Supplementary Material for: Beyond Self-Reports: Integrating Cortisol Measurement in Psychotherapy Process Research among Adolescents with Borderline Personality Pathology

Psychotherapy is the primary treatment for adolescent Borderline Personality Pathology (BPP), yet its mechanisms remain unclear. Given potential self-report biases due to alexithymia and impaired interoception, this study examined cortisol responses as a physiological stress marker alongside session ratings from adolescent BPP patients and their therapists to assess its potential as a complementary measure in psychotherapy process research.

N = 56 adolescents (94.6% female) with BPP (≥ 3 DSM-IV BPD criteria) receiving Adolescent Identity Treatment or Dialectical Behavioral Therapy and their therapists provided pre- and post-session salivary cortisol samples and completed the Session Evaluation Questionnaire. Residual Dynamic Structural Equation Modelling examined associations between cortisol responses and session ratings, with moderation by age, depression, BPP severity, identity diffusion, and trauma.

Cortisol responses did not correlate with session ratings in patients. In therapists, higher cortisol responses were associated with lower session smoothness (r = -.164; p < .001) and deepness (r = -.086; p = .004), as well as with lower positivity (r = -.145; p < .001) and higher arousal (r = .072; p = .012) post-session. Higher depression levels moderated the association between cortisol responses and session deepness in patients (β = -.009, p = .007).

While session ratings and cortisol responses correlated in therapists, no such correlation was found in patients. Possible mechanisms include altered interoceptive abilities, dysregulated hypothalamic-pituitary-adrenal (HPA) axis function, or increased variability in self-ratings or cortisol due to BPP. While physiological markers and self-reports offer complementary insights into psychotherapy processes, future studies should include healthy and clinical controls and baseline investigation of HPA axis function (i.e. stress reactivity) in both groups, along with additional hormonal markers.