Treatment - CBT
Perceived social support, anxiety sensitivity, and symptoms of anxiety and depression: longitudinal mediation analyses within a randomized controlled trial for anxiety disorders
Allison V. Metts, M.A.
PhD Candidate
University of California, Los Angeles
LOS ANGELES, California
Peter Roy-Byrne, M.D.
Professor of Psychiatry
University of Washington School of Medicine
Seattle, Washington
Cathy Sherbourne, Ph.D.
Senior Health Policy Analyst
RAND Corporation
Santa Monica, California
Murray Stein, M.P.H., M.D.
Distinguished Professor of Psychiatry and Public Health
University of California San Diego
San Diego, California
Greer Sullivan, M.D., Other
Professor
University of Arkansas for Medical Sciences
Little Rock, Arkansas
Alexander Bystritsky, M.D., Ph.D.
Professor of Psychiatry and Biobehavioral Sciences
University of California, Los Angeles
Los Angeles, California
Michelle G. Craske, Ph.D.
Distinguished Professor
University of California Los Angeles
1285 Franz Hall, California
Background: Methods to enhance efficacy of cognitive behavioral therapy (CBT) for anxiety disorders are needed. Social support is theorized to reduce psychological and physiological stress responses. Because stressors can mitigate coping-style reappraisals, individuals under stress often turn to others for assistance. In the case of anxiety, somatic symptoms themselves are stressors and negative beliefs about these symptoms worsen and maintain symptoms. Social support may encourage reappraisal of symptoms and thereby changes in anxiety sensitivity. Because reappraisal skills are a key component of CBT, these processes may be strengthened as a function of receiving CBT. The current study evaluated whether anxiety sensitivity mediated the association between perceived social support and anxiety and depression symptoms in a randomized controlled trial for anxiety disorders that compared evidence-based intervention (CALM) to usual care (UC).
Method: Data from 961 outpatients with anxiety disorders over an 18-month follow-up period were examined in a series of multilevel mediation models to predict anxiety and depression symptoms. Models examined indirect effects of perceived social support on symptoms through anxiety sensitivity within-person and between-person and compared the magnitude of effects between CALM and UC. Tests of specificity were performed.
Results: Within-level indirect effects were significant in CALM such that anxiety sensitivity decreases mediated the relationship between perceived social support increases and lower symptoms of anxiety (Est. = -1.09, SE = .15, p < .001) and depression (Est. = -.67, SE = .11, p < .001). Specificity tests indicated that within-person associations were specific to anxiety. Between-level indirect effects were significant in CALM and UC such that lower average anxiety sensitivity mediated the relationship between higher average perceived social support and lower anxiety symptoms (CALM: Est. = -1.61, SE = .28, p < .001; UC: Est. = -.89, SE = .24, p < .001) and depression symptoms (CALM: Est. = -.96, SE = .17, p < .001; UC: Est. = -.50, SE = .13, p < .001) in both groups. Between-person effects were larger in CALM and lacked specificity.
Discussion: Within-level mediation findings suggest that increases in social support that occur in CALM may lead to changes in symptoms through reductions in anxiety sensitivity. Specificity test results suggest that depression effects are likely due to comorbidity of depression and anxiety in our sample. Between-level mediation findings suggest that higher social support and lower anxiety sensitivity overall are associated with less general distress and are strengthened over the course of and following CBT. This study adds to previous work by suggesting that perceived social support may also enable more effective reappraisal of somatic symptoms. Thus, CBT for anxiety may benefit from increased attention to perceived social support to strengthen one’s ability to reappraise somatic symptoms. Strategies include careful assessment of social support at intake and over the course of treatment, psychoeducation on the utility of social support for beliefs about symptoms, and challenging erroneous perceptions of social support increase one’s resources.