This cross-sectional study used latent profile analysis to identify subgroups of adults with moderate-to-severe social anxiety (N=406) based on their beliefs about what caused their social anxiety (aetiological beliefs), then compared the subgroups on clinical outcomes (social anxiety, life interference, depression) and cognitive-behavioural maintaining processes (anticipatory processing, self-focus, attention to social threat, maladaptive beliefs, safety behaviours, post-event processing).
Four profiles emerged: Low Diffuse Factors (n=85, 20.9%), High Diffuse Factors (n=122, 30.0%), Parenting Factor (n=86, 21.2%), and Dispositional and Bad Social Experiences Factors (n=113, 27.8%). The High Diffuse Factors profile—those attributing anxiety to many causes—had the worst outcomes across all 9 variables (all ANOVAs p<.001, partial η² ranging 0.13–0.28), while the Low Diffuse Factors profile had the most adaptive outcomes.
- Cross-sectional design prevents causal inference between aetiological beliefs and outcomes. - Community sample without formal SAD diagnosis limits generalisability to clinical populations. - Sample was disproportionately female, reducing generalisability to the broader population.
Ask patients with social anxiety what they believe caused their condition—those who attribute it to many different causes show the most severe symptoms and the highest cognitive-behavioural maintaining processes, suggesting these beliefs warrant direct attention in assessment and treatment planning. Addressing or narrowing a patient's aetiological beliefs (e.g., during psychoeducation) may be a practical adjunct to standard CBT or pharmacotherapy.
Explore related topics
