There are nearly 20,000 registered nurses in the

Predicting the Risk of Compassion Fatigue A Study of Hospice Nurses Maryann Abendroth, MSN, RN Jeanne Flannery, DSN, ARNP, CNRN, CRRN, CCH There is a growing interest in the clinical phenomenon of compassion fatigue and its impact on healthcare providers; however, its impact on hospice nurses is basically unknown. This study investigated the prevalence and the relationships between nurse characteristics and compassion fatigue risk. It also provided a model for predicting compassion fatigue risk. A non-experimental descriptive design using cross-sectional data and descriptive and inferential statistics was used. Nurses (N = 216) from 22 hospices across the state of Florida participated in the study. Findings revealed that 78% of the sample was at moderate to high risk for compassion fatigue, with approximately 26% in the high-risk category. Trauma, anxiety, life demands, and excessive empathy (leading to blurred professional boundaries) were key determinants of compassion fatigue risk in the multiple regression model that accounted for 91 % (P<.001) of the variance in compassion fatigue risk. Knowledge of these variables may help organizations identify nurses at risk and provide interventions and preventions to maintain optimal nursing care. K W O R D burnout compassion fatigue hospice nurses secondary traumatic stress There are nearly 20,000 registered nurses in the hospice setting in the United States, and they are part of the aging nursing workforce. In 2000, the national average age of all registered nurses was 45.2 years, which, incidentally, has climbed steadily in recent years.^ Hospice nurses are not immune to the aging nursing workforce, general population growth, and work-related stress that may be factors in future nursing shortages.'^' ^ In addition to these outside factors, caring for dying patients induces considerable stress, which includes the challenge of providing comfort care to patients with complex disease processes and being empathic to families in psychosocial and Maryann Abendroth, MSN, RN, is on faculty as Assistant Professor in Nursing, Florida State University College of Nursing, Tallahassee, FL. Jeanne Flannery, DSN, ARNP, GNRN, GRRN, GGH, is a Professor, Florida State University College of Nursing, Tallahassee, FL. Address correspondence to Maryann Abendroth, MSN, RN, Florida State University College of Nursing, 419 Duxbury Hall, Tallahassee, FL 32306 (e-mail: mabendroth@nursing.fsu.edu). 346 JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006

spiritual crisis.'*'^ This study focused initially on a random sample of Florida's hospice nurses and their risk of compassion fatigue (CF), which is a secondary traumatic stress reaction resulting from helping, or desiring to help, a person suffering from traumatic events.^ Second, the study created models to predict CF risk among this population. The results of this study revealed that nurses in this specialty are especially vulnerable to the risk of CF. Variables linked to work-related stressors, demographics, and health factors reflected the impact that stress has on CF risk, which is exacerbated when nurses exhibit a lack of self-care. Many of these variables provided strong correlational and predictive value in targeting individuals at risk for this clinical phenomenon. Because the risk is apparent, CF is a preventable and treatable phenomenon. Understanding the prevalence of CF risk and implementing a model that predicts this risk could empower nurses and administrators to use preventive measures tbat promote self-care, improve patient outcomes, and enhance productivity. REVIEW OF THE LITERATURE The literature reviewed for this study included numerous findings and recommendations related to tbe impact of CF on healthcare workers.''"^ Studies also bave been conducted on stress and burnout, which may be precursors to CF in many professions, including hospice nursing. ^^'^^ Nonetheless, nurses wbo work in various settings and wbo are engaged in various levels of trauma work have received little attention regarding the effects of CF on their profession.^'^"* Qualitative studies revealed themes of stress wben nurses were exclusively involved witb tbeir terminal patients and experienced a true sense of loss wben tbeir patients died.^^'^^ Tbe results of quantitative studies noted tbat bigb education levels, stress from continuously facing difficult family dynamics, and multiple patients dying witbin a sbort period of time were related to high burnout scores. Otber studies also confirmed tbat tbese factors had a cumulative effect on nurses' coping abilities.^^'^^'^^"^^ The finding tbat hospice nurses use bigbly structured defense mecbanisms was supported by Payne,^^ wbo noted tbat bospice is a positive environment in wbicb to work, altbougb tbe nature of tbe work is difficult and may lead to burnout. Tbis finding is supported furtber by Wrigbt,^ wbo researcbed tbe qualities of bospice nurses and noted tbat tbese nurses bave a true sense of calling and are compassionate and accepting of tbe cboices made by patients and tbeir families. Tbe accumulated factors are tbere to predict CF risk; bowever, nurses' coping abilities, sucb as bealtby professional distancing learned from years of nursing experience, balanced out many of tbose stressors. Healtby distancing may be acbieved by a sense of selfcare, as evidenced by consciously taking time off from work, especially wben stress begins to accumulate. In tbis study, 17% {n = 37) of tbe nurses indicated tbat tbey received no support from tbeir own defined professional support system (ie, peers, supervisors, administrators) after a patient's traumatic death. Of tbose participants, 83% («= 31) were in tbe moderateto-bigb risk CF category. An inability to debrief after a traumatic event diminisbed internal coping mechanisms. Tbis finding confirms tbe results of earlier studies by Dean^^ and Mallet et al,^^ wbicb demonstrated tbat lack of support after experiencing a patient's traumatic deatb bad an impact on tbese nurses. More generally, tbe literature supports tbe belief tbat lack of support during occupational stress and trauma leads to psycbological distress.^^'^^ Bebaviors emanating from occupational stress and trauma revealed a central tbeme of tbis study: tbat nurses wbo become overly empatbic witb tbeir patients are most at risk for CF. Tbis central tbeme was higbly supported by Riggio and Taylor,^'*'''^^^' wbo identified empatby as an essential aspect of bospice nursing wben it takes tbe form of "perspective taking" and "empatbic concern"; bowever, once empatby becomes unbealtby it leads to "personal distress," wbicb negatively affects nursing care and leads to stress and burnout. Tberefore, unbealtby empatby leading to blurred professional boundaries appeared as a major stressor in bospice nursing. Tbe gaps in tbe literature were evident because virtually no studies evaluated bospice nurses and tbeir risk for CF; however, tbe fact tbat CF may exist among tbis population was verified by literature, wbicb reported tbat nurses wbo work in bospice care, emergency room settings, and psychiatric units are engaged in trauma work.'''^'* Risk of CF among chaplains and otber respondents after tbe World Trade Center disaster on September 11, 2001 in New York City was studied^ and it was reported tbat 55% of tbe sample {n = 403) \vere in tbe moderate-to-bigb CF risk category, accordirig to Roberts et al.^ Anotber finding of tbe study noted tbat workplace proximity to Ground Zero and lengtb of time volunteering for a relief agency bad no effect on JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006 347

w-> t CF risk.^ Tbese findings were similar to tbis study, wbicb reported tbat demograpbic and work-related variables did not bave a major discriminating value on CF risk; bowever, 79% {n = 170) of participants (n = 216) in tbis study had a bigber probability of being in tbe moderate-to-bigb risk category tban individuals in tbe aforementioned study. Tbe nurses in tbis study were exposed to ongoing career stressors, sucb as patient deatb and family crisis, on a continuous basis, wbereas participants in tbe previous study bad a controlled period of episodic acute stress. PROBLEM AND SIGNIFICANCE Tbere is a growing interest in understanding tbe clinical pbenomenon of CF and its impact on bealtbcare providers according to Huggard.^^ Altbougb tbis pbenomenon bas received considerable attention witbin otber bealtbcare populations, bospice caregivers bave not been evaluated for risk. These end-of-life (EOL) caregivers are especially vulnerable because tbey can become overinvolved due to tbe degree and lengtb of interaction witb their patients, their emotional investments, and frequent exposure to loss.^^ A study of CF risk in the population of bospice nurses was not only warranted but also essential to tbe viability of tbe profession and tbe future of optimal EOL care for an aging US populace. According to tbe National Hospice and Palliative Care Organization (NHPCO) 885,000 patients were served in 2002 by hospice affiliates, compared to 246,000 served in 1992. Tbis number represented a 259% increase in tbe 10-year period.^^ Because tbe hospice census is expected to increase, it is important to attract and retain nurses in this specialty area; bowever, it is projected that the nursing population will fall 29% below demand in the year 2020.^^ According to the NHPCO,^ hospice nurses will not escape this shortage, which will have a direct impact on FOL care in the US. Studies have shown that stress from continuously being faced witb difficult family dynamics and multiple patients dying witbin a sbort period of time bas a cumulative effect on nurses' coping abilities.^ '^^'^ Consequently, this stress may lead to burnout and eventually CF, which can challenge a caregiver's ability to provide services and maintain personal and professional relationships.^** It is important to note that burnout is tbe physical, emotional, and mental exhaustion caused by long-term exposure to emotionally demanding situations and can be considered a precursor or risk factor for CF. According to symptoms of CF and burnout are similar; bowever, CF bas a more sudden and acute onset that results from specific exposure to trauma and suffering. Reports from the National Center for Health Workforce Analysis^^ indicate that between 2000 and 2020, the nation's population is expected to grow 18%, and the subgroup of individuals aged 65 years and older is projected to grow 54%. As a result, tbere will be an increased demand for nurses, especially in geriatrics and areas such as hospice care, across the nation. An example of this demand was seen in 2002, when 81% of hospice admissions were patients aged 65 years and older.^*^ Therefore, selecting Florida for the venue of this study was an optimal choice because of the state's aging population; however, all nurses could benefit because of tbe nation's aging populace projection. Tbis study calls attention to tbe risks of secondary traumatic stress reactions in bospice care in order to prevent an impact on tbe integrity of nursing and compromise patient care. PURPOSE OF THE STUDY Tbe purpose of the study was to describe the prevalence of the risk of CF in hospice nurses in Florida and explore the relationship between various nurse cbaracteristics and CF risk. This study also sought to provide a model for predicting the risk of CF from knowledge of demographic and work-related factors. Research Question 1: What are the demographic and work-related characteristics of hospice nurses sampled for this study? Research Question 2: What is the prevalence of the risk of CF among the hospice nurses in the state of Florida? Research Question 3: What is the nature of the relationship between demographic, hospice workrelated factors and the risk of CF? Research Question 4: What demographic and bospice work-related factors predict the risk of CF? METHODS Design A non-experimental, correlational (descriptive) design was chosen utilizing cross-sectional data. The design 348 JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006

also was predictive, because factors found to be strongly associated with the risk of CF were used as independent variables in a linear regression analysis. Setting At the time of this study there were 40 not-for-profit and for-profit hospice organizations in the state of Florida, all of which were queried for participation in the research. Initially, 17 not-for-profit hospices agreed to participate through mailed surveys. Nurses from an additional four not-for-profit and one for-profit hospice participated later through surveys completed during the State Hospice Symposium in December 2004, which provided an overall 55% facility participation rate across the state of Florida. Sample The target population for the present inquiry included registered nurses (RNs), advanced registered nurse practitioners (ARNPs), and licensed practical nurses (LPNs) who met the following inclusion criteria: (1) are 18 years of age or older, (2) are employed by a hospice organization in Florida, and (3) interact directly with patients and their families. This population worked in any area of the organization, such as (1) freestanding inpatient hospice facility care, (2) home care, and (3) hospice admissions. The state of Florida is statutorily divided into 11 health service planning districts that provide a framework for projections of need for beds or health services. Nurses from all 11 districts were eligible to participate in the study. Study Variables Several groups of variables, including demographic/ informational, health, and work-related information, were used in this study. The demographic/informational independent variables included age, gender, ethnicity, marital status, children in the home, and the responsibilities of caring for an elderly or disabled parent or loved one. Health-related variables measured stressors related to finances and the death of a loved one. Other health-related variables questioned headache symptomology and diagnoses of hypertension, depression, or posttraumatic stress disorder (PTSD) and the tendency to self-sacrifice for the needs of patients. Work-related variables included level of education, licensure, certification, years in the profession and in hospice nursing, work setting, level of care (ie, primary care, charge nurse), hours of weekly work, shift work, case load, patient-to-nurse ratios, exposure to traumatic patient death, and number of cases of (direct involvement with patient deaths per month. Finally, another set of variables measured participants' behaviors that could lead to CF risk. Examples of these variables were "I feel as though I am experiencing the trauma of someone I have helped" and "As a result of my helping, I have sudden unwanted frightening thoughts."^*^ Data Collection Instruments and Procedures Two instruments were used in data collection. A demographic questionnaire developed by the researcher focused on demographic, work-, and health-related information that used theoretical concepts from nursing, medicine, and the social sciences.^'^^'^^ The second instrument was the Professional Quality of Life Compassion Satisfaction and Fatigue Subscales: Revision-III (ProQOL-CSF-R-III) developed by Stamm,^ :who established its reliability with a score of.80 for Cronbach's alpha. The ProQOL used 30 statements that assessed behaviors that could lead to CF ' risk, which was the dependent variable. Participants were asked to rate each statement as it applied to their current situation. Each statement was an independent variable in the study that was analyzed and grouped to produce a CF risk score. Designated contact persons from each of the 17 initially participating hospice organizations distributed packets containing the instruments, a cover letter, and a postage-paid return envelope to the randomly selected nurses in their respective agencies. The number of packets each facility received was based on the results of a proportional sampling procedure. Each hospice provided the principal investigator with the total number of eligible nurses in the organization. From these reports, each hospice organization's proportion of the total Florida eligible population was calculated, adjusted for response rates, and constituted each organization's contribution to the study sample. Hospice administrators had no access to the completed surveys because the participants were able to return the documents directly to the researcher. Data collection at the Hospice Symposium involved distributing a cover letter and the two instruments to the 150 nurse JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006 349

attendees. They were asked to complete the instruments and place them in a designated locked box at the conference site, where they were collected by the researcher and secured. Each facility was identified by a coding process on the survey forms; however, precautions were taken to maintain the confidentiality of these mailings and conference surveys, and any identifying information was deleted during data entry. Precautions were taken during the Symposium to prevent duplicate responses from nurses attending the conference who might already have completed the instruments in the mail. Outcomes were analyzed and reported as regionspecific results and aggregate group results. Individual facilities were not identified in outcome measures, reports, or publications. RESULTS A total of 583 surveys were distributed through the combined venues. The mailed surveys yielded a response rate of 38.3% {n - 166), whereas the symposium surveys yielded a response rate of 33% (n = 50). Data obtained from the two venues (mail versus conference) were reviewed using independent samples t tests, and contingency tables using Chi-square and Fisher's exact tests, to determine the extent to which the data collection venue was differentially associated with all relevant variables. Results of these tests indicated that no factors could potentially confound the final results. Civen the foregoing results, data from the two venues were combined into a single stratified random sample of 216 usable instruments. This 37% {n = 216) response rate was representative of all 11 healthplanning districts across the state of Florida, giving the study results good generalizability. Participants (N = 216) were predominantly female (n = 204, 94%) and white {n = 177, 81.9%). The participants' ages ranged from 23 to 76 years, with the median age being 50.5 years (M = 53.9, SD = 9.05). Most participants in the sample were married (59.7%, n = 129), and approximately a quarter (23.1%, n = 50) were divorced. The average number of years of nursing experience for the entire sample was 20.19 {md = 20, SD = 11.37), and the average number of years of hospice experience was 5.65 {md - 4, SD = 5.03). Of all participating nurses, 101 (46.8%) of the nurses had Associate Degrees in Nursing, and the RNs {n - 183, 84.7%) were the largest group represented in the sample, which was not unusual, because this group encompassed nurses with the associate, diploma, baccalaureate, and master's degrees. Other licensed nurses included 13.4% {n = 29) LPNs and approximately 2% {n = 4) ARNPs. The sample in this study had a higher proportion of RNs compared to the licensure distribution of all nurses in Florida, which was RNs {n = 173,000, 62%), LPNs {n = 97,000, 35%), and approximately 9700 (3.4%) ARNPs.-'^'-''* One would expect more RNs than LPNs in this field of nursing than in general nursing because of the heavy emphasis on case management within a multidisciplinary team setting. The sample in this study can be considered nationally representative of hospice nurses, because nationally, approximately 86% {n = 16,716) of hospice nurses are registered nurses, whereas 14% {n = 2628) are LPNs.^^ More than half {n - 130, 60.2%) of the nurses were field nurses who provided hospice care in private homes and made visits to hospice patients in nursing homes or were continuous care nurses. Approximately one quarter of the nurses in the study [n = 50, 23.1%) worked in in-patient hospice units. These nurses noted that their average care ratio was one nurse to approximately six patients (M = 5.59, md = 6, SD = 1.31). The nurseto-patient ratios were higher in this study than the one-to-four ratios recommended by the Center to Advance Palliative Care.^^ The nurses who did not work in an inpatient unit reported having an average weekly caseload of 19.8 {md = 13, SD = 54) patients. Caseloads for this sample of nurses were higher than the national median hospice nurse case load {md = 10) as reported by the National Trend Summary Report for 2000-2003.^^ The participants' (N = 216) average hours worked per week were 40.5 {md = 40, SD - 10.6); however, the weekly time worked ranged from 8 to 90 hours. Hospice nurses, in general, encounter more patient deaths than nurses in many other specialties. Within a 30-day period, they were exposed to an average of seven deaths {md = 5.00, SD 8.85), and within the last year almost 42% {n = 90) experienced the death of someone close to them. Risk of CF was operationalized with the ProQOL, as evidenced by a process by which score ranges could be categorized into minimum (<7), moderate (8-17), and high (>18) levels for CF risk.-'" For the entire sample, the median score for CF risk was 14 (M = 13.6, SD = 6.59). The minimum score was 1, and the maximum 350 JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006

score was 39. Results revealed that 57 (26.4%) of the participants were in the high-risk category, 113 (52.3%) were at moderate risk, and 21.3% (n = 46) were at low risk. Consequently, nearly 80% (n = 170) of the sampled hospice nurses were at moderate to high risk for CF. The study analyzed the prevalence of CF risk for the entire sample and identified specific subgroups for which such risk exists. Table 1 identifies several subgroup characteristics and relates them to low, moderate, and high CF risk levels. Some of the characteristics exhibited little or no variability among the three risk levels. Personal health factors and their relationship with CF risk revealed that there was virtually no difference in CF risk categories between cigarette smokers and individuals who cared for a loved one. However, 35% (M = 17) of individuals diagnosed with depression or PTSD were in the high-risk CF category, as opposed to 24% {n - 40) among persons who did not have depression or PTSD. A greater proportion of individuals who had financial stress {n = 36, 31%) and persons If \«I a B I e " '- Compassion Fatigue Risk Levels and Nurse Characteristics! I Characteristics Age: years Years of professional nursing Years of hospice experience Average hours worked per week Gender Male («= 12) Female {ft = 204) Ethnicity White/Non-Hispanic («= 177) African American (n = 20) Hispanic (n = 6) Asian (n = 5) Marital status Single {«= 24) Married («= 129) Divorced (n = 50) Widowed («= 8) Nursing licensure LPN («= 29) RN (n = 183) Nursing education Practical («= 25) Diploma («= 36) Associate (n = 101) Baccalaureate (n = 45) Professional setting Field {n = 130) Inpatient (n = 50) Personal characteristics Self-sacrifice for others («= 136) Diagnosed with depression/ptsd («= 48) Compassion Fatigue Risk Levels Low Risk Moderate Risk HighlRisk 52 [10.3] 16 [10.8] 4 [5.3] 40 [15.3] 3 (25.0) 43 (21.1) 37 (20.9) 7(35.0) 0(0) 0(0) 6 (25.0) 27 (20.9) 9 (18.0) 2 (25.0) 6 (20.7) 38 (20.8) 6 (24.0) 6 (16.7) 23 (22.8) 8 (17.8) 25 (19.2) 11 (22.0) 21 (15.4) 4 (8:.3) 51 [8.7] 20 [11.5] 4.5 [4.8] 40 [9.6] 5 (41.7) 108 (52.9) 95 (53.7) 9 (45.0) 4 (66.7) 1 (20.0) 16 (66.7) 67 (51.9) 24 (48.0) 4 (50.0) 15 (51.7) 98 (53.6) 11 (44.0) 22 (61.1) 55 (54.5) 22 (48.9) 70 (53.8) 26 (52.0) 68 (50.0) 27 (56.3) 50' 44 6'.7j^ A I 4(33.3) 53 (2^6.0) 45 (25.f) 4 (2:0.0) 2 ( 3.3) 4 (80.0) i \ 2 {i2\ 35 (%7.1) 17 (3,4.0) 2 (25.0) 8 (2^7.6) 47 (2:5:7) 8(i2.0) 8 (2^.0) 23 {±.8} 15(43.3) f i 35 (Z6.^) 13 (2%.O) 47 m. Values are presented as median [SD] or n (%). Percent is within each individual characteristic. JOURNAL OF HOSPICE AND PALLIATIVE NURSING VoL 8, No. 6, November/December 2006 351

who experienced frequent headaches {n = 22, 36.1%) also was in the high-risk category for CF. Nurses who are inclined to put their patients' needs ahead of their own may be prone to paying attention to their own needs last in other areas of their lives. This characteristic can lead to stress. The subgroup of participants who had a tendency to sacrifice their own psychological needs to satisfy the needs of their patients (Table 1) had a much higher proportion of nurses in the high-risk category for CF («high risk = 47, 34%) than those nurses who answered "no" to the self-sacrifice question (whigh risk = 10, 13%). The ProQOL also measured burnout risk and provided percentile bands for low-, moderate-, and high-risk classifications. The results of the prevalence data indicated that Florida hospice nurses are an atrisk population for burnout and CF, with 91% of nurses in the moderate- to high-risk category for burnout also being classified in the moderate- to high-risk category for CF. See Table 2 for score summaries among various percentile bands. Another component of the study analyzed the relationship between demographic, hospice workrelated factors, personal health factors, and the risk of CF. There was virtually no correlation between participants' age and their scores on CF risk (r = -0.04). Marital status (r - 0.15) and ethnicity {n = 0.20) also revealed low correlations. Correlations did exist between CF risk and whether nurses had experienced a patient's traumatic death (r = 0.24). Although this value was less than expected, the categorical (;^ = 3) nature of the traumatic death variable could have been partially responsible for such a value. Studies have noted that stress levels rise when nurses do not receive psychological/social support after this type of stressor. ^^ The investigator's review of the literature'''^^ revealed that nurses' work-related variables, such as long work hours, high patient caseloads, multiple deaths occurring within a short period of time, and shift work. have stressful effects on the individual, which then can lead to burnout and CF. The relationships between these variables and CF risk in this study were positively correlated but smaller than expected. Five of the work-related variables were selected within the ProQOL instrument and analyzed for their associations with the three ProQOL subscale variables. These first five variables, shown in Table 3, revealed higher correlations than the other demographic, work-, and health-related variables. See Table 3 for correlational values. A final component of this study analyzed demographic, hospice work-related, and personal health factors to predict CF risk. An Ordinary Least Squares multiple regression analysis was conducted using variables obtained through a stepwise variable selection process. All variables (demographic, personal health, and work related) were considered for inclusion, and zero-order and semi-partial correlations were used as selection criteria. The regression analyses initially generated three regression models to predict CF risk among hospice nurses. Subsequently, a total of seven variables representing each of the first three models were chosen to build a composite multiple regression model to predict CF risk. All four models were statistically significant (P <.001), and the seven independent variables of the composite model provided 91% {Radn - -907) of the information needed to perfectly predict CF risk with a prediction error of 2.01 units. Subglobal tests of the unique contributions of each of the seven variables in the composite multiple regression model indicated that they were all statistically significant (P <.001) except for "feelings of being overwhelmed" (P =.433). Table 4 summarizes the regression models. The study analysis revealed that stress, trauma, anxiety, life demands, and excessive empathy (leading to blurred professional boundaries) were key determinants of CF risk. These basic themes were threaded throughout the variables. The variables of each of the four regression models represented these themes, which T a b l e 2 Burnout and Compassion Fatigue Risk Frequencies for Florida Hospice Nurses Burnout Compassion fatigue Low Risk" «= 84 38.9% 21.3% Moderate to High Risk* It = 132 tt = 170 6L1".. 78.7".. 1 > Hmh " Score delineations: Low risk: < 7, moderate to high risk > 8, high risk > 18. 352 JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006

a b I e 3 Correlations Between lndepen'deh /a' and ProQOL Subscale Variabfes-^^ Independent Variables Correlations "Infected" by traumatic stress Being "on edge" due to helping Losing sleep over patients' trauma Feelings of being overwhelmed by work and caseload Feelings of being "bogged down" by the system Self-sacrifice for patients' needs** Financial stress** Fxperienced patient traumatic death* Headaches** Depression/PTSD diagnosis** Ethnicity* Marital status* Average hours work per week Work setting* Average nurse/patient ratio *Eta correlation coefficients. **Point biserial correlation coefficients. All others are Pearson Product Moment Correlation Coefficients. also were linked to theories from nursing, psychology, and biology. DISCUSSION It is not uncommon for nurses in this specialty of nursing to experience secondary traumatic stress; therefore, it was not surprising that Florida hospice nurses are an at-risk population for CF. Nearly 80% (n - 170) of the total sample (N = 216) were in the moderate-tohigh risk category for CF. The researcher found that demographics and work-related factors were of little discriminating value for describing the prevalence of risk for this phenomenon. This finding implies that these nurses possess intricate defense mechanisms to cope with caring for terminally ill patients and their families, who are often in a state of crisis. These coping abilities may be inherent or may have been learned from years of nursing experience. Many nurses indicated that they were undergoing financial stress or were diagnosed with depression/ PTSD. Because PTSD is closely linked to CF,^ it was not unusual that participants exhibited higher risk scores, because their coping mechanisms may have been affected by their own trauma and health conditions. The sampled nurses (n = 136, 64%) who tended to sacrifice their own personal and psychological needs for the needs of their patients had a greater percentage of smoking behavior, financial stress, headaches, 'and hypertension than nurses who responded negatively to that question. This result confirms similar findings in the literature with respect to the personal health behaviors of selfless caregivers. From a different perspective, 83% (n - 47) ofthe sampled participants who were classified in the highrisk category for CF responded positively that they selfsacrifice for others' needs. This finding was important because nurses who indicated that they self-sacrifice also admitted to other health problems resulting from physiological stress. This behavior of nurses caring more for their patients' needs than for their own needs reflects an unhealthy level of empathy, which is a risk factor for CF. Not only were the nurses who selfsacrificed in a high-risk category but they also exhibited high-risk behaviors. The researcher in the current study was able to claim the alternate hypothesis because the results of this study revealed that one or more of the independent variables provided a significant contribution to the prediction of CF risk. These demographic, work-, and healthrelated variables reflected the impact that stress has on CF risk, which is exacerbated when nurses exhibit a lack of self-care. The seven variables that predicted CF risk in i the composite model were linked to behaviors that cbuld have been influenced by the multiple independent variables in this study. These behaviors also revealed a central theme of the negative effects of overly identifying with patients, which resulted in unintentionally, vicariously experiencing their pain and anxiety. This finding was important because unhealthy levels of empathy, coupled with life demands and health factors, are directly related to CF risk and distress. According to Selye,''^ unremitting stress is a cause of additional physiological and psychological health concerns. Being able to predict CF risk via these risk factors is paramount to the health and welfare of hospice nurses and their ability to have effective therapeutic relationships. JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006 353

T a b l e 4 Multiple Regression Model Summaries MODEL 1* Independent variables (predictors) MODEL 2** Independent variables (predictors) MODEL 3"* Independent variables (predictors) COMPOSITE MODEL* Independent variables (predictors) Multiple Regression Model Summaries Full Model Full Model Adjusted Standard Error R R^ R^ of the Estimate Full Model r Value 0.702 0.493 0.481 4.78 39.705 <.001 "Preoccupied with those I help"; feelings of being overwhelmed; bogged down by the system, self-sacrifice for others' needs; stress from finances 0.847 0.718 0.711 3.556 105.785 <.001 "On edge" due to helping; frightening thoughts due to work; memory loss; experience frequent headaches; full or part-time work 0.965 0.932 0.930 1.755 401.250 <.001 Being "infected" by traumatic stress; "on edge" due to helping; depressed due to helping; frightening thoughts due to work; situational avoidance; difficulty with personal and professional separation; "Preoccupied with those I help" 0.954 0.910 0.907-2.01 295.874 <.001 "Preoccupied by those I help"; difficulty with personal and professional separation; being "infected" by traumatic stress; depressed due to helping; feelings of being overwhelmed; "on edge" due to helping; frightening thoughts due to work Independent variable theoretical influences: * = Selye,''^ ** = Neuman,^ In this study, 17% (n = 37) of the nurses indicated that they received no support after a patient's traumatic death, and 83.3% (n - 31) of those participants were in the moderate- to high-risk CF category. An inability to debrief after a traumatic event diminished internal coping mechanisms. This finding confirms the results of earlier studies,^^'^^ which demonstrated that lack of support after experiencing a patient's traumatic death had an itnpact on these nurses. More generally, the hterature supports the belief that lack of support during occupational stress and trauma leads to psychological Nurses who worked shifts (n = 86) in this study had lower CF risk scores {md = 12.5) than nurses who did not work shifts {md = 15.00). There was virtually no difference between CF risk scores of nurses who worked rotating shifts {md = 13.89) and nurses who did not rotate {md - 13.62). This finding contradicted Selye,^^ who indicated that working shifts predisposes different occupational groups, including nurses, to physiological stress and health conditions. One of the main reasons why the findings in this study may have turned out differently is the unique, highly structured coping mechanisms that hospice nurses may possess, which cannot be explained in a quantitative study. They may have lifestyles that lend themselves better to evening shift work, which otherwise can disrupt *** = Figley, ' = Selye, Neuman and Figley. families with children. These nurses may be better organized to balance the stress of long, 12-hour shifts and perhaps a rotating shift schedule, still find time to maintain a household, and sustain their relationships with significant others. Stress also is muitifactoriai; each factor builds upon others differently in different people, so an impact on CF may not be observed easily by one or two variables alone. This study measured the risk of CF at one point in time. There is a possibility that perceptions may have changed over time due to individual circumstances. There also may have been a general disinterest in the study due to the focus on the aftermath of hurricanes in 2004. All Florida hospice facilities were invited to participate in this study; however, no effort was made to follow up with hospices in areas that may have been most adversely affected by these storms. Although the overall response rate was 37%, it was not a true limitation according to Norwood,^^ because the average return rate for mailed surveys is 20%. This research has shown that hospice nurses are at a moderate to high risk for CF. They experience, on average, seven patient deaths per month and must communicate compassionately and professionally with distraught families before, during, and after the dying process. Therefore, these nurses may be at risk for increased absenteeism and an exodus from the 354 JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006

profession, which results in lost revenue for the organization. More difficult to measure is the loss when an experienced hospice nurse leaves the profession. On average, hospice nurses have more than 20 years' nursing and hospice nursing experience. Not only is there a financial impact on the organization when they leave the profession but there is also a huge loss of mentorship to newer nurses and a loss of established relationships with physicians.''^ This study was able to fill some of the gaps in literature; however, others still exist, such as comparison studies of larger populations, analysis of existing policies in relation to CF risk, and implementation of qualitative studies to provide greater depth to quantitative findings. CF is a preventable and treatable phenomenon. Hospice organizations with policies, interventions, and evaluation methodologies that address CF risk may result in substantial employee benefit cost savings, uninterrupted professional nursing care, and increased patient family satisfaction and may continue to be regarded highly in communities as an optimal choice in EOL care. Acknowledgments The authors would like to thank Mr. Paul Ledford and the staff at Florida Hospices and Palliative Care, Inc. for their support by promoting the study and providing a venue for this research. They also would like to thank Dr. Charles Figley, PhD for his support and encouragement throughout this process and Mr. John Abendroth for editorial feedback. References 1. National Center for Health Workforce Analysis. The registered nurse population: findings from the 2000 national sample survey of registered nurses. US Department of Health and Human Services. 2002;1-135. 2. United States General Accounting Office. Testimony before the committee on health, education, labor and pensions, US Senate. Nursing workforce: recruitment and retention of nurses and nurse aides is a growing concern. GAO-01-750T. 2001 ;1-8. 3. Hospice and Palliative Nurses Association. HPNA position paper: value of the professional nurse in end-of life care. Journal of Hospice and Palliative Nursing. 2004;6(l):65-66. 4. Kulbe J. Stressors and coping measures of hospice nurses. Home Healthcare Nurse. 2001;19(ll):707-711. 5. Sontag MA. Hospices as providers of total care in one western state. HospJ. 1996;ll(3):71-94. 6. Figley CR, ed. Compassion Fatigue: Coping with Secondary Traumatic Stress Disorder in Those who Treat the Traumatized. New York: Brunner/Mazel; 1995. 7. Clark ML, Gioro S. Nurses, indirect trauma, and prevention. Image J Nurs Sch. 1998;30(]):85-87. 8. Figley GR. Gompassion fatigue: psychotherapists' chronic lack of self care. JCLP/ln Session: Psychotherapy in Practice. 2002; 58(11):1433-1441. 9. Roberts SB, Flannelly KJ, Weaver AJ, Figley GR. Gompassion fatigue among chaplains, clergy, and other respondents after September 11*. J Nerv Ment Dis. 2003;19](]l):756-758. 10. Wastell GA. Exposure to trauma: the long-term effects of suppressing emotional reactions. / Nerv Ment Dis. 2002;! 90(12): 839^845. 11. Bene B, Foxall M. Death anxiety and job stress in hospice and medical surgical nurses. HospJ. 199];7(3):25-41. 12. Dean RA. Occupational stress in hospice care: causes and coping strategies. Am J Hosp Palliat Care. ]998;15(3):15M54. 13. Payne N. Occupational stressors and coping as determinants of burnout in female hospice nurses. J Adv Nurs. 2001;33(3): 396-405. 14. Schwam K. The phenomenon of compassion fatigue in perioperative nursing. AORN J. 1998;68(4):642-645. 15. Gray-Toft PA, Anderson JG. Sources of stress in nursing terminal patients in a hospice. OM CA. 1986-87;17{l):27-39. 16. Maytum JC, Heiman MB, Garwick AW. Gompassion fatigue and burnout in nurses who work with children with chronic conditions and their families. J Pediatr Health Care. 2004;] 8: 171-179. 17. Adams JP, Hershatter MJ, Moritz DA. Accumulated loss phenomenon among hospice caregivers. Am J Hosp Palliat Care. 1991;8(3):29-37. 18. Keidel GG. Burnout and compassion fatigue among hospice caregivers. Am J Hosp Palliat Care. 2002;]9(3):200-205. 19. Masterson-Allen S, Mor V, Laliberte L, Monteiro L. Staff burnout in a hospice setting. Hosp J. ]985;1(3):]-14. 20. Wright DJ. Researching the qualities of hospice nurses. Journal of Hospice and Palliative Nursing. 2002;4(4):210-216. 21. Mallett K, Price JH, Jurs SG, Slenker S. Relationships among burnout, death, anxiety, and social support in hospice and critical care nurses. Psychol Rep. 1991;68:]347-1359. 22. AbuAlRub RF. Job stress, job performance, and social support among hospital nurses. Image J Nurs Sch. 2004;34(l):73-78. 23. Olofsson B, Bengtsson G, Brink E. Absence of response: a study of nurses' experience of stress in the workplace./ Nurs Manag. 2003; 11:351-358. 24. Riggio RE, Taylor SJ. Personality and communication skills as predictors of hospice nurse performance. / Bus Psychol. 2000;]5(2): 351-359. 25. Huggard P. Gompassion fatigue: how much can I give? Med Educ. 2003;37:163-164. 26. National Hospice and Palliative Gare Organization (NHPGO). NHPGO facts and figures. Available at: http://www.nhpco.org/ files/public/facts7o20figures%20feb04.pdf. Accessed April 5, 2004. 27. Hospice and Palliative Nurses Association. HPNA position statement: shortage of registered nurses. 2003. Available at: http:// www.hpna.org/position_nurseshortage.asp. Accessed February 17, 2004. 28. Gollins S, Long A. Too tired to care? The psychological effects of working with trauma. / Psychiatr Ment Health Nurs. 2003;] 0:17-23. JOURNAL OF HOSPICE AND PALLIATIVE NURSING Vol. 8, No. 6, November/December 2006 355

I'T 29. National Center for Health Workforce Analysis. Projected supply, demand and shortages of registered nurses: 2000-2020. US Department of Health and Human Services. 2002;!-22. 30. Stamm BH. Research information on the ProQOL-CSF-R-III. 2002. Available at: http://www.isu.edu/~bhstamm/testsproqol_ psychometric.htm. Accessed March 22, 2004. 31. Neuman B. The Neuman systems model. In: The Neuman Systems Model. 3rd ed. Norwalk: Appleton and Lange; 1995: 3-76. 32. Selye H. The Stress of Life. Revised edition. New York: McGraw- Hill; 1976. 33. Florida Center for Nursing. Statewide strategic plan for nursing workforce in Florida, 2004. Available at: http://www.flcenter fornursing.org/strategic/stratplan-november04.pdf. Accessed February 14, 2005. 34. Gregg A, Brunell ML. Nursing supply and demand in Florida: analysis of nursing licensure data. Florida Genter for Nursing, 2003. Available at: http://www.flcenterfornursing.org/research/ FLSupply.pdf. Accessed July 10, 2004. 35. Hospice Association of America. Hospice facts and statistics, 2002. Available at: http://www.nahc.org/gonsumer/hpcstats. html. Accessed February 14, 2004. 36. Genter to Advance Palliative Gare. GAPG manual: acuity, 2002. Available at: http://64.85.16.230/educate/content/elements/inpatient unitacuity.html. Accessed March 1, 2005. 37. National Hospice and Palliative Gare Organization. National trend summary report, 2004. Available at: http://www.nhpco.org/ files/pubiic/nds00_03trendsstatsl01904.pdf. Accessed March 1, 2005. 38. Norwood SJ. Research Strategies for Advanced Practice Nurses. Upper Saddle River: Prentice Hall Health; 2000. 39. Abendroth, M. Predicting the risk of compassion fatigue: an empirical study of hospice nurses executive summary (unpublished document]. Tallahassee: 2005:1-6. STATEMENT OF OWNERSHIP, MANAGEMENT, AND CIRCULATION (Act of August 12, 1970; Section 3685; Title 39 United States Code) Date of Filing October 1, 2006. Title of Publication /owma/ of Hospice & Palliative Nursing; Frequency of Issue Bi-Monthly; Annual Subscription Price $59.00; Location of Known Office of Publication Lippincott Williams & Wilkins, Inc., 16522 Hunters Green Parkway, Hagerstown, MD 21740-2116; Location of the Headquarters or General Business Offices of the Publisher Lippincott Williams & Wilkins, Inc., 530 Walnut Street, Philadelphia, PA 19106; Publisher Lippincott Williams & Wilkins, Inc., 530 Walnut Street, Philadelphia, PA 19106; Editor Leslie H. Nicoll, PhD, MBA, RN, loa Beach Street, Ste 2, Portland, ME 04101. Managing Editor^Randi Davis, Lippincott Williams &c Wilkins, 530 Walnut Street, Philadelphia, PA 19106; Owner Lippincott Williams & Wilkins, Inc., 530 Walnut Street, Philadelphia, PA 19106, 351 West Camden Street, Baltimore, MD 21201; Wolters Kluwer, US, 333 Seventh Avenue, New York, NY 10001; Wolters Kluwer nv (owns 100% of stock), Stadouderskade 1, 1054 FS Atnsterdam, The Netherlands; Known Bond Holders, Mortgagees, and other security holders owning or holding 1 percent or more of the total amount of bonds, mortgages, or other securities None. A. Total no. of copies printed (net press srun), average 9,583, actual 9,600. B. Paid and/or requested circulation 1. Paid/requested outside-county mail subscriptions stated on form 3541, average 8,553, actual 8,531; 2. Paid in-county subscriptions, none; 3. Sales through dealers and carriers, street vendors, counter sales, and other non-usps paid distribution, average 51, actual 49; 4. Other classes mailed through the USPS, none. C. Total paid and/or requested circulation [sum of B (1), (2), (3), and (4)], average 8,604, actual 8,580. D. Free distribution by mail (samples, complimentary, and other free). Outside-county as stated on form 3541, average 156, actual 392; 2. In-county as stated on form 3541, none; 3. Other classes mailed through the USPS, none. E. Free distribution outside the mail (carriers or other means), average 1, actual 5. F. Total free distribution (sum of D and E), average 157, actual 397. G. Total distribution (sum of C and F), average 8,761, actual 8,977. H. Copies not distributed, average 822, actual 623.1. Total (sum of G and H), average 9,583, actual 9,600. Percent paid and/or requested circulation, average 98.21%, actual 95.58%. I certify that the statements made by me above are correct and complete. Jeffrey Brown, Manager, Periodicals Operations. 356 JOURNAL OF HOSPICE AND PALLIATIVE NURSING VoL 8, No. 6, November/December 2006