Registered Psychiatric Nurses, Partnering with People

Factors Affecting Post-Traumatic Stress Disorders and Depression in Psychiatric Nurses




There has been very little research into psychiatric nursing and the psychological implications for individuals employed in the field. In the psychiatric milieu assaults, by clients, against staff and other clients are not uncommon. Furthermore, psychiatric nurses are often encounter frequent vicarious trauma through the course of daily interactions with clients. The purpose of the current study was to examine variables affecting the incidence of post-traumatic stress disorder (PTSD) and depression in psychiatric nurses in Saskatchewan. The Beck Depression Inventory (BDI), Davidson Trauma Scale (DTS), Eysenck Personality Questionnaire (EPI), two sub-scales of the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q), and an open ended questionnaire and demographic sheet were mailed out to 600 psychiatric nurses in Saskatchewan. A cut off of 212 responses were used in the data analysis. Results indicate that many psychiatric nurses (48%) reported experiencing a traumatic event in the workplace. The overall incidence of PTSD (24.53%) found in this sample was twice the rate expected in normal community-based samples. Incidence of depression were also increased (25%). Statistical analysis suggests that variables related to increased PTSD are: lower quality of life both in and out of the workplace, lack of employer support, and higher levels of neuroticism.

Most people will spend a large portion of their lives in the workplace. Unfortunately, some work settings perpetuate an unhealthy environment and have a greater likelihood, than others, of contributing to stress, depression and post-traumatic stress. The American Psychiatric Association (1994) has defined post-traumatic stress disorders (PTSD) as being caused by witnessing or experiencing an event that involves actual or threatened death in relation to self or others, combined with personal fear or helplessness. Afterwards, the victim may reexperience the event, exhibit avoidance behaviors, and show signs of increased arousal (i.e., irritability, sleep problems, hypervigilance). Depression is an affective disorder characterized by depressed mood and loss of interest or pleasure and may include insomnia or hypersomnia, weight loss or gain, diminished concentration, and fatigue (American Psychiatric Association, 1994).

Return to top

Literature Review

Literature Review

Previous research has suggested that institutions, such as hospitals, could qualify as milieus that may perpetuate psychological symptoms in employees. The psychiatric ward in particular can be extremely problematic in this regard (Wykes & Whittington, 1991). More than any other area of the hospital, this setting has been found to have employees who are verbally and physically assaulted (Wykes & Whittington, 1998). Studies have consistently demonstrated that in cases of in-hospital assaults 90% are directed at nursing staff (Whittington & Wykes, 1994). Furthermore, nurses may experience vicarious trauma (i.e. as a result of seeing or hearing about traumatic events), on a regular basis (Figley, 1995).

PTSD predisposes individuals to developing other psychological problems, it has been estimated that up to 80% of victims suffer from some type of comorbid disorder. Numerous studies describe that depression, in particular, has long been recognized as being comorbid with PTSD (Lange, Lange & Cabaltica, 2001; Stein, McQuaid, Pedrelli Lenox & McCahill; Teegan & Muller, 2000). Demographic variables, such as age and gender, as well as environmental variables like the availability of social support have also been identified as related to differing levels of PTSD (Brewin, Andrews & Valentine, 2000; King, King & Foy, 2000).

The clinical course of the disease varies and symptoms that appear immediately or later can either disappear over a period of months or may last indefinitely. Of those having clinical levels of pathology one third meet the DSM criteria for post traumatic stress disorder after 10 years (Lange, 2000), making it a potentially long term treatment problem. Coping strategies are associated with arousal of the sympathetic nervous system, which in turn caused biological and psychological phenomena, through the release of neurotransmitters, ultimately concluding in maladaptive responses by the individual. These maladaptive responses eventually end in frustration, exhaustion, burn out and, long term symptoms. Of concern, for the individual and society, is the potential for unresolved or repeated stress and trauma to contribute to the development of chronic psychiatric disorders (Valent, 1995). Previous studies have shown that stressful jobs involving exposure to trauma may lead to PTSD (Lange, 2000).

The present study examined traumatic incidents in the workplace, as precursors to symptomatology indicative of PTSD and further examined the possibility of a diagnosis of comorbidity of depression and PTSD among psychiatric nurses. In addition, workplace quality of life, employee perception of the work environment, and personality type, as outlined by the Eysenck Personality Inventory, were considered in an attempt to define whether these variables, when juxtaposed with a traumatic event, were predictive of the development of PTSD.

Return to top


Stress, Depression and Trauma in Psychiatric Nurses

Historically, research into problematic work environments has focused on levels of stress experienced by employees, since the mid 1970’s, burnout, in various occupations has been the topic of many research projects. One occupation that has not been extensively researched but that indicated a high level of burnout was nursing (Malach-Pines, 2000). Several studies have concluded that health care staff are more prone to experience an increased incidence of stress and depression related to workplace experiences than are those in other occupations (Caldwell, 1992; van Servellen, Fopf, & Leake, 1994; Weinberg & Creed, 2000).

Murray and Snyder (1991) noted that nurses experience a range of complications related to workplace assaults including physical, social, family, and emotional reactions. Following critical workplace incidents, psychiatric nurses have been found to manifest avoidance behaviors with family and others as well as to withdraw from intense relationships (Scott & Stradline, 1994; Jones, Janman, Payne, & Rick, 1987). Also, they experience a heightened sense of anger and vulnerability and tend to subsequently view the world as unpredictable and unjust. Murray and Snyder’s (1991) study reported that 67% of 61 assaulted staff members at the University of California, Los Angeles “had a significant response to the assault within 1-week (p. 25)”. Reactions, however, were not necessarily immediately apparent and delayed responses were noted to occur six months to 1 year, or even longer, after the incident. Although staff members were not screened for or identified as having PTSD, it was ascertained through questionnaires that these individuals were, in fact, experiencing PTSD like symptoms (Murray & Snyder, 1991).

Patient suicide and violence against psychiatric staff have been recognized as contributing factors to the types of behavioral and cognitive changes seen in post- traumatic stress disorder (Cooper, 1987). These changes affect the ability of individual nurses to function both in and out of the workplace. The long term implications are burnout and dysfunctional behaviours resulting in compromised patient care and increased conflicts between psychiatric nurses and their families and co-workers (Jones, 1987). Eventually, certain individuals may develop symptomatology indicating elevated levels of both PTSD and depression further compromising professional and personal relationships (Figley, 1995).

Another compounding factor is the degree to which secondary or vicarious traumatic stress impacts on psychiatric staff (Figley, 1995). Valet (1995), claims that secondary traumatic stress is a response of the initial, or primary, victim that generates a resulting secondary response in the caregiver. Anxiety and depression are commonly recognized responses to such stress. Valet further proposes that care givers attempt to nullify their responses through survival strategies.

A study of ambulance personnel who were exposed to critical incidents revealed that approximately one third had symptoms of burnout and PTSD culminating in lower job satisfaction (Alexander & Klein, 2001). This type of level was also observed in personnel working in other jobs where trauma exposure is not unusual such as firefighting, emergency room nursing, police work, military combat and military nursing (Cudmore, 1996; Lange, et al, 2000; Wagner, Heinrich, Ehert, 1998)). Despite the fact that psychiatric nurses have been found to have high levels of job stress, little is known about the incidence of job related PTSD in this population.

One study investigated the incidence of traumatic events experienced by psychiatric hospital personnel. Caldwell (1992) examined the incidence of PTSD in a sample of 300 hospital staff employed at two separate psychiatric facilities. Of these, 138 of the 224 clinical staff (62 %) reported experiencing a critical event in the workplace and 61% also reported symptoms of PTSD while 10% (n = 23) were felt to have merited an actual clinical diagnosis of PTSD (Caldwell, 1992).

Return to top


Demographic and Personality Factors

A number of demographic variables have been found to be predictors of stress and depression in psychiatric nurses. Gender, which has often been identified as playing an important role in the rate of occurrence of various types of psychopathology, has also surfaced as a factor in disorders in psychiatric nurses. Female psychiatric nurses exhibit slightly more psychological distress than male psychiatric nurses (Cooper, 1995). Females in general also exhibit a greater prevelance of PTSD than males in virtually all studies (Stein, Walker & Ford, 2000, Brelau, Davis, Andreski, Peterson, & Schultz, 1997). An exception was a meta-analysis of a military sample which found no significant difference between genders as a risk factors for developing PTSD (Brewin, Andrews & Valentine, 2000).

Although Cooper (1995), found no difference between the number of assaults on male or female psychiatric nurses on the ward, she concluded that women are held responsible by the employer for assaults more often than are men. In spite of this, females showed greater job satisfaction (Cooper, 1995; Jones, 1987).

Psychopathology if left untreated may have serious consequences for the affected individual regardless of gender. For example, while nurses in general are reported as being at higher risk for suicide, there is no difference in numbers of attempted suicides between the sexes (Katz, 1983). Samuelsson (1997) supported this conclusion in a study of 197 psychiatric nurses. Out of 138 women and 53 men, in the study a total of 13% had attempted suicide earlier than the previous year with no difference between the sexes (Samuelsson, Gustavsson, Peterson, Arnetz & Asberg, 1997).

Low levels of personality hardiness and self-esteem have also been found to be predictors of both anxiety and depression (van Servellen, Fopf & Leake, 1994; Thomsens, Arnetz, Soares & Dallander, 1999). Personality hardiness is described as the inter-relatedness of three factors exercised by the individual in their lifestyle: control of environment, commitment to self-fulfilling goals and reasonable levels of challenge in daily life. These traits serve as buffers and seem to protect the individual from the psychological repercussions of stress (Thomsens et al., 1999). Self-esteem encompasses positive and negative attitudes concerning the self (Rosenberg, 1965). Both hardiness and self-esteem have been correlated with how the workplace is perceived by workers as well as with personal health (Thomsen et al., 1999). Van Servellen (1994) found that hardiness decreased work-related stress, depression, and anxiety in hospital nurses. Thomsen et al., (1999) in a corroborative study found that both self-esteem and hardiness correlated with less psychological distress.

Other studies using personality scales have examined the role of personal disposition in psychological symptomatology. The personality trait of neuroticism has been positively correlated with increased levels of numerous psychiatric pathologies including physical and psychological distress, anxiety, depression, and PTSD (Jones, et al, 1987; Jorm, Christensen, Henderson, Jacomb, Korten, Rodgers, 2000, Samuelsson et al., 1997). Piedmont (1993), utilizing the Maslach Burnout Inventory, found that emotional exhaustion and depersonalization are strongly correlated with neuroticism and may predispose the individual to burnout. High scores on neuroticism have also been reported to intensify the individual’s perception of psychological and physical symptoms (Feldman, Cohen, Gwaltney, Doyle, Skoner, 1999; Payne, 1986). Jones (1987) compared 349 psychiatric nurses to a normative population of 2,000 people and found nurses to score significantly lower in neuroticism but higher in job related psychological stress.

Being younger and having less education have also been found to be positively correlated with more psychological symptoms. A mixed sample of 126 Greek nurses (79 general and 47 psychiatric) completed The Symptom Check List-90-R, which measures a variety of psychopathological parameters. As a group, nurses scored higher in most areas than a standardized population sample matched for sex and age. The younger and less experienced nurses showed more psychological symptoms than those who were older and had more experiences (Garyfallos, Adamopoulou, Moutzoukis, Panakleridou, Kapsala & Lanara, 1993).

Finally, working in larger centers was shown to decreases job satisfaction and contributes to overall stress. Jones (1997) explained that the greater the number of nurses, or the larger the center, the more social distance there is between workers which contributes to increased stress levels (Jones, 1997).

Return to top


Work Environment

When examining the impact of stress or trauma on any individual within a given workplace, environmental factors must be considered. Some of the factors which have been found to influence workers level of stress are: the environment itself, interpersonal relationships, degree of employer responsibility for a positive work environment, and availability of support programs for employees. Moreover, work related stress in hospital nurses was found to increase both the reported incidence of emotional exhaustion (Oehler, Davidson, Starr & Lee 1991) and reports of poorer health (van Servellen et al., 1994).

Cooper’s (1995) meta analysis identified employer reactions to critical incidents such as psychiatric patient suicide and assault of psychiatric staff as key in facilitating or impeding the integration of the incident. The employer often failed to provide support on behalf of employees when, for example, psychiatric nurses face a suicide on the ward. Institutional reactions include panic, denial, and the blaming of staff victims for incidents such as assault. While the staff member involved may react by manifesting PTSD symptoms, the employer compounds the problem through their reaction. A further difficulty arises when staff attempt to cope with such incidents themselves and to preserve a good worldview and sense of personal invulnerability. This psychological coping can lead to a tendency to “blame the victims” and in incidents of assaults “the victims blame themselves” (Cooper, 1995).

Previous research suggests that co-workers and employers are often the first to be aware of staff victims. Murray and Snyder (1991) surveyed a small sample of nurses who became victimized. They defined a staff victim as a worker who has been assaulted in the workplace by a patient. They found that the reaction of co-workers is influential in the staff victims recovery from traumatic assaults. When co-workers minimize or are critical of the traumatic incident experienced by the victimized staff member, their maladaptive response ultimately becomes more important to the staff victim than the incident itself. Employers have been reported to downplay the importance of such events while nurses tend to shoulder the blame (Cooper, 1995). This is particularly relevant in that symptomatology, particularly of PTSD, may be delayed for weeks following the incident (Murray & Snyder, 1991). Ten or even more years after the fact, health care professionals have often been found to recall incidences of assaults with great clarity (Kellog & Triffleman, 1998). Although Murray’s (1991) sample was very small and could not, therefore, be used to generalize to the population of psychiatric nurses, other studies confirm the importance of responses from both the workplace and the employer following critical incidents.

Lack of management support and role problems are serious considerations within the hospital milieu (Weinberg, 2000). A sample of 300 mental health staff employed at two sites (a state hospital and private psychiatric facility) returned questionnaires assessing the incidence of trauma and PTSD (Caldwell, 1994). A large number of participants displayed symptoms of PTSD, however, even more significantly, employees at both locations indicated that the employer provided little support of victimized staff. In debriefing sessions review questions focused on how the employee had erred rather than exploring psychological implications for the traumatized individuals (Caldwell, 1992). This type of workplace response rather than supporting positive recovery contributes to the tendency of the employees to blame themselves for the incident.

Return to top


Purpose of Research

There is a need for further research investigating potential levels of traumatic stress and depression in the psychiatric nursing milieu. There has been no such study within the province of Saskatchewan to date. Previous research suggests that the workplace influences development of stress related disorders in psychiatric nurses and indicates that they experience elevated symptomatology with regard to both depression and traumatic stress. The often maladaptive responses from the workplace impact on nurses employment and their home environment (Cooper, 1995; Jones, et al, 1987; Scott, et al, 1994; Skinner, et al, 1993).

Employer programs have been shown to contribute to the development of psychiatric symptomatology in nurses when they do not provide the needed support for recovery from traumatic events. Debriefing, in critical situations, is often superficial, conducted quickly and may, in fact, be more harmful than helpful in that scapegoating and blaming frequently occur. These are important findings since the employer/co-worker response has been shown to be more important, in recovery, than even the actual critical situation (Caldwell, 1992; Murray & Snyder, 1991; Weinberg et al., 2000). Finally, studies indicate that demographic and personality traits of health care professionals affect the impact of the individual response to stress and traumatic circumstances. Some employees appear to be more resilient due to personality hardiness and low levels of neuroticism while other less hardy individuals frequently display increased levels of psychopathology (Dudley, Langluddecke & Tennante, 1988; Samuelsson, et al., 1997; Thomsens, 1999; van Servellen, 1994).

Research Q

Research Questions

This study adds to and expands on existing literature by investigating the relationships between workplace stress, depression, and trauma levels in psychiatric nurses. This study hypothesizes that:

  1. Psychiatric nurses show high levels of depression as compared to the normal population.
  2. Many psychiatric nurses have experienced trauma in the workplace and a larger number show symptoms of post-traumatic stress disorder.
  3. Quality of life in the work environment and in general activities is lower for psychiatric nurses who suffer from PTSD symptomatology.
  4. Lack of employee access to adequate critical incident debriefing is related to elevated PTSD scores.
  5. Neuroticism scores will be higher for those experiencing symptomatology indicative of PTSD.
  6. Female psychiatric nurses will have higher levels of PTSD than male psychiatric nurses.

Return to top


Method – Data Collection

Participants. The participants in this study were 212 registered psychiatric nurses who responded to a mail survey. Following approval from the University of Regina ethics board (Appendix A) six hundred packages were sent to Saskatchewan nurses by the Saskatchewan Registered Psychiatric Nurses Association. Names were randomly selected from the overall membership by the secretary of the association. Participants reside in various cities and towns throughout Saskatchewan and are employed in a variety of areas within the field of psychiatric nursing giving the sample a broad range of treatment settings.

Instruments. Two questionnaires and 4 instruments were used (See Appendices C-E). A demographic and background questionnaire asked for information regarding education and work histories including non-identifying work related information. Workplace questions determined employee job satisfaction and perception of the work environment.

Confidentiality was assured by assigning a number to each mailed package. These numbers appeared on completed packages in lieu of names. Completed packages were returned to the supervisor at Luther College. The researcher did not receive any signed documents.

The Davidson Trauma Scale (DTS) is a seventeen-item scale developed to assess symptoms of post-traumatic stress. Responses are recorded using three clusters of sub- scores that are computed from 5-point rating scales. A total score is generated using the totals of the three sub scales. Reliability and validity have been established (Multi-Health System Inc., 1996). Test-retest reliability is r =0.86, and split-half reliability is r =. 95. Validity was determined comparing DTS’s similarity to the conceptually related Symptom Checklist 90-R and its difference from conceptually different measures. The cut- off score of 20, as recommended by Davidson (1996) was used to separate participants having probable clinical pathology from those less likely to.

The Beck Depression Inventory (BDI) is a 21- item self-report instrument used to measure the severity of depression from minimal, mild moderate through severe. Each item is rated on a 4-point scale from 0-3. The BDI is one of the most commonly used instruments for measuring levels of depression and both reliability and validity have been repeatedly established. Test-retest correlation is r = .93. In construct validity the correlation between BDI-IA and BDI-II is .93 (The Psychological Corporation, 1996). A cut-score of 14 as recommended by Beck (1996) was used to separate participants likely to merit a clinical diagnosis of depression from those without depression.

The Quality of Life Satisfaction Questionnaire (Q-LES-Q) consists of 8 sub-scales that examine quality of life in a variety of areas. It is used to test all categories or to isolate specific sections. The 13-question work sub-scale of the Q-LES-Q was used to determine participant satisfaction within and enjoyment of the work place environment over the last week of employment. The 16-question general activities sub-scale was used to measure perceptions of satisfaction with daily life. Questions are responded to on a 5- point Likert scale. Test-retest reliability is r=. 82. Internal consistency is alpha=. 92 (Endicott, Nee, Harrision, Blumenthal, 1993).

The Eysenck Personality Inventory (EPI) inventory measures three dimensions of personality traits: Extraversion-Introversion, Neuroticism-Stability, and Psychoticism. A lie scale was incorporated into the questionnaire to measure the possible tendency of some participants to “fake good” (Eysenck & Eysenck, 1975). EPI has been one of the most commonly used instruments to assess personality traits and all questions require either yes or no answers. Reliability and validity have been previously established for this instrument (Jones, Janman, Payne & Rick, 1987). Analysis of the sub-scales of the Eysenck Personality Inventory indicated standardized item alphas of .4077 for psychoticism, .8424 for introversion, .8873 for neuroticism, and .8179 for the lie scale. Procedure. Following Ethics Board approval of the study, the questionnaire packages were prepared and delivered to the Psychiatric Nurses Association. Each package contained

  1. Two letters of introduction
  2. Demographic and background questionnaire
  3. Davidson Trauma Scale
  4. Beck Depression Inventory
  5. Two Quality of Life Satisfaction Questionnaire sub-scales
  6. Eysenck Personality Inventory. (Appendices B through F respectively).

Self- addressed stamped envelopes were provided for participants to return completed instruments. These items constituted a complete package and were mailed to randomly selected members of the association. The letter of introduction addressed all consent form guideline issues and further explained that return of completed material constituted agreement to participate in the study. Telephone numbers and e-mail addresses of the researcher, thesis supervisor, and Chair of the Research Ethics Board were provided for potential participants to contact should they have had any concerns. A second letter, from the Executive Director of the Registered Psychiatric Nurses Association requested support for the study and explained the relevance to the association. (Appendix B).

Return to top

Data Analysis

Data Analysis

The Statistical Package for Social Sciences (SPSS-7.5) was used to analyze the data. Means, standard deviations for scales, sub-scales, and demographic information were tabulated

The overall sample of 212 was divided into those who reported a work-place traumatic event (trauma event group; n = 100) and those who reported no work-place traumatic event (no trauma event group; n = 112). The traumatic event group was further divided into those who scored 20 and above on the Davidson Trauma scale (Davidson, 1996); this group was referred to as the PTSD group (n = 52). Participants who experienced work-place trauma but score less than 20 on the Davidson Trauma scale were assigned to the non-PTSD group (n = 48).

A further division was made between psychiatric nurses working in various areas of employment within the psychiatric nursing milieu, as identified on the demographic sheet including; psychiatric wards, acute inpatient settings, elderly care, correctional centers, and other unspecified locations. Open-ended qualitative questions were analyzed using thematic analysis and responses were recorded by identified categories. These questions were also converted into quantitative form for analytical purposes. Questions four, through seven and questions nine and ten were assigned numerical values. Questions one through three received one point for each response. The numerical values were utilized in chi square analysis.

Return to top


Results – Demographic Information

Due to time restrictions a cut off for analysis of returned questionnaires was implemented at which time 212 (33.5%) of individuals, out of the 600-person mail out, had responded. Analysis of demographic variables presents a description of participants (Table 1). Out of this sample 82.1% (n = 174) were female, 14.2 % (n=30) were male, and 3.7% (n = 8) did not identify gender. Mean age of participants was 42.7 years (SD = 9.33). Mean income was $47, 150 per anum. Average years of post secondary education history relevant to psychiatric nursing, was 2.4 years. Reported average for psychiatric nursing work history was 18.35 years (Table 1). Participants reported a variety of populations they were working in for example 3.2% (n = 28) work on psychiatric wards; 11.3% (n = 24) work in acute inpatient care; 29.2 % (n =62) work with the elderly; 10.8

Table 1: Demographic Characteristics

Variable N Mean SD
Age 212 42.7 9.33
Income 212 47,150 2.48*
Education 212 2.44 0.97
Work History 212 18.35 1.37**
*Based on $5,000 increments
**Based on 5 year increment
% (n = 23) work in correctional centers; 34.4% (n = 73) worked in other locations; and
.9% (n = 2) did not identify workplace.

Return to top



Hypothesis One

Results indicate some support for hypothesis one that hypothesized nurses would have a higher level of depression than the general population (Table 2). In the current study, participants were identified as probable candidates for a clinical diagnosis of depression if they obtained a score of 14 or over on the BDI II (Beck et al., 1996). T-test analysis indicated that a higher percentage of participants in this study experienced depression (25%) than has been found in the general population (19.9 %) (Kirsling et al, 1989). Participants in this study, however, had a lower overall group mean on the BDI II (M = 9.47) than a previous normative study of college students (M = 12.56) (Beck, et al, 1996).

Table 2: Means, Standard Deviations, and t Test Results

Event With PTSD Event Without PTSD
Measure M SD M SD t df
BDI-II 14.10 9.36 6.51 6.70 -4.57 95*
DTS 42.65 19.68 8.13 6.22 -11.63 98
Q-LES-Q Work 48.10 7.31 54.52 5.95 4.76 96*
Q-LES-Q General Activities 49.80 11.12 59.21 8.40 4.72 97*
EPI Neuroticism 12.83 5.38 7.30 5.13 -5.13 93*
EPI Psychoticism 1.98 1.32 2.10 2.05 .31 94
EPI Extroversion 12.26 6.19 13.26 5.47 .83 94
EPI Lie 8.10 3.69 8.13 3.82 .03 96
Age 40.31 8.16 42.63 9.05 1.35 98
Education 2.50 .74 2.40 .74 -.60 98
Work History 5.38 1.44 5.67 1.32 1.01 9
* Significant at the p < 0.001 level

Hypothesis Two

Results support hypothesis two in that many psychiatric nurses experienced traumatic events in the workplace and showed symptoms of PTSD (Table 2). One hundred (48%) of the sample of 212 psychiatric nurses reported experiencing a traumatic event or critical incident within the workplace in the last 18 months. Davidson (1996) suggests a cut-off score of 20 on the Davidson Trauma Scale (DTS) as a “flag” to indicate PTSD when it is not known what level of trauma to expect. Of the 100 nurses who had experienced a traumatic event, 52 (52%) reported PTSD symptoms at a level high enough to indicate probable pathology (Table 4).

Hypothesis Three

Statistical analysis supports hypothesis three that quality of life in the work environment would be lower for those suffering from PTSD. The work sub-scale of the Q-LES-Q was utilized to assess quality of life experienced in the workplace. As predicted, a lower mean score on the work sub-scale was found for the PTSD group (M= 48.10, SD = 7.31) compared to the non-PTSD group (M = 54.52, SD = 5.95), t (96 ) = 4.75 , p < .0001 (Table 2).

Hypothesis Four

Hypothesis four, that employee access to critical incident debriefing would be related to PTSD status was supported. Provision of critical incident debriefing by employers was indicated on the workplace-related questionnaire by checking either yes or no. Fifty participants in the PTSD group provided a response to this item. Of those 50 participants, 37 indicated receiving a critical incident debriefing and 13 did not. Forty-five participants in the non-PTSD group provided a response to this questionnaire. Of those 45 participants, 42 reported receiving a critical incident debriefing and 3 did not.

Table 3: Relationship of PTSD to Gender and Workplace Questionnaire in Those Experiencing a Critical Event.

Variables PTSD No PTSD x2
Gender – Male 6 11 2.44
Gender – Female 46 36 2.44
WRQ 4 – Debriefing availability 50 45 6.32*
WRQ 5 – Debriefing adequacy 35 39 7.64**
WRQ 7 – Comfort in accessing EAP 37 32 3.53
* p < .01
**p < .001

chi-square test of these frequencies was significant (x2 (1, N = 95) = 6.32, p < .01) (Table 3). This finding suggests that participants with PTSD received less support in the workplace following critical incidents since their employers did not provide a debriefing (Table 3).

Further support was provided for the fourth hypothesis in that participants with elevated trauma levels did not find the debriefing to be adequate. Thirty-five of the PTSD group responded to item five of the workplace related questionnaire. Of those, 13 found the debriefing to be adequate and 22 did not. Thirty-nine of the non-PTSD group responded to item five. Of those, 27 found the debriefing to be adequate and 12 did not. The relationship between PTSD and debriefing adequacy was significant, (x2 (1, N= 74)= 7.65, p < .005). Therefore, more participants with PTSD indicated not receiving adequate debriefing (Table 3).

Table 4: Gender Differences in PTSD in Psychiatric Nurses For Total Sample and Traumatic Event

Traumatic Event
n = 52 52% n = 48 48%
Females 46 56.1 36 43.9
Males 6 35.3 11 64.7
Total Sample
n = 52 24.5% n = 160 75.5%
Females 46 25.8 32 74.2
Males 6 19.4 31 81.6


Hypothesis Five

Hypothesis five, that neuroticism scores would be different for participants with and without PTSD symptomatology,was supported. Neuroticism was assessed by one sub-scale of the EPI. Higher scores on neuroticism were found for the PTSD group (M = 12.83 , SD = 5.38) compared to the non-PTSD group (M = 7.30, SD = 5.13 ), t (93) = -5.13, p < .001.This result suggests that those experiencing post-traumatic stress also reported increased scores for neuroticism (Table 2).

Hypothesis Six

The sixth hypothesis that female nurses would be over-represented in the PTSD group relative to the non-PTSD group was not supported. Of the 52 participants in the PTSD group, 46 were female and 6 were male. Of the 47 participants in the non-PTSD (one participant in this group did not report their gender), 36 were female and 11 were male. The relationshipbetween gender and PTSD status was not significant, 2 (1, n = 99) = 2.44, p < .12 (Table 3).

There was however, some support for a difference played by the role of gender in the development of PTSD. In the overall sample of 212 nurses, 25.8% of the women (n= 46 of 174) as compared to 19.4% of the men (n= 6 of 31) had symptoms of PTSD (a score of 20 or above on the DTS). When only those women having experienced a traumatic event are considered 56.1% (n = 46 out of 82) had elevated levels of PTSD whereas males indicated a rate of 35.3% (n = 6 out of 17) (Table 4).


Other Findings

Although not hypothesized, depression was found more frequently in the PTSD group ( M = 14.10 SD = 9.36) than in the non PTSD group ( M = 6.51 SD = 6.70), t (95) = -4.57, p < .001 with high levels of PTSD (Table 2). There was also a great deal of difference in the incidence of PTSD in relation to the work environments that psychiatric nurses were employed in. Certain areas of employment were shown to be much more likely than others to have nurses who were suffering from PTSD symptoms. The incident breakdown for PTSD in each area was: 1) psychiatric ward = 35.71; acute inpatient care = 20.83; elderly care = 22.56; not identified = 16.44 and Corrections = 47.83 (Table 5).

Age, education, income and the availability of employee assistance programs as well as whether or not psychiatric nurses were comfortable utilizing them were found not to be significant in relation to levels of PTSD.

Table 5: Incidence of PTSD in Psychiatric Nurses Working With Various Populations.

1 (N = 28) 2 (N = 24) 3 (N = 62) 4 (N = 23) 5 (N = 73)
Diagnosis N % N % N % N % N %
PTSD 10 35.7 5 20.8 14 22.6 11 47.8 12 16.4
No PTSD 18 64.3 19 79.2 48 77.4 12 52.2 61 83.6

Return to top



Overall prevalence of PTSD, at any given time, in US populations is estimated to be between 1 and 12 percent (Lange et al., 2000) while lifetime prevalence is between 8 and 12 percent (Davidson, 1996). Although participants in this study have an elevated level of PTSD (24.53%), that exceeds twice that found in community based samples, other studies have suggested that it is not unusual for certain populations to report significantly higher levels of post traumatic stress disorder.

Results in such studies parallel or exceed those found in the current study. Conclusions indicate that traumatic events in daily life or in the workplace seem to contribute to higher symptom scores on trauma scales. Wagner (1998) using a sample of 402 German professional firefighters found a prevalence rate of 18.2% for PTSD symptoms. Another German study of post traumatic stress disorder and trauma exposure in intensive care nursing staff conducted in 15 intensive care units found 41% of the 144 person sample to have developed clinical levels of PTSD (Teegan & Muller, 2000). Frombach (1997) found that 27% of female adult cancer patients and 10% of adult male cancer patients qualified for a diagnosis of post traumatic stress disorder (Frombach & Hampton, 1999). Clohessy identified 21% of a sample of 56 ambulance service workers as suffering from post traumatic stress in a London based ambulance service (Clohessy & Ehlers, 1999).

Susceptible populations vary in psychopathology and have been shown in samples to have diverse levels of PTSD extending anywhere from 34 % in survivors of motor vehicle accidents, to 48 % in female rape victims, to 67% in prisoners of war and finally, to 20% of men and 35% of women following victimization related to violent crime (Brewin et al., 2000, Seedat & Stein, 2000; Travis-Lange, 2000).

Individuals may encounter critical events during the performance of job duties that lead to the development of PTSD. This is particularly true when work responsibilities include an unusual or severe exposure to violent events not normally encountered within the workplace. The results of the present study suggest that many psychiatric nurses in Saskatchewan fall into this category and, not surprisingly, tend to experience symptoms indicative of PTSD.

Of this sample of 212 psychiatric nurses in Saskatchewan, 100 (47.17%) reported experiencing a traumatic event at work in the last 18 months. Fifty-two percent of these reported substantial levels of psychopathology. These findings suggest that 24.53 % of the 212 respondents, who would be likely to qualify for a diagnosis of PTSD according DSM-IV criteria (American Psychiatric Association, 1994). Many of these also suffered from symptoms of depression and the relationship between PTSD and depression was statistically significant consistent with the findings in other studies.

Clearly, this survey represents only an initial investigation into the incidence of PTSD in psychiatric nurses. It is possible that some of those who did not respond may have qualified for a PTSD diagnosis but, as, as part of the avoidance behavior associated with the disorder may not have been capable of responding without exacerbating symptoms, in which case PTSD, in this sample would be underrepresented. However, since this is the only opportunity that psychiatric nurses in Saskatchewan have had to address psychopathology in the professional environment and to determine employer responsibility PTSD could also be over represented.

This study hypothesized that high levels of PTSD and depression in psychiatric nurses would be significantly related to: 1) a lower quality of life in and out of the workplace; 2) a lack of support in the work environment; 3) gender, with females displaying more PTSD than males; and 4) higher levels of neuroticism within those with PTSD symptoms. Corroboration for these hypotheses was provided by the results of this study. The current findings also agree with previous research which suggested high levels of PTSD in professionals who are employed in occupations where unusual stress and traumatic events are common occurrences. Given that the current sample represents nearly 18% of the population of psychiatric nurses in Saskatchewan and that individuals were randomly selected, the findings could be expected to be indicative of the population as a whole.

PTSD develops following a perceived life threatening trauma in relation to the self or when witnessed happening to another (American Psychiatric Association, 1994). Individuals may display increased susceptibility to developing PTSD through a combination of variables including: previous traumatic experiences, age, gender, trauma severity and type, lack of social support, personality factors and other pre-trauma conditions (Brewin, Andrews, Valentine, 2000; Lowe & Northcott, 1988; Seedat & Stein, 2000).

PTSD has been found to cause sufferers to experience less job satisfaction , greater health related difficulties, and may lead to other psychiatric complications, such as depression. (Alexander & Kline, 2001; Landau & Litwin, 2000; Stein et al., 2000). The impact on the individual and within the workplace can be severe and lead to consequences such as decreased work performance, lost work time, and other dysfunctional behavior including the possibility of physical, social, family and emotional implications (Murray & Snyder, 1991). Debriefing, when properly conducted, may help to reduce abnormal stress reactions and afford the affected individuals an opportunity to share feelings about critical situations and learn effective coping mechanisms. Unfortunately, in workplace related trauma, debriefing is not always available or is inadequate or untimely (Cudmore, 1996; Fitzgerald, et al., 1993).

Results of this study suggest that quality of life in and out of the workplace were negatively affected in nurses who suffer from PTSD symptomatology. In this case, lower Q-LES-Q quality of life scores in individuals with PTSD symptoms was indicative of trauma exposure on the job. This corroborates findings in previous studies that have indicated lower scores on quality of life scales in the presence of psychopathology. The New York Psychiatric Institute used the Q-LES-Q as part of the intake and outcome portions of several studies and concluded that client scores improved as psychopathology were alleviated (Endicott, et al., 1993). On the job, decision making, problem solving and communication with coworkers are affected by PTSD symptomatology. Daily activities are also affected and problems may manifest in physical health, daily functioning, sexual drive and social as well as family relationships.

Of greater concern, from the perspective of prevention, is the finding indicating that the PTSD group reported either no access to or inadequate debriefing processes following acute workplace occurrences. Nurses who did not experience a critical event in the workplace or did not show clinical levels of pathology, reported availability of debriefing programs and / or satisfaction with them. That is to say that individuals who did not require debriefing are satisfied with the available programs, while those most in need fail to receive adequate support.

Personality factors such as neuroticism are a concern in attempting to understand many types of psychopathology. Neuroticism has been repeatedly correlated with both depression and PTSD (Casella & Motta, 1990; Jorm ,et al.,2000). The Saskatchewan sample was interesting in that neuroticism was significantly related to PTSD symptomatology at the DTS cut-off score of 20. However, neuroticism was not significantly related when the cut-off score was raised to 24 or greater. By changing the cut-off score, a probable clinical diagnosis increases from 67.47% at 20 to 79.52 at 24. The question is why is neuroticism correlated with lower levels of PTSD and not with higher levels?

Several questions arises from the finding that up to one half of those qualifying for a diagnosis of PTSD will fall below diagnostic level within a one year period following a traumatic event while one third of those suffering from PTSD will still meet the criteria for the disorder after 10 years. Is it that those who continue to have diagnostic levels of PTSD, over an extended period subsequently begin to show neurotic personality traits or is it that those with neurotic tendencies plateau at a lower symptomatic level because of these tendencies? Also, is it possible that those who experience PTSD and had neurotic tendencies prior to the critical incident are not able to recover for a very long time if ever because of the pre-existing neuroticism? Finally, if those with neurotic personality traits tend not to recover from the disorder as quickly, if at all, why are they not represented at the higher symptom levels? Further studies should examine the relationship between neuroticism and post traumatic stress disorder over time.

Gender was found not to be significant to PTSD levels, contrary to previous research. Although there was a difference in DTS scores between females and males, particularly following exposure to workplace trauma, these did not reach significance. Interestingly, the only other study, which found that gender was not significant, was a military sample cited in Brewins (2000) meta-analysis. Brewin considered previous studies on post traumatic stress disorder in the military. There was no significant difference for gender in the incidence of PTSD. Female combat veterans would be considered as occupying non- traditional work roles as would males employed within the nursing profession. Further studies need to examine PTSD levels in those employed in non-traditional jobs which predispose the worker to high levels of trauma. A qualitative study could further assist in determining whether personality traits or other demographic factors within individuals employed in non-traditional work roles affect their response to trauma.

Finally, the different work environments where psychiatric nurses are employed were found to produce dramatically differing levels of post traumatic stress disorder. In each workplace the incidence of PTSD was as follows: not identified = 16.44% ; psychiatric wards = 35.71%; acute inpatient care = 22.56 %; and Corrections = 47.83 (Table 4). Although all areas of psychiatric nursing in this sample are substantially higher than what is expected in community based samples those who worked in corrections displayed a rate that is 23.33 % higher than the overall nursing sample. Corrections psychiatric nurses work with a complex population where up to 15% of inmates suffer from severe or chronic mental illness. They are actively involved in a cycle of care “… that includes crisis intervention, early identification of problems, acute inpatient psychiatry, and rehabilitation of offenders with long-term mental illness, distinct from other psychiatric nurses (Peternelj-Taylor & Johnson, 1995 p. 15). Violence amongst inmates is common and nurses are often feel the conflict between care and custody. The perpetual tension leads to lack of trust and inmates often turn to nurses as a sounding board for their frustration (Bernier, 1986).

The prison environment has an inherent level of anxiety for inmates, guards and nurses and it is in this environment, more than others, that dependability on employer provided support programs is crucial. There is an obvious failure in this system for the provision of appropriate recourse for employees who are exposed to frequent critical situations. Further studies must examine stress, depression and PTSD levels in psychiatric nurses. Available employee assistance programs must be considered in light of their effectiveness and staff suggestions for viable options should be considered.

In summation, the present study did not find age, income level, number of years of education or gender to be statistically significant in relation to PTSD. The personality trait of neuroticism was higher in the PTSD group than in the no PTSD group. Also it was found that individuals who did not receive employer support following a critical incident in the form of debriefing or adequate debriefing showed higher levels of PTSD symptomatology. The work environment and population psychiatric nurses worked in and with also played a role in the incidence of PTSD.

Return to top



In conclusion, the level of PTSD found in psychiatric nurses needs to be studied further. Qualitative data, in particular, could add substantially to the body of knowledge. The input of the nurses themselves is vital to any future recommendations that are needed to improve their workplace. It is crucial for all employees to have a work environment that enhances their self-esteem, personal goals and dignity. In a work environment where nearly one quarter of the employees are reporting pathological levels of symptomatology indicative of PTSD, it is prudent and necessary for the employer to take an interest in the etiological factors surrounding the disorder.

The consequences related to this unusually high incidence of PTSD symptoms are emotionally, physically and psychologically expensive and are an unnecessary financial encumbrance. It is well known that prevention relieves this financial encumbrance but, it is also well known that prevention has been last on the agenda in mental health care. Perhaps it should not be surprising that those who are so integrally important in providing services for persons with psychiatric disorders are themselves not receiving the consideration they need. Further studies into PTSD in psychiatric nurses within diverse work populations are needed. Qualitative studies could assist in identifying predictors, as well as potential solutions for the problems raised by this study.

Return to top



Alexander, D., Klein, S. (2001). Ambulance personnel and critical incidents: Impact of accident and emergency work on mental health and emotional well being. British Journal of Psychiatry. 1, 76-81.

Allen, D. (1993). Prison nursing: Doing time the hard way. Nursing Standard. 7, 20-22.

American Psychiatric Association. (1994). DSM IV Diagnostic and statistical manual of mental disorders. Washington, D.C.: American Psychiatric Association Beck, A., Steer, R. & Brown, G. (1996). Beck Depression Inventory 2nd Edition. San Antonio, Texas: Harcourt Brace & Company.

Brewin, C. Andrews, B. & Valentine, J. (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology. 68. 748-766.

Casella, L. & Motta, W. (1990). Comparisons of characteristics of Vietnam veterans with and without posttraumatic stress disorder. Psychological Reports.67. 595-607.

Cudmore, J. (1996). Preventing post traumatic stress disorder in accident and emergency nursing. A review of the literature. Nursing in Critical Care. 3. 120-126.

Cooper, C. (2000). Patient suicide and assault: their impact on psychiatric hospital staff. Journal of Psychosocial Nursing, 33, 26-29.

Eyesenck, H. & Eyesenck S. (1996). Eyesenck Personality Questionnaire. San Diego, California: Educational and Industrial Testing Services.

Feldman, P., Cohen, J., Gwaltney, J., Doyle, W. & Skoner, D. (1997). The impact of personality on the reporting of unfounded symptoms of illness. Journal of Personality and Social Psychology.77. 370-379.

Figley, C. (1995) Compassion Fatigue: Coping With Secondary Traumatic Stress Disorder in Those Who Treat the Traumatized. Portland, OR: Book News.

Frombach, I. & Ruckles Hampton, M. (1999). Gender differences in post-traumatic stress disorder in a sample of cancer patients. In: Fernandez-Heber, L. & George, T. (1999): International Perspectives on Women, Health and Culture: A World Wide Anthology. Salisbury, England: Quay Books.

Garyfallos, G., Adamopoulou, A., Moutzoukis, C., Panakleridou, T., Kapslan, T., & Linara, A. (1993). Mental health status of Greek female nurses. Personality and Individual Differences, 15, 199-204.

Jorm, A., Christensen, H., Henderson, A., Jacomb,P., Korten, A. & Rodgers, B. (2000). Predicting anxiety and depression from personality: Is there a synergistic of neuroticism and extraversion? Journal of Abnormal Psychology.109. 145.

Jones, J., Janman, D., & Payne, R. (1987). Some determinants of stress in psychiatric nurses. International Journal of Nursing Studies, 24, 129-144.

Katz, R. (1983). Causes of death among registered nurses. Journal of Occupational Medicine, 25, 760-762.

Kellog, S., & Triffleman, E. (1998). Treating substance-abuse patients with histories of violence: Reactions, perspectives, and interventions. Psychotherapy, 35, 405-413.

King, D., King, L. & Foy, D. (2000). Post-traumatic stress disorder- Care and treatment. Harvard Mental Health Letter. 16 18-19.

Kirsling, R., Kochar, M. & Chan, C. (1989). An evaluation of mood states among first-year residents. Psychological Reports.65. 355-367.

Landau, R. & Litwin H. (2000). The effects of extreme early stress in very old age. Journal of Traumatic Stress.3. 473-487.

Lange, J, Lange, C. & Cabaltica, R. (2000). Primary care treatment of post-traumatic stress disorder. American Family Physician. 62. 1035-1042.

Lowe, G. & Northcott, H. The impact of working conditions, social roles, and personal characteristics on gender differences in distress. In: Lowe, G. & Northcott, H.(1988): Work and Occupations. Thousand Oaks, California: Sage Publications. Malach-Pines, A. (2000). Nurses burnout: An existential psychodynamic perspective. Journal of Psychosocial Nursing, 38, 23-31.

Murray, M., & Snyder, J. (1991). When staff are assaulted. Journal of Psychosocial Nursing, 29, 24-29.

Peternelj-Taylor, C. & Johnson, L. (1995). Serving time: Psychiatric mental health nursing in corrections. Journal of Psychosocial Nursing and Mental Health Services 33. 12-19.

Seedat, S., & Stein, D. (2000). Trauma and post-traumatic stress disorder in women: A review. International Clinical Psychopharmacology. 15 s25-s33.

Stein, M., McQuaid, J., Pedrelli, P., Lenox, R. & McCahill, M. (2000). Posttraumatic stress disorder in the primary care medical setting. General Hospital Psychiatry. 4. 261-269.

Piedmont, R. (1993). A longitudinal analysis ob burnout in the health care settings: the role of personal disposition. Journal of Personality Assessment, 61, 457-473. Samuelsson, M. Gustavsson, J., Petterson, I., Arnetz, B., & Asberg, M. (1997). Suicidal feelings and work environment in psychiatric nursing personnel. Social Psychiatry and Psychiatric Epidemiology, 32, 391-397.

Skinner, K., & Scott, D. (1993). Depression among female registered nurses. Nursing Management, 24, 42-45.

Thomsens, S., Arnetz, B., Nolan, P., Soares, J., & Dallander, J. (1999). Individual and organizational well-being in psychiatric nursing. Journal of Advanced Nursing, 30, 749-757.

Valent, P. (1995). Survival strategies: A framework for understanding secondary traumatic stress and coping in helpers. Portland, OR: Book News.

van Servellen, G., Fopf, M., & Leake, B. (1994). Personality hardiness, work related stress, and health in hospital nurses. Hospital Topics, 72, 34-39.

Weinberg, A., & Creed, F. (2000). Stress and psychiatric disorder in health care professionals and hospital staff. Lancet, 12, 533-537.

Whittington, R. & Wykes, T. (1994). Violence in psychiatric hospitals: Are certain staff prone to be assaulted? Journal of Advanced Nursing, 19, 219-225. Wykes, T., & Whittington, R., Coping strategies used by staff following assault by a patient: an exploratory study. Work and Stress, 5, 37-48.

Return to top