2 - (SYM 13) Using Stochastic Search Variable Selection (SSVS) to Identify Key Neural and Psychological Correlates of PTSD Symptoms in a Large Nonclinical Sample

A multitude of factors have been associated with post-traumatic stress symptoms, including neural, psychological, and behavioral correlates. However, using traditional statistical approaches, even large samples are insufficiently powered to determine which factors are most important for predicting post-traumatic stress symptoms and adequately adjust for the large number of comparisons. This is because with multiple correlated predictors, the strength of effects changes depending on which other predictors are controlled in the model, and as a result, it is unclear which factors robustly predict outcomes. Further, research on psychological factors and neural substrates of trauma-related psychopathology is usually conducted in isolation. It is not known for example whether brain network connectivity and psychological measures explain overlapping or distinct variability in symptoms. Stochastic search variable selection (SSVS) is a Bayesian method for selecting a subset of important predictors, controlling for uncertainty in which predictors are included in the model. By accounting for model uncertainty, SSVS can increase power and protect against false-positive results.
We used SSVS to identify the most important predictors for overall PTSD symptoms and individual symptom dimensions in a representative sample of 569 adults (18-85 years of age). Candidate predictors included psychological measures, behavioral measures, and resting state functional connectivity among previously identified brain regions. In a follow-up analysis we compared results controlling for current depression symptoms. Results supported the importance of measures across domains in explaining PTSD symptoms. Poor sleep quality and dimensions of temperament and impulsivity were consistently associated with greater PTSD symptom severity. In addition to self-report measures, brain functional connectivity among regions commonly ascribed to the default mode network, central executive network, and salience network explained unique variability of PTSD symptoms. This study demonstrates the unique contributions of psychological measures and neural substrates of PTSD symptoms.