Hair concentrations of cortisol and dehydroepiandrosterone (DHEA) are increasingly recognized as non-invasive, retrospective biomarkers of hypothalamic-pituitary-adrenal axis function. The ratio of these hormones may reflect the balance between catabolic and anabolic activity, potentially serving as a composite marker of chronic stress. Despite its theoretical relevance, the clinical utility of the hair cortisol: DHEA ratio remains unclear. This PROSPERO-registered systematic review, done according to PRISMA guidelines, aimed to evaluate the association between the hair cortisol: DHEA ratio and perceived stress in adults with chronic stress exposure. A comprehensive search of five electronic databases and grey literature was conducted up to March 5, 2025. Studies were eligible if they included adults (≥18 years) with at least 4 weeks of documented chronic stress exposure and reported both hair cortisol and DHEA concentrations alongside validated perceived stress measures. Meta-analyses were performed for correlation coefficients and group comparisons. Risk of bias was assessed using a modified Downs and Black checklist. Eight studies (five cross-sectional, three cohort; n = 60-210) met inclusion criteria. A meta-analysis of two studies using validated stress scales showed no significant association between the hair cortisol/DHEA ratio and perceived stress (ρ = 0.09; 95% CI: -0.07 to 0.26; I² = 0%). However, a separate analysis of three studies comparing high- versus low-stress groups revealed a moderate, statistically significant pooled effect (Hedges' g = 0.69; 95% CI: 0.14-1.24; p = .01) with substantial heterogeneity (I² = 77.17%). Risk of bias ranged from moderate to low, but the overall certainty of evidence was rated very low due to small sample sizes and inconsistent findings. The hair cortisol: DHEA ratio has biological plausibility as a marker of chronic stress, but current evidence is insufficient to support its clinical application. Further standardized, longitudinal studies are needed to establish its diagnostic and prognostic value. CRD420251003364.
Huchegowda et al. (2025) studied this question.