although the efficacy of group cognitive-behavioural therapy (gcbt) is well documented, the mechanisms by which this therapy works are not well understood. evaluating mediators and moderators of symptom change during therapy is an important step in the process of identifying the mechanisms that contribute to symptom change (kazdin, 2007). two studies were conducted in order to elucidate the process of symptom change within gcbt and to better understand how mediation of change during gcbt was evaluated in recent literature. the first study consisted of adults (n = 15) attending transdiagnostic gcbt for anxiety, whereby clients with different anxiety presentations were treated in the same group. the efficacy of this therapy group was evaluated, and cognitive change was examined as a potential mediator of symptom change. results did not indicate significant symptom improvement, and the mediation hypothesis was not supported. the small sample size is a prominent limitation that may have contributed to the lack of statistically significant findings. the second study consisted of a systematic review of recent literature to determine how mediation is assessed during gcbt. a total of 30 studies met inclusion criteria, and were rated based on when potential mediators and outcomes were assessed. the most frequently used design measured outcomes and mediator variables at pre- and post-treatment only, which is a design that cannot determine the time sequence of change. though many studies have investigated mediation, research designs that can truly identify a mediator have rarely been used. together, the findings of the present studies highlight the need for future research to investigate mediation with more rigorous designs, while also using samples that are sufficient enough in size and representativeness to provide important information about how gcbt for anxiety works.