Post-traumatic stress disorder (PTSD) constitutes a debilitating psychiatric disease due to a trauma or a stress factor. Its diagnosis is based on the presence of various symptoms (including intrusion, avoidance, and hyperarousal) after exposure to severe injuries or death (1). PTSD is defined as the presence of symptoms greater than 1 month after the trauma; before this, it is an acute stress disorder according to Diagnostic and Statistical Manual of Mental Disorders (DSM) 5 (1). PTSD can become chronic and be frequently associated with frequent severe psychiatric comorbidities, including not only depression, anxiety disorders, addictions, or suicidality (2,3), but also somatic complications such as cardiovascular diseases (4,5). PTSD is under-diagnosed and an early identification constitutes a major issue in terms of prognosis.
Over the last few years, certain studies have investigated the psychological effects induced by an ICU stay. The development of PTSD in these patients has been reported to range from 4% to 62% over a 1-year follow-up, according to the tool used, its cutoff, and the time of screening (6); with the Impact Event Scale-Revisited (IES-R) it was estimated to be 13% at 1 month (7), 8% at 2 months (8), 13% (PTSD), and 14% (partial PTSD) at 3 months (9), using a cutoff of 33, 30, and 34. However, there is still no consensus for PTSD prevention in this population. Identifying risk factors could be used for screening, and many authors have investigated these, but published data related to risk factors in ICU patients are often contradictory (6,10–12). As in the general population, the presence of acute stress disorder is a known risk factor for PTSD (1), it could therefore be interesting to use this factor to identify patients at risk of PTSD after ICU discharge, but a systematic psychiatric evaluation is not feasible in real-life practice.
To date, there is no validated scale to detect acute stress disorder in ICU patients, but some scales used to assess PTSD after ICU stay could be used for this screening (6), as previously investigated by some authors; for instance in a preliminary study with a limited sample size (13), and another but which used a questionnaire based on an outdated version of the DSM (DSM-III) (14). DSM-III included fewer items than the most recent versions (persistent negative emotional states, irritable behavior or angry outbursts, and reckless or self-destructive behavior have since been added), possibly leading to the diagnosis of fewer patients (15). The IES-R is a more recent validated tool for screening post-traumatic stress symptoms (16) or to follow patients with PTSD (17); it only requires 5 to 10 minutes to complete (18) and is one of the most frequently used scales (6,18). This self-administered questionnaire includes 22 items assessing avoidance, hyperarousal, and intrusion symptoms. Its psychometric characteristics have a sensitivity of 100%, a specificity of 85%, a positive predictive value (PPV) of 50%, and a negative predictive value (NPV) of 100% as compared to the gold standard scale (Clinician-Administered PTSD Scale [CAPS]) for PTSD diagnosis in the ICU population (9). However, no study has evaluated the effectiveness of this scale used at ICU discharge to predict the later development of PTSD. It could constitute an easy, quick, and useful tool for nonspecialists to detect patients at risk of developing PTSD symptoms.
The aim of the study was, therefore, to evaluate the diagnostic accuracy of the IES-R assessed within 8 days following the ICU discharge, to predict the presence of PTSD symptoms at 3 months. The secondary objective was to explore risk factors for PTSD symptoms at 3 months.
MATERIALS AND METHODS
A prospective study was conducted in three medical/surgical ICU of the Lyon university hospital (France) between September 2017 and April 2018. The study was approved by the Ethics Committee (Sud-Méditerranée IV, number 2017-A02083-50) and was registered on ClinicalTrials.gov (NCT03278171). Written informed consent was waived; oral consent was considered as sufficient by the ethics committee according to French law.
The inclusion criteria were: age greater than or equal to 18 years, hospitalization greater than or equal to 2 nights in the participating ICU, transfer to a ward of the Lyon university hospitals, and agreement to participate. Exclusion criteria were: language barrier, confusion at the first assessment, and being under legal protection. Patients were identified by the intensivists before they left the ICU.
PTSD is defined by the presence from 1 month after exposure to a stressor (death, threatened death, actual or threatened severe injuries/sexual violence) of symptoms of intrusion, avoidance, negative alterations in cognitions and mood, and alterations in arousal and reactivity. The presence of symptoms in each cluster is necessary to make the diagnosis; if symptoms of one or more clusters are missing, the diagnosis is partial PTSD (1).
Acute stress disorder is defined by the presence of several symptoms among those of PTSD (and also included possible peritraumatic dissociation), in the first month after the trauma, irrespective of cluster (1).
At any time from 1 to 8 days after ICU discharge, a psychiatry resident or a clinical research associate met the patient in the hospital ward to collect data regarding their psychiatric history (patients answered if they had history of depression, anxiety disorder, PTSD or alcohol/tobacco/drugs addiction, and ongoing psychotropic treatments), delusional memories and socio-professional situation (used to calculate Economic Position Index, Indice de Position Socio-Economique or IPSE ), and to give them validated French versions of two self-administered questionnaires (IES-R and Peritraumatic Dissociative Experiences Questionnaire [PDEQ]) to complete. If the patient could not complete the questionnaires themselves, the clinician helped them; seated beside them, they read it word-for-word as if they had filled it out themselves. IES-R scores were analyzed and reported as sum of all items. As it was measured in the first month after the trauma, an IES-R greater than or equal to 12 within 8 days after ICU discharge defined the presence of acute stress symptoms (20,21). A PDEQ greater than or equal to 15 defined the presence of peritraumatic dissociation experiences (22,23), a symptom that is a risk factor for PTSD and can be present in acute stress disorder but which is not evaluated by the IES-R (1).
Data about ICU stay were collected retrospectively from medical files by a clinical research associate: Simplified Acute Physiology Score (SAPS) II at admission (24), cause of admission (medical, surgical, traumatic), unexpected hospitalization, administration of benzodiazepines (cumulated dose for midazolam and cumulated dose in diazepam equivalent for others), vasopressor requirements, opioid requirements (converted to equivalent dose of oral morphine), invasive treatments (extracorporeal membrane oxygenation, mechanical ventilation, central venous catheter), duration of mechanical ventilation, and ICU length of stay.
Three months after ICU discharge, a trained psychiatric clinician called patients by phone to administer the IES-R questionnaire. A previous study showed that phone and face-to-face methods were comparable for assessment of PTSD symptoms (25). Other investigators have also used phone assessment to administer the IES-R after ICU discharge (7,8).
At 3 months, an IES-R greater than or equal to 35 was used to define patients with PTSD symptoms; an IES-R between 12 and 34 was used to define patients with partial PTSD symptoms (20,21).
Characteristics of the population were described using count and percentages for categorical variables and median (interquartile range [IQR]) for continuous variables. Comparisons between groups were performed using the Wilcoxon rank-sum test for continuous variables and the chi-square test or Fisher exact test (as appropriate) for categorical variables. The diagnostic accuracy of the IES-R within 8 days after ICU discharge to predict PTSD symptoms at 3 months (IES-R ≥ 35) was evaluated by the area under the receiver operating characteristic curve (AUC), with its 95% CI; the sensitivity, specificity, as well as PPV and NPV for the conventional threshold of greater than or equal to 12 points, but also for various other thresholds, were calculated.
For the secondary objective, risk factors for PTSD symptoms (IES-R ≥ 35 at 3 mo) were investigated using univariate and multivariate logistic regression. The choice of risk factors for the analysis was made by combining those identified in the general population (gender, age, history of depression, anxiety disorder, PTSD, or addiction, as well as peritraumatic dissociation and acute stress disorder) (1) and frequently identified in the literature and in a recent meta-analysis (delusional memories and administration of benzodiazepines) (6); (excluding the administration of corticosteroids and the presence of ICU diaries because these were not used in the three ICU that participated in the study). The effect of the factors was quantified by odds ratios (ORs) with the associated 95% CI. A backward selection approach was used to select variables included in the final model among those analyzed in univariate analyses, using a p value of less than 0.05 to retain variables in the model.
Assuming an AUC of 0.90 and a prevalence of patients with PTSD symptoms at 3 months of 35%, 144 patients were considered necessary to obtain a precision of ± 0.06 of the 95% CI of the AUC. A p value of 0.05 was considered significant. Statistical analyses were performed using R software Version 3.4.4 (26).
Between September 2017 and April 2018, a total of 208 patients were screened, 174 included of whom 145 were reassessed at 3 months (Fig. 1). The IES-R was assessed as soon as the patients were able to respond within 8 days after ICU discharge. There was no clinically significant difference between the patients assessed at ICU discharge and those who were reassessed 3 months later: the majority were men, the median age was between 61 and 62 years, a quarter had an education level greater than high school, and nearly two-thirds of patients had a psychiatric history (Table 1). The only significant difference between patients assessed at 3 months and the 29 who were not reassessed was a lower severity score (SAPS II) at admission (p = 0.03; Table 1).
TABLE 1. -
Characteristics of Patients Assessed Within 8 Days and at 3 Months After ICU Discharge and Patients Assessed Within 8 Days But Not at 3 Months
||Patients Assessed Within 8 d After ICU Discharge, n = 174
||Patients Assessed Within 8 d and at 3 mo After ICU Discharge, n = 145
||Patients Assessed Within 8 d After ICU Discharge But Not at 3 mo, n = 29
|Education level > high school
| Anxiety disorder
| Post-traumatic stress disorder
|Ongoing psychotropic treatment
|Reason for admission
|Length of stay (d)
|Duration of mechanical ventilation
|Simplified Acute Physiology Score II
|IES-R score ≥ 12 within 8 d after ICU discharge
|IES-R score within 8 d after ICU discharge
|Peritraumatic Dissociative Experiences Questionnaire score ≥ 15 within 8 d after ICU discharge
|Socio-economic position index (Indice de Position Socio-Economique)
IES-R = Impact Event Scale-Revisited.
p: comparison of those evaluated at 3 mo to those not evaluated using χ2, Wilcoxon rank-sum, or Fisher exact tests.
Data are presented as n (%) or median (interquartile range).
At 3 months, 19 of 145 patients (13%) had an IES-R greater than or equal to 35 (PTSD symptoms) and 25 (17%) had a score between 12 and 34 (partial PTSD symptoms); the median (IQR) IES-R was 4.0 (1.0–19.0). Concerning the changes in IES-R from 8 days to 3 months after the ICU discharge: IES-R score increased for 47 patients (32.4%), decreased for 80 (55.2%), and did not change for 18 (12.4%).
The AUC of the IES-R performed within 8 days after the ICU discharge to predict PTSD symptoms at 3 months was 0.90 (95% CI, 0.80–0.99; Fig. 2). An IES-R greater than or equal to 12 (corresponding to 40% of the patients) had a sensitivity of 0.90 (95% CI, 0.67–0.99), a specificity of 0.71 (95% CI, 0.63–0.80), a PPV of 0.32 (95% CI, 0.25–0.39), and a NPV of 0.98 (95% CI, 0.94–0.99).
An IES-R greater than or equal to 35 within 8 days after ICU discharge had a sensitivity of 0.63 (95% CI, 0.38–0.84), a specificity of 0.95 (95% CI, 0.90–0.98), a PPV of 0.67 (95% CI, 0.45–0.82), and a NPV of 0.95 (95% CI, 0.91–0.97).
Using the simple rule of a PDEQ greater than or equal to 15 and an IES-R greater than or equal to 12 to predict PTSD symptoms at 3 months, sensitivity was 0.79 (95% CI, 0.54–0.94), specificity 0.79 (95% CI, 0.70–0.85), PPV 0.36 (95% CI, 0.22–0.52), and NPV 0.96 (95% CI, 0.90–0.99).
In univariate analysis, factors associated with an increase in the risk of PTSD symptoms at 3 months were previous psychiatric history (depression, anxiety disorder, PTSD, and/or addiction), an ongoing psychotropic treatment, a PDEQ greater than or equal to 15, and an IES-R greater than or equal to 12 (Table 2).
TABLE 2. -
Univariate and Multivariate Analysis of Risk Factors for Post-Traumatic
Stress Disorder at 3 Months (Impact Event Scale-Revisited
|OR (95% CI)
||OR (95% CI)
| 40–60/≤ 40
| > 60/≤ 40
| Anxiety disorder
| Post-traumatic stress disorder
|Ongoing psychotropic treatment
|Administration of benzodiazepines
|Peritraumatic Dissociative Experiences Questionnaire score within 8 d after ICU discharge ≥ 15
|Impact Event Scale-Revisited score within 8 d after ICU discharge ≥ 12
OR = odds ratio.
aMultivariate analysis after backward selection.
In multivariate analysis, previous history of anxiety disorders (OR, 3.70; 95% CI, 1.24–11.05; p = 0.02) and IES-R greater than or equal to 12 within 8 days after ICU discharge (OR, 16.57; 95% CI, 3.59–76.46; p < 0.001) were significantly associated with PTSD symptoms at 3 months.
The present study found that the IES-R used early after ICU discharge has a good ability to predict PTSD symptoms 3 months later (AUC, 0.90; 95% CI, 0.80–0.99) and that an IES-R less than 12 within 8 days after ICU discharge excludes the risk to develop PTSD symptoms (NPV, 0.98; 95% CI, 0.94–0.99). Patients having acute stress symptoms (IES-R ≥ 12) at ICU discharge were significantly more at risk of PTSD symptoms. In the analysis sample, only 10% of patients with PTSD symptoms would not have been detected, indicating that this tool has good detection properties. Furthermore, only a third of the population had an IES-R greater than or equal to 12 within 8 days after ICU discharge, potentially avoiding the reassessment of the remaining two-thirds of patients.
The results of the present study are concordant with those reported by Milton et al (14), who found an AUC of 0.90 for an early assessment with the Post-Traumatic Symptom Scale-10, based on an outdated version of the DSM referencing fewer symptoms, which can induce underdiagnoses of PTSD. Other studies have investigated the early detection of patients at risk for PTSD using other scales such as the PTSD Checklist-Civilian version (27), the Post-Traumatic Stress Syndrome 14-Questions Inventory (PTSS-14) (13,28), while others developed their own screening tool (29,30); however, these are not generalizable (small sample size, specific population, scales not validated on an ICU population). Furthermore, regarding the IES-R, this tool being quick, easy to administer by nonspecialists, and validated in many languages, it makes it practical to use in many countries; furthermore, the present study included patients after a minimum stay of two nights in medical and surgical ICU which allows the generalization of the results to the majority of the ICU population.
Among the patients followed-up at 3 months, nearly a third had symptoms of PTSD 3 months after the ICU stay (nearly half of these developed PTSD symptoms and the remaining had significant symptomatology). This is concordant with previously reported data (6) and highlights the potential psychologic impact of ICU stay. It is of note that at the 3-month telephone interview, certain patients having PTSD symptoms reported orally that they had approached a psychiatrist since the ICU discharge and some had a prescription for an antidepressant. In addition, many patients suffering from PTSD symptoms often did not know this could be treated, believing that it was “normal” after such an experience (these data were not formally collected).
Regarding the identification of risk factors in multivariate analysis, only the presence of acute stress symptoms at ICU discharge and history of anxiety disorder were significantly associated with PTSD symptoms at 3 months; gender, age, and the presence of delusional memories were not significantly associated. Regarding the risk factors of PTSD, the literature is contradictory (6,12,31,32). In this study, we chose to study the risk factors that were the most frequently identified in the literature.
Interestingly, half of the patients remembered they had experienced delirium during their ICU stay, and many patients mentioned during the telephone interview having traumatic memories of this experience even if they did not present PTSD symptoms at 3 months. This clinical side effect could constitute a traumatic experience for patients, and some authors have suggested its possible implication in the development of future PTSD (33). However, no significant association was found (OR, 1.95; 95% CI, 0.72–5.27), which could be due to the small sample size of the study.
The majority of patients had psychiatric history (64%) and/or ongoing psychotropic treatment before admission (37%). van der Kuur et al (34) recently reported that 19.7% of ICU patients presented with a premorbid psychiatric disorder, which seems to be much lower than that found herein. One aspect that could explain this difference is that the authors collected data from hospital electronic health records, whereas in the present study, we interviewed directly the patients, specifying each diagnosis (34). It is important to note that psychiatric disorders can change perceptions and experiences of the hospitalization, and therefore it seems important to take this factor into account in the care of these ICU patients.
We did not use the gold standard (CAPS) to assess diagnosis of PTSD at 3 months because its administration requires a 1-hour specialized consultation. However, the value of IES-R has been previously demonstrated for the detection of PTSD (9,18) and would be more useful for a nonexperienced healthcare provider regarding PTSD. Regarding the usefulness of IES-R for PTSD diagnosis after 3 months, this scale is not validated but its administration by a psychiatry resident enhanced its reliability. We could have chosen other tools, such as PTSD Checklist for DSM-5 or PTSS-14, that are based on the DSM-5, but the former has not been validated in the ICU population, and the latter has only been validated in a preliminary study on a small ICU population, and there is no validated French version (13). More recently, a shortened version of IES-R, the Impact of Event Scale-6, has been validated with good reliability versus the CAPS in acute respiratory distress syndrome survivors (35). Future research is needed to ascertain whether its administration during the hospital predicts PTSD at 3 months.
According to the NPV of the IES-R threshold of 12, it would be possible to propose a follow-up only to patients with a score greater than or equal to 12, hence to 40% of the study population and to thus avoid evaluations and specialized follow-up for 60% of the patients at ICU discharge. Another point to consider is that sample size was not sufficient to analyze the implications of the different drugs administrated in ICU which may be of importance as it has been suggested that opioids could be a protective factor contrary to inotropes, vasopressors, and benzodiazepines (10,11,32,33,36), and therefore further studies are necessary to explore this point. We did not exclude the four patients who had been rehospitalized during the 3 months, which could be a source of confounding bias.
The results presented herein imply that screening of acute stress symptoms with IES-R in the 8 days after ICU stay could identify patients at risk of later PTSD symptoms. Screening could avoid psychologic follow-up for 57% of patients who are at low risk, in order to preserve limited ICU follow-up resources. Conversely, patients with an IES-R greater than or equal to 12 could benefit from a second assessment by a psychiatrist and possibly, psychologic care. There is no clear guideline regarding treatment of the acute stress symptoms, but most committees recommend short sessions of behavioral-cognitive therapy (37,38). Furthermore, this second assessment by a psychiatrist could avoid the initiation of benzodiazepine treatment in patients at risk of PTSD, as it has been found that exposure to such drugs favors the development of PTSD (39,40); benzodiazepines are frequently prescribed by practitioners after ICU discharge to those with anxiety or sleep disturbance (due to the stress of the hospitalization), without considering that it might integrate into a more complex disorder. More generally, the IES-R questionnaire could also be attached to an information letter about PTSD after ICU stay, explaining symptoms and the possibility to contact a specialist or the family doctor if they cannot benefit of it in hospital.
IES-R assessed at ICU discharge has a good ability for the detection of patients at risk of PTSD symptoms. Patients having history of anxiety disorder and those presenting acute stress symptoms at ICU discharge are more at risk for PTSD symptoms.
We would like to acknowledge all intensivists from G, P, and N ICU of the Edouard Herriot Hospital of Lyon: Dr. Guillaume Marcotte, Dr. Alexandre Faure, Dr. Baptiste Hengy, Dr. Benjamin Delwarde, Dr. Amélie Mazaud, Dr. Claire Delsuc, Dr. Sophie Debord-Peguet, Dr. Bertrand Devigne, Dr. Thomas Jouot, Dr. Martin Bonnassieux, Dr. Jullien Crozon-Clauzel, Dr. Frédérique Bayle, Dr. Arnaud Gregoire, Dr. Laetitia Huriaux, Dr. Veronique Leray, Dr. Thomas Uberti, Pr. Laurent Argaud, Dr. Romain Hernu, Dr. Marie Simon, Dr. Thomas Baudry, Dr. Vincent Jahandiez, Dr. Thomas Madeleine, Dr. Adeline Grateau, Dr. Martin Cour, and Dr. Laurent Bitker. We also would like to acknowledge Pr. Jean-Christophe Richard for his advices in this article redaction, Valérie Cerro, Justine Dubreuil, and Frederic Moll for their hard work in collecting data.
1. American Psychiatric Association: Chapter 10. Trauma- and Stressor-Related Disorders. In: Diagnostic and Statistical Manual of Mental Disorders (DSM-5®). 2013, Ann Arbor, MI, American Psychiatric Association Publishing, University of Michigan p 1519
2. Panagioti M, Gooding PA, Tarrier N. A meta-analysis of the association between posttraumatic stress disorder and suicidality: The role of comorbid depression. Compr Psychiatry. 2012; 53:915–930
3. Lee DJ, Liverant GI, Lowmaster SE, et al. PTSD and reasons for living: Associations with depressive symptoms and alcohol use. Psychiatry Res. 2014; 219:550–555
4. Ahmadi N, Hajsadeghi F, Mirshkarlo HB, et al. Post-traumatic
stress disorder, coronary atherosclerosis, and mortality. Am J Cardiol. 2011; 108:29–33
5. Dedert EA, Calhoun PS, Watkins LL, et al. Posttraumatic stress disorder, cardiovascular, and metabolic disease: A review of the evidence. Ann Behav Med. 2010; 39:61–78
6. Parker AM, Sricharoenchai T, Raparla S, et al. Posttraumatic stress disorder in critical illness survivors: A metaanalysis. Crit Care Med. 2015; 43:1121–1129
7. Wallen K, Chaboyer W, Thalib L, et al. Symptoms of acute posttraumatic stress disorder after intensive care. Am J Crit Care. 2008; 17:534–543; quiz 544
8. Samuelson KA, Lundberg D, Fridlund B. Stressful memories and psychological distress in adult mechanically ventilated intensive care patients - a 2-month follow-up study. Acta Anaesthesiol Scand. 2007; 51:671–678
9. Bienvenu OJ, Williams JB, Yang A, et al. Posttraumatic stress disorder in survivors of acute lung injury: Evaluating the Impact of Event Scale-Revised. Chest. 2013; 144:24–31
10. Jones C, Bäckman C, Capuzzo M, et al. Precipitants of post-traumatic
stress disorder following intensive care: A hypothesis generating study of diversity in care. Intensive Care Med. 2007; 33:978–985
11. Girard TD, Shintani AK, Jackson JC, et al. Risk factors for post-traumatic
stress disorder symptoms following critical illness requiring mechanical ventilation: A prospective cohort study. Crit Care. 2007; 11:R28
12. Morrissey M, Collier E. Literature review of post-traumatic
stress disorder in the critical care population. J Clin Nurs. 2016; 25:1501–1514
13. Twigg E, Humphris G, Jones C, et al. Use of a screening questionnaire for post-traumatic
stress disorder (PTSD) on a sample of UK ICU patients. Acta Anaesthesiol Scand. 2008; 52:202–208
14. Milton A, Brück E, Schandl A, et al. Early psychological screening of intensive care unit survivors: A prospective cohort study. Crit Care. 2017; 21:273
15. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, DSM-III-R. 1980Ann Arbor, MI, American Psychiatric Association Publishing, University of Michigan
16. Brunet A, St-Hilaire A, Jehel L, et al. Validation of a French version of the Impact of Event Scale-Revised. Can J Psychiatry. 2003; 48:56–61
17. Haute Autorité de la Santé Affections Psychiatriques de longue durée -Troubles Anxieux Graves. 2007. Available at: https://www.has-sante.fr/upload/docs/application/pdf/guide_medecin_troubles_anxieux.pdf
. Accessed July 29, 2020
18. Elhai JD, Gray MJ, Kashdan TB, et al. Which instruments are most commonly used to assess traumatic event exposure and posttraumatic effects?: A survey of traumatic stress professionals. J Trauma Stress. 2005; 18:541–545
20. Weiss DS, Marmar CR. The Impact of Event Scale—Revised. Assessing Psychological Trauma and PTSD. 1997New York, NY, Guilford Press399–411
21. The Douglas Mental Health Institute and McGill University: Rapid Diagnosis Toolkit for PTSD.. 2009. Available at: http://www.infotrauma.org/flash/media-e/diagnosisToolkit.pdf
. Accessed April 3, 2020
22. Birmes P, Brunet A, Benoit M, et al. Validation of the Peritraumatic Dissociative Experiences Questionnaire self-report version in two samples of French-speaking individuals exposed to trauma. Eur Psychiatry. 2005; 20:145–151
23. Bui E, Brunet A, Olliac B, et al. Validation of the Peritraumatic Dissociative Experiences Questionnaire and Peritraumatic Distress Inventory in school-aged victims of road traffic accidents. Eur Psychiatry. 2011; 26:108–111
24. Le Gall JR, Lemeshow S, Saulnier F. A new Simplified Acute Physiology Score (SAPS II) based on a European/North American multicenter study. JAMA. 1993; 270:2957–2963
25. Aziz MA, Kenford S. Comparability of telephone and face-to-face interviews in assessing patients with posttraumatic stress disorder. J Psychiatr Pract. 2004; 10:307–313
26. R Core Team. R: A Language and Environment for Statistical Computing. 2018Vienna, Austria, R Foundation for Statistical Computing
27. Davydow DS, Zatzick D, Hough CL, et al. A longitudinal investigation of posttraumatic stress and depressive symptoms over the course of the year following medical-surgical intensive care unit admission. Gen Hosp Psychiatry. 2013; 35:226–232
28. Warlan H, Howland L, Connelly C. Detection of posttraumatic stress symptoms in patients after discharge from intensive care. Am J Crit Care. 2016; 25:509–515
29. Wade DM, Hankins M, Smyth DA, et al. Detecting acute distress and risk of future psychological morbidity in critically ill patients: Validation of the intensive care psychological assessment tool. Crit Care. 2014; 18:519
30. Schandl A, Bottai M, Hellgren E, et al. Developing an early screening instrument for predicting psychological morbidity after critical illness. Crit Care. 2013; 17:R210
31. Long AC, Kross EK, Davydow DS, et al. Posttraumatic stress disorder among survivors of critical illness: Creation of a conceptual model addressing identification, prevention, and management. Intensive Care Med. 2014; 40:820–829
32. Davydow DS, Gifford JM, Desai SV, et al. Posttraumatic stress disorder in general intensive care unit survivors: A systematic review. Gen Hosp Psychiatry. 2008; 30:421–434
33. Wade DM, Howell DC, Weinman JA, et al. Investigating risk factors for psychological morbidity three months after intensive care: A prospective cohort study. Crit Care. 2012; 16:R192
34. van der Kuur A, Bethlehem C, Bruins N, et al. Impact of a premorbid psychiatric disorder on the incidence of delirium
during ICU stay, Morbidity, and long-term mortality. Crit Care Res Pract. 2019; 2019:6402097
35. Hosey MM, Leoutsakos J-MS, Li X, et al. Screening for posttraumatic stress disorder in ARDS survivors: Validation of the Impact of Event Scale-6 (IES-6). Crit Care. 2019; 23:276
36. Huang M, Parker AM, Bienvenu OJ, et al.; National Institutes of Health, National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome Network. Psychiatric symptoms in acute respiratory distress syndrome survivors: A 1-year national multicenter study. Crit Care Med. 2016; 44:954–965
37. Forbes D, Creamer M, Bisson JI, et al. A guide to guidelines for the treatment of PTSD and related conditions. J Trauma Stress. 2010; 23:537–552
38. Department of Veterans Affairs and Department of Defense: VA/DoD Clinical Practice Guideline for Management of Post-Traumatic
Stress. 2010. Available at: https://www.healthquality.va.gov/guidelines/MH/ptsd/cpgPTSDFULL201011612c.pdf
. Accessed June 7, 2020
39. Gelpin E, Bonne O, Peri T, et al. Treatment of recent trauma survivors with benzodiazepines: A prospective study. J Clin Psychiatry. 1996; 57:390–394
40. Mellman TA, Bustamante V, David D, et al. Hypnotic medication in the aftermath of trauma. J Clin Psychiatry. 2002; 63:1183–1184