Florida State University Libraries

Florida State University Libraries Electronic Theses, Treatises and Dissertations The Graduate School 2005 The Effect of a Debriefing on Compassion Fatigue Levels in Public Health Nurses after Hurricanes Charley, Frances, Ivan, & Jeanne Julie Ann Sherrod Follow this and additional works at the FSU Digital Library. For more information, please contact lib-ir@fsu.edu

THE FLORIDA STATE UNIVERSITY SCHOOL OF NURSING The Effect of a Debriefing on Compassion Fatigue Levels in Public Health Nurses After Hurricanes Charley, Frances, Ivan, & Jeanne by Julie Ann Sherrod, RN, BSN A Thesis submitted to the School of Nursing In partial fulfillment of the Requirements for the degree of Master of Science in Nursing Degree Awarded: Fall Semester, 2005

The members of the committee approve the thesis of Julie Sherrod defended on October 21, 2005. Deborah Frank Professor Directing Thesis Sandra Faria Committee Member Denise Tucker Committee Member Approved: Linda Sullivan, Director, School of Nursing Graduate Program Katherine P. Mason, Dean, School of Nursing The Office of Graduate Studies has verified and approved the above named committee members. iii

TABLE OF CONTENTS List of Tables List of Figures Abstract vi vii viii Chapter One: Introduction of Study Introduction 1 Statement of Problem 1 Significance of Problem 2 Conceptual Framework 3 Operational Definitions 5 Purpose 6 Hypotheses 6 Chapter Two: Review of Literature Introduction 7 Compassion Fatigue 7 Debriefing 8 Summary 9 Chapter Three: Methodology Introduction 10 Design 10 Procedure 10 Instrumentation 11 Statistical Analysis 11 Chapter Four: Results Introduction 12 Research Questions 13 Summary 15 Chapter Five: Discussion Introduction 16 Review of Findings 16 Literature Review 17 Conceptual Framework 18 Limitations 18 Implications for Practice 19 Future Research 20 Generalizability 20 Summary 21 iv

Appendices 22 References 42 Biographical Sketch 44 v

LIST OF TABLES Table 1 Means and standard deviations for the week of hurricanes 34 Table 2 Means and standard deviations after the hurricanes 35 vi

LIST OF FIGURES Figure 1. Two way interaction STS X Debriefing 36 Figure 2. Two way interaction Burnout X Debriefing 37 Figure 3. Two way interaction Compassion Fatigue 38 Figure 4. Main effect for STS 39 Figure 5. Main effect for Burnout 40 Figure 6. Main effect for Compassion Fatigue 41 vii

ABSTRACT The purpose of this study was to examine the effect a debriefing had on Compassion Fatigue levels as measured by the Compassion Fatigue Self Test (CFST) in public health nurses employed by the Florida Department of Health who provided care to the victims of the hurricanes that occurred in the fall of 2004. 500 surveys were mailed to these nurses. While 120 returned the survey, 89 completed the entire survey and thus, were the sample for this study. This study also examined the effect the debriefing had on the CFST subscales of STS and burnout. A 2X2 Factorial ANOVA was used to analyze the data generated from the sample. This revealed there was no significant change in the CFST scores related to the debriefing. There was, however, a decrease in Compassion Fatigue scores overall and in the subscale of STS from the week of the hurricanes to the time the test was administered post hurricanes, regardless of participation in debriefing. Implications for nursing practice and research are presented. viii

CHAPTER 1 INTRODUCTION Introduction In the fall of 2004 four hurricanes hit the coast of Florida. It was a challenging time for all Floridians. It was the largest relief effort ever undertaken by FEMA. The nurses employed by Department of Health deployed for relief efforts found shell-shocked victims when they reached their posts. The areas hit by the hurricanes were out of water and electricity on an average of 14 days, leaving plumbing and waste problems, contaminated water and dehydrated, ailing victims. Besides the utility issues, the nurses also had to contend with patients not having access to their usual medications or even medication lists (Blaney-Brouse, 2005). The observation of the effects of trauma on the human psyche is not new; in fact it has existed for centuries and been told through historical text and interpreted in art. It can be dated back thousands of years as it was described in a Kunus Papyrus found in Egypt dated 1900 B.C. (Figley, 1995). The study of this phenomenon gained much attention in the 20 th Century due to American involvement in four major wars during that century. More recently, increasing media attention on mass casualty tragedies such as the Columbine shootings and the September 11 th, 2001 attacks and natural disasters such as earthquakes and the recent devastating hurricane season, the study of trauma, its consequences, and treatment have again become important and relevant for the general population. The scientific community has also become aware that stress related to exposure to trauma not only affects veterans of war, but can affect any human being who has experienced a traumatic event and can even affect those who provide care to victims of trauma in the form of Compassion Fatigue. The following is a description of a proposed research study that will explore Compassion Fatigue levels in the Public Health Nurses who were deployed to the areas in Florida most devastated by the hurricanes to provide health care services to the victims of the hurricanes. The proposed study will examine what effect, if any, a debriefing intervention, provided to the nurses after deployment, had on their levels of Compassion Fatigue. Statement of Problem During the fall of 2004, four hurricanes hit the coast of Florida. Hundreds of nurses employed by the Department of Health in the state of Florida were deployed to care for the victims of the hurricanes. The nurses worked shifts that extended up to 24 hours and many were deployed up to four times providing care during all four of the hurricanes. Compassion Fatigue 1

is described by Figley (1995) as a natural consequence of caring for traumatized people. Compassion Fatigue develops when a person is secondarily exposed to trauma by providing direct care to a person who has been traumatized. Compassion Fatigue has many sequelae that can interfere with job performance and job satisfaction. After each deployment the nurses were offered a non-mandatory debriefing session. Debriefings are commonly employed after such events, but have recently been under scrutiny as an effective treatment for trauma exposed individuals. This study identified public health nurses that were deployed in the hurricane season of 2004, who are at risk for Compassion Fatigue. The study examined the effects of Compassion Fatigue, such as signs and symptoms of Secondary Traumatic Stress and Burnout, and whether or not a debriefing intervention reduced the risk for Compassion Fatigue or the signs and symptoms of Secondary Traumatic Stress and Burnout. Significance of Problem Compassion Fatigue, a form of traumatic stress, is a psychological consequence of providing care for victims of traumatic events, such as natural disasters. Symptoms of Compassion Fatigue include insomnia, gastrointestinal disturbances, fatigue, psychogenic amnesia, decreased concentration, impaired judgment and reasoning, thought disturbances and can result in anxiety disorder, depression and substance abuse. Untreated, Compassion Fatigue can lead to difficulty in one s personal and professional life, such as low self-esteem, decreased job satisfaction, and poor work performance. Indirectly health care organizations can be effected by decreased retention of employees and decreased patient satisfaction related to poor work performance (Figley, 1995). Currently, Compassion Fatigue is not recognized as a working diagnosis in the Diagnostic and Statistical Manual (DSM), therefore the diagnosing criteria and treatment modalities for the disorder are the same as Post Traumatic Stress Disorder (PTSD). Although the two disorders have similar symptomology, they occur under different circumstances (Figley, 1995). The most common treatment used to initially treat PTSD and Compassion Fatigue is Psychological Debriefing (Bisson, McFarlane, & Rose, 2000). There are several types of Psychological Debriefing, some more structured than others, but they all follow the same basic format. All Psychological Debriefings are a group intervention, all contain a venting phase in which the participants are allowed to share thoughts and feelings related to the traumatic event, and all contain an information phase in which the participants are given information on stress reactions and coping mechanisms (McCammon & Allison, 1995). 2

In recent years, the efficacy of psychological debriefing has been questioned. Dichotomous research findings have made it unclear how successful Psychological Debriefings are in treating either PTSD or Compassion Fatigue. Some of the research even goes so far as to suggest the debriefings may be harmful to the psyche and impede the recovery process. While other research, which supports the debriefing process, suggest that the timing, length, and repetition of the debriefing process may be the keys to its success. With all of the conflicting information on the use of Psychological Debriefing in circulation and the naysayers not providing suggestions for alternatives, it is difficult to know how to treat Compassion Fatigue. Research supports the fact that without treatment, symptoms of Compassion Fatigue will worsen (Maytum, Bielski, Heiman, & Garwick, 2004). With worsening symptoms and continued exposure to physical, emotional, and psychological stressors in the work environment, care providers experience Burnout, a related phenomenon that is a part of Compassion Fatigue, but can also exist independently of Compassion Fatigue. Burnout in nurses has been associated with unhappiness, job dissatisfaction, stress and decreased staff retention, which in turn exacerbates the nursing shortage (Department for Professional Employees, 2003). Conceptual Framework Compassion Fatigue Figley s Theory of Compassion Fatigue was developed by Charles Figley (1995) and evolved from his work in Traumatology. The theory began as STS; a type of vicarious traumatization experienced by and originally recognized in mental health counselors who, through counseling and therapy, treated victims of traumatic events. Figley noted, through research, that these care providers were experiencing symptoms similar to those of PTSD and proposed a revision of PTSD to include those that were vicariously traumatized. Figley identifies four classifications of trauma, including 1) simultaneous trauma, trauma that takes place when all members of a system are directly affected at the same time, such as that of natural disaster; 2) vicarious trauma, trauma that occurs when a single member of a system is affected out of contact with other members; 3) intrafamilial trauma, trauma that occurs when a member of a system inflicts harm upon another member of the same system, such as in abuse; and 4) chiasmal trauma, trauma that is secondary in nature and appears to infect an entire system when originally appearing to affect a single member of the system (Figley, 1995). Figley differentiates STS from countertransference (the process by which counselors overidentify with a client or meet their needs through the client), an unnatural and neurotic reaction, by identifying STS as a natural consequence of caring for traumatized people. In 3

addition, Figley noted there also were similarities to Burnout in those who were vicariously exposed to trauma including chronicity of the disorder and the occurrence of symptoms such as depersonalization and sense of helplessness. This created an evolution of the original theory of STS to include Burnout and thus the theory became identified as Compassion Fatigue (Figley, 1995). Burnout Maslach developed burnout theory in 1981. Burnout is a three-part syndrome that consists of depersonalization, emotional exhaustion, and a low sense of personal accomplishment. Emotional Exhaustion is described as loss of interest in professional life and fatigue and weakness. Depersonalization is defined as behavior toward a patient without emotion, as if the patient is not a unique individual. Personal accomplishment is described as feelings of productivity and adequacy in one s professional life (Maslach, 1982). Burnout, as opposed to general stress and depression, may affect a person only in their work life and not generally. Burnout in nurses has been noted to cause anger, short attention span, physical, emotional, and intellectual exhaustion, loss of interest in work, and lack of concern for patients (Sherman, 2004). Debriefing The process of Psychological Debriefing is used worldwide as an initial treatment for those exposed to trauma both directly and indirectly. Mitchell (1983) was the first to conceptualize the debriefing process and provided structure to the process, which has been used and modified by many professionals (Dyregov, 1997). Therefore the debriefing process known as Critical Incident Stress Debriefing (CISD) will be used as a theoretical framework for the purposes of this study. CISD is described as a form of crisis intervention as opposed to psychiatric treatment and is intended to be a part of a series of interventions used to treat traumatic stress and not as a stand-alone treatment for traumatic stress. CISD was designed to decrease the initial distress and prevent future sequelae of traumatic stress by initiating and promoting emotional processing. CISD allows participants to ventilate and normalize reactions to traumatic stress and allows participants to prepare for future reactions and emotional experiences. CISD also provides early support in a group environment. The intervention focuses on present actions and emotions and discourages expressions of previous experiences that might have shaped individual reactions to the current incident (Bisson, McFarlane, & Rose, 2000). 4

The basic structure of the CISD is a seven-phase group process, lead by a person trained in debriefing. The first phase of the process is the introduction in which the leader sets ground rules. In this phase the tone is set for the rest of the intervention. The second phase is the fact phase. In this phase the leader reviews the facts of the event with the group. The third phase is the thought phase in which the participants are allowed to share thoughts about the event and discuss decisions. The reaction and symptom phase is the fourth phase in the debriefing process. In this phase the participants relate emotional reaction and share sensory impressions of the images of the event. The fifth phase is normalizing, this is the phase in which normalization occurs. In the sixth phase, the teaching phase, the participants are given verbal and written information on the recovery process of traumatic stress and coping strategies for dealing with future intrusive material and emotions. The last phase of the process is called disengagement. The participants are given access for future help, if needed and the leader concludes and closes the debriefing session (Dyregrov, 1997). The debriefing process can take anywhere from one to three hours and is generally held immediately following the traumatic incident, however can take place up to ten days later. The type of group being debriefed can have an impact on the proceedings as well as the strength of the leadership skills of the group leader and the extent of their training (Dyregrov, 1997). Operational Definitions Compassion Fatigue: a psychological phenomenon with rapid onset of symptoms that occurs when a human being provides care to another human being who has been traumatized and is comprised of two separate, but related phenomena, Burnout and STS. For the purposes of this study Compassion Fatigue is measured by the CFST. Compassion Fatigue Self Test: The Compassion Fatigue Self Test (CFST) is a 30 question survey scored on a 10-point Lickert-type scale with 1 being interpreted as rarely/never and 10 being interpreted as very often. The CFST is broken into subscales with items 1-22 representing Secondary Traumatic Stress and 23-30 representing Burnout. Secondary Traumatic Stress: natural consequent behaviors and emotions that are the result of knowing about a traumatic event experienced by a person and the stress resulting from helping or wanting to help that person with rapid onset of symptoms. For the purposes of this study Secondary Traumatic Stress will be measured by the STS subscale on the CFST. Burnout: a chronic state of emotional, physical, and mental exhaustion that intensifies over time and is caused by long term involvement in emotionally demanding situations that has 5

psychosomatic symptoms and may effect one s personal and professional life. For the purposes of this study Burnout will be measured by the Burnout subscale on the CFST. Debriefing: a single session, structured intervention used to treat victims of traumatic events or those providing care for victims of a traumatic event in which participants are allowed to share feelings and thoughts on the traumatic event and are given information regarding stress reaction and stress coping. For the purpose of this study debriefing is defined as a single session, voluntary intervention used to treat the nurses who were deployed to work during the hurricanes that occurred in Florida during the fall of 2004. Public Health Nurse: a Registered Nurse or Advanced Registered Nurse Practitioner employed by the State of Florida Department of Health who provided care to victims of hurricanes Charley, Frances, Ivan, and/or Jeanne. Purpose The purpose of this study is to examine the effect a debriefing had on the Public Health Nurses that were deployed during the hurricane season of 2004 to provide care to the victims of hurricanes Charley, Frances, Ivan, and Jeanne. The Compassion Fatigue levels, as scored on the Compassion Fatigue Self Test (CFST), of the nurses who participated in the debriefing were compared with the Compassion Fatigue levels of the nurses who did not participate in the debriefing. The CFST also contains two subscales, Burnout and Secondary Traumatic Stress (STS), which were also examined to determine any effect the debriefing may have had on the individual scores of either subscale. Hypotheses Based upon the research questions does a debriefing affect Compassion Fatigue levels and are the subscales of STS and Burnout on the CFST affected by a debriefing, the following hypotheses were developed: 1. Compassion Fatigue levels, as scored on the Compassion Fatigue Self Test, will not be affected by a debriefing intervention. 2. The subscale levels on the Compassion Fatigue Self Test of Secondary Traumatic Stress and Burnout will not be affected by a debriefing intervention. 6

CHAPTER 2 REVIEW OF LITERATURE Introduction Numerous research studies on Compassion Fatigue and the related concepts of STS and Burnout have been conducted, however there are few studies on Compassion Fatigue in nurses. Pertinent studies that discuss the effects of Compassion Fatigue and the effects Psychological Debriefings may have been reviewed by the researcher and will be discussed in this section. Compassion Fatigue In an attempt to describe triggers and coping mechanisms for Compassion Fatigue, a qualitative study conducted in Minnesota involving 20 pediatric nurses identified work related and personal triggers for Compassion Fatigue and Burnout, as well as coping mechanisms the nurses utilized for dealing with Compassion Fatigue and Burnout. The researchers found work overload, lack of organizational support and system issues as common triggers, as well as the witnessing of painful procedures, feeling sad about the condition of their patients, and being the receiver of parents verbalization of feeling as work related triggers. Personal triggers common to the group included over involvement with patients, unrealistic expectations of self and family crises. Methods of coping included formally or informally debriefing, supportive work colleague relationships, and the development of a planned approach to Compassion Fatigue (Maytum, Heiman, & Garwick, 2004). A study of 233 nurses working in two hospitals in Turkey sought to identify factors influencing Burnout. The study related certain factors to the subscale levels of Emotional Exhaustion (EE) and Depersonalization (DP). EE levels were found to be higher in nurses who worked night shifts, even if only occasionally. EE levels were also associated with shortage of staff and sanitation of facilities. DP and EE levels were influenced by lack of rest facilities in the hospitals, personal health problems, and poor relationships with supervisors (Demir, Ulusoy, & Ulusoy, 2003). In a study of 437 ambulance workers researchers sought to answer two questions. The first was to identify what defense mechanisms the participants used to control emotional reactions to stress and trauma utilized and the second was to measure the consequences of using emotional control as a coping mechanism. The researchers found that the inability of the participants to articulate emotions was positively correlated with higher levels of traumatic stress 7

and that suppression or control of emotion was the largest predictor of high stress scores (Watsell, 2002). In a study conducted about a year after the September 11 th, 2001 disaster, 403 clergy and mental health workers were surveyed to examine Compassion Fatigue levels. The participants were divided into groups; the first group only worked for the American Red Cross, who regularly debriefed the workers, one group worked for the American Red Cross and other organizations, and the third group did not work for the American Red Cross at all. The researchers reported the participants who only worked for the American Red Cross had the lowest levels of Compassion Fatigue, suggesting that regular debriefing lowers compassion fatigue levels (Roberts, Flannelly, Weaver, & Figley, 2003). However, the reader does not know if the other groups were or were not debriefed by the organizations in which they worked. In the literature reviewed, much was said and implied relating lack of support to high levels of Compassion Fatigue and Burnout. There was one study that was reviewed that attempted to directly relate social support to levels of STS. The study was conducted in Israel and the participants were employed by the police department. The researchers actually found that higher levels of STS were positively correlated with higher perceived social support (Hyman, 2004). Psychological Debriefing This section is a discussion of the literature reviewed on debriefing process. There was a wide variety of studies available both supporting and not supporting debriefing interventions as an early intervention for traumatic stress. As with the available literature on Compassion Fatigue, there is little literature available on the use of debriefing with nurses. Again, the most pertinent studies were selected for the literature review. Regehr (2001) suggested the reason debriefings work in some situations is the perceived organizational support. The author stated crisis debriefing could be an indicator that the organization supports its workers by purchasing mental health services (p.92). The author also stated that many studies on the efficacy of debriefing that supported its usefulness relied upon anecdotal evidence, that is to say they asked the participants if they felt the debriefing helped. When studies on similar populations used a tool to measure STS levels before and after debriefing, levels of STS either did not change or they worsened (Regehr, 2001). Van Emmerick, Kamphius, Hulsbosch, and Emmelkamp (2002) sought to assess the efficacy of single session debriefings. The study was a meta-analysis, which after imposing several restrictions on the studies to be included; the researchers narrowed the meta-analysis to 8

seven studies. The interventions that were evaluated in the study were divided in to two groups, CISD and non-cisd. The study found that the non- CISD group had a greater reduction in posttraumatic stress symptoms than did the CISD group. The authors suggested that CISD might interrupt the natural psychological process of recovery from exposure to trauma (Van Emmerick, Kamphius, Hulsbosch, & Emmelkamp, 2002). Another study linked the timing of debriefing with its success. The study was of a quasiexperimental, cross-sectional research design. In the study, non-violent robbery victims were randomly assigned to two groups. One group received the debriefing within 10 hours of the robbery; the other group received debriefing 48 hours after the robbery. The participants were reevaluated at two days, four days, and at two weeks. The researchers found that at the time of debriefing there was no statistically significant difference in the levels of posttraumatic stress. However, they did find that at each time interval there was a statistically significant difference in posttraumatic stress levels in the participants (Campfield & Hills, 2001). One study, conducted with British soldiers returning from Bosnia, found while debriefings may not have reduced the actual posttraumatic stress levels, it did reduce one very important symptom. As discussed previously, alcohol and drug abuse are common sequelae of posttraumatic stress. This study found that there was a reduction in CAGE questionnaire scores after the debriefing. This study suggests that, while debriefings may not effectively reduce levels of traumatic stress, it can reduce the risk of important after effects of being traumatized, such as alcohol and drug abuse (Deahl, Srinivasan, Jones, Neblett, & Jolly, 2001). Summary From all of the literature reviewed, it was very difficult to find consensus on the efficacy of debriefings as way to begin the process of recovery from traumatization. There is much disagreement in the mental health community about why or how debriefings work when they do. In fact the current available literature suggests everything from the helpfulness of debriefings to the potential harm they may cause. This conflicting information places importance on research, such as the proposed study, to guide professionals and organizations toward the safest and most effective means of treating traumatized individuals. 9

CHAPTER 3 METHODOLOGY Introduction This section will review the design, procedure, and instruments that will be used in the conduction of the current research. It will also present the proposed plan for analysis of the data and rationale for choosing the analyses. Design This research study is a quantitative, non-experimental, retrospective study of a previously existing set of data. Procedure Permission was obtained from the original researchers to use the data from the primary study for analysis in this study. Approval and permission for the study and data analysis was obtained by the Florida State University Institutional Review Board (FSU IRB) prior to data analysis. Collection of the data in the original study was as follows: Permission was obtained from the FSU IRB and the Florida Department of Health to conduct the research. The names and addresses of 500 nurses identified by the Florida Department of Health as having been likely to be deployed to care for the victims of hurricanes Charley, Frances, Ivan and Jeanne were provided to the original researchers. Packets including a letter of consent, a letter containing the description and purpose of the study, a demographic data sheet, and two CFSTs (one to measure Compassion Fatigue levels at the time of the hurricanes and one to measure Compassion Fatigue levels at the time of receipt of the packet, approximately three to four months after deployment) were coded and sent to the 500 nurses identified. The surveys were sent approximately 3-4 months after the hurricane deployment and debriefing. The nurses were asked not to include any identifying information and were provided with means to obtain results from the study while still maintaining confidentiality. Data were entered into an excel database protected by password, with only the primary researchers having access to the password and database. The secondary researcher only had access to the data that pertained to the proposed research questions. The data utilized for the analysis of the proposed secondary research was available to the researcher without any identifying information. The data was sent to a 10

statistician for entry into SPSS and statistical analysis. The data obtained by the researcher for this study will be kept in a locked file for five years after the study is complete. Instrumentation The instrument used in the study is the Compassion Fatigue Self Test (CFST) developed by Figley. The CFST is a 30-question survey with a 10-point Likert-type scaling with one interpreted as never and 10 interpreted as very often. The CFST contains two subscales with items one through 22 measuring STS and items 23-30 measuring Burnout. The CFST is scored by totaling the points given to each item. The scores are interpreted as follows: <94= low risk for Compassion Fatigue, 95-128= some risk of Compassion Fatigue, 128-172= moderate risk for Compassion Fatigue, and >173= high risk for Compassion Fatigue (Appendix A). The CFST was developed in 1993 and has been used in numerous studies to measure Compassion Fatigue. In a recent validational study the CFST as a whole was found to have a Cronbach s alpha of 0.90 and the subscales of STS and Burnout had Cronbach s alphas of 0.84 and 0.83, respectively (Jenkins & Baird, 2002). Statistical Analyses The statistical analyses used for this study were descriptive statistics, including frequency distributions, means, and standard deviations for demographic data. A 2x2 Factorial ANOVA was performed to explore the interactional effect the debriefing had on levels of Compassion Fatigue, STS, and burnout. 11

CHAPTER 4 RESULTS Introduction This chapter is a summary of the results of the study. It includes a description of the sample and descriptive statistics for the demographic data. A 2x2 Factorial ANOVA was used to analyze the effect and interaction the debriefing intervention had on levels of Compassion Fatigue, STS, and burnout. Descriptions of the results of the ANOVA are also included in this chapter. Sample The sample used for this study was taken from an existing data set. The original study produced 119 subjects. For the purposes of this study 89 subjects were provided complete data, these subjects answered all questions regarding the debriefing and having completed both CFST surveys; one for the week of the hurricanes and one for the week they completed the survey. The participants were all employed by the Florida Department of Health and were deployed to work after the hurricanes in the fall of 2004. Of the valid participants there were 86(96.6%) females and 3 (3.4%) males. Eighty (89.9%) of the participants indicated they were non-hispanic white, four (4.5%) indicated they were African American, two (2.2%) indicated they were Hispanic, and three (3.4%) indicated other as their ethnic origin. The participants ranged in age from 24-65. The mean age was 51 (SD =7.9). Of the participants 67 (75.3%) of the participants reported they were married, 13 (14.6%) were divorced; six (6.7%) reported they were single, while three (3.4%) of the participants reported other as their marital status. The years of nursing experience of the nurses who participated in the study ranged from 2 to 41 years. The mean years of experience was 24.9 (SD =.45). Years of experience in public health nursing ranged from one year to 34 years. The mean for years of public health experience was 10.8 (SD =8.6). Eighty-two (92.1%) of the participants were Registered Nurses (RNs) and 7 (7.9%) were Advanced Registered Nurse Practitioners (ARNPs). Of the nurses who participated, 55 (61.8%) had never had experience working in hurricane relief efforts, while 33 (37.1%) had experience working in hurricane relief efforts, one participant did not report hurricane experience. Twenty-six (29.2%) of the participants had 12

provided care in other natural disasters besides hurricanes and 63 (70.8%) had never provided care during any other type of natural disaster. The nurses who participated in the survey reported being deployed up to five times during the hurricane season of 2004. Thiry-three (37.1%) were deployed only once, 12 (13.5%) were deployed twice, 17 (19.1%) were deployed three times, 12 (13.5%) were deployed four times, and 3 (3.4%) of the participants were deployed five times. Of the 89 nurses who completed the questions regarding debriefing 43 (48.3%) reported participating in the debriefing and 46 (51.7%) reported that they did not participate in any debriefing. Of the nurses who participated in the debriefing, 32 (74.4%) stated the debriefing was helpful, while 8 (18.6%) stated it was not helpful and 3 (6.9%) reported no answer to the question. All but one nurse (n=44, 98%) indicated the debriefing occurred over one month ago, one nurse (2%) indicated debriefing had occurred within the last month. Of the nurses who participated in the debriefing 32 (74.4%) stated they found the debriefing helpful, while 8 (18.6%) did not find it helpful. Three (6.9%) did not submit a response to the question. Research Questions To answer the research questions, Does a debriefing affect compassion fatigue levels? and Are the subscales of STS and burnout affected by a debriefing?, a 2x2 Factorial ANOVA was performed with one within subjects variable (compassion fatigue) and one between subjects variable (debriefing). Mean scores for the entire CFST and for the subscales of STS and burnout were also calculated for the sample. Tables and graphs for the following data can be found in Appendix D. Compassion Fatigue Research question 1 investigated the effectiveness of a debriefing on compassion fatigue levels as measured by the CFST. The mean scores for the CFST for the week of the hurricane on all participants, regardless of participation in debriefing was 75.4 (SD=41.9), which falls in the low risk category, with the lowest score at 30 and the highest score at 208. Mean scores for Compassion Fatigue in the group that was debriefed were 74.1 (SD=40.1) and in the group that was not debriefed were 76.6 (SD=43.8). Both of these mean scores fall in the low risk category (Appendix D, table 1). Scores for CFST post hurricane were 62.79 (SD=37.9), which fall in the low risk category, with the lowest score at 30 and the highest score at 205. Means scores the group that was debriefed were 59.5 (SD=36.7) and were 65.8 (SD=39.2) for the group that was not debriefed. Both of these scores also fall in the low risk category (Appendix D, table 2). 13

The CFST is divided into categories or levels of Compassion Fatigue as follows: 94 or less is low risk for Compassion Fatigue, 95-128 is some risk for Compassion Fatigue, 129-172 is moderate risk for Compassion Fatigue, and a score of 173 or greater is high risk for Compassion Fatigue. The overall scores for both groups were spread across the levels of Compassion Fatigue for the week of the hurricanes were as follows: 65 (73%) participants were low risk, 14 (15.7%) were some risk, 7 (7.9%) were moderate risk, and 3 (3.4%) were high risk. Scores for the group of participants that were debriefed for the week of the hurricanes were as follows: 31(72%) were low risk, 8 (18.6%) were some risk, 3 (6.9%) were moderate risk, and 1 (2.3%) was high risk. Scores for the week of the hurricanes in the group that was not debriefed were as follows: 34 (73.9%) were low risk, 6 (13%) were some risk, 4 (8.7%) were moderate risk, and 2 (4.3%) were high risk. Overall scores for both groups post hurricane were: 74 (83.1%)were low risk, 7 (7.9%)were some risk, 6 (6.7%)were moderate risk, and 2 (2.2%) were high risk. In the group that was debriefed the scores were as follows: 37 (86%) were low risk, 2 (4.7%) were some risk, 3 (7%) were moderate risk, and 1 (2.3%) was high risk. In the group that was not debriefed scores were as follows: 37 (80.4%) were low risk, 5 (10.9%) were some risk, 3 (6.5%) were moderate risk, and 1 (2.2%) was high risk. Results from the 2x2 Factorial ANOVA found that the two-way interaction (CF x Debriefing) was not significant [F(1,87)=.56, p=.456] (Appendix D, figure 3). Investigation of the main effects found that the main effect of compassion fatigue was significant [F(1,87)=23.24, p<.05], in that levels of compassion fatigue were significantly higher during the week of the hurricane deployment than they were post hurricane (Appendix D, figure 6). The main effect of debriefing was not significant [F(1,87)=.307, p=.581]. STS and Burnout Research question two examined the effect, if any, the debriefing intervention had on the subscales of STS and burnout on the CFST. The mean score of the STS subscale for the week of the hurricanes, regardless of participation in debriefing, was 55.6 (SD=29.3) with the lowest score being 22 and the highest score being 149. The mean score of the STS subscale for the past week, regardless of participation in debriefing, was 44.8 (SD=25.5) with the lowest score being 22 and the highest score being 132. Results from the ANOVA found that the two-way interaction (STS x Debriefing) was not significant [F(1,87)=.609, p=.437] (Appendix D, figure 1). Investigation of the main effects found that the main effect of STS was significant [F(1,87)=30.49, p=<.05] (Appendix D, figure 4), in that levels of STS were significantly higher during the week of the hurricane than they were in the past week. The mean score for burnout 14

for the week of the hurricane, regardless of participation in debriefing, was 19.8 (SD=15.1) with the lowest score being 8 and the highest score being 73. The mean score for burnout during the past week, regardless of participation in debriefing, was 18.0 (SD=14.1) with 8 being the lowest score and the highest score being 73. The results from the AVOVA found that the two-way interaction (Burnout x Debriefing) was not significant [F(1,87)=.178, p=.674] (Appendix D, figure 2). Investigation of the main effects found that burnout was not significant [F(1,87)=3.096, p=.082] in that levels of burnout were not significantly higher during the week of the hurricane than they were in the past week (Appendix D, figure 5). Summary In regard to research question 1, the results of the statistical analysis of this data found that overall levels of compassion fatigue, as measured by the CFST, were higher during the week of the hurricane as compared to the past week, or the time at which the nurses completed the survey, regardless of participation in debriefing. That is to say that there was no significant difference in the change in scores between the nurses who participated in debriefing and those who did not participate in debriefing. The second research question examined the effect of a debriefing on the CFST subscales of STS and burnout. STS levels were found to be significantly higher in the week of the hurricane than they were during the past week regardless of participation in debriefing, however there was no significant difference in the change in scores between the two groups of nurses, those who did participate in debriefing and those who did not. Burnout scores were neither significantly higher during the week of the hurricane as compared to the past week, nor were the burnout scores significantly different between those participating in the debriefing and those who did not. A detailed discussion of the findings of this study is presented in chapter 5. The following discussion will relate the findings to the theoretical framework of this study and to current literature. Implications for future research will also be discussed. 15

CHAPTER 5 DISCUSSION Introduction Chapter 5 is a discussion of the findings of the research. This chapter compares the findings of this study to those of current research and discusses the findings in the context of theoretical framework used to support this study. A discussion of the limitations of the study, implications for practice and suggestions for further research are discussed. Review of Findings The analyses of the data found that most of the participants, regardless of participation in debriefing, had scores in the low risk categories for Compassion Fatigue, both during the week of the hurricanes and during the past week or time of completion of the survey. Sixty-five participants scored in the low risk category for the week of the hurricanes, while 74 participants scored in the low risk category in the past week time frame. There was not much difference between the two groups in the frequency distributions among the various risk levels (low, some, moderate, and high risk) of Compassion Fatigue. In essence, the scores were relatively low, suggesting both group experienced Compassion Stress, but not Compassion Fatigue (for tables see Appendix D). For research question 1, the findings were that overall levels of risk for compassion fatigue, as measured by the CFST, were higher during the week of the hurricane as compared to the past week, or the time at which the nurses completed the survey, regardless of participation in debriefing. That is to say, there was no significant difference in the change in scores between the nurses who participated in debriefing and those who did not participate in debriefing. Although there was no statistical difference in the two groups, 32 (74.4%) the 43 nurses who did participate in debriefing responded in the comments section that they felt the debriefing was helpful. The second research question examined the effect of a debriefing on the CFST subscales of STS and burnout. STS levels were found to be significantly higher in the week of the hurricane than they were during the past week regardless of participation in debriefing. 16

However, there was no significant difference in the change of burnout scores between the two groups of nurses, those who did participate in debriefing and those who did not, at either time. Burnout scores were neither significantly higher during the week of the hurricane as compared to the past week, nor were the burnout scores significantly different between those participating in the debriefing and those who did not at either time. Literature Review The literature used to guide this study was mixed in its findings of the usefulness and helpfulness of debriefing. Though a few of the studies reviewed found debriefing beneficial, most of the literature reviewed did not find debriefing an effective tool in reducing stress levels or symptoms of traumatic stress. Van Emmerick, Kamphius, Hulsbosch, and Emmelkamp (2002) found in a meta-analysis that patients who did not participate in debriefing after traumatic incident actually had a greater reduction in traumatic stress symptoms. Similarly, in another study, Deahl, et al (2001) found that post-traumatic stress levels were not decreased after a debriefing. However, the study did find that there was a reduction in other symptoms and in CAGE questionnaire scores after the debriefing. In the current study, though scores for Compassion Fatigue and STS were lower in both groups for the past week, no significant difference was found between the two groups on scores on the CFST, indicating the debriefing had no effect on the decrease in scores. Regehr (2001) found that anecdotal evidence was the main support for the use of debriefing. That is to say, participants were asked if the debriefing was helpful and their responses indicated that the participants did find debriefing helpful. Likewise, in the current study, respondents were asked if they felt the debriefing was helpful and the majority of the nurses indicated it was. However, the data indicates that the debriefing did not have a statistically significant effect on the CFST scores. In contrast to the findings of this study, there were two studies that found that a debriefing was helpful in reducing Compassion Fatigue levels. In a study done after the World Trade Center bombings, the researchers found that the relief workers who were regularly debriefed had lower Compassion Fatigue levels than did the workers who were not regularly debriefed (Roberts, Flannelly, Weaver, & Figley, 2003). This finding suggested multiple interventions may be needed for relief workers whose work is continued over time, such as the relief work of the public health nurses deployed last fall. And, in a study that evaluated the timing of debriefing, Campfield and Hills (2001) found that debriefing victims within 10 hours 17

of the event was more effective than a later debriefing, which occurred 48 hours the event. The study found no difference in scores at the time of debriefing, but found that scores were lower in the 10-hour group than in the 48-hour group. Both groups had lower scores in general. This study suggests that timing of the debriefing can play a role in how effective the debriefing is. Conceptual Framework Compassion Fatigue Compassion Fatigue is composed of two components, STS and burnout. In describing the development of the theory Figley (1995) stated STS was the basis for the theory. Burnout was added to Compassion Fatigue secondarily because of a similarity in the chronicity of the disorders and in symptomology. In the analysis of the data for this study it was found that overall Compassion Fatigue scores and the subscale scores for STS were significantly lower during the past week time period as compared to the week of the hurricanes. Burnout scores did not change significantly. This could suggest that the event caused some stress to the nurses, however the stress did not persist. Thus, it did not lead to more chronic conditions, such as Compassion Fatigue and even Burnout. Further support for this indication is that the majority of the nurses relatively low on the CFST, indicating relatively low levels of stress that diminished over time. Debriefing In debriefing theory, Mitchell (1983) described debriefing as a series of interventions used to treat traumatic stress, a crisis intervention and not a stand-alone treatment. Debriefing allows the participants to normalize stress reactions and provides support. A CSID, as proposed by Mitchell (1983) should contain seven phases and should provide resources for future mental health needs. In the comments section of the survey, the nurses indicated that the debriefings did not all follow this format. In fact many of the comments indicated the debriefing was a performance improvement type of activity. One nurse commented, Good opportunity to relate difficulties of special needs shelter, but wasn t helpful in terms of personal benefit. Another similar comment was, Helpful suggestions for future disaster planning and to give positive, helpful suggestions how to improve utilization of volunteers and improve communication between specialty teams out in the field. One nurse commented that she was given a pamphlet in lieu of debriefing. Thus, the format of the debriefing may have been a partial cause for its lack of significance in reducing CF risk levels. 18

Limitations This study was not without limitations. The first being, this study is a secondary analysis on an existing data set. The original study produced 120 subjects with completed demographic data, but the valid sample for this study was much smaller. For the purposes of this study only 89 of these 120 subjects submitted completed CFST, a necessary step as the study is based on a comparison of the scores for the week of the hurricane and the scores for the past week or time of completion of the survey. Another limitation of the study is the timing of the survey. Both CFSTs were given to the subject at the same time. The nurses were asked to recall the feelings they had during the week of the hurricane. While it would have been more desirable to have nurses complete the survey as they lived the experience, this can only be accomplished if the research had already been planned at the time of the hurricanes. The final limitation that is noted is that of the consistency of the debriefing. By definition debriefings are to follow a structured format and contain certain components. Based upon the comments mentioned in the conceptual framework section of this chapter it is unclear if the debriefings followed the standard format and therefore the integrity of the debriefing effectiveness could have been compromised. Implications for Practice Though the statistical analyses did not show a significant change in scores from the week of the hurricanes to the past week time period related to debriefing, there are many implications from this research for the assessment and treatment of traumatic stress and Compassion Fatigue in disaster relief workers. One area is disaster preparation. Health care and disaster relief providers need properly trained mental health professionals who follow a structured debriefing format. Aside from the need for adequately prepared mental health professionals, the entire theory of CSID needs to be applied to disaster situations, not just a portion of it. CSID was never intended as a stand-alone treatment, but it was intended as an initial intervention and referral for further services. Another area that could be addressed is risk assessment of the nurses and relief workers for Compassion Fatigue prior to deployment. Personnel should be screened for risk for Compassion Fatigue before they provide services and those found to be most at risk for Compassion Fatigue should have follow-up services. One possibility for accomplishing this task is to screen employees who are eligible for deployment as a part of their yearly evaluation. 19