Posttraumatic stress disorder in the emergency room: exploration of a cognitive model

https://doi.org/10.1016/S0005-7967(01)00123-1Get rights and content

Abstract

Ehlers and Clark (Behav. Res. Ther., 38 (2000) 319) recently proposed a cognitive model of posttraumatic stress disorder (PTSD). In this study, we examined two facets of this model, appraisal and peritraumatic dissociation, in the context of a hospital emergency room. Fifty-one emergency room personnel completed questionnaires measuring posttraumatic stress symptoms, interpretations of traumatic events experienced while working in the emergency room and subsequent intrusive recollections, and peritraumatic dissociation. Twelve percent of participants met formal diagnostic criteria for PTSD, and 20% met PTSD symptom criteria. As predicted, both negative appraisals of the trauma and of intrusive recollections were associated with increased PTSD severity. Although peritraumatic dissociation did not correlate with overall PTSD symptom severity, it was associated with the reexperiencing symptom cluster. Discussion focuses on the factors associated with PTSD in emergency room professionals and implications for intervention.

Introduction

Posttraumatic stress disorder (PTSD) is classified as an anxiety disorder, and typically with anxiety the person fears an impending threat. With PTSD, however, the distressing event has already occurred. One ongoing puzzle with this condition is why patients continue to behave as though a past event is an impending event. To explain this phenomenon, Ehlers and Clark (2000) proposed a cognitive model in which they assert that persistent PTSD occurs when people process the traumatic event in ways that lead to a sense of current, serious threat. According to these writers, this sense of threat arises as a result of: (a) inordinately negative appraisals of the trauma and its sequelae; and (b) disturbances in autobiographical memory. Regarding memory disturbance, one factor that is hypothesized to contribute to the poor contextualism and elaboration that characterize memory disturbance in individuals with PTSD is dissociation during the event, and this will be our focus here.

Studies by Ehlers and her colleagues have provided preliminary support for this model, particularly with regard to the association between negative appraisals and severity of PTSD. Excessively negative appraisals of traumatic events, initial PTSD symptoms, and trauma-induced changes in self have been found to correlate with both PTSD severity and persistence in victims of various traumas (Dunmore, Clark, & Ehlers, 1997; Ehlers et al., 1998a; Ehlers, Maercker, & Boos, 2000; Ehlers, Mayou, & Bryant, 1998b). In a similar vein, negative appraisals of intrusive thoughts and recollections of the event have also been shown to be associated with PTSD (Clohessy & Ehlers, 1999; Ehlers et al., 1998b), even after partialing out accident severity, intrusion frequency, or general anxiety-related cognitions (Steil & Ehlers, 2000). Negative appraisals of intrusions were also found to be associated with the use of maladaptive strategies, such as rumination, suppression and dissociation, to control spontaneous recollections. This association holds even when controlling for intrusion frequency. These strategies, in turn, have been linked to more severe PTSD symptomology (Clohessy & Ehlers, 1999; Ehlers et al., 1998b; Steil & Ehlers, 2000).

There is also empirical support for the role of dissociation in PTSD. Peritraumatic dissociation refers to derealization, memory disturbances, depersonalization, and altered body image and time sense experienced at the time of the trauma (Marmar, Weiss, Metzler, Ronfeldt, & Foreman, 1996b). Consistent with the cognitive model, peritraumatic dissociation has been found to predict symptomatic stress and PTSD and to remain strongly predictive even after controlling for factors such as adjustment and social support (Marmar et al., 1996b, Marmar et al., 1999; Weiss, Marmar, Metzler, & Ronfeldt, 1995). Moreover, Ursano et al. (1999) reported that those who experienced peritraumatic dissociation were more than four times as likely to develop PTSD than those who did not.

Although the cognitive model is quite promising, more work is needed to replicate and extend these findings. Relatively few studies have been conducted and many of these focus on only one aspect of the model. For example, only one study has measured appraisal and dissociation simultaneously. It is also important to examine the model in other populations and settings to determine whether findings suggesting that cognitive processes contribute to PTSD are robust to changes in context.

Studies of cognitive processes in PTSD have focused primarily on individuals who are the victims of traumatic events that are out of the ordinary for them, for example, the woman who was raped, or the man who survived a terrible motor vehicle accident. Less work has addressed the development of PTSD in individuals who routinely deal with horrifying events as part of their jobs, for example, emergency service workers such as paramedics or firefighters. This emphasis may reflect the earlier DSM-III requirement that an event be “outside the normal range of human experience” (American Psychiatric Association, 1980). In the DSM-III-R and DSM-IV, criterion A, which defines what constitutes a traumatic event, has been broadened to include witnessing or being confronted with an event that threatens one's own or others' lives and produces a response of “intense fear, helplessness, or horror” (American Psychiatric Association, 1987, American Psychiatric Association, 1994). This change increased the boundaries for diagnosing PTSD. It also points to a need to examine the frequency with which PTSD develops in populations where distressing events are routine rather than atypical and more likely to be witnessed than directly experienced, such as emergency service workers.

The most commonly studied group of emergency service workers is disaster workers, typically a combination of paramedics, police officers, and firefighters. Whereas epidemiological studies investigating past week to six month rates in the general population have found prevalences ranging from 0.4 to 4.6% (Bernat, Rondfelt, Calhoun, & Arias, 1998; Davidson, Hughes, Blazer, & George, 1991; Kessler, Sonnega, Bromet, Hughes, & Nelson, 1995; Resnick, Kilpatrick, Dansky, Saunders, & Best, 1993), prevalence rates in disaster workers range from 2–17% (Anderson, Christensen, & Petersen, 1991; Durham, McCammon, & Allison, 1985; Ersland, Weisaeth, & Sund, 1989; McCammon, Durham, Allison, & Williamson, 1988; Ursano, Fullerton, Kao, & Bhartiya, 1995; Weiss et al., 1995). For ambulance attendants it has been reported at 15–22%, and for firefighters alone between 17–32% (Clohessy & Ehlers, 1999; Corneil, Beaton, Murphy, Johnson, & Pike, 1999; DeAngelis, 1995, Grevin, 1996, McFarlane, 1989; Rentoul & Ravenscroft, 1993; Wagner, Heinrichs, & Ehlert, 1998). These results suggest that emergency workers are indeed at risk of developing PTSD and other distress symptoms (Marmar et al., 1996b, Wagner et al., 1998).

In terms of the cognitive model, many emergency service workers report intrusive memories of work-related traumatic events (Thompson and Suzuki, 1991). For example, Durham et al. (1985) reported that among the rescue personnel in their study, intrusive thoughts about the disaster were the most frequently endorsed PTSD symptom. Similarly, Genest, Levine, Ramsden, and Swanson (1990) found that among volunteer ambulance attendants involved in unsuccessful cardiopulmonary resuscitation attempts, fewer than 5% reported never thinking about the experience again. Moreover, for most of the participants, their recollections were not voluntary. This suggests that these individuals share one of the hallmark cognitive characteristics of PTSD.

Only a few studies have directly examined the association between the cognitive processes implicated in the Ehlers and Clark model and PTSD in the context of the emergency services. One study indicated that negative interpretations of intrusive recollections were predictive of PTSD severity in a sample of ambulance attendants (Clohessy & Ehlers, 1999). As well, a series of publications on one sample of rescue workers revealed that peritraumatic dissociation predicted symptomatic stress and PTSD (Marmar et al., 1996b, Marmar et al., 1999, Weiss et al., 1995). However, formal tests of the cognitive model in emergency workplaces have been few and replication of these findings is necessary. In addition, the emergency service populations studied to date, disaster workers, firefighters, police officers and paramedics, are unique in that they face external physical challenges such as bad weather or burning buildings, which might be expected to elicit fear and arousal. For these individuals, the line between direct threat to the self and more routine workplace witnessing of threat to others may be blurred.

Here, we propose to study hospital emergency room personnel. These workers routinely witness life-threatening situations experienced by their patients. However, they work in a more controlled environment than disaster workers, ambulance attendants and firefighters. They do not face the added external physical dangers of outside workers and are equipped to deal with a wider variety of medical difficulties, having at their disposal a broader array of diagnostic tools as well as experts, specialists and advanced facilities on site. Despite these differences, no empirical studies of PTSD in emergency room professionals have been conducted. In fact, the only studies on PTSD in hospital staff involved a cross-section of hospital personnel, and comparisons between this group and on-scene rescue workers yielded inconsistent results (Durham et al., 1985, McCammon et al., 1988). The importance of emergency room professionals to medical care and reports of high rates of burnout and job transfer in this group underscores the need to understand anxiety-related conditions in this population.

Emergency room professionals are not only in a unique position to provide information on the boundaries of work-related PTSD, they also offer a challenging context in which to evaluate the Ehlers and Clark cognitive model. Because these are, for the most part, well-trained, experienced professionals working under controlled conditions, they operate under circumstances that might reduce the toxic cognitive processes that give rise to PTSD. For example, one might ask whether these individuals are as likely to experience dissociation and subsequent memory disturbance as disaster workers. It may also be that their professional training reduces the frequency or impact of intrusive memories and negative appraisals of the events they handle.

One important limitation of previous studies of emergency workers is the failure to measure full PTSD criteria. None of the studies noted above have measured criterion A, which refers to witnessing or experiencing an event involving actual or threatened death or serious injury, where the individual experienced intense fear, helplessness or horror. Due to their training, emergency service workers may not be as likely to have a subjective response of intense fear, helplessness or horror. If so, previous work may have overestimated the prevalence of PTSD in this population. Furthermore, no one has assessed criterion F, i.e. whether the event causes significant impairment in life functioning, as required in the DSM-IV.

Several studies based PTSD prevalence rates on measures such as the Impact of Event Scale (IES), which only assesses two out of the three symptom clusters, or the Posttraumatic Stress Symptom Scale (PSS), which only measures the three symptom clusters. While researchers have found that high scores on scales such as these are good predictors of PTSD (Foa, Riggs, Dancu, & Rothbaum, 1993; Perry, Difede, Musngi, Frances, & Jacobsberg, 1992) neither the IES nor PSS assess all the criteria necessary to warrant a diagnosis of PTSD. Again, such practices may mean that current PTSD prevalence estimates are overestimated.

We will also address a phenomenon that has yielded conflicting evidence, namely the relationship between years of experience in the emergency services and PTSD severity. Some studies have found positive relationships wherein more experienced staff experienced greater symptomatology (Corneil, 1995; Hodgkinson & Shepherd, 1994; Moran & Britton, 1994; Wagner et al., 1998). These findings point to a sensitization, or accumulation, effect whereby exposure to repeated distressing events results in greater pathology. Conversely, other researchers found negative relationships (Hytten & Hasle, 1989; Marmar et al., 1999, Weiss et al., 1995) or no relationship between these variables (Beaton, Murphy, Johnson, Pike, & Corneil, 1999; Clohessy & Ehlers, 1999; Grevin, 1996, Marmar et al., 1996b; Thompson & Suzuki, 1991; Ursano et al., 1995). In light of this very mixed evidence, the issue requires further study.

This study had three goals. First, to formally assess the prevalence of PTSD in emergency room professionals using full DSM-IV criteria. We predicted that emergency room professionals would exhibit a higher incidence of PTSD than that found in the general population. Second, to examine the relationship between PTSD and key cognitive processes implicated in the Ehlers and Clark model. Here, we predicted that appraisal and dissociation would be associated with increased PTSD symptom severity. Finally, we examined the impact of years of workplace experience on PTSD.

Section snippets

Participants

Participants consisted of 53 emergency room workers at a major hospital in a large urban center in British Columbia. These participants were recruited from staff meetings and by word of mouth. Of the 79 people who picked up a questionnaire package, 53 (67%) completed it,1 which represents approximately 44% of people who

Results

This study examined psychological difficulties in a normal population. Therefore, as expected, scores on most measures in this study were significantly non-normally distributed. Log transformations were completed to normalize the data; however, the differences in results between the transformed and untransformed data were not large enough to warrant transformation of all variables. Consequently, the results reported below are based on untransformed data and these results are thus conservatively

Discussion

The results suggested that hospital emergency room workers were at increased risk for developing PTSD, whether the disorder was diagnosed on the basis of core symptom clusters alone or full PTSD criteria. The findings also indicated that cognitive processes played a key role in the disorder. How the individual appraised the traumatic event and its sequelae and peritraumatic dissociation predicted more severe symptoms. These results provide independent confirmation of key elements of the Ehlers

Acknowledgements

This study was supported by a Social Sciences and Humanities Research Council grant to the second author, and a National Sciences and Engineering Research Council scholarship and British Columbia Medical Services Foundation scholarship to the first author.

We would like to thank Anke Ehlers for providing the Response to Intrusions Questionnaire, and the hospital for supporting this study. We would also like to thank the following people for their assistance in conducting this research and for

References (54)

  • H.S Anderson et al.

    Post-traumatic stress reactions amongst rescue workers after a major rail accident

    Anxiety Research

    (1991)
  • R Beaton et al.

    Coping responses and posttraumatic stress symptomatology in urban fire service personnel

    Journal of Traumatic Stress

    (1999)
  • J.A Bernat et al.

    Prevalence of traumatic events and peritraumatic predictors of posttraumatic stress symptoms in a nonclinical sample of college students

    Journal of Traumatic Stress

    (1998)
  • C Burns et al.

    Emergency nurses' perceptions of critical incidents and stress debriefing

    Journal of Emergency Nursing

    (1993)
  • S Clohessy et al.

    PTSD symptoms, response to intrusive memories and coping in ambulance service workers

    British Journal of Clinical Psychology

    (1999)
  • W Corneil

    Traumatic stress and organizational strain in the fire service

  • W Corneil et al.

    Exposure to traumatic incidents and prevalence of posttraumatic stress symptomology in urban firefighters in two countries

    Journal of Occupational Health Psychology

    (1999)
  • J Cudmore

    Preventing posttraumatic stress disorder in accident and emergency nursing: A review of the literature

    Nursing in Critical Care

    (1996)
  • J.R Davidson et al.

    Posttraumatic stress disorder in the community: An epidemiological study

    Psychological Medicine

    (1991)
  • T DeAngelis

    Firefighter's PTSD at dangerous levels

    APA Monitor

    (1995)
  • E Dunmore et al.

    Cognitive factors in persistent versus recovered PTSD after physical or sexual assault: A pilot study

    Behavioural and Cognitive Psychotherapy

    (1997)
  • A Ehlers et al.

    Predicting response to exposure treatment in PTSD: The role of mental defeat and alienation

    Journal of Traumatic Stress

    (1998)
  • A Ehlers et al.

    Posttraumatic stress disorder following political imprisonment: The role of mental defeat, alienation, and perceived permanent change

    Journal of Abnormal Psychology

    (2000)
  • A Ehlers et al.

    Psychological predictors of chronic posttraumatic stress disorder after motor vehicle accidents

    Journal of Abnormal Psychology

    (1998)
  • A Ehlers et al.

    Maintenance of intrusive memories in PTSD: A cognitive approach

    Behavioral and Cognitive Psychotherapy

    (1995)
  • S Ersland et al.

    The stress upon rescuers involved in an oil rig disaster. “Alexander L. Kielland” 1980

    Acta Psychiatrica Scandinavica

    (1989)
  • C.M.B Fernandes et al.

    Violence in the emergency department: A survey of health care workers

    Canadian Medical Association Journal

    (1999)
  • Cited by (116)

    • A first remotely-delivered guided brief intervention to reduce intrusive memories of psychological trauma for healthcare staff working during the ongoing COVID-19 pandemic: Study protocol for a randomised controlled trial

      2022, Contemporary Clinical Trials Communications
      Citation Excerpt :

      During the pandemic, high and increased levels of symptoms related to PTSD have been documented in healthcare workers globally [6–9], e.g. 40% of ICU staff [9]. PTSD symptoms have been linked to impaired work functioning [10], job termination [11], and burnout [12]. Whilst empirically-supported treatments for PTSD exist, we lack evidence-based interventions suitable for individuals experiencing ongoing/recurrent trauma (such as healthcare staff), as well as interventions targeting specific symptoms [13].

    View all citing articles on Scopus
    View full text