Associations between PTSD and psychotic symptoms: A network analysis of patients with psychotic disorders in Uganda Psychosis-PTSD network analysis

Abstract

Introduction: Individuals with psychosis often endorse a history of trauma, yet the symptom-level interplay between trauma-related and psychotic symptoms is understudied globally and largely absent in Sub-Saharan Africa. We used network analysis to examine associations within and between these domains. Methods: We conducted a network analysis of 807 participants with a psychotic disorder and trauma history in the Ugandan cohort of the NeuroGAP-Psychosis program. Psychotic symptoms were assessed using the standard Mini International Neuropsychiatric Interview (MINI) version 7.0.2's module K for psychotic disorders and mood disorder with psychotic features, and trauma-related symptoms were assessed using the Post Traumatic Stress Disorder Checklist for DSM-5 (PCL-5). We estimated a mixed graphical model with Extended Bayesian Information Criterion regularization (gamma = 0.25) and pairwise interactions (k = 2) to examine PTSD and psychosis symptom networks, computed two-step bridge expected influence, and bootstrapped (n = 1000) to assess network stability and accuracy. Results: Participants' average PCL-5 score was 28.0 (SD = 23.3). Network analysis revealed clustering in both PTSD and psychosis symptoms; psychotic symptoms clustered more strongly together than PTSD symptoms did. Four cross-domain edges, 1) negative symptoms with trouble recalling; 2) odd or unusual beliefs with difficulty concentrating; 3) disorganized speech with negative feelings; 4) odd or unusual beliefs with flashbacks, were identified. Trouble recalling showed the highest mean 2-step bridge expected influence. Discussion: Memory-related symptoms such as trouble recalling or flashbacks emerged as potential bridge symptoms between PTSD and psychosis in this population. These symptoms could serve as transdiagnostic intervention targets and warrant longitudinal investigation.