Publication: Hair cortisol concentrations in war-affected adolescents: A prospective intervention trial


Dajani, Rana; Hadfield, Kristin; van Uum, Stan; Greff, Michael; Panter-Brick, Catherine (2018) Hair cortisol concentrations in war-affected adolescents: A prospective intervention trial. Psychoneuroendocrinology. 89: 138-146. doi: 10.1016/j.psyneuen.2017.12.012


  • Three trajectories describe cortisol production of refugee and non-refugee youth.
  • Insecurity and female gender predicted a pattern of cortisol hypersecretion.
  • Trauma exposure predicted within-individual cortisol dysregulation.
  • The stress-attunement intervention decreased hair cortisol concentrations by a third.
  • Effective interventions may normalize cortisol production in war-affected adolescents


Temporal examinations of the biological signature of stress or trauma in war-affected populations are seldom undertaken. Moreover, few studies have examined whether stress biomarkers track biological sensitivity to brief interventions targeting the improvement of psychosocial wellbeing. Our study is the first to prospectively examine, in war-affected adolescents, the associations between hair cortisol concentrations (HCC) and self-reports of stress, insecurity, posttraumatic reactions, and lifetime trauma. We conducted a randomized controlled trial to test the impact of an 8-week intervention based on profound stress attunement. We collected data for a gender-balanced sample of 733 Syrian refugee (n=411) and Jordanian non-refugee (n=322) adolescents (12–18 years), at three time-points. We used growth mixture models to classify cortisol trajectories, and growth models to evaluate intervention impact on stress physiology. We observed three trajectories of HCC: hypersecretion, medium secretion, and hyposecretion (9.6%, 87.5% and 2.9% of the cohort, respectively). For every one percent increase in levels of insecurity, adolescents were 0.02 times more likely to have a trajectory of hypersecretion (95% CI: 1.00, 1.03, p=0.01). For each additional symptom of posttraumatic stress reported, they were 0.07 times less likely to show hyposecretion (95% CI: 0.89, 0.98, p=0.01). Indeed, stronger posttraumatic stress reactions were associated with a pattern of within-individual cortisol dysregulation and medium secretion. Overall, HCC decreased by a third in response to the intervention (95% CI: −0.19, −0.03, p=0.01). While the intervention decreased HCC for youth with hypersecretion and medium secretion, it increased HCC for youth with hyposecretion (95% CI: 0.22, 1.16, p=0.004), relative to controls. This suggests a beneficial regularization of cortisol levels, corroborating self-reports of improved psychosocial wellbeing. We did not find evidence to suggest that gender, resilience, or posttraumatic stress disorder influenced the strength or direction of responses to the intervention. This robust impact evaluation exemplifies the utility of biomarkers for tracking physiological changes in response to interventions over time. It enhances the understanding of trajectories of endocrine response in adverse environments and patterns of stress responsivity to ecological improvement.

Full Text:

The full text is available via the publisher.

Media Attention: