judge watched. Saliva samples were collected before, after, and during recovery to measure cortisol. Usable data were available for 145 participants. The team distinguished two measures: total cortisol output during the task and the rise in cortisol from baseline, arguing these may reflect different facets of the body’s stress system.
Among youths with early trauma exposure, higher overall cortisol output during the task was linked with more externalising symptoms at the time of measurement and with externalising problems being more prominent than anxiety. Interestingly, higher overall cortisol output was associated with less anxiety in both groups.
Over the next two years, the trajectories diverged. For those with early trauma, a smaller cortisol increase during the task predicted a shift toward externalising problems relative to anxiety, while a larger increase predicted a shift toward anxiety. By contrast, in youths with no trauma history, a larger cortisol rise forecast a decrease in anxiety over time.
The study found that the results depended on which cortisol measure and symptom comparison the researchers used. Overall cortisol output tended to relate to current symptoms, whereas the rise in cortisol was more informative about future changes in symptom patterns. Several associations among trauma-exposed youths emerged only when the analysis focused on which type of problem was more prominent, rather than on anxiety and externalising symptoms separately.
Participants in the two groups did not differ in their overall cortisol responses. Although most youths in the early-trauma group had experienced only one or two such events, the researchers noted that exposure level did not alter the links between cortisol and symptoms. The trauma group, however, tended to come from households with lower average incomes and were less likely to be white.
Perhamus and colleagues summarised the complexity of the findings: “the specific pattern of findings was complex and depended on whether models examined concurrent or longitudinal associations, and whether they examined symptom domains or directionality.”
Several limitations were acknowledged. Trauma histories were reported retrospectively, and the study did not assess whether participants later encountered additional traumatic events or how puberty may have influenced the results. The researchers measured cortisol alone, used broad measures of anxiety and externalising problems, and drew on a relatively small community sample, meaning replication is needed to confirm the findings.
The authors, including Gretchen R. Perhamus, Rachel E. Siciliano, Tiffany Phu, and colleagues, emphasise that while the results are not conclusive, they point to a nuanced relationship between early experiences, stress physiology, and mental health trajectories in young people.
