The impact of reduced worry on general functioning: A mediation analysis from a randomized trial

Abstract

Previous lab findings have indicated that excessive worry may impair cognitive performance and problem solving capabilities but it is unclear if excessive worry also leads to broader impairments in general functioning. We report a secondary process data analysis of a large randomized waitlist-controlled trial (N = 670) of a self-guided online psychological intervention for dysfunctional worry related to the Covid-19 pandemic. Specific aims were to investigate (1) if improvements in general functioning were mediated by reductions in worry related to Covid-19 during the acute intervention phase, and (2) if reduced worry related to Covid-19 during acute intervention phase had a positive long-term impact on general functioning up to 1 year after the end of the intervention. To address aim 1, we used a mediation analysis framework where outcome (general functioning measured with an adapted version of the Work and Social Adjustment Scale) and the hypothesized mediator (worry measured with an adapted version of the Generalised Anxiety Disorder 7-Item Scale) were administered weekly during the controlled phase of the trial of 3 weeks. To address aim 2, we investigated if reductions in worry during the 3-week treatment period predicted improved general functioning at 1- and 12 months after treatment completion. Results showed that improvements in general functioning at week 3 were mediated by reductions in worry during the first 2 weeks of treatment (indirect effect estimate −0.08; 95% CI −0.15, −0.02). A sensitivity analysis indicated that the mediation effects dropped significantly when the residual correlation values between the mediator and the outcome exceeded r = 0. A reversed causation model was not significant. Additionally, reductions in worry during treatment predicted subsequent improvements in general functioning at both 1- and 12-month follow-ups (p