Negative affect and intolerance of uncertainty as potential mediators of change in comorbid depression in transdiagnostic CBT for anxiety

Alexander M. Talkovsky, Peter J. Norton

Research output: Contribution to journalArticleResearchpeer-review

23 Citations (Scopus)


Background: Transdiagnostic group cognitive behavioral therapy for anxiety (TGCBT) has demonstrated effectiveness in improving comorbid conditions, including depression. Mediators of change in reductions in comorbid depression have yet to be established following this anxiety-focused treatment. Negative affect (NA) and intolerance of uncertainty (IU) have demonstrated contributions to anxiety reduction in TGCBT, as has positive affect (PA) on depression reductions in other cognitive behavioral treatments. IU is also associated with depression and anxiety as a transdiagnostic vulnerability. This study evaluated the mediational effect of NA, IU, and PA on reduction in comorbid depressive symptoms in TGCBT. We hypothesized the indirect effect of the treatment through IU would be significant even when controlling for indirect effects through NA, PA, anxiety symptoms, and diagnostic severity. Method: Data collected as a part of previous clinical trials at a clinic at a large university in the southwestern United States were used. Participants included 61 individuals from a mixed anxiety-disordered sample for whom pre- and post-treatment data were available. Mediation analyses were conducted using ordinary least squares regression in path-analytic form. Results: The total effect of the model was significant (F(10, 50) = 8.40, p <.001). When controlling for potential mediators, the direct effect of treatment was not significant (B = -1.01, t = -0.40, p =.688). Analysis of indirect effects revealed significant effects of NA (B = 0.50, SE = 0.17, p =.005, 95%CI [0.02, 0.83]) and IU (B = 0.13, SE = 0.05, p =.010, 95%CI [0.03, 0.23]), but not diagnostic severity (B = 1.39, SE = 0.86, p =.114, 95%CI [-0.35, 3.12]), anxiety symptoms (B = 0.19, SE = 0.12, p =.136, 95%CI [-0.06, 0.44]), or PA (B = -0.25, SE = 0.13, p =.051, 95%CI [-0.51, 0.01] on depression symptoms when modeled with TGCBT). Conclusion: These results suggest IU contributes uniquely to the improvement of comorbid depression in TGCBT above negative affect, an established change process. Implications for treatment and theory are discussed.

Original languageEnglish
Pages (from-to)259-265
Number of pages7
JournalJournal of Affective Disorders
Publication statusPublished - 15 Aug 2018


  • Anxiety
  • Cognitive behavioral therapy
  • Depression
  • Intolerance of uncertainty
  • Transdiagnostic

Cite this