Journal of Traumatic Stress · 2019 · 57 citations · 33 references
MedicinePsychiatryPsychologyAbnormal PsychologyBehavioral HealthTrauma-related DisordersPost-traumatic Stress DisorderStress-related DisordersStressor-related DisordersTrauma AssessmentTrauma DisordersInternational Classification Of DiseasesPartial Correlation Network InferencePartial Correlation NetworksSymptom Cluster InteractionsSymptom Interactions MeasurementSymptom InterdependenceTrauma-related Disorders DiagnosisTrauma-related Psychopathology
The 11th revision of the World Health Organization's International Classification of Diseases (ICD-11) includes a new disorder, complex posttraumatic stress disorder (CPTSD). The network approach to psychopathology enables investigation of the structure of disorders at the symptom level, which allows for analysis of direct symptom interactions. The network structure of ICD-11 CPTSD has not yet been studied, and it remains unclear whether similar networks replicate across different samples. We investigated the network models of four different trauma samples that included a total of 879 participants (M age = 47.17 years, SD = 11.92; 59.04% women) drawn from Austria, Lithuania, and Scotland and Wales in the United Kingdom. The International Trauma Questionnaire was used to assess symptoms of ICD-11 CPTSD in all samples. The prevalence of PTSD and CPTSD ranged from 23.7% to 37.3% and from 9.3% to 53.1%, respectively. Regularized partial correlation networks were estimated and the resulting networks compared. Despite several differences in the symptom presentation and cultural background, the networks across the four samples were considerably similar, with high correlations between symptom profiles (ρs = .48-.87), network structures (ρs = .69-.75), and centrality estimates (ρs = .59-.82). These results support the replicability of CPTSD network models across different samples and provide further evidence about the robust structure of CPTSD. The most central symptom in all four sample-specific networks and the overall network was "feelings of worthlessness." Implications of the network approach in research and practice are discussed.
33
R: A Language and Environment for Statistical Computing
R Core Team · 2000 · 353K citations · Full text
Natural Sciences, Computer Science, Statistical Computing +10