TY - JOUR
T1 - Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data
T2 - A machine learning multicenter cohort study
AU - Schultebraucks, Katharina
AU - Sijbrandij, Marit
AU - Galatzer-Levy, Isaac
AU - Mouthaan, Joanne
AU - Olff, Miranda
AU - van Zuiden, Mirjam
N1 - Funding Information: The TraumaTIPS study was supported by ZonMw, the Netherlands Organization for Health Research and Development (grant # 62300038 ), the Hague, the Netherlands, and by Stichting Achmea Slachtoffer en Samenleving (SASS) , Aid to Victims, Zeist, the Netherlands. The current project was additionally supported by an Amsterdam Public Health Research Institute Alliance Grant, Amsterdam, the Netherlands , within the Mental Health Program. Katharina Schultebraucks was supported by the German Research Foundation (SCHU 3259/1-1 ). Mirjam van Zuiden was supported by a Veni grant from ZonMw, the Netherlands organization for Health Research and Development (# 91617037 ), the Hague, the Netherlands.. The funders were not involved in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. Publisher Copyright: © 2021 The Authors
PY - 2021/5/1
Y1 - 2021/5/1
N2 - The necessary requirement of a traumatic event preceding the development of Posttraumatic Stress Disorder, theoretically allows for administering preventive and early interventions in the early aftermath of such events. Machine learning models including biomedical data to forecast PTSD outcome after trauma are highly promising for detection of individuals most in need of such interventions. In the current study, machine learning was applied on biomedical data collected within 48 h post-trauma to forecast individual risk for long-term PTSD, using a multinominal approach including the full spectrum of common PTSD symptom courses within one prognostic model for the first time. N = 417 patients (37.2% females; mean age 46.09 ± 15.88) admitted with (suspected) serious injury to two urban Academic Level-1 Trauma Centers were included. Routinely collected biomedical information (endocrine measures, vital signs, pharmacotherapy, demographics, injury and trauma characteristics) upon ED admission and subsequent 48 h was used. Cross-validated multi-nominal classification of longitudinal self-reported symptom severity (IES-R) over 12 months and bimodal classification of clinician-rated PTSD diagnosis (CAPS-IV) at 12 months post-trauma was performed using extreme Gradient Boosting and evaluated on hold-out sets. SHapley Additive exPlanations (SHAP) values were used to explain the derived models in human-interpretable form. Good prediction of longitudinal PTSD symptom trajectories (multiclass AUC = 0.89) and clinician-rated PTSD at 12 months (AUC = 0.89) was achieved. Most relevant prognostic variables to forecast both multinominal and dichotomous PTSD outcomes included acute endocrine and psychophysiological measures and hospital-prescribed pharmacotherapy. Thus, individual risk for long-term PTSD was accurately forecasted from biomedical information routinely collected within 48 h post-trauma. These results facilitate future targeted preventive interventions by enabling future early risk detection and provide further insights into the complex etiology of PTSD.
AB - The necessary requirement of a traumatic event preceding the development of Posttraumatic Stress Disorder, theoretically allows for administering preventive and early interventions in the early aftermath of such events. Machine learning models including biomedical data to forecast PTSD outcome after trauma are highly promising for detection of individuals most in need of such interventions. In the current study, machine learning was applied on biomedical data collected within 48 h post-trauma to forecast individual risk for long-term PTSD, using a multinominal approach including the full spectrum of common PTSD symptom courses within one prognostic model for the first time. N = 417 patients (37.2% females; mean age 46.09 ± 15.88) admitted with (suspected) serious injury to two urban Academic Level-1 Trauma Centers were included. Routinely collected biomedical information (endocrine measures, vital signs, pharmacotherapy, demographics, injury and trauma characteristics) upon ED admission and subsequent 48 h was used. Cross-validated multi-nominal classification of longitudinal self-reported symptom severity (IES-R) over 12 months and bimodal classification of clinician-rated PTSD diagnosis (CAPS-IV) at 12 months post-trauma was performed using extreme Gradient Boosting and evaluated on hold-out sets. SHapley Additive exPlanations (SHAP) values were used to explain the derived models in human-interpretable form. Good prediction of longitudinal PTSD symptom trajectories (multiclass AUC = 0.89) and clinician-rated PTSD at 12 months (AUC = 0.89) was achieved. Most relevant prognostic variables to forecast both multinominal and dichotomous PTSD outcomes included acute endocrine and psychophysiological measures and hospital-prescribed pharmacotherapy. Thus, individual risk for long-term PTSD was accurately forecasted from biomedical information routinely collected within 48 h post-trauma. These results facilitate future targeted preventive interventions by enabling future early risk detection and provide further insights into the complex etiology of PTSD.
KW - Biomarkers
KW - HPA axis
KW - Machine learning
KW - PTSD
KW - Pharmacotherapy
KW - Prognosis
KW - Thyroid hormones
KW - Traumatic stress
UR - http://www.scopus.com/inward/record.url?scp=85099915098&partnerID=8YFLogxK
U2 - https://doi.org/10.1016/j.ynstr.2021.100297
DO - https://doi.org/10.1016/j.ynstr.2021.100297
M3 - Article
C2 - 33553513
SN - 2352-2895
VL - 14
JO - Neurobiology of Stress
JF - Neurobiology of Stress
M1 - 100297
ER -