Skip to main content
Download PDF
  • Study Protocol
  • Open access
  • Published:

Towards accurate screening and prevention for PTSD (2-ASAP): protocol of a longitudinal prospective cohort study

BMC Psychiatry volume 24, Article number: 688 (2024) Cite this article

Abstract

Background

Effective preventive interventions for PTSD rely on early identification of individuals at risk for developing PTSD. To establish early post-trauma who are at risk, there is a need for accurate prognostic risk screening instruments for PTSD that can be widely implemented in recently trauma-exposed adults. Achieving such accuracy and generalizability requires external validation of machine learning classification models. The current 2-ASAP cohort study will perform external validation on both full and minimal feature sets of supervised machine learning classification models assessing individual risk to follow an adverse PTSD symptom trajectory over the course of 1 year. We will derive these models from the TraumaTIPS cohort, separately for men and women.

Method

The 2-ASAP longitudinal cohort will include N = 863 adults (N = 436 females, N = 427 males) who were recently exposed to acute civilian trauma. We will include civilian victims of accidents, crime and calamities at Victim Support Netherlands; and who were presented for medical evaluation of (suspected) traumatic injuries by emergency transportation to the emergency department. The baseline assessment within 2 months post-trauma will include self-report questionnaires on demographic, medical and traumatic event characteristics; potential risk and protective factors for PTSD; PTSD symptom severity and other adverse outcomes; and current best-practice PTSD screening instruments. Participants will be followed at 3, 6, 9, and 12 months post-trauma, assessing PTSD symptom severity and other adverse outcomes via self-report questionnaires.

Discussion

The ultimate goal of our study is to improve accurate screening and prevention for PTSD in recently trauma-exposed civilians. To enable future large-scale implementation, we will use self-report data to inform the prognostic models; and we will derive a minimal feature set of the classification models. This can be transformed into a short online screening instrument that is user-friendly for recently trauma-exposed adults to fill in. The eventual short online screening instrument will classify early post-trauma which adults are at risk for developing PTSD. Those at risk can be targeted and may subsequently benefit from preventive interventions, aiming to reduce PTSD and relatedly improve psychological, functional and economic outcomes.

Peer Review reports

Background

Exposure to potentially traumatic events (PTE) is common, with the overlarge majority of the global population being exposed to at least one PTE throughout their lives [1]. Similarly in the Dutch general civilian population, lifetime exposure is estimated at 80% [2]. Approximately 8% of Dutch civilians develop posttraumatic stress disorder (PTSD) at some point throughout their lives as a result of trauma exposure [2]. PTSD is a psychiatric disorder characterized by involuntary trauma re-experiencing, avoidance of trauma-related reminders; negative alterations in mood and cognitions; hyperreactivity; and hyperarousal [3]. PTSD is associated with a high risk for comorbid psychological symptoms and psychiatric disorders, including mood-, anxiety-, substance abuse- and sleep/wake disorders [4]. Furthermore, the presence of PTSD is associated with impaired well-being, daily functioning, and (health-related) quality of life, increased morbidity and mortality; and relatedly high societal costs due to for example health care use and productivity loss at work [5,6,7,8,9,10,11,12]. Although several evidence-based treatments are available, 40% of those who are affected are never treated; average time to treatment start is 4.5 years post-trauma; and approximately one third of treated patients does not respond adequately to treatment [13, 14]. Thus, once PTSD has developed it has a high likelihood for chronicity and negative consequences for affected individuals and larger society. As PTSD by definition has its onset after traumatic events, the first period post-trauma provides a unique opportunity for preventive interventions to reduce PTSD and related adverse psychological, functional and economic outcomes. Over the past decade there has been increasing evidence that such preventive interventions should not be administered in a universal manner to all trauma-exposed individuals. Instead, selective or targeted preventive intervention strategies are more promising (e.g., [15,16,17]). Herein one strategy could be to offer preventive interventions to those for whom can be established early following trauma that they are at high risk for developing PTSD. However, existing validated prognostic risk screening instruments for PTSD classify a high proportion of individuals who do not develop PTSD as at risk (e.g., low specificity ranging between 59–72% with sensitivity at 80% for available instrument for Dutch general civilian population; [18]). Thus, there is a need for more accurate prognostic risk screening instruments that can be used to determine an individuals’ risk for PTSD early post-trauma.

Previous research reported a broad range of risk and protective factors for PTSD, including demographic; socio-economic; psychiatric; psychosocial; biological; trauma history; and environmental domains [19,20,21,22]. Recent studies with a computational approach strongly indicate that these risk and protective factors for PTSD interact in dynamic non-linear ways and seem to differ between PTSD symptom trajectories (e.g., [23, 24]). Increasing empirical evidence supports the existence four common distinct courses or trajectories of PTSD symptoms following trauma exposure [25]. Of these, resilient and recovery trajectories are considered adaptive as they are associated with minimal long-term symptoms. Chronic and delayed onset trajectories reflect adverse outcome as they are associated with long-term high symptoms and the presence of diagnostic PTSD. Thus, there is considerable heterogeneity in PTSD symptom courses following trauma, even when comparing those with similar long-term outcome.

A growing number of studies support the potential of machine learning in prognostic risk classification. These studies achieved good accuracy in classifying recently traumatized individuals into their subsequent PTSD diagnostic status and/or symptom trajectory (e.g., [23, 24, 26,27,28,29,30]). However, most of these existing prognostic risk classification models cannot be applied beyond acute medical care settings, as they mainly include acute biomedical and hospital patient record information. To promote large-scale applicability within a broader population, it would be preferable to develop prognostic screening instruments based on self-report data. This approach would enable recently trauma-exposed individuals to fill out the instrument without requiring involvement of a health care professional. Moreover, this could promote user empowerment, engagement and sense of self-control and thereby increase its uptake and acceptability [31]. However, the currently available studies using self-report data only prognostically classify PTSD outcomes in the first 3 months post-trauma, for example 1 month post-trauma in adults after emergency department (ED) admission and 3 months post-trauma in family members of intensive care unit (ICU [32, 33]). Thus, there is a lack of studies using self-report data to predict PTSD over a longer time period post-trauma. Moreover, prior to large-scale implementation of derived prognostic screening instruments, external validation in an independent sample to investigate generalizability of model performance is required [34]. To date there are few studies performing external validation of machine learning models: for example in ED patients assessing PTSD course at 6 months (Area Under the Receiver Operator Characteristic [ROC] curve [AUC] of external validation set ranging between 0.46 [low] and 76 [fair]; [35]), and at 12 months (AUC of external validation set ranging between 0.78 [fair] to 0.86 [good]; [24]). To date, no studies have performed external validation of prognostic risk screening instruments within a broader civilian population.

There is a growing focus on examining sex (i.e., biology) and gender (i.e., social identity) in traumatic stress studies, with twice as many studies incorporating sex or gender in the last five years compared to earlier periods [36, 37]. This is considered especially relevant as women are commonly found to have a higher conditional risk of PTSD development than men following comparable trauma exposure [20, 38,39,40]. Moreover, we recently found differences between men and women in trajectories assignment and within-trajectory differences [41]. We observed that women were more often assigned to the recovery trajectory while men were more often assigned to the delayed trajectory, and women had higher symptom severity in the resilient trajectory than men [41]. This emphasizes the need for adequate methodological approaches such as stratification or disaggregation when establishing classification models (see also SAGER guidelines in [42]) to prevent considerable bias and incorrect trajectory assignment in the underrepresented group (in this case women). Yet, although gender and/or sex have been included as prognostic features, only few studies investigated differential prognostic value of risk and protective factors for later PTSD outcomes between men and women (e.g., [36, 43,44,45]). Furthermore, no studies have investigated whether deriving prognostic risk screening instruments separately for women and men is relevant for improving early PTSD risk detection.

In summary, there is a need for accurate prognostic risk screening instruments for PTSD, validated both internally and externally, that can be widely implemented in recently trauma-exposed civilians. We will derive full machine learning classification models for men and women separately. We will also derive classification models using a minimal feature set to eventually develop a prognostic risk screening instrument for men and women separately. These models will be derived using the existing TraumaTIPS prospective cohort of N = 852 adults (65% men) with (suspected) acute serious injury [18]. This involves training and internally validating supervised machine learning classification models for women’s and men’s risk to follow an adverse PTSD symptom trajectory. This will be measured over the course of 1 year following acute civilian trauma. These models will be based on self-report information on known PTSD risk and protective factors collected early post-trauma. Subsequently, we will externally validate the classifications models in a broader population using the newly established 2-ASAP (Towards Accurate Screening and Prevention for PTSD) cohort consisting of recently trauma-exposed adults, of which the study protocol is described below.

Goals of the current study

Primary study objectives

The primary objective is to perform external validation of both full and minimal feature set supervised machine learning-based classification models assessing individual risk to follow an adverse PTSD symptom trajectory over the course of 1 year following acute civilian trauma. The classification models will be derived from the TraumaTIPS cohort in men and women separately.

Secondary study objectives

First, to investigate the predictive value of new potential risk and protective factors based on self-report information for PTSD risk in men and women separately. Second, to derive an updated prognostic classification model assessing individual risk to follow an adverse PTSD symptom trajectory by adding information on potential risk and protective factors to the supervised machine learning classification models in men and women separately. Third, we will assess differences in common co-morbid psychological symptoms to PTSD and functional and economic outcomes between recently trauma-exposed individuals classified as low versus high risk based on the derived and validated prognostic screening instrument. Last, we will compare the accuracy of the prognostic screening instrument with current best-practice screening instruments for PTSD.

Methods

Participants and study design

The 2-ASAP cohort will include N = 863 adults (stratified by sex irrespective of self-identified gender for consistency with the existing TraumaTIPS cohort data [50% women, 50% men], taking previously established PTSD trajectory prevalence and dropout rates in the TraumaTIPS cohort into account [18], N = 436 females, N = 427 males) who were recently exposed to a (traffic) accident with injuries or violence (i.e., assault, threat, robbery or theft by force). We will include civilian victims of accidents, crime and calamities who have been reached out by Victim Support Netherlands (Slachtofferhulp Nederland), the largest institute in the Netherlands to provide emotional and practical support upon (mostly police) referral; or who were presented for medical evaluation of (suspected) traumatic injuries by emergency transportation to the ED of an Urban Level-1 Trauma Center in Amsterdam (Amsterdam University Medical Center [UMC]). Inclusion criteria of the 2-ASAP cohort are: age 18 years or older; experience of a traumatic event according to Diagnostic and Statistical Manual of Mental Disorders 5th edition (DSM-5) PTSD A criterion (i.e., exposure to actual or threatened death, serious injury or sexual violence [3]) maximally 2 months post-trauma at baseline; the traumatic events need to be directly experienced by the participant themselves, have an acute onset and external cause of a civilian nature; and have the potential to lead to serious physical injury (i.e., not a preceding medical condition). Exclusion criteria are: evidence of homocidality; suicidality, injuries due to intentional self-inflicted injury; evidence of ongoing or repeated trauma exposure, such as ongoing domestic violence; evidence of an inability to understand study procedures, risks or being otherwise unable to give informed consent; evidence of being unable to follow protocol (due to any reason, including visual or cognitive or physical impairment precluding completion of protocol); impairment in ability to use or no regular access to e-mail and mobile phone required for completion of online informed consent and internet-connected smartphone, tablet or computer for completion of online assessments; insufficient understanding of Dutch language to follow protocol. The 2-ASAP cohort was approved by the institutional review board of Amsterdam UMC (2022.0030).

Procedures

Participants will be recruited in collaboration with Victim Support Netherlands and ED of Amsterdam UMC. Potential participants will be identified via electronic client or patients records within 18–24 days post-trauma and invited for study participation by the recruitment sites. Potential participants who have contacted us for study participation will receive a participant’s information letter, and will be called for an eligibility screening (T0). Upon meeting all inclusion criteria and none of the exclusion criteria, we will obtain informed consent online or through postal services. After inclusion, a baseline (T1) and 4 follow-up assessments (T2-T5) will be performed via online self-report questionnaires at 3, 6, 9 and 12 months post-trauma (see Fig. 1). The baseline assessment needs to be completed within 2 months after their traumatic event and includes self-report questionnaires on demographic, medical and traumatic event characteristics; potential risk and protective factors for PTSD; PTSD symptom severity and other adverse outcomes; and current best-practice PTSD screening instruments (see Table 2; measures section). The follow-up assessments include PTSD symptom severity and other adverse outcomes (see Table 2; measures section).

Fig. 1
figure 1

Study design of the 2-ASAP cohort study

Full size image

Measures

PTSD symptom severity

PTSD symptom severity over the past month will be measured at all assessments using the Dutch validated version of the PTSD checklist for DSM-5 (PCL-5; [46, 47]. This self-report questionnaire consists of 20 items corresponding to the DSM-5 diagnostic symptoms of PTSD, measuring how much participants have been bothered by each symptom, ranging from not at all (0) to extremely (4) on a 5-point Likert scale. Domain scores will be calculated based on summing the corresponding DSM-5 symptom clusters: intrusions (5 items), avoidance (2 items), negative alterations in cognitions and mood (7 items), and hyperarousal (6 items). The total score will be calculated by summing all items, resulting in a total score ranging from 0 to 80, with higher scores indicated higher PTSD symptom severity.

Risk and protective factors

Both known and potential new risk and protective factors for following an adverse trajectory of PTSD symptoms will be measured at baseline. The following known risk and protective factors will be included for the primary objective: demographic and health characteristics; medical and psychiatric history; current trauma characteristics; peri-traumatic distress; psychological and physical symptoms; alcohol use; social support; post-traumatic cognitions; prior trauma [21, 23, 48, 49].

For the secondary objective we will assess potential new risk and protective factors for following an adverse trajectory of PTSD symptoms. These factors are known to be associated with PTSD, but are not yet investigated prospectively in relation to PTSD symptom course or without adequate sized samples. These include pre-trauma and immediate post-trauma chronotype; sleep characteristics; perceived psychological resilience to cope with adversity; and female hormonal status; acute pre-trauma and acute post-trauma general health and related impairments; and acute pre-trauma psychological symptoms [50,51,52,53,54]. See Table 1 for an overview of all measures.

Table 1 Measurements Conducted at Baseline Assessment (T1) and Follow-up Assessments (T2, T3, T4, T5)
Full size table

Secondary outcomes

Secondary outcomes include co-morbid psychological, functional and economic outcomes; and the current best-practice PTSD screening instruments. See Table 1 for an overview of all measures and assessment timing.

Covariates

Throughout the assessments, we will assess various factors with known potential confounding effects on PTSD symptom severity and other adverse psychological outcomes, including new traumatic events; received psychological treatment and medication use; pre-existing and new medical conditions; pain and health impairments; and Body Mass Index (BMI), nicotine use and caffeine intake. See Table 1 for an overview of all measures and assessment timing.

Sex and gender

We will measure both sex and gender at baseline. Sex will be based on their reported assignment at birth, categorizing individuals into male, female or other. We will ask participants their gender, i.e., whether they self-identify as men, women or otherwise.

Statistical analyses

PTSD Symptom Trajectories

First, PTSD symptom trajectories will be determined using an unsupervised machine learning technique, specifically latent growth mixture modeling (LGMM) on the repeatedly assessed PCL-5 total scores across the 1 year follow-up period. Hereby we empirically derive the optimal latent (unobserved) trajectories based on similarities in initial symptom severity and change over time across participant subgroups. In order to increase model accuracy in the context of the expected imbalance in the prevalence of the expected trajectories (i.e., with the adverse outcome trajectories being relatively small compared to the adaptive trajectories), we will apply informative priors for the model parameters to be estimated (i.e., the expected number and proportions of latent trajectories, and the expected symptom severity immediately after trauma [intercept] and change over time [slope] for a given trajectory; [80]). This approach has already been established and applied to the existing TraumaTIPS cohort [41], resulting in an optimal model containing 4 latent PTSD symptom trajectories. The prior parameters of the current analyses are derived from similarly observed parameters (i.e., CAPS-IV) of the 4 trajectories in the TraumaTIPS cohort [41]. Subsequently, we will determine participants’ probability of belonging to each observed PTSD symptom trajectories and assign participants to one of these trajectories based on the highest probability. Missing data will be estimated using maximum likelihood estimation (MLE).

Accuracy of classification models

In order to externally validate the derived classification models from the TraumaTIPS cohort, we will perform the same supervised machine learning classification techniques as within development and internal validation of the model. To select the optimal machine learning technique for model building, we compared the accuracy of several classification techniques in a meta-modelled simulation study using data from population characteristics of the TraumaTIPS cohort. Overall, ensemble techniques performed better than multinominal techniques regarding the sample size. For model building we start with eXtreme Gradient Boosting (XGBoost), in case of insufficient accuracy we will continue with a Naïve Bayes classifier, followed by Support Vector Machine classifiers [81,82,83]. To guard against hyperparameter overfitting we will use repeated cross-validation. In addition, we will use random oversampling of the PTSD symptom trajectory with the smallest sample size to ensure a balanced set for model training. As the number of available cases for the delayed trajectory in the existing TraumaTIPS data is too small to perform reliable machine learning analyses, we will oversample the chronic trajectory. This way the model will not only focus on correctly classifying participants of adaptive PTSD symptom trajectories, but those with an adverse trajectory as well [84]. These analyses will be performed in men and women separately, to prevent similar bias for the underrepresented sex as observed for the unsupervised machine learning analyses. Moreover, to eventually derive a novel prognostic screening instrument, we will extract the minimal feature (item) set necessary for good accuracy from the training set. The primary metric for assessing optimal model selection based on training performance will be the AUC. The AUC reflects the accuracy in correctly classifying participants into their assigned PTSD symptom trajectories. Individuals are considered at risk if the model classifies them into an adverse trajectory versus adaptive (i.e., resilient or recovering) trajectory. Identical pre-processing steps will be repeated in the test set, which is subsequently used for internal validation to test whether the prognostic accuracy for the selected best model holds. We will assess the following accuracy parameters: AUC (primary parameter); cumulative gain AUC (penalizing incorrect classification of smallest class); f1 score; precision and recall [82, 85].

We will perform external validation in the 2-ASAP cohort in men and women separately by testing the accuracy of the full derived models in a first external test set and the minimal feature sets in a second external test set. The classification models are generalizable, reliable and robust when we achieve good accuracy, that is, AUC of preferably ≥ 0.8 (good) and acceptable if ≥ 0.7 (fair; [86, 87]). Moreover, we will descriptively compare the derived models in men and women by applying explainable machine learning to investigate the relative contributions of the most important features and to interpret the decision rules incorporated in the classification models. We will use SHAP (SHapley Additive exPlanation) values for decision tree-based non-linear models [88]. State-of-the-art guidelines will be followed in reporting results of prognostic models [89].

New potential risk and protective factors

We will first examine the predictive value of the new potential PTSD risk and protective factors per above mentioned domains, separately for men and women, and adjusting for relevant covariates. The predictive value will be examined using multinominal logistic regression modelling following a recommended 3 step approach to add predictors to latent class models such as the applied LGMM [90]. To account for the number of regression analyses performed with interrelated data, we will correct the false discovery rate for multiple comparisons. Subsequently we will update the full feature classification models in men and women separately by adding new potential risk and protective factors. The same machine learning classification techniques as in the TraumaTIPS cohort will be used for training and internal validation of the updated model.

Secondary outcomes

We will compare differences in co-morbid psychological, functional and economic outcomes between the obtained risk classifications of PTSD trajectory memberships, using either ANCOVA’s or linear mixed model (LMM) depending on the number of assessments for the specific outcome. Separate analyses will be performed for men and women, and for each investigated outcome. To examine at which time point potential differences of adverse outcomes between PTSD trajectories occur, we will perform false discovery rate-corrected post-hoc pairwise comparisons of estimated means by the LMM. Furthermore, we will examine the accuracy for correctly classifying participants in their assigned PTSD symptom trajectory based on total scores of the 3 current best-practice PTSD screening instruments (i.e., GPS, PC-PTSD-5, MIRROR). The same accuracy parameters as within our previous analyses will be used. Subsequently, we will statistically test differences in the AUC of each existing screening instrument with the derived PTSD risk classification models, using a z-test approximation accounting for paired data, with alpha = 0.05 (2 sided).

Power calculation

We calculated the required sample size for sufficiently powered accuracy for developing and updating the prognostic screening instruments. The required number of participants in the test sets are calculated for sufficiently powered discriminatory accuracy (primary parameter = AUC). Power calculations were performed in EasyROC (version 1.3.1; [91]) for men and women separately, using the approach of Obuchowski for single test regarding AUC of 2 machine learning classifiers [92], using a type-1 error of 0.05, and the allocation ratio based on previously determined prevalence of PTSD symptom trajectories for men in the TraumaTIPS cohort (see Table 2, [41]). For the purpose of external validation of the derived machine learning classification models we will require 2 independent test sets: a full feature and minimal feature test set. We used an AUC of 0.7 (fair) as the minimally acceptable accuracy of our classification models instead of 0.8 (good), considering that external validation samples contain less homogenous population characteristics than training and internal validation samples and will thus likely result in lower accuracy. We require n = 13 cases; n = 150 non-cases for men; total n = 163 per test set, resulting in a total sample size of N = 338 (163*2 + n = 12 for 3.5% expected to delayed trajectory) for men. We require n = 13 cases; n = 159 non-cases for women; total n = 172) per test set, resulting in a total sample size of N = 344 (172*2) for women.

Table 2 Previously Determined Sample Size and Prevalence for the 4 PTSD Symptom Trajectories in the TraumaTIPS cohort in van Zuiden et al. [41]
Full size table

Furthermore, we calculated the required sample size for sufficiently powered accuracy for updating the prognostic screening instruments. For this purpose, we used an AUC of 0.8; and require an 80:20 split of the participants into a training and a test set. The test sets require a sample size of n = 75 (n = 6 cases; n = 69 non-cases) for men, and n = 82 (n = 6 cases; n = 74 non-cases) for women. In order to meet the proposed rule of thumb of at least 25 cases available in the training set, this requires to include a minimum of 31 cases in both samples [24]. Expecting minimally the same allocation ratio of the chronic trajectory relative to the resilient and recovery trajectory as in the TraumaTIPS cohort, this translates into a total sample size of n = (74.964*5) * 1.035 (to account for delayed trajectory) = 388 men, and n = (79.17*5) = 396 women. This sample size was increased by 10% to account for anticipated drop-out resulting in a required sample of n = 427 (388*1.10) men and n = 436 (396*1.10) women. This leads to a total sample size of N = 863 participants.

Discussion

We presented the protocol for the 2-ASAP longitudinal prospective cohort study of recently trauma-exposed adults in the Netherlands. The primary goal of this study is to externally validate machine learning classification models (i.e., full and minimal feature sets) predicting early post-trauma the individual risk to follow an adverse PTSD symptom trajectory over the course of 1 year following trauma, that we will derive from the TraumaTIPS cohort [18]. The secondary goals are to update the classification model by examining new potential risk and protective factors; and compare the classification models with current best-practice screening methods for PTSD. We will also assess differences in adverse psychological, functional and economic outcomes between adults classified as low versus high risk.

This is the first study to perform external validation of machine learning prognostic models (full and minimal feature sets) based on self-report data in a broad recently trauma-exposed adults, for men and women separately. External validation of prognostic classification models is necessary for generalizability to new and different individuals beyond those in which the model was developed [93]. Hence, the 2-ASAP cohort will include trauma exposed adults at the ED of Amsterdam UMC similar to those in the acute injury population of the TraumaTIPS cohort in which the model was developed [18], as well as adults from a broader trauma population with a higher proportion of interpersonal traumatic events at Victim Support Netherlands. Moreover, it is also recommended to have separate models for men and women to prevent bias and incorrect trajectory assignment for the underrepresented group (in this case women). To enable future large-scale implementation, self-report data to inform the prognostic models will be used; and a minimal feature set of the classification models will be derived. These sets will be transformed into a short online screening instrument assessing the risk of developing PTSD that is user-friendly for recently trauma-exposed adults to fill in.

The eventual new screening instrument will classify early post-trauma which individuals are at risk for developing PTSD. Subsequently, these recently trauma-exposed adults at risk for developing PTSD can benefit from preventive interventions, aiming to reduce PTSD prevalence and relatedly improve psychological, functional and economic outcomes [16, 17]. This will contribute to our ultimate goal to improve accurate screening and prevention for PTSD in recently trauma-exposed civilians.

Study status

Currently recruiting participants since July 2022.

Data availability

To promote open science, statistical data analyses plans (SAP) related to the study’s objectives will be pre-registered to Open Science Framework (OSF); manuscripts containing the results will be published in peer-reviewed Open Access journals; data of the study and the code to produce the results will be made publicly available upon study completion at OSF. Earlier requests for access to the data should be directed to the 2-ASAP principal investigator dr. M. van Zuiden. The study will be registered in the FAIR Traumatic Stress Data Sets library of the Global Collaboration on Traumatic Stress (GCTS).

Abbreviations

2-ASAP:

Towards Accurate Screening and Prevention for PTSD

APA:

American Psychological Association

AUC:

Area Under the ROC Curve

BMI:

Body Mass Index

DSM-5:

Diagnostic and Statistical Manual of Mental Disorders 5th edition

ED:

Emergency Department

LGMM:

Latent Growth Mixture Modelling

LMM:

Linear Mixed Model

MLE:

Maximum Likelihood Estimation

PTE:

Potentially traumatic event

PTSD:

Posttraumatic stress disorder

ROC:

Receiver Operator Characteristic

SHAP:

Shapley Additive explanation

UMC:

University Medical Center

XGBoost:

EXtreme Gradient Boosting

References

  1. Kessler RC, Aguilar-Gaxiola S, Alonso J, Benjet C, Bromet EJ, Cardoso G, Koenen KC. Trauma and PTSD in the WHO world mental health surveys. Eur J Psychotraumatol. 2017;8(sup5):1353383.

    Article  PubMed  PubMed Central  Google Scholar 

  2. de Vries GJ, Olff M. The lifetime prevalence of traumatic events and posttraumatic stress disorder in the Netherlands. J Trauma Stress. 2009;22(4):259–67.

    Article  PubMed  Google Scholar 

  3. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders Fifth Edition. Am Psychiatr Assoc. 2013;21(21):591–643.

    Google Scholar 

  4. Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB. (2013). Posttraumatic stress disorder in the National Comorbidity Survey. In Fear and Anxiety (pp. 22–34). Routledge.

  5. Bisson JI, Wright LA, Jones KA, Lewis C, Phelps AJ, Sijbrandij M, Roberts NP. Preventing the onset of post traumatic stress disorder. Clin Psychol Rev. 2021;86:102004.

    Article  PubMed  Google Scholar 

  6. Haagsma JA, Scholten AC, Andriessen TM, Vos PE, Van Beeck EF, Polinder S. Impact of depression and post-traumatic stress disorder on functional outcome and health-related quality of life of patients with mild traumatic brain injury. J Neurotrauma. 2015;32(11):853–62.

    Article  PubMed  Google Scholar 

  7. Haagsma JA, Polinder S, Olff M, Toet H, Bonsel GJ, van Beeck EF. Posttraumatic stress symptoms and health-related quality of life: a two year follow up study of injury treated at the emergency department. BMC Psychiatry. 2012;12(1):1–8.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Kessler RC. Posttraumatic stress disorder: The burden to the individual and to society. J Clin Psychiatry. 2000;61:4–14.

    PubMed  Google Scholar 

  9. Pacella ML, Hruska B, Delahanty DL. The physical health consequences of PTSD and PTSD symptoms: A meta-analytic review. J Anxiety Disord. 2013;27(1):33–46.

    Article  PubMed  Google Scholar 

  10. Pagotto LF, Mendlowicz MV, Coutinho ESF, Figueira I, Luz MP, Araujo AX, Berger W. The impact of posttraumatic symptoms and comorbid mental disorders on the health-related quality of life in treatment-seeking PTSD patients. Compr Psychiatry. 2015;58:68–73.

    Article  PubMed  Google Scholar 

  11. Spinhoven P, Penninx BW, Van Hemert AM, De Rooij M, Elzinga BM. Comorbidity of PTSD in anxiety and depressive disorders: Prevalence and shared risk factors. Child Abuse Negl. 2014;38(8):1320–30.

    Article  PubMed  Google Scholar 

  12. van der Vlegel M, Polinder S, Toet H, Panneman MJ, Geraerds AJLM, Haagsma JA. Anxiety, depression and post-traumatic stress symptoms among injury patients and the association with outcome after injury. Eur J Psychotraumatol. 2022;13(1):2023422.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Bradley R, Greene J, Russ E, Dutra L, Westen D. A multidimensional meta-analysis of psychotherapy for PTSD. Am J Psychiatry. 2005;162(2):214–27.

    Article  PubMed  Google Scholar 

  14. Goldstein RB, Smith SM, Chou SP, Saha TD, Jung J, Zhang H, Grant BF. The epidemiology of DSM-5 posttraumatic stress disorder in the United States: results from the National Epidemiologic Survey on Alcohol and Related Conditions-III. Social psychiatry and psychiatric epidemiology. 2016;51:1137–48.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Bisson JI, Ariti C, Cullen K, Kitchiner N, Lewis C, Roberts NP, Simon N, Smallman K, Addison K, Bell V, Brookes-Howell L, Cosgrove S, Ehlers A, Fitzsimmons D, Foscarini-Craggs P, Harris SRS, Mckenna M, McNamara R, Nollett C, Pickles T, Williams-Thomas R. Guided, internet based, cognitive behavioural therapy for post-traumatic stressdisorder: pragmatic, multicentre, randomised controlled non-inferiority trial (RAPID). BMJ. 2022;377:e069405.

  16. Ennis N, Sijercic I, Monson CM. Internet-delivered early interventions for individuals exposed to traumatic events: Systematic review. J Med Internet Res. 2018;20(11):e280.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Garcia MA, Delahanty DL. Oxytocin and other pharmacologic preventive interventions for posttraumatic stress disorder: Not a one-size-fits-all approach. Biol Psychiat. 2017;81(12):977–8.

    Article  PubMed  Google Scholar 

  18. Mouthaan J, Sijbrandij M, Reitsma JB, Gersons BP, Olff M. Comparing screening instruments to predict posttraumatic stress disorder. PLoS ONE. 2014;9(5):e97183.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Brewin CR, Andrews B, Valentine JD. Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. J Consult Clin Psychol. 2000;68(5):748.

    Article  CAS  PubMed  Google Scholar 

  20. Olff M, van Zuiden M. Neuroendocrine and neuroimmune markers in PTSD: pre-, peri and post-trauma glucocorticoid and inflammatory dysregulation. Curr Opin Psychol. 2017;14:132–7.

    Article  PubMed  Google Scholar 

  21. Ozer EJ, Best SR, Lipsey TL, Weiss DS. Predictors of posttraumatic stress disorder and symptoms in adults: a meta-analysis. Psychol Bull. 2003;129(1):52.

    Article  PubMed  Google Scholar 

  22. Sayed S, Iacoviello BM, Charney DS. Risk factors for the development of psychopathology following trauma. Curr Psychiatry Rep. 2015;17:1–7.

    Google Scholar 

  23. Schultebraucks K, Sijbrandij M, Galatzer-Levy I, Mouthaan J, Olff M, van Zuiden M. 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. 2021;14: 100297.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Schultebraucks K, Shalev AY, Michopoulos V, Grudzen CR, Shin SM, Stevens JS, Galatzer-Levy IR. A validated predictive algorithm of post-traumatic stress course following emergency department admission after a traumatic stressor. Nat Med. 2020;26(7):1084–8.

    Article  CAS  PubMed  Google Scholar 

  25. Galatzer-Levy IR, Huang SH, Bonanno GA. Trajectories of resilience and dysfunction following potential trauma: A review and statistical evaluation. Clin Psychol Rev. 2018;63:41–55.

    Article  PubMed  Google Scholar 

  26. Galatzer-Levy IR, Karstoft KI, Statnikov A, Shalev AY. Quantitative forecasting of PTSD from early trauma responses: A machine learning application. J Psychiatr Res. 2014;59:68–76.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Karstoft KI, Galatzer-Levy IR, Statnikov A, Li Z, Shalev AY, members of the Jerusalem Trauma Outreach and Prevention Study (J-Tops) group. Bridging a translational gap: using machine learning to improve the prediction of PTSD. BMC Psychiatry. 2015;15:1–7.

    Article  Google Scholar 

  28. Papini S, Iturralde E, Lu Y, Greene JD, Barreda F, Sterling SA, Liu VX. Development and validation of a machine learning model using electronic health records to predict trauma-and stressor-related psychiatric disorders after hospitalization with sepsis. Transl Psychiatry. 2023;13(1):400.

    Article  PubMed  PubMed Central  Google Scholar 

  29. Papini GB, Fonseca P, Eerikäinen LM, Overeem S, Bergmans JW, Vullings R. Sinus or not: A new beat detection algorithm based on a pulse morphology quality index to extract normal sinus rhythm beats from wrist-worn photoplethysmography recordings. Physiol Meas. 2018;39(11):115007.

    Article  PubMed  Google Scholar 

  30. Ziobrowski HN, Kennedy CJ, Ustun B, House SL, Beaudoin FL, An X, AURORA Consortium. Development and validation of a model to predict posttraumatic stress disorder and major depression after a motor vehicle collision. JAMA Psychiat. 2021;78(11):1228–37.

    Article  Google Scholar 

  31. Huygens MW, Vermeulen J, Swinkels IC, Friele RD, Van Schayck OC, De Witte LP. Expectations and needs of patients with a chronic disease toward self-management and eHealth for self-management purposes. BMC Health Serv Res. 2016;16:1–11.

    Article  Google Scholar 

  32. Dupont T, Kentish-Barnes N, Pochard F, Duchesnay E, Azoulay E. Prediction of post-traumatic stress disorder in family members of ICU patients: a machine learning approach. Intensive Care Med. 2024;50(1):114–24.

    Article  PubMed  Google Scholar 

  33. Wshah S, Skalka C, Price M. Predicting posttraumatic stress disorder risk: a machine learning approach. JMIR mental health. 2019;6(7):e13946.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Altman MB, Stinauer MA, Javier D, Smith BD, Herman LC, Pytynia ML, Aydogan B, Pelizarri CA, Chmura SJ, Roeske JC. Validation of temporal optimization effects for a single fraction of radiation in vitro. Int J Radiat Oncol Biol Phys. 2009;75(4):1240–6.

    Article  PubMed  Google Scholar 

  35. Tomas CW, Fitzgerald JM, Bergner C, Hillard CJ, Larson CL, deRoon-Cassini TA. Machine learning prediction of posttraumatic stress disorder trajectories following traumatic injury: Identification and validation in two independent samples. J Trauma Stress. 2022;35(6):1656–71.

    Article  PubMed  Google Scholar 

  36. Haering S, Seligowski AV, Linnstaedt SD, Michopoulos V, House SL, Beaudoin FL, Stevens JS. Disentangling sex differences in PTSD risk factors. Nat Mental Health. 2024;2:605–15.

    Article  Google Scholar 

  37. Langeland W, Olff M. Sex and gender in psychotrauma research. Eur J Psychotraumatol. 2024;15(1):2358702.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Olff M, Langeland W, Draijer N, Gersons BP. Gender differences in posttraumatic stress disorder. Psychol Bull. 2007;133(2):183.

    Article  PubMed  Google Scholar 

  39. Tolin DF, Foa EB. (2008). Sex differences in trauma and posttraumatic stress disorder: a quantitative review o f 25 years of research. Psychological Trauma: Theory, Research, Practice, and Policy, S(1):37–85.

  40. Shalev AY. Posttraumatic stress disorder and stress-related disorders. Psychiatric Clinics. 2009;32(3):687–704.

    PubMed  Google Scholar 

  41. van Zuiden M, Engel S, Karchoud JF, Wise TJ, Sijbrandij M, Mouthaan J, van de Schoot R. Sex-differential PTSD symptom trajectories across one year following suspected serious injury. Eurn J Psychotraumatol. 2022;13(1):2031593.

    Article  Google Scholar 

  42. van Epps H, Astudillo O, Martin YDP, Marsh J. The Sex and Gender Equity in Research (SAGER) guidelines: Implementation and checklist development. Eur Sci Edit. 2022;48:e86910.

    Article  Google Scholar 

  43. Haering S, Seligowski AV, Linnstaedt SD, Michopoulos V, House SL, Beaudoin FL, Swor R. A. (2024). Sex-dependent differences in vulnerability to early risk factors for posttraumatic stress disorder: results from the AURORA study. Psychological Medicine, 1–11

  44. Haering S, Schulze L, Geiling A, Meyer C, Klusmann H, Schumacher S, Engel S. Higher risk—less data: A systematic review and meta-analysis on the role of sex and gender in trauma research. J Psychopathol Clin Sci. 2024;133(3):257.

    Article  PubMed  Google Scholar 

  45. Haering S, Kooistra MJ, Bourey C, Chimed-Ochir U, Doubková N, Hoeboer CM, de Haan A. Exploring transdiagnostic stress and trauma-related symptoms across the world: a latent class analysis. Eur J Psychotraumatol. 2024;15(1):2318190.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Blevins CA, Weathers FW, Davis MT, Witte TK, Domino JL. The posttraumatic stress disorder checklist for DSM-5 (PCL-5): Development and initial psychometric evaluation. J Trauma Stress. 2015;28(6):489–98.

    Article  PubMed  Google Scholar 

  47. Boeschoten MA, Bakker A Jongedijk, RA, Olff M. (2014). PTSS checklist voor de DSM-5 (PCL-5) [PTSD checklist for the DSM-5 (PCL-5)]. Diemen: Arq Academy.

  48. DiGangi JA, Gomez D, Mendoza L, Jason LA, Keys CB, Koenen KC. Pretrauma risk factors for posttraumatic stress disorder: A systematic review of the literature. Clin Psychol Rev. 2013;33(6):728–44.

    Article  PubMed  Google Scholar 

  49. Maes M, Delmeire L, Mylle J, Altamura C. Risk and preventive factors of post traumatic stress disorder (PTSD): alcohol consumption and intoxication prior to a traumatic event diminishes the relative risk to develop PTSD in response to that trauma. J Affect Disord. 2001;63(1–3):113–21.

    Article  CAS  PubMed  Google Scholar 

  50. Bonanno GA, Field NP, Kovacevic A, Kaltman S. Self-enhancement as a buffer against extreme adversity: Civil war in Bosnia and traumatic loss in the United States. Pers Soc Psychol Bull. 2002;28(2):184–96.

    Article  Google Scholar 

  51. Engel S, van Zuiden M, Frijling JL, Koch SB, Nawijn L, Schumacher S. Patterns of recovery from early posttraumatic stress symptoms after a preventive intervention with oxytocin: hormonal contraception use is a prognostic factor. Biol Psychiat. 2019;85(12):e71–3.

    Article  PubMed  Google Scholar 

  52. Germain A, McKeon AB, Campbell RL. Sleep in PTSD: conceptual model and novel directions in brain-based research and interventions. Curr Opin Psychol. 2017;14:84–9.

    Article  PubMed  Google Scholar 

  53. Landgraf D, McCarthy MJ, Welsh DK. Circadian clock and stress interactions in the molecular biology of psychiatric disorders. Curr Psychiatry Rep. 2014;16:483.

    Article  PubMed  Google Scholar 

  54. Pineles SL, Hall KAA, Rasmusson AM. Gender and PTSD: different pathways to a similar phenotype. Curr Opin Psychol. 2017;14:44–8.

    Article  PubMed  Google Scholar 

  55. Weiss DS. The impact of event scale: revised. In: Cross-cultural assessment of psychological trauma and PTSD. Boston, MA: Springer, US; 2007. p. 219–38.

    Chapter  Google Scholar 

  56. Prins A, Bovin MJ, Smolenski DJ, Marx BP, Kimerling R, Jenkins-Guarnieri MA, Tiet QQ. The primary care PTSD screen for DSM-5 (PC-PTSD-5): development and evaluation within a veteran primary care sample. J Gen Intern Med. 2016;31(10):1206–11.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Frewen P, McPhail I, Schnyder U, Oe M, Olff M. Global Psychotrauma Screen (GPS): psychometric properties in two internet-based studies. Eur J Psychotraumatol. 2021;12(1):1881725.

    Article  PubMed  PubMed Central  Google Scholar 

  58. van Herpen MM, Boeschoten MA, Te Brake H, van der Aa N, Olff M. Mobile Insight in Risk, Resilience, and Online Referral (MIRROR): psychometric evaluation of an online self-help test. J Med Internet Res. 2020;22(9):e19716.

    Article  PubMed  PubMed Central  Google Scholar 

  59. Boeschoten MA, Van der Aa N, Bakker A, Ter Heide FJJ, Hoofwijk MC, Jongedijk RA, Olff M. Development and evaluation of the Dutch clinician-administered PTSD scale for DSM-5 (CAPS-5). Eur J Psychotraumatol. 2018;9(1):1546085.

    Article  PubMed  PubMed Central  Google Scholar 

  60. CBS (2005). Permanent onderzoek leefsituatie. Voorburg/Heerlen: Statistics Netherlands (CBS).

  61. Marmar CR, Metzler TJ, Otte C. The peritraumatic dissociative experiences questionnaire. The Guilford Press; 2004.

    Google Scholar 

  62. Brunet A, Weiss DS, Metzler TJ, Best SR, Neylan TC, Rogers C, Marmar CR. The Peritraumatic Distress Inventory: a proposed measure of PTSD criterion A2. Am J Psychiatry. 2001;158(9):1480–5.

    Article  CAS  PubMed  Google Scholar 

  63. van Sonderen, E. (1997). Sociale Steun Lijst–Interacties (SSL-I) en Sociale Steun Lijst-Discrepanties (SSL-D). Centrum voor Gezondheidsvraagstukken Rijksuniversiteit Groningen.

  64. Alberts M, Smets EMA, Vercoulen JHMM, Garssen B, Bleijenberg G. Verkorte vermoeidheidsvragenlijst’: een praktisch hulpmiddel bij het scoren van vermoeidheid. Nederlands Tijdschrift Geneeskunde. 1997;141(31):411–525.

    Google Scholar 

  65. Alberts M, Smets EMA, Vercoulen JHMM, Garssen B, Bleijenberg G. Verkorte vermoeidheidsvragenlijst’: een praktisch hulpmiddel bij het scoren van vermoeidheid. Nederlands Tijdschrift Geneeskunde. 1997;141(31):411–525.

    Google Scholar 

  66. Foa EB, Ehlers A, Clark DM, Tolin DF, Orsillo SM. The posttraumatic cognitions inventory (PTCI): Development and validation. Psychol Assess. 1999;11(3):303.

    Article  Google Scholar 

  67. de Beurs E, Van Dyck R, Marquenie LA, Lange A, Blonk RW. De DASS: een vragenlijst voor het meten van depressie, angst en stress. Gedragstherapie. 2001;34(1):35–54.

    Google Scholar 

  68. Spinhoven PH, Ormel J, Sloekers PPA, Kempen GIJM, Speckens AE, van Hemert AM. A validation study of the Hospital Anxiety and Depression Scale (HADS) in different groups of Dutch subjects. Psychol Med. 1997;27(2):363–70.

    Article  CAS  PubMed  Google Scholar 

  69. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA, Ambulatory Care Quality Improvement Project (ACQUIP). The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Arch Intern Med. 1998;158(16):1789–95.

    Article  CAS  PubMed  Google Scholar 

  70. Skevington SM, Lotfy M, O’Connell KA. The World Health Organization’s WHOQOL-BREF quality of life assessment: psychometric properties and results of the international field trial. A report from the WHOQOL group. Qual Life Res. 2004;13:299–310.

    Article  CAS  PubMed  Google Scholar 

  71. Brooks R, Group E. EuroQol: the current state of play. Health Policy. 1996;37(1):53–72.

    Article  CAS  PubMed  Google Scholar 

  72. Topp CW, Østergaard SD, Søndergaard S, Bech P. The WHO-5 Well-Being Index: a systematic review of the literature. Psychother Psychosom. 2015;84(3):167–76.

    Article  PubMed  Google Scholar 

  73. Bouwmans C. (2013) Handleiding iMTA Medical Cost Questionnaire (iMCQ). Manual 2013.

  74. Bouwmans C, Krol M, Severens H, Koopmanschap M, Brouwer W, Hakkaart-van Roijen L. The iMTA productivity cost questionnaire: a standardized instrument for measuring and valuing health-related productivity losses. Value in health. 2015;18(6):753–8.

    Article  PubMed  Google Scholar 

  75. van der Meer CA, Te Brake H, Van der Aa N, Dashtgard P, Bakker A, Olff M. Assessing psychological resilience: Development and psychometric properties of the English and Dutch version of the Resilience Evaluation Scale (RES). Front Psych. 2018;9:169.

    Article  Google Scholar 

  76. Wassing R, Benjamins JS, Dekker K, Moens S, Spiegelhalder K, Feige B, Van Someren EJ. Slow dissolving of emotional distress contributes to hyperarousal. Proc Natl Acad Sci. 2016;113(9):2538–43.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  77. Buysse DJ, Reynolds CF III, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28(2):193–213.

    Article  CAS  PubMed  Google Scholar 

  78. Morin CM, Belleville G, Bélanger L, Ivers H. The Insomnia Severity Index: psychometric indicators to detect insomnia cases and evaluate treatmentresponse. Sleep. 2011;34(5):601–8.

    Article  PubMed  PubMed Central  Google Scholar 

  79. Juda M, Vetter C, Roenneberg T. The Munich chronotype questionnaire for shift workers (MCTQShift). J Biol Rhythms. 2013;28(2):130–40.

    Article  PubMed  Google Scholar 

  80. van de Schoot R, Sijbrandij M, Depaoli S, Winter SD, Olff M, Van Loey NE. Bayesian PTSD-trajectory analysis with informed priors based on a systematic literaturesearch and expert elicitation. Multivar Behav Res. 2018;53(2):267–91.

    Article  Google Scholar 

  81. Chen, T., & Guestrin, C. (2016, August). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining (pp. 785–794).

  82. Kuhn M, Johnson K. Applied predictive modeling. New York: Springer nature; 2013.

    Book  Google Scholar 

  83. Meyer, M. (2015). Machine learning to detect online grooming. Digitala Vetenskapliga Arkivet.

  84. López V, Fernández A, García S, Palade V, Herrera F. An insight into classification with imbalanced data: Empirical results and current trends on using data intrinsic characteristics. Inf Sci. 2013;250:113–41.

    Article  Google Scholar 

  85. Luo G. A review of automatic selection methods for machine learning algorithms and hyper parameter values. Network Modeling Analysis in Health Informatics and Bioinformatics. 2016;5(1):16.

    Article  Google Scholar 

  86. Collins GS, Ogundimu EO, Altman DG. Sample size considerations for the external validation of a multivariable prognostic model: a resampling study. Stat Med. 2016;35(2):21426.

    Article  Google Scholar 

  87. Mandrekar JN. Receiver operating characteristic curve in diagnostic test assessment. J Thorac Oncol. 2010;5(9):1315–6.

    Article  PubMed  Google Scholar 

  88. Lundberg, S. (2017). A unified approach to interpreting model predictions. arXiv preprint arXiv:1705.07874.

  89. Moons KG, de Groot JA, Bouwmeester W, Vergouwe Y, Mallett S, Altman DG, Collins GS. Critical appraisal and data extraction for systematic reviews of prediction modelling studies: the CHARMS checklist. PLoS medicine. 2014;11(10):e1001744.

    Article  PubMed  PubMed Central  Google Scholar 

  90. Vermunt JK. Latent class modeling with covariates: Two improved three-step approaches. Polit Anal. 2010;18(4):450–69.

    Article  Google Scholar 

  91. Goksuluk D, Korkmaz S, Zararsiz G, Karaagaoglu AE. easyROC: an interactive web tool for ROC curve analysis using R language environment. R J. 2016;8(2):213.

    Article  Google Scholar 

  92. Obuchowski NA, McCLISH DK. Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices. Stat Med. 1997;16(13):1529–42.

    Article  CAS  PubMed  Google Scholar 

  93. Ramspek CL, Jager KJ, Dekker FW, Zoccali C, van Diepen M. External validation of prognostic models: what, why, how, when and where? Clin Kidney J. 2021;14(1):49–58.

    Article  PubMed  Google Scholar 

Download references

Acknowledgements

We sincerely thank all participants in this study, as well as the preceding panel of experts-by-experience that provided their input to the study design for their valuable time and effort. Furthermore, we thank the 2-ASAP consortium partners and advisors for their collaboration and input. In addition, we would like to thank our recruitment sites Victim Support Netherlands (Slachtofferhulp Nederland) and the Emergency Department of Amsterdam University Medical Center (UMC), as well as the 2-ASAP research assistants for their efforts in participant recruitment.

Protocol version

Version 1: original, issue date 11–03-2022; Version 2: amendment adding the possibility for e-consent in coherence with new regulations by the Dutch Central Committee on Research Involving Human Subjects (CCMO), issue date 04–04-2023; Version 3: amendment adding Amsterdam UMC ED as recruitment site, issue date 02–01-2024.

Funding

The 2-ASAP cohort study is part of the 2-ASAP consortium project funded by the Netherlands Organization for Health Research and Development (ZonMw; #636340004).

Author information

Authors and Affiliations

  1. Amsterdam UMC, University of Amsterdam, Psychiatry, Amsterdam Public Health, Amsterdam, The Netherlands

    Jeanet F. Karchoud, Chris M. Hoeboer, Greta Piwanski, Miranda Olff & Mirjam van Zuiden

  2. Department of Public Health, Erasmus MC, Rotterdam, The Netherlands

    Juanita A. Haagsma

  3. ARQ National Psychotrauma Centre, Diemen, The Netherlands

    Miranda Olff

  4. Department of Methods and Statistics, Utrecht University, Utrecht, The Netherlands

    Rens van de Schoot

  5. Department of Clinical Psychology, Utrecht University, Utrecht, The Netherlands

    Mirjam van Zuiden

Authors
  1. Jeanet F. Karchoud
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Chris M. Hoeboer
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Greta Piwanski
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Juanita A. Haagsma
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Miranda Olff
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Rens van de Schoot
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Mirjam van Zuiden
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

M.v.Z. is the principal investigator and project leader of the 2-ASAP consortium and initiated the 2-ASAP cohort study as part of this consortium. M.v.Z. developed the study design together with R.v.d.S., M.O., J.A.H., and the related initial study protocol and amendments together with J.F.K, G.P. and C.H.. J.F.K. and G.P. are responsible for coordinating the study, supervised by M.v.Z. and C.H.. J.F.K. is responsible for writing the main manuscript text and figures with input from all co-authors. All authors reviewed the draft manuscript and approved submission of the final manuscript.

Corresponding author

Correspondence to Mirjam van Zuiden.

Ethics declarations

Ethics approval and consent to participate

The 2-ASAP cohort study was approved by the Institutional Review Board of the Amsterdam University Medical Centers on April 25, 2022 (NL80296.018.22; 2022.0030). The study is conducted according to the principles of the Declaration of Helsinki (1964, last updated 2013) and the Medical Research Involving Human Subjects Act (WMO). All researchers are trained in good clinical practice. Participants are extensively informed about the study both through information letters and orally and provide written informed consent prior to inclusion into the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Karchoud, J.F., Hoeboer, C.M., Piwanski, G. et al. Towards accurate screening and prevention for PTSD (2-ASAP): protocol of a longitudinal prospective cohort study. BMC Psychiatry 24, 688 (2024). https://doi.org/10.1186/s12888-024-06110-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12888-024-06110-6

Keywords

BMC Psychiatry

ISSN: 1471-244X

Contact us