Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study
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...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2021-05-01
|
Series: | Neurobiology of Stress |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2352289521000059 |
id |
doaj-bcb1a3dedc854563b2e0bcdd6ba59eab |
---|---|
record_format |
Article |
spelling |
doaj-bcb1a3dedc854563b2e0bcdd6ba59eab2021-05-16T04:23:39ZengElsevierNeurobiology of Stress2352-28952021-05-0114100297Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort studyKatharina Schultebraucks0Marit Sijbrandij1Isaac Galatzer-Levy2Joanne Mouthaan3Miranda Olff4Mirjam van Zuiden5Vagelos School of Physicians and Surgeons, Department of Emergency Medicine, Columbia University Medical Center, New York, NY, United States of America; Data Science Institute, Columbia University, New York, New York, USAVrije Universiteit, Department of Clinical, Neuro- and Developmental Psychology; Amsterdam Public Health Research Institute, World Health Organization Collaborating Centre for Research and Dissemination of Psychological Interventions, Amsterdam, the NetherlandsDepartment of Psychiatry, New York University School of Medicine, New York, New York, USADepartment of Clinical Psychology, Institute of Psychology, Faculty of Social and Behavioural Sciences, Leiden University, Leiden, the NetherlandsARQ National Psychotrauma Centre, Diemen, the Netherlands; Department of Psychiatry, Amsterdam University Medical Centers, Location Amsterdam Medical Center, University of Amsterdam, Amsterdam Public Health Research Institute and Amsterdam Neuroscience Research Institute, Amsterdam, the NetherlandsDepartment of Psychiatry, Amsterdam University Medical Centers, Location Amsterdam Medical Center, University of Amsterdam, Amsterdam Public Health Research Institute and Amsterdam Neuroscience Research Institute, Amsterdam, the Netherlands; Corresponding author. Amsterdam University Medical Centers, location Amsterdam Medical Center, University of Amsterdam, Department of Psychiatry, Meibergdreef 9, 1105, AZ, Amsterdam, the Netherlands.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.http://www.sciencedirect.com/science/article/pii/S2352289521000059PTSDTraumatic stressBiomarkersPrognosisMachine learningHPA axis |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Katharina Schultebraucks Marit Sijbrandij Isaac Galatzer-Levy Joanne Mouthaan Miranda Olff Mirjam van Zuiden |
spellingShingle |
Katharina Schultebraucks Marit Sijbrandij Isaac Galatzer-Levy Joanne Mouthaan Miranda Olff Mirjam van Zuiden Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study Neurobiology of Stress PTSD Traumatic stress Biomarkers Prognosis Machine learning HPA axis |
author_facet |
Katharina Schultebraucks Marit Sijbrandij Isaac Galatzer-Levy Joanne Mouthaan Miranda Olff Mirjam van Zuiden |
author_sort |
Katharina Schultebraucks |
title |
Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study |
title_short |
Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study |
title_full |
Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study |
title_fullStr |
Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study |
title_full_unstemmed |
Forecasting individual risk for long-term Posttraumatic Stress Disorder in emergency medical settings using biomedical data: A machine learning multicenter cohort study |
title_sort |
forecasting individual risk for long-term posttraumatic stress disorder in emergency medical settings using biomedical data: a machine learning multicenter cohort study |
publisher |
Elsevier |
series |
Neurobiology of Stress |
issn |
2352-2895 |
publishDate |
2021-05-01 |
description |
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. |
topic |
PTSD Traumatic stress Biomarkers Prognosis Machine learning HPA axis |
url |
http://www.sciencedirect.com/science/article/pii/S2352289521000059 |
work_keys_str_mv |
AT katharinaschultebraucks forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy AT maritsijbrandij forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy AT isaacgalatzerlevy forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy AT joannemouthaan forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy AT mirandaolff forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy AT mirjamvanzuiden forecastingindividualriskforlongtermposttraumaticstressdisorderinemergencymedicalsettingsusingbiomedicaldataamachinelearningmulticentercohortstudy |
_version_ |
1721440070182895616 |