Prospective prediction of PTSD diagnosis in a nationally representative sample using machine learning

Abstract

Background Recent research has identified a number of pre-traumatic, peri-traumatic and post-traumatic psychological and ecological factors that put an individual at increased risk for developing PTSD following a life-threatening event. While these factors have been found to be associated with PTSD in univariate analyses, the complex interactions of these risk factors and how they contribute to individual trajectories of the illness are not yet well understood. In this study, we examine the impact of prior trauma, psychopathology, sociodemographic characteristics, community and environmental information, on PTSD onset in a nationally representative sample of adults in the United States, using machine learning methods to establish the relative contributions of each variable. Methods Individual risk factors identified in Waves 1 of the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) were combined with community-level data for the years concurrent to the NESARC Wave 1 (n = 43,093) and 2 (n = 34,653) surveys. Machine learning feature selection and classification analyses were used at the national level to create models using individual- and community-level variables that would best predict the new onset of PTSD at Wave 2. Results Our classification algorithms yielded 89.7 to 95.6% accuracy for predicting new onset of PTSD at Wave 2. A prior diagnosis of DSM-IV-TR Borderline Personality Disorder, Major Depressive Disorder or Anxiety Disorder conferred the greatest relative influence in new diagnosis of PTSD. Distal risk factors such as prior psychiatric diagnosis accounted for significantly greater relative risk than proximal factors (such as adverse event exposure). Conclusions Our findings show that a machine learning classification approach can successfully integrate large numbers of known risk factors for PTSD into stronger models that account for high-dimensional interactions and collinearity between variables. We discuss the implications of these findings as pertaining to the targeted mobilization emergency mental health resources. These findings also inform the creation of a more comprehensive risk assessment profile to the likelihood of developing PTSD following an extremely adverse event.

One Year after the Flood: Prevalence and Correlates of Post-Traumatic Stress Disorder among Residents in Fort McMurray

“…Although compared to males and older age groups, females and the younger age group expressed a higher prevalence of PTSD symptoms, the difference was slight and not significant. This conflicts with previous studies, where females and younger age groups were indicated to be more likely to express mental health problems, including likely PTSD symptoms, compared to males and older age groups [44][45][46][47][48]. Since our study sample size was relatively small, it may not have elicited that difference as reported in other studies.…”

“…The majority of respondents who had a depression history (67.3%) reported having likely PTSD symptoms after the flood. This finding is highly consistent with research on the main effects of prior mental health disorders on PTSD [3,45,[49][50][51][52]. Much research has determined a relationship between depression and PTSD among those who have experienced trauma, including natural disasters, childhood maltreatment, serious accident, or terrorist act [53][54][55].…”

“…Much research has determined a relationship between depression and PTSD among those who have experienced trauma, including natural disasters, childhood maltreatment, serious accident, or terrorist act [53][54][55]. Worthington's research had identified past borderline personality disorder diagnosis and past depression as two of the most significant prediction for the onset of PTSD after the disaster [45]. In addition, a study with 1820 respondents who supported the mission in Afghanistan [51] showed that the prevalence of PTSD was highest among those with a pre-deployment history of depression [51].…”

Psychometric properties and validity of the German version of the Post-Traumatic Diagnostic Scale for DSM-5 (PDS-5)

“…Obviously plausible is, of course, the notion of diagnostic overlap between PTSD and frequent comorbid diagnoses, as depression or anxiety. Recent progress in prospective research designs (DiGangi et al, 2013 ; Worthington, Mandavia, & Richardson-Vejlgaard, 2020 ) additionally supports the pre-dispositional model (McFarlane, 2004 ) assuming that especially individuals suffering from specific pre-existing disorders may develop symptoms of posttraumatic stress following the experiences of potentially traumatic events.…”

Uncovering heterogeneous associations of disaster‐related traumatic experiences with subsequent mental health problems: A machine learning approach

“…16,17 Although several studies have applied ML algorithms to predict PTSD, our study is distinct because these studies focused on pure prediction of PTSD rather than estimating effect heterogeneity, which is our focus. [18][19][20][21] We leveraged a unique natural experiment setting stemming from the 2011 Great East Japan Earthquake and Tsunami, where a longitudinal cohort study of Japanese older adults established 7 months before the earthquake onset offered an opportunity to collect rich predisaster data of the survivors. 22…”