Objective: This study aimed to determine some fundamental factors specific to posttraumatic embitterment disorder (PTED) using deep machine learning (ML) and network analysis techniques.
Method: Sociodemographic data form, Buss–Perry Aggression Questionnaire, Brief Symptom Inventory (BSI), PTED Self-Rating Scale (PTED Scale), and list of stressful life events were administered to 557 people who applied to the outpatient anxiety clinic. ML method and network analysis were applied with the 33 most significant variables.
Results: PTED was found in urban areas (p=0.006), individual health problems (p=0.029), early separation from their families (p=0.040), previous trauma (p=0.021), describing childhood sexual abuse (p<0.001), and those with the illness for more than 10 years (p<0.001) were detected at a higher rate than those without. The PTED score was higher in those with an anxiety disorder (p=0.043) and a personality disorder (p<0.001). Almost all life stressors were higher in the PTED group. There was a statistically significant difference between the groups in all subscales of the BSI. When the ML procedure was applied, sullenness was identified as the main symptom of PTED. The factors most associated with sullenness were well-being, hopelessness, and painful event experience.
Conclusion: The higher rate of chronic trauma in the group with PTED and the detection of sullenness as the main symptom have been important data for understanding the psychopathological process.