COMPASSION FATIGUE IN HOSPICE SOCIAL WORK: POTENTIAL MODERATING FACTORS By Sally Beth Pelon A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Social Work - - D octor of Philosophy 2015 ABSTRACT COMPASSION FATIGUE IN HOSPICE SOCIAL WORK: POTENTIAL MODERATING FACTORS By Sally Beth Pelon As part of the inter professional team of hospice caregivers, social workers are exposed to multiple stressors, both in their work with dying patients and their families and in their functioning within rapidly changing health care organizations. Ongoing exposure to these professional stressors prompts consideration of the emotional and psychological impact hospice social work may have on those who do it, how hospice social workers evaluate and interpret the costs and benefits of their work, how they process and cope with the consistent themes of dying, death, loss, and grief that pervade their everyday lives, and what intervention s may assist in mitigating th eir stressors. Th is descriptive, cross - sectional research study explore d the prevalence of compassion fatigue w ithin the context of hospice social work and provided information regarding the nature of professional stress expe rienced by social workers in end - of - life care settings. Professional stress factors and worker - related factors were considered. Fifty - five hospice social workers from 35 Hospice organizations in Michigan completed an online survey for this study. Result s suggested that compassion fatigue is a genuine concern among hospice social workers and that compassion fatigue is correlated with wok overload and a depression diagnosis within the past 12 months. In addition, compassion fatigue and compassion satisfa ction were found to be negatively correlated , suggesting that as compassion fatigue increases, compassion satisfaction decreases. Copyright by SALLY BETH PELON 2015 ACKNOWLEDGEMENTS This document has been the product of a long and arduous journey and I am keenly aware that t his journey would not have been possible without the support and assistance of a number of remarkable individuals. I would like to thank Dr. Anne Hughes, my disse rtation committee chair , who has been incredibly patient and supportive, answering all of my questions and spending hours reading drafts and offering helpful feedback. I want to thank Dr. Rena Harold, Dr. Amanda Woodward and Dr. Linda Keilman who have so willingly and capably served on my dissertation committee. I have had the privilege to work with over the years and who have taught me what it means to live and die with grace and dignity. I also want to express my appreciation to my hospice co - workers who have been a source of support, hope, inspiration, and humor. Finally, I owe my deepest gratitude to my family . They have enveloped me with unwavering love and support and have made ma ny sacrifices that have sustained me through this process - thank you does not seem quite enough. iv TABLE OF CONTENTS LIST OF TABLES vii CHAPTER ONE : INTRODUCTION 1 Introduction and Background 1 Hospice and Social Work 3 Hospice Social Work and Stress 4 Constructs of Occupational Stress 6 Compassion Satisfaction 10 Hospice Social Work and Professional Supervision 12 Theoretical Orientation: Attachment Theory and Supervision 13 Statement of the Problem 14 Purpose of the Study 16 CHAPTER TWO : LITERATURE REVIEW 17 Compassion Fatigue and End of Life Caregiving 17 Professional Stress Factors: Exposure to Death 18 Professional Stress Factors: High Cas eloads 19 Worker - Related Factors: High Empathetic Attunement 20 Worker - Related Factors: History of Loss 21 Worker - Related Factors: Diagnosis of Clinical Depression 21 Social Work Supervision and Compassion Fatigue 23 Compassion Sat isfaction and Compassion Fatigue 25 Limitations of Prior Research 27 Review of Findings and Research Questions 29 CHAPTER THREE : METHODOLOGY 32 Study Design 32 Sample 32 Procedures 33 Measures 34 Dependent Variable 35 Compassion Fatigue 35 Independent Variables 37 Professional Stress 37 Exposure to Death 37 Work Overload 37 Number of Years Working in Hospice Setting 38 Employment Status 39 Independent Variables 39 Worker - Related Variables 39 Personal Experience of the Dea th of a Family Member or Friend 40 v Current Diagnosis or Diagnosis within the Past Year of Depression 40 Empathy 40 Moderat ing Variables 42 Professiona l Supervision 42 Compassion Satisfaction 45 Demographic Variables 45 Data Analysis 46 C HAPTER FOUR : RESULTS 48 Response Rate 48 Sample Characteristics 48 Descriptive Statistics 50 Compassion Fatigue and Compassion Satisfaction 50 Professional Stress 52 Worker - Related Factors 53 Supervision 55 Bivariate Analysis 57 C HAPTER FIVE : DISCUSSION 61 Major Fi ndings 61 Additional Findings 67 Study Limitations 70 Implications for the Social Work Profession 73 Recommendations for Future Research 77 Conclusion 79 APPENDICES 80 APPENDIX A R equest for Partic ipation Email Communication 81 APPENDIX B Institutional Review Board Approval 82 APPENDIX C Email Contacts with Participating Hospices 84 APPENDIX D Study Survey 86 REFERENCES 94 vi LIST OF TABLES Table 1 Sample Characteristics 49 Table 2 Average Daily Census of Participating Hospices 50 Table 3 Compassion Fatigue Scores 51 Table 4 Compassion Satisfaction Scores 51 Table 5 Professional Stress 53 Table 6 Worker - Related Factors 54 Table 7 Supervisor Characteristics 56 Table 8 Bivariate Analysis of Compassion Fatigue and Professional Stress and Compassion Fatigue and Worker - Related Factors 58 Table 9 Comparative Compassion Fatigue and Compassion Satisfaction Scores 5 9 vii 1 CHAPTER ONE: INTRODUCTION Introduction and Background The past several decades have witnessed dramatic advancement s in medical and scientific t echnology intended to extend life and to postpone death. Indeed, these advancements have been successful in promoting the diagnosis, treatment, and cure of many once fatal diseases , and in extending life expectancy within the United States (U.S.) by several years ( Centers for Disease Control and Prevention, [CDC ], 2014 ). These advancements have also c reated a reality in which death is often no longer perceived as part of the natural order of life but as an event to be feared and to be avoided, no matter the cost ( Cicirelli, 2006 ; Feifel, 1977) ; a reality in which death frequently occurs in sterile institutional settings, removed from public view and from the familiarity and comfort of home (Bern - Klug, Gessert, & Forbes, 2001) ; a reality often perceived as failure of the health care system, of medi cal technology, and of the health care team (Clark, 1989). However, while scientific and medical advances over the past decades were shaping the manner in which death is perceived and changing the places where death most o ften occur s , a small but influential grassroots movement sought to challenge and temper these perceptions and to humanize care to the dying . hospitium, which means hospitality, and hospis meaning guest or s tranger (Conner, 2007). The notion of providing hospitality for the sick and disabled dates back to 17 th century Europe, where monasteries often served as places of refuge for travelers and those in need of care (Noe, Smith, & Younis, 2012). Hospice as a reference to specialized care for the dying was first used by Dr. Cicely Saunders , who began 2 working extensively with the terminally ill in England in 1948. In 1967, Saunders established St. , the first modern hospice and the catalyst for the hospice and palliative care movement (Saunders, 2001). From the beginning, Saunders challenged the traditional medical model of care in which th ose with advanced illness we re treated as individuals with physical deficits to be aggressively managed or (DiTullio & MacDonald, 1999) , and a dvocate d instead f or the emotional, s piritual , and physical comfort of the dying through a holistic and interprofessional approach to care (Noe et al., 2010) and the recognition of death as the natural and final life stage (DiTullio & MacDonald, 1999). Saunders introduced the hospice concept to the U .S. in 1963 , during a speech to medical students, nurses, social workers, and chaplains at Yale University. Howeve r, it was not until 1974 that the first hospice was established in the U . S. in Branford, Connecticut by Florence Wald, Dean of the School of Nursing at Yale University, along with two pediatricians and a chaplain (National Hospice and Palliative Care Organization [NHPCO], 2013). The hospice movement experienced significant growth and made important strides toward becoming a recognized and valid health care entity over the next decades, particularly in the 1980s, with the creation of a federal reimbursement structure. In 1982, Congress included a provision as part of the Tax Equity and Fiscal Responsibility Act to provide federal reimbursement for hospice services through the Medicare Hospice Benefit (Conner, 2007 ). Reimbursement opportunities through Medicare and commercial health insurance companies coupled with increasing societal awareness of and desire , created an environment ripe for hospices to grow beyond their humble beginnings as small, all - volunteer agencies into large, recognized, and profitable players in the health care biosphere. In 2011, an estimated 1.65 million term inally ill individuals received Hospice services nationwide and 3 approximately 45% of those who died in the U .S. were under the care of a hospice program (NHPCO Facts and Figures, 2012). The mission of hospice is to provide necessary and appropriate phy sical, emotional, and spiritual care to individuals in the final stages of terminal illness in order to ensure that the dying process is as comfortable and as rich as possible (N HPCO , 2013). Hospice specializes in exceptional pain and symptom management, available 24 hours a day through medical intervention as well as emotional and spiritual support. This holistic care is provided through an inter professional team of care providers which includes physicians, nurses, home health aides, chaplains, voluntee rs , and social workers . Hospice emphasizes care to patient family members as well, offering emotional and spiritual support to families throughout the duration of the patient s involvement with hospice and p roviding bereavement support services for 13 months following the patient s death. Nursing staff comprise the largest percentage of hospice caregivers (30%) followed by home health aides (19%). Social workers represent approximately nine percent of Hospice caregivers (NHPCO Facts and Figures, 2012). Hospice and Social Work With the implementation of the Medicare Hospice Benefit in 1982, social workers became some extent , with all hospice patients and families (N HPCO , 2013). This requirement dying includes not just attention to physical needs and bodily symptoms, but also to social, psycholo gical, emotional , and spiritual well - being. Social workers contribute a unique set of skills and professional values that are essential to the hospice inter professional team and to the holistic care of dying individuals and their 4 families. Social workers are educated to practice from a person - in - environment perspective , which allows them to offer a distinctive and helpful perspective to the inter professional team regarding patients and families (Christ & Blacker, 2005). educational focus on human behavior in the social environment, human diversity, as well as communication and counseling skills (C ouncil on Social Work Education [C SWE ] Educational Policy and Accreditation Standards, 2010) uniquely equips social workers to assess and address the psychosocial needs of the dying and their families and to facilitate communication between patients and their families and among inter professional team members (Christ & Blacker, 2005; tment to patients - determination and informed consent (N ational Association of Social Workers [NASW] Code of Ethics, 2008) prepares them to assist patients and families as well as inter professional team members with ethical issues regarding m edical treatment and end - of - values of social justice and the dignity and worth of each person (NASW Code of Ethics, 2008) equip them to advocate for humane care for the vulnerable and underserved members of societ y and for the provision of end - of - pay. Hospice Social Work and Stress As part of the inter professional team of h ospice caregivers, hospice social workers routinely encounter concerns and needs that lie beyond the conventional boundaries of both normal life experiences and social services work. dying and their families require them to be intimately connected with death on a daily basis. Ro utine exposure to the physical sights, sounds, and smells of death can prompt both fear and repulsion within social workers, and then consequent guilt for feeling frightened and repulsed 5 (Pilsecker, 1979). In addition, hospice social workers are repeated ly reminded of their own human limits and mortality and of the brevity of life, both for themselves and for their loved ones (Davidson & Foster, 1995; Leon et al., 1999). Hospice social workers also frequently find themselves working within stressf ul family situations in which emotions are raw, coping is minimal, and predictable patterns of living and relating dissolve into uncertainty and confusion (DiTulio & MacDonald, 1999; Keidel, 2002; Rynde, 1977). Prolonged or ongoing exposure to this distr ess by hospice professionals , uncertainty and distress (Rynde, 1977). In addition, the life and death nature of hospice work often creates a sense of urg ency and need for crisis - driven intervention that can create a distinctive stress for hospice social workers (DiTulio & MacDonald, 1999). In addition to these patient - related stressors, hospice social workers may also be exposed to various organization al stressors as part of an inter professional team in a primarily medical setting. Other disciplines within the team, most often nursing, frequently claim similar competency in addressing psychosocial concerns and providing psychosocial interventions, thus ction within the team (Dane & Chachkes, 2001; Vachon, 1995). In addition, social workers are frequently not credited for their knowledge and skills in medical settings due to lack of understanding by medical staff of their training and expertise (Dane & C hachkes, 2001). Hospice social workers also struggle with burgeoning caseloads that add to their work stress. In a 2004 national review of 330 patient charts within 66 hospices, there were 18 nurses for every 100 patients, while there were just 6 soc ial workers for every 100 patients (Reese, 6 average of 11.3 patients for nurses and 11 patients for home health aides (NHPCO, Facts and Figures, 2013). Ongoing exposure to these professional stressors within the hospice setting prompts consideration of the emotional and psychological impact hospice social work may have on those who do it, how hospice social workers evaluate and interpret the costs and benefits of their work, and how they process and cope with the consistent themes of dying, death, loss, and grief that pervade their everyday lives. The stress of working in a professional capacity with the suffering of others has been understood and described in a variety of ways. Constructs of Occupational Stress Historically, there has been a lack of clarity and consensus within the professional literature regarding the constructs utilized to describe the adverse emotional, physical, and social impacts of professional caregiving ( Baird & Kracen, 2006; Najjar, Davis, Beck - Coon, & Doebbling, 2009). Descriptive terms such as burnout, compassion fatigue, vicarious traumatization, and secondary traumatic stress are often used interchangeably, which can create difficulty in understanding the nuances of why and how professional caregivers are impacted by their work and in identifying practices that may prevent or lessen their harmful effects. While these constructs share significant similarities, there are impor tant differences as well. Freudenberger (1974) first used the term burnout to describe a range of behavioral and physical symptoms including exhaustion, irritation, depression, and cynicism commonly experienced by those whose work required them to resp ond to the needs of others. Freudenberger (1974) proposed that these symptoms develop gradually and become progressively more paralyzing for caregivers. Maslach (1976; 1978) expanded on of forms in care providers 7 whose work involves direct and ongoing interaction with individuals facing various needs and problems. B urnout was described as the gradual loss of caring for the individuals with whom the care - provider works, emotional exhausti on, cynicism, and loss of positive feelings. Maslach and Jackson (1981) further developed the concept of burnout and created the widely used Maslach Burnout Inventory (MBI), which suggests a multi - dimensional model of burnout consisting of three critical components: emotional exhaustion, depersonalization, and reduced personal accomplishment. These early conceptualizations of burnout focused on the relationships between care providers and the distressed individuals with whom they worked as a necessary e lement in the development of burnout. In particular, the care provider - care receiver relationship was seen as contributing to emotional exhaustion , which was believed to be the root cause of burnout (Sabo, 2011). Additional research has expanded these ea rly conceptualizations and suggests that burnout is not specific to care provider - care recipient relationships. Instead, issues such as work overload, organizational dysfunction, lack of control, lack of reward, lack of support, lack of fairness, and valu e conflicts are more predictive of burnout, and consequently, burnout can be found in all work environments, regardless of whether care provider - care recipient relationships are central to the work ( Harrison & Westwood, 2009; Leiter & Maslach, 2004; Sabo, 2011). Figley (1988) introduced the concept of secondary traumatic stress to suggest that individuals traumatized by abuse, crime, war, or natural disaster se ek support and share their experiences with family, friends, and professional caregivers, traumatic symptoms are transferred to these supporters (Figley, 1988). Consequently, supporters of those who have been traumatized have been shown to develop sympt oms that mimic Post - Traumatic Stress 8 Disorder ( American Psychiatric Association, 2013 ) such as hypervigilance, exaggerated startle response, intrusive thoughts and images, diminished affect, and avoidance behaviors (Figley, 1988). McCann and Pe arlman (1990) introduced the concept of vicarious traumatization as an occupational hazard specifically for therapists working with traumatized individuals. Pearlman and colleagues ty, and great loss through empathetic engagement in the therapy relationship can lead to feelings of fear, pain, and suffering similar to those of their clients (McCann & Pearlman, 1990; Pearlman & MacIan, 1995; Pearlman & Saakvitne, 1995). Ongoing exposur e may gradually lead to relationships, and beliefs about the self and the world (Pearlman & Saakvitne, 1995). Compassion fatigue as a similar construct for desig nating the costs of professional caregiving was first used in the literature by Joinson (1992) in discussing burnout symptoms among emergency room nurses. This concept was further developed and applied to a range of professions in which professional careg iving with distressed individuals is central (Adams, Boscarino, & Figley, 2006; Figley, 1995; Keidel, 2002; Radey & Figley, 2007). Figley (1995) , expanding on his work with secondary traumatic stress, defines compassion fatigue in the context of compassio n, or the deep sense of knowing or awareness of the suffering of another together with the desire to relieve that suffering. The ability to demonstrate compassion, then, is a central component of compassion fatigue, in which ongoing exposure to and empat hetic concern for the suffering of others may lead to deep physical, emotional, and spiritual exhaustion, hopelessness, disconnection from others, and decreased capacity for and interest in empathetic attunement with clients (Radey & Figley, 2007). 9 Compa ssion fatigue and related constructs describe a natural rather than a pathological, process of human caring (Figley, 1995; Stamm, 2010) and represent preventable and treatable consequences of empathetic engagement with suffering individuals (Harrison & Wes twood, 2009). In addition, it is suggested that compassion fatigue is an occupational hazard that reflects neither pathology in the therapist nor intention on the part of the client (Pearlman & Saakvitne, 1995). Theoretically, it has been suggested tha t t he risk for compassion fatigue increases relative to empathetic aptitude, with individuals who exhibit high levels of empathy and empathetic al., 2006; Figley, 200 2). Empathy is defined as a personality characteristic that designates the capacity to perceive, understand, and respond to the emotional state of another (Barker, 2003). This definition may be further delineated by noting both the emotional element of empathy, the cognitive element, the capacity to respond appropriately to these emotions and circumstances through words and action (Davis, Soderland, Cole, Gadol, K ute, & Myers, 2004). A final aspect to perceive, understand, and respond to the emotions and experiences of another while maintaining sufficient emotional separation to prevent the loss of self (Corcoran, 1989). Ongoing research in to the costs of professional caregiving ha s not yet provided adequate tools to consistently differentiate between these various forms of occupational stress (Najjar et al., ssion fatigue in which ongoing exposure to and empathetic concern for the suffering of others may lead to deep physical, emotional, and spiritual exhaustion, hopelessness, disconnection from others, and decreased 10 capacity for and interest in empathetic att unement with clients appears to be the most commonly used construct (Najjar et al., 2009). In addition, b ased on this researcher of c ompassion appears to reflect the centrality of the mission of hospice to provide holistic and tender care to dying individuals (N HPCO , 2013) and to the deep connection hospice social workers often feel for their work and for their patients and families. C onsequently, compassion fatigue is the construct used throughout this study to denote and describe the occupational stress of social workers within the hospice setting. Compassion Satisfaction Research on the well - being of practitioners working with suffering individuals has focused almost exclusively on the negative costs of caring (Abendroth & Flannery, 2006; Badger et al., 2008; Figley, 1995; Linley & Joseph, 2007; Tehrani, 2007). More recent discussions of compassion fatigue expand considera tions of professional caregiving not only as a source of depletion but also as a source of fulfillment and satisfaction. Compassion satisfaction is from contribu ting to the well - being of others (Radey & Figley, 2007; Stamm, 2010). The focus of compassion satisfaction is on the powerful experience of emotional engagement, compassionate helping, and the outcomes of worker - clie nt interactions on both an individual a nd societal level, despite the risks and the costs of caring (Larsen & Stamm, 2008; Slatten, Carson, & Carson, 2011; Stamm, 2010). The compassionate engagement of work with suffering individuals makes the pleasure and fulfillment of compassion satisfactio n unique in comparison with other types of vocational satisfaction. Larsen and Stamm (2008) suggest that the ability to 11 embrace the benefits of work with suffering individuals may serve as a buffer against the negative costs of such work. Compassion satis faction as a positive effect of professional caregiving and as a potential protective mechanism against probable risks may be considered within the context of one of the hallmarks of sound social work practice, the strengths - based perspective. This orientation asserts that individuals, families, and communities are to be seen through the lens of their capacities, abilities, motivations, knowledge, experience, and possibilities, no matter how distorted or damaged they may have become through circumstance, oppression, or trauma (Saleeb e y, 1996). While not denying or negating very real issues and struggles faced by clients, the strengths - based perspective requires social work practitioners to also uncover and highlight resources and strengths existing within and around the individuals, families, and communities with whom they work , rather than simply focusing on the issues and struggles themselves ( Saleeby, 1996). More recently, the strengths perspective has expanded beyond social interactions and interventions with clients and has been applied to social workers themselves. Bell (2003) first suggested that stressors of professional caregiving might potentially be mitigated by incorporating a strengths - based perspective in t o research, prevention, and treatment of these stressors. Clark (2011) further posited that pr ofessional resilience may be developed and strengthened within social work clinicians when they are encouraged to draw up on the strengths perspective to locate a nd expand on the positive aspects of their work, even in the face of the many stressors they routinely face . Rather than avoiding compassion fatigue by attempting to protect professional caregivers from the more difficult and painful aspects of 12 working wi th suffering individuals, greater awareness of and emphasis on the positive aspects of caregiving , as endorsed by the strengths - based perspective, may be a more helpful approach. The reality of compassion fatigue for hospice social work clinicians a nd the potential for compassion satisfaction to serve as a protective mechanism against this professional vulnerability suggests the need for increased awareness of these issues and improved mindfulness in addressing them. Professional social work supervision within the hospice setting may prove helpful in this regard . Hospice Social Work and Professional Supervision Social work supervision is a building block for effective social work pr actice, professional growth, and agency accountability. It has been identified as a key factor in job satisfaction and high quality service delivery in the social service field (Harkness, 1995; Tsui & Ho, 1997). Social work supervision historically has reflected three interrelated functions: administration, education, and support (Kadushin, 1992; Munson, 2002). The administrative function of supervision includes managerial tasks such as staff recruitment and selection, work planning and assignment, moni toring and evaluating work, and advocating for staff with upper management and other agency entities (Kadushin, 1992; Kadushin & Harkness, 2002; Munson, 2002). Educational activities in social work supervision include instruction and teaching in skills an d desired outcomes specific to agency function (Kadushin, 1992; Kadushin & Harkness, 2002). The crucial objective of both educational and administrative supervision is to assist social workers in providing the best possible service to clients (Harkness & Hensley, 1991; Kadushin, 2002). Administrative supervision affords the organizational structure and resources, while educational supervision provides the preparation and ongoing training to achieve this objective. 13 Support is the third major component of supervision , assist s social workers in managing the stress of their work and to increas ing positive feelings about their work and about who they are as professionals (Kadushin, 1992; Kadushin & Harkness, 2002). Excessive stress may adversely impact soc ial workers capacity to provide effective and helpful services to clients (Greene, 1991). Consequently, social work supervisors act to prevent, reduce, and mitigate this stress through attentive listening, validating stressors and concerns, offering comf ort, and situations and cases (Kadushin, 1992; Slatten et al., 2011). Kadushin and Harkness (2002) also posit that supportive supervision is essential in maintainin intrinsic commitment to and motivation for their work, particularly when there is limited access to extrinsic rewards such as high salaries or social status. Theoretical Orientation : Attachment Theory and Supervision Consideration of the benefits of professional social work supervision may be more fully understood in the context of attachment theory. Based upon concepts introduced by Bowlby (1969) and Ainsworth (1978), attachment theory posits that humans are born wi th an innate need for attachment to significant others. The central theme of attachment theory suggests that primary caregivers who are available and responsive to the needs of their infants, establish a sense of security within their children. When chi ldren trust that their caregivers are dependable, they develop a secure base from which to explore their environment. In addition, if their environment is perceived as threatening or frightening, children trust that they may return to their caregivers for comfort and soothing (Newman & Newman, 2009). More recent empirical research has expanded this original childhood attachment hypothesis, suggesting that early childhood attachment patterns continue into adulthood and 14 provide a basis for adult relationsh ips (Fonagy, 2001). Most often, the primary intimate relationship provides both the secure base from which adults are able to interact with their environment and the safe haven to which they may return for comfort and support (Bennett, 2008). Pi stole and Watkins (1995) first suggested that the professional supervisory relationship of his or her professional world. Pistole and Watkins (1995) further indicate that the supervisory relationship functions to affirm for clinicians that they are not alone in their efforts with clients and that they have a ready resource for support and information when needed. Similarly, Newswald - McCalip (2001) posits that creates sufficient confidence necessary for social workers to talk honestly about the struggles of their work and to prompt help - seeking behavior in times of crisis. B ennett (2008) propose s similar interpretations in her work with field supervisors and social work students. Bennett (2008) notes that the challenges of field placement may activate attachment responses, in which field instructors provide the secure base for students to explore an unfa miliar professional world and to seek the safe haven of the field instructor during times of vocational stress. Attachment theory suggest s that access to supervision as well as a strong supervisory relationship has the potential to assist social workers in recognizing and coping with the demands of their work. This theoretical orientation undergirds and informs this inquiry in to the role of supervision in moderating compassion fatigue in hospice social workers. Statement of the Problem The deleteri - being have been well documented in the literature. Compassion fatigue has been associated with 15 decreased productivity (Leon, Altholz, & Dziegielewski, 1999), diminished quality of c are to clients (Davidson & Foster, 1995; Clark et al. , 2007; Swetz, Harrington, Matsuyama, Shanafelt, 1999), job dissatisfaction (Lloyd et al., 2002; Clark et al., 2007), poor work quality ( H arr & M oore, 2011); and personal health issues (Trippany, White - Kress, & Wilcoxon, 2004; Swetz et al., 2009). In addition, symptoms of compassion fatigue are frequently int erpreted as reflective of personal incompetence, and consequently, may result in those struggling with this issue to leave their chosen profession (Bell, Kulkarni, & Dalton, 2003; Leon et al., 1999). Although t he detrimental impact of compassion fatigue on professional caregivers working with suffering individuals is well - documented ( Clark et al., 2007; Davidson & Foster, 1995; Harr & Moore, 2011; Lloyd et al., 2002; Swetz et al., 2009 Trippany, et al., 2004) , the prevalence and impact of compassion fatigue on social workers within the hospice setting has been largely unexplored. In addition, there is very limited research that considers if and how various interventions such as professional supervision and a fo cus on compassion satisfaction , may serve as protective mechanisms against compassion fatigue among hospice social workers. This is problematic for hospice social work er s who may struggle to recognize, to understand , and to cope with the emotional and psychological demands of their work. I t is problematic for hospice organizations that may struggle to retain qualified social work clinicians. And i t is also problematic for dying individuals and their families who may receive less than the highest quality, most compassionate care from social workers who are overwhelmed by the stress and demands of their work. It is imperative to address these issues on ethical grounds, as social work clinicians, educators, and resear chers alike must endeavor to provide appropriate and effective care and support for hurting individuals. 16 Purpose of the Study In order to assist hospice social work ers to thrive both professionally and personally and to maintain the highest standards of service in the hospice care environment, it is essential that compassion fatigue be recognized, understood, and addressed. While previous research on compassion fatigu e has been helpful in understanding the detrimental and harmful impact of suffering , there has been very little investigation of this professional hazard within the co ntext of hospice social work . Additionally, there is a paucity of empirical studies examining what sorts of intervention s and resources may serve to protect professional caregivers against the experience of compassion fatigue. The purpose of this resear ch study is , first, to expand current research by exploring the prevalence of compassion fatigue specifically within the context of hospice social work and by providing detailed knowledge of the nature of professional stress experienced by social workers i n end - of - life settings. Second, this study seeks to investigate potential moderating factors in the experience of compassion f atigue among hospice social work ers . Specifically, this study considers both compassion satisfaction and social work supervision a s potential moderating influences on compassion fatigue. 17 CHAPTER TWO: LITERATURE REVIEW In order to explore the issue of compassion fatigue within the hospice social work setting and to consider potential moderating factors, a comprehensive review of relevant literature is necessary. Pertinent empirical studies are examined related to compas sion fatigue and end - of - life caregiving, the role of empathy related to compassion fatigue, professional social work supervision and its impact on professional stress, and compassion satisfaction as a potential moderating factor in compassion fatigue. Re search gaps and limitations are noted as well. Compassion Fatigue and End o f Life Caregiving Due to a very limited number of research studies related to hospice social work and compassion fatigue, studies contained within this discussion have been expand ed to include compassion fatigue and end - of - life professional caregivers from all disciplines in hospice settings as well as palliative care and oncology settings Empirical research regarding end - of - life professional caregiving and compassion fatigue has m ost often been considered from the perspective of nurses (Abendroth & Flannery, 2006; Melvin, 2012; Potter et al., 2010) or from the perspective of the hospice interprofessional team with no distinction made between disciplines (Alkema et al., 2008; Mor & Laliberte, 1984; Slocum - Gori et al., 2011; Thompson, 2007). O ne dissertation study addresses the issue of compassion fatigue and its impact on hospice social workers (Lawson, 2008) and two studies speak to professional stress from the perspective of socia l workers in oncology settings (Simon et al., 2005; Supple - Diaz & Mattison, 1992). Despite this scarcity of research, compassion fatigue appears to be a genuine risk for end - of - life caregivers. In a cross sectional quantitative study with 216 hospice nu rses from 22 Florida 18 hospices, Abendroth and Flannery (2005) found that nearly 80 percent of respondents were at moderate to high risk for compassion fatigue as measured by the Professional Quality of Life (ProQOL) Compassion Fatigue su bscale (Stamm, 200 2 ). Similarly, Melvin (2012) found that all participants in her descriptive qualitative study with six highly experienced, purposively sampled end - of - life care nurses had experienced or were currently experiencing some aspects of compassion fatigue in the ir work with dying patients. Thompson (2007) found that three of the six hospice staff members in her mixed methods study scored high or very high on the C ompassion F atigue portion of the P roQOL (Stamm, 2010). Similarly, in a cross sectional quantitative study with 153 oncology nurses, Potter and colleagues (2010) found that 37 percent of inpatient oncology nurses and 35 percent of outpatient oncology nurses were at high risk for compassion fatigue, based on the Compassion Fatigue subscale of the ProQOL ( Stamm, 2002) . Finally, in a national survey of 503 health care pro fessionals in hospitals, care homes, and community - based health services, Slocum - Gori and colleagues (2011) found that those health care workers who provide psychosocial support to patien ts and families demonstrated higher levels of compassion fatigue (1.91 of 5 versus 1.72 of 5) and burnout (2.14 of 5 versus 1.93 of 5) when compared to those who provide other types of health care services to patients and families. In addition to t he noted prevalence of compassion fatigue among end - of - life care professionals, various correlates of compassion fatigue and end - of - life caregiving have been identified. These correlates may be understood under two central themes of professional stress fa ctors and worker - related factors. Professional Stress Factors: Exposure to Death Frequent and ongoing exposure to death by end - of - life caregivers has been identified as a significant correlate in the development and experience of compassion fatigue by va rious 19 researchers. Abendroth and Flannery (2006) noted a significant correlation ( r = .24) between compassion fatigue and multiple deaths of patients in a short period of time and with exposure to particularly traumatic patient deaths in their study with hospice nurses in Florida . This hospice nurses in the Northeast U. S. Respondents in this study consistently described multiple deaths of patients as significant in their self - assessed feelings of compassion fatigue. Finally, Supple - Diaz and Mattison (1992) noted similar results in their mixed method study with 27 identif ied repeated exposure to death and multiple deaths in a short time as their greatest job - related stress. Professional Stress Factors: High Case loads H igh caseloads in end - of - life care have been correlated with compassion fatigue as well. Abendroth and Flannery (2006) noted that consistently high caseloads were correlated with increased risk of compassion fatigue ( r = .475) in their cross - sectional stud y with 216 hospice nurses in Florida . Similarly, respondents in Supple - Diaz and Mattison (1995) mixed method study identified high caseloads as a significant stressor in their work when questioned about sources of stress in their work with oncology patients. Finally, more than half of respondents in a qualitative study with 38 hospice caregivers in upstate New York cited high work volume as a significant and distinct source of work stress (DiTullio & MacDonald, 1999). In addition, 71 650), or the progres sive constriction of time available for attention to patient needs and emotional processing as a primary source of work stress 20 Worker - Related Factors: High Empathetic Attunement Empirical data regarding the connection between empathy and compassion fati gue is somewhat limited. Consequently, little is known regarding if and how empathy may act as a gateway to this type of professional vulnerability in end - of - life caregivers. Various studies do explore empathy and its relation not only to compassion fati gue, but also to related professional stress constructs including burnout and secondary traumatic stress. These studies are included in this discussion in order to offer a more extensive understanding of the relationship between empathy and work related s tress in end - of - life care workers. Slocum - Gori, Hemsworth, Chan, Carson, and Kazanjian (2011) utilized a self - report questionnaire with a self - selected sample (n = 503) to investigate compassion fatigue and burnout among hospice and palliative care wo rkers. Results suggest those hospice and palliative care workers who self - identified as providing emotionally intense psychosocial support to dying patients and their families had significantly higher levels of compassion fatigue (1.91 of 5 versus 1.72 of 5) and burnout (2.14 of 5 versus 1.93 of 5) than those who did not provide this type of service. Abendroth and Flannery (2006) noted similar findings in their study with hospice nurses in Florida. H ospice nurses in this study wh ose self - reported use of excessive empathy and blurring of professional boundaries with their patients and families identified this as a key determinant in the risk for compassion fatigue. Badger, Royce, and Craig (2008) reported similar findings in their descriptive, cross - sectional study with 121 self - selected hospital social workers. The researchers found a weak but positive correlation ( r = .19 ; p < .05) between higher levels of empathy as measured by the Interpersonal Reactivity Index [IRI] ( Davis, 1983) and symptoms of secondary traumatic stress. In addition, the researchers noted a strong negative correlation ( r = - .63 ; p < .001) between emotional separation and increased likelihood of 21 secondary traumatic stress, suggesting that difficulty emotionally differentiating from another while being empathetic is associated with increased secondary stress. These findings support who exhibit high levels o Worker - Related Factors: History of Loss Research has suggested that in addition to factors encountered within the hospice work setting, there are personal stress factors and experiences of hospice workers themselves that may create an increased vulnerability to compassion fatigue. One such factor is a personal history of loss. In their cross sectional study with 21 oncology social workers recruited through the Association of Oncology Soci al Workers in six southeastern states, Simon, Pryce, Roff, and Klemmack (2005) found that personal experiences with loss w ere correlated with higher burnout scores. Respondents in this study who had experienced the death of a loved one in the previous fou r years had statistically significant (p = .05) higher mean burnout scores (31.6) than those respondents who had not experienced a recent loss (24.1). This finding is similar to Supple - Diaz and Mattison (1992), who utilized a mixed method study to explor e sources of stress and reward in oncology social workers. Respondents identified a personal history of loss or serious illness as both a factor in their decision to pursue a social work career in the oncology setting and also as an issue that at times c ompromised their capacity to cope with the demands of their work. Worker - Related Factors: Diagnosis of Clinical Depression Personal mental health concerns have been demonstrated to be an additional personal stress factor that may increase vulnera bility to compassion fatigue in professional end - of - life caregivers. In a cross sectional, quantitative study with a self - selected sample of 212 hospice social workers recruited through the Hospice Foundation of America (HFA) , Lawson (2008) 22 found that 53 respondents reported a current diagnosis of depression. These self - identified depressed respondents scored significantly higher ( M = 12.25, SD = 5.21) on the compassion fatigue subscale of the P roQOL R - III (Stamm, 2005) than those respondents who did not report a diagnosis of depression ( M = 10.31, SD = 5.72). Similarly, Abendroth and Flannery (2006) found that of the 48 hospice nurses in their study with a self - reported diagnosis of depression or Post Tra umatic Stress Disorder (PTSD) , approximately 91 percent of these 48 were at moderate or high risk for compassion fatigue. Finally, Whitebird and colleagues (2013) noted similar results in their cross sectional, quantitative study with 547 self - selected ho spice workers from Minnesota. The researchers found moderate correlations between depression scores on the Patient Health Questionnaire - Eight (Kroenke et al., 2009) and both compassion fatigue ( r = .48) and burnout ( r = .51), as measured by the Pro QOL R - III (Stamm, 200 8 ). Clearly, there is a dearth of research in this area and existing research is consistently limited by small sample size (Abendroth & Flannery, 2006; Lawson, 2008; Melvin, 2012; Supple - Diaz & Mattison, 1992; Thompson, 2007) , self - selection sampling methods (Abendroth & Flannery, 2006; Badger et a l ., 2008; Melvin, 2012; Simon et al., 2005; Slocum - Gori et al., 2011; Supple - Diaz & Mattison, 1992), focus on specific geographic regions (Abendroth & Flannery, 2006; Melvin, 2012; S upple - Diaz & Mattison, 1992; Thompson, 2007; Whitebird et al., 2013), and cross - sectional designs (Abendroth & Flannery, 2006; Badger et al., 2008; Simon et al., 2005; Slocum - Gori et al., 2011; Supple - Diaz & Mattison, 1992; Whitebird et al., 2013) . Howev er, despite these limitations, it does appear that compassion fatigue is indeed a valid and troublesome issue for those working in end - of - life care, including hospice social workers. This suggests the need for investigation in to possible mechanisms and p ractices that may serve to protect hospice social workers from compassion fatigue and other forms of work - related stress. 23 Social work supervision and an improved emphasis on compassion satisfaction may be two such mechanisms. Social Work Supervision and Compassion Fatigue The benefits of professional social work supervision have been well - documented within the literature. Supervisory support has been found to be associated with job satisfaction (DeLoach, 2003; Grasso, 1994; Mena & Bailey, 2007; Pouli n &Walter, 1992; Scannapieco & Connell - Carrick, 2006), with decisions regarding whether to leave or to stay with an organization (Rycraft, 1994; Samantrai, 1992; Scannapieco & Connell - Carrick, 2006), and with work manageability (Samantrai, 1992; Rycraft, 1994). However, there is a surprisingly weak evidence base for supervision as a protective mechanism for work - related stress broadly, or compassion fatigue specifically, in social work practice. In addition, no studies were found that consider professional supervision in hospice social work as a means to moderate and to cope with compassion fatigue or that include professional supervision as a correlate in the study of compassion fatigue in hospice social work. This gap is notable and prompts the need to expand the range of empirical research contained within this inquiry and to include studies regarding the use of professional supervision to assist social workers in coping with various forms of work related stress in all fields of social work practice within the U .S . Professional supervision as a moderating factor in assisting social workers to cope with their work stress was first noted by Davidson (1985). Utilizing an exploratory, cross sectional survey with 36 purposively sam pled social workers working with cancer patients within five U . S . hospitals, Davidson considered various supports to assist these social workers in coping with the emotional impact of their work. Thirty five of the 36 respondents in this study indicated t hat their work with cancer patients was especially stressful. Along with team and peer support, 24 a number of respondents (no percentage provided) identified supervision as an important element in coping with the demands of their work with cancer patients a nd the stress their work generated. Similarly, Dane and Chachkes (2001) utilized focus groups to explore the effects of chronic exposure to illness and preferred coping mechanisms among hospital social workers at a medical center in New York. The resear chers found that all 12 hospital social workers included in their study , cited support from both supervisors and peers as very important in their ability to cope with the demands of their work. Whitebird and colleagues (2013) utilized a cross - sectional survey with 547 hospice professionals, including 61 social workers, from several hospices in Minnesota to explore the factors that might help to mitigate the profes sional stress of these workers. In addition to noting that 60 percent of respondents reported moderate to high levels of stress, the researchers found that 81 percent of respondents also identified both personal and professional social support as a vital factor in coping with their work stress. Similarly, in a cross - sectional study with a random sample of 211 California social workers employed in health and mental health care agencies, Kim and Lee (2009) explored the effects of various types of supervi sory communication on burnout and turnover intention. Utilizing the Maslach Burnout Inventory - Human Services Survey (Maslach & Jackson, 1986), the researchers found moderate negative correlations between the emotional exhaustion facet of burnout and posit ive supervisory relationships ( r = - .31. p < .05) as well as job - relevant supervisory communication ( r = .29, p<.05). This suggests that positive supervisory emotiona l stress and burnout in their work . 25 The limited research available indicates that social work supervision may be a promising intervention in assisting social workers in recognizing and coping with their work - related stress. However, additional empirical research is clearly needed. Compassion Satisfaction and Compassion Fatigue Research on the well - being of end - of - life caregivers has focused almost exclusively on the negative costs of caring (Abendroth & Flannery, 2006; Alkema et al., 2008; Me lvin, 2012; Mor & Laliberte, 1984; Potter et al., 2010; Slocum - Gori et al., 2011; Thompson, 2007), with very little attention paid to the enrichment and satisfaction these caregivers may experience as a result of their work. Empirical research has just re cently begun to explore these enriching and satisfying aspects using a variety of constructs, both in the end - of - life health care field and in the mental health field. Literature from both fields are included in this discussion in order to provide a broad er understanding of the potential positive effects of professional caregiving and their capacity to act as a protective mechanism against professional stress. Alkema, Linton, and Davies (2008) utilized a cross - sectional survey with 37 purposively sampled hospice workers from two Midwest hospice agencies to investigate the relationship between self - care, compassion fatigue, burnout, and compassion satisfaction. Results from this study suggest a strong negative correlation between compassion satisfa ction and burnout ( r = - .612, p <.05) as well as a moderate negative correlation between compassion satisfaction and compassion fatigue ( r = - .30, p <.05), indicating that increased feelings of compassion satisfaction are associated with decreased feelings o f burnout and compassion fatigue. These researchers further suggest that increasing compassion satisfaction may enhance quality of work and care provided to patients and families (Alkema et al., 2008). Slocum - Gori and colleagues (2011) noted similar results in their study with 503 hospice and palliative care workers. These 26 researchers found a negative correlation between compassion satisfaction and burnout ( r = - .531 , p < .001) as well as between compassion satisfaction and compassion fatigue ( r = - .208, p < .001). Harr, Brice, Riley, and Moore (2014) explored the impact of compassion fatigue and compassion satisfaction on 480 social work students in field placements as compared with 186 employed human service professionals. The Pro QOL R - IV (St amm, 200 8 ) was utilized to Results indicate a statistically significant negative relationship between compassion satisfaction and both compassion fatigue ( r = - .175 , p < .001) and burnout ( r = - .489, p < .01) among social work students. Respondents with higher levels of compassion satisfaction exhibited lower levels of both compassion fatigue and burnout . O n the other hand, respondents with higher levels of compass ion fatigue and burnout experienced lower levels of compassion satisfaction. While not specifically labeled as compassion satisfaction, Davidson (1985) noted that 67 percent of the 36 hospital social workers in her cross sectional study reported exper iencing their work as especially satisfying for them. Similarly, Dane and Chachkes (2001) found a recurrent theme of positive work - related feelings in a qualitative study with 12 purposively sampled full time social workers in a New York medical facilit y. Study participants consistently reported that their work often felt significant and helpful and that these feelings helped them to cope with the more stressful aspects of their work. Harrison and Westwood (2009) noted comparable findings in their qu alitative study with six purposively sampled therapists working with individuals dealing with some sort of trauma. All participants shared that they take satisfaction in being effective in their work, in making meaningful contributions through their work efforts, and in being highly skilled in their work. Participants further describe their work 27 as an extraordinary privilege and that this is, in part, what sustains them in their professional efforts (Harrison & Westwood, 2009). Hernandez and colleague s (2007) explored the potential positive impacts of working with hurting individuals in their exploratory, qualitative study with 12 purposively sampled mental health clinicians. In addition to suggesting that work with suffering individuals does indeed h ave a positive impact on mental health clinicians, these researchers also noted this effect may be strengthened by bringing conscious attention to it. This finding may be suggestive of the importance of professional supervision to assist clinicians in rec ognizing and celebrating the positive aspects of their work and the positive change they can influence in the lives of their clients. Additional empirical studies are clearly necessary in order to strengthen and clarify the evidence base regar ding how best to respond to the issue of compassion fatigue in professional caregiving. However, despite this scarcity of research and its limitations, it certainly appears that professional social work supervision and an improved focus on compassion sati sfaction may be important and promising interventions in combatting compassion fatigue and other forms of professional stress. Limitations of Prior Research While the studies cited are helpful in understanding the impact of compassion fatigue on hospic e social workers and in recognizing potential moderating factors, the limited number of empirical studies available clearly suggests a significant gap in the literature regarding this issue. In addition, the studies cited are consistently limited by a num ber of methodological issues. Many studies cited are limited by small sample size (Abendroth & Flannery, 2006; Alkema et al., 2008; Dane & Chachkes, 2001; Davidson, 1985; Harrison & Westwood, 2009; 28 Hernandez et al., 2007; Melvin, 2012; Supple - Diaz & Matt ison, 1992; Thompson, 2007), which creates suspicion regarding sample representativeness and generalizability (Thyer, 2001). In addition, many of the studies utilize self - selection sampling methods (Abendroth & Flannery, 2006; Badger et a., 2008; Melvin, 2012; Simon et al., 2005; Slocum - Gori et al., 2011; Supple - Diaz & Mattison, 1992) which raises concerns about selection bias (Engel & Schutt, 2009). One must consider if certain types of end - of - life care workers are more likely to respond to surveys genera lly , for example, those who are over - achieving or new to the field. One must also consider if the self - selecting respondents in these studies relating specifically to compassion fatigue and work stress are perhaps those end - of - life care workers who are s truggling with symptoms of compassion fatigue and find responding to a survey a means to safely communicate their struggles. In this case, prevalence of compassion fatigue would be over - represented in the studies. However, it may also be true that those end - of - life care workers experiencing symptoms of compassion fatigue may be too overwhelmed or dispirited to respond to a survey, and consequently, these studies under - represent the extent of compassion fatigue in end - of - life care workers. Many of the studies cited focus on specific geographic regions (Abendroth & Flannery, 2006; Dane & Chachkes, 2001; Harrison & Westwood, 2009; Hernandez, et al., 2007; Kim & Lee, 2009; Melvin, 2012; Sudeck, 2012; Supple - Diaz & Mattison, 1992; Thompson, 2007; Whitebird , et al., 2013). This suggests that one must utilize caution in generalizing study results from end - of - life care workers in Michigan (Supple - Diaz and Mattison, 1992), or Florida (Abendroth & Flannery, 2006), or California (Kim & Lee, 2009 ; Sudeck, 2012 ) or Minnesota (Whitebird et al., 2013), to end - of - life care workers in all regions of the United States. 29 An additional limitation of these studies is the consistent use of the cross - sectional designs which measure compassion fatigue at just one point in time. This certainly suggests the possibility that perceptions and experiences of compassion fatigue may change over time due to changing individual and organizational circumstances. In addition, the cross - sectional designs indicate that results must be considered with caution as causality cannot be assumed (Engel & Schutt, 2009). Review of Findings and Research Questions The nature of hospice work requires those social workers engaged in this role to be intimately connected with death on a daily ba sis, to be in perpetual touch with their own mortality, and to be exposed each work day to potentially psychologically stressful situations involving dying patients and grieving families. Compassion fatigue, defined as physical, emotional, and spiritual exhaustion, hopelessness, disconnection from others, and decreased capacity for empathetic attunement (Adams et al., 2006; Figley, 1995; Radey & Figley, 2007), is a useful construct for understanding the potential impact of this work on hospice social work ers. It is truly unfortunate that this knowledge has prompted so little investigation in to the prevalence and impact of compassion fatigue among hospice social work clinicians . It is also unfortunate that this lack of research has preempted considerati on of possible mitigating interventions such as professional social work supervision or compassion satisfaction. However, the limited studies available for review in this discussion do afford helpful insight and a solid foundation for the current inquiry . Compassion fatigue does appear to be a genuine risk for end - of - life caregivers (Abendroth & Flannery; 2005; Melvin, 2012; Potter et al., 2010; Thompson, 2007). In addition, various correlates identified in the existing literature suggest there are professional stress factors and 30 worker - related factors to consider in the discussion of compassion fatigue and end - of - life caregiving. Professional stress factors noted include multiple deaths of patients in a brief period of time (Abendroth & Flannery, 2005; Melvin, 2012; Supple - Diaz & Mattison, 1992) and high patient caseloads (Abendroth & Flannery, 2006; Supple - Diaz & Mattson, 1995; Sze & Ivker, 1986; DiTullio & MacDonald, 1999) . Worker - related factors noted as correlates to compassion fatigue in end - of - life care workers include the use of empathy and emotionally intense contact with patients and families (Abendroth & Flannery, 2006; Slocum - Gori et al., 2001), a personal history of loss (Simon et al., 2005; Supple - Diaz & Mattison 1992), and personal me ntal health concerns such as depression (Abendroth & Flannery, 2006; Lawson, 2008). These stressors certainly prompt the need to explore interventions that may act as protective mechanisms. The use of social work supervision appears promising in this r egard (Davidson, 1985; Dane & Chachkes, 2001; Kim & Lee, 2007; Whitebird et al., 2013). In addition, existing research suggests that assisting professional caregivers to recognize, celebrate, and strengthen the satisfying and positive aspects of their wor k with the suffering of others may serve as a protective mechanism against the more stressful and painful facets of their work (Alkema et al., 2008; Dane & Chachkes, 2001; Harr et al., 2014; Harrison & Westwood, 2009; Hernandez et al., 2007; Slocum - Gori et al., 2011) These findings prompt the following research questions and hypotheses that act to guide this research study: 1. What are the relationship s between compassion fatigue and professional stress and between compassion fatigue and worker - related fa ctors among hospice social workers? 31 H 1 : Professional stress has a positive association with compassion fatigue among hospice social workers. H 2 : Worker - related factors have a positive association with compassion fatigue among hospice social workers. 2. What is the impact of supervision on compassion fatigue and professional stress and on compassion fatigue and worker - related factors among hospice social workers? H 3 : Supervision moderates the relationship between professional stress and compassion fatigue among hospice social workers. H 4 : Supervision moderates the relationship between worker - related factors and compassion fatigue among hospice social workers. 3. What is the impact of compassion satisfaction on compassion fatigue and professional stress and on compassion fatigue and worker - related factors among hospice social workers? H 5 : Compassion satisfaction moderates the relationship between professional stress and compassion fatigue. H 6 : Compassion satisfaction moderates the relationship between work er - related factors and compassion fatigue. 32 CHAPTER THRE E: METHODOLOGY Study Design This study utilized a descriptive, cross - sectional design that included a web - based survey for data collection . This format offered the most ideal method to conduct the study within a restricted time period and with minimal resources. This design was also a reasonable option to explore compassion fatigue among hospice social workers as there is currently very little empirical knowledge about this issue wit hin this population (Engle & Schutt, 2009). Sample Social work clinicians employed with a hospice organization within the state of Michigan served as potential participants in this research study. S tudy participants were recruited using the Hospi hospice organizations within each state, including Michigan. HFA identified 140 hospice organizations located within the state of Michigan (HFA, 2015). However, a number of hospices were eliminated from the list by the researcher for various reasons . Twenty - two hospices listed individually were discovered to be part of larger hospice conglomerates operating under the same management; four were found to have offices in Michigan but primarily served patients in the bordering states of Indiana, Ohio and Northern Wisconsin ; and finally , 24 hospices were noted to be licensed to provide hospice care but offered primarily home care services. Con sequently, 90 Michigan hospice organizations were identified and invited to participate in this study. 33 Individual social workers from each hospice were eligible to participate if they met the following inclusion criteria : 1) a baccalaureate or graduate d egree in social work from an accredited School of Social Work; 2) current employment as a social worker within a Michigan hospice targeted by this study; 3) involvement in the direct care of dying patients and their families as part of the social work job description; and 4) willingness to read and comply with the provided Institutional Review Board (IRB) consent form. Procedures The procedures for this inquiry were based on the Dillman, Smyth, and Christian (2009) Tailored Design Method (TDM). A not able modification to the TDM for this web - based survey wa s attention to survey brevity. Multiple studies have suggested that time - consuming, complex surveys in the web environment have high dropout rates (Sue & Ritter, 2012). This qualification informed t he design of the survey itself as well as the content of all communication with participants. The study procedures were reviewed and approved by the Michigan State University Human Research Protection Program Institutional Review Board (IRB) as an exempt study (See Appendix B) . The survey was administered through the SurveyMonkey web site. This site was chosen as it permitted ease and freedom in survey construction and formatting and the ability to return data in a usable format. More importantly, Surve yMonkey afforded simple navigation and usability, which are essential in decreasing respondent frustration and drop out and increasing response rate (Sue & Ritter, 2012). Contact with the 3 5 Michigan hospices who agreed to participate in the study includ ed three separate email communications (See Appendix C ) timed in a manner that has been demonstrated to be ideal for optimal survey response (Sue & Ritter, 2012). An initial pre - notice 34 email outlined the purpose of the study, highlighted the importance of and contribution, and requested participation in the study. This initial communication was included as response to electronic surveys has been shown to be greatest when respondents are prenotified of the upcoming survey request (Sue & Ritter, 2012). A second email communication followed three days later. This communication again included the purpose of the participation. In addition, the link and the password to the survey website were provided. A final, reminder email communication was sent one week thereafter that again contained the link to the survey website and request ed participation. This single reminder strategy wa s recommended by the survey design literature, which suggest ed that additional reminders afford only very slight increases in response rates (Sue & Ritter, 2012). Neither the individual participants nor the hospices which they represented were identifiable from participation in the study or from data collected. Preserving this level of anonymity is preferable in a web - based survey, particularly when potentially sensitive questions are asked (Sue & Ritter, 2012). Since this approach also prevented tracking of response rates, the final reminder email was sent to each contact person at each participating hospice. Email communications as well as the introductory page of the survey highlighted the voluntary and anonymous nature of the survey and assured respond ents that lack of participation or survey response answers in no way negatively impacted their standing as hospice social work professionals. At the conclusion of the survey, respondents were invited to submit an email address in order to be entered into a lottery for electronic gift cards. A drawing for one of ten $20 Amazon electronic gift cards was offered as an incentive as the use of incentives has been demonstrated to 35 be an effective strategy for increasing survey response rates (Dillman et al. , 200 9; Ritter & Sue, 2012). Measures The survey for this study consisted of 62 questions (See Appendix D ) . In addition to soliciting demographic and general program information, the survey inquired about experiences of compassion fatigue, professional stre ss, empathy, compassion satisfaction, and the scope and perceived helpfulness of professional supervision. These variables were measured using both standardized instruments as well as measures created by the researcher and pre - tested with colleagues repre senting both practicing social work clinicians and research authorities as recommended by Dillman and associates (2009) and Sue and Ritter (2012). Dependent Variable Compassion Fatigue Following careful review of instruments available to measure compassion fatigue, the Professional Quality of Life Scale Version 5 (ProQOL - V5; Stamm, 2010) was selected for use in this inquiry. The ProQOL is a revision of the Compassion Fatigue Self - Test (CFST) developed by Figley in 1995. The CFST had proven to have various psychometric issues and to be quite time - consuming and cumbersome (Stamm, 2010). The ProQOL is a shorter, more user - friendly instrument without the psychometric issues of the CFST. Su rvey statements are framed in a positive manner, which eliminates concerns regarding the potentially adverse and damaging impact of negative statements about professional caregiving on respondents (Stamm, 2010). The ProQOL has been widely used in many co untries and is available in eleven languages (Stam m, 2010). Version five of this instrument has been utilized in past studies to assess burnout, compassion fatigue, and/or compassion satisfaction with professional counselors 36 (Lawson & Myers, 2011), correc tional mental health workers (Gallavan & Newman, 2013), health care workers including nurses, physicians, and nursing assistants (Smart, English, James, Wilson, & Daratha, 2014), heart and vascular nurses (Young, Derr, Cicchillo, & Bressler, 2011), and onc ology social workers (Joubert, Hocking, & Hampson, 2013). The ProQOL - V5 (Stamm, 2010) is a 30 - item, self - administered assessment designed for use with professionals exposed to primary and/or secondary trauma in the work setting. It is not a diagnostic tool and results are not associated with any diagnoses contained within the International Statistical Classification of Diseases and Related Health Problems 10 th Revision (ICD 10) or the Diagnostic and Statistical Manual of Mental Disorders (DSM - 5 , 2013). Each item of the ProQOL is measured on a five point Likert scale consisting of Never ( 1 ), Rarely ( 2 ), Somet imes ( 3 ), Often ( 4 ), and Very Often ( 5 ). The instrument contains three subscales of 10 items each: Burnout (B), Secondary Trauma/Compassion Fatigu e (CF), and Compassion Satisfaction (CS). The instrument affords a score for each subscale ranging from 10 to 50 with higher scores indicating higher levels of each construct. The ProQOL does not yield a total score a scores noted by the author (Stamm, 2010) for the subscales of Version 5 are .75, .81, and .88 respectively for the burnout, compassion fatigue, and compassion satisfaction subscales. These are consistent with various studies that have utilized the ProQ OL Version 5 . Gallavan and Newman ( 2013 ) noted a .8 8 compassion fatigue and .8 1 for compassion satisfaction in their study with correctional mental health professionals. Lawson and Myers ( 2011 compassion fatigue and .87 for compassion satisfaction, and .78 for burnout in their study with professional counselors. And finally, Smart and colleagues ( 2014) scores for compassion fat igue, compassion satisfaction, and burnout as .70, .90, and .81 37 respectively in their study with U. S. health care workers. Construct validity for the ProQol - V5 has been established in the literature as well, with low subscale intercorrelations, suggestin g that the subscales do indeed measure three distinct constructs (Lawson & Myers, 2011). The Compassion Fatigue/Secondary Trauma (CF) subscale measures the impact of work - related, secondary exposure to traumatically stressful events and includes such c oncerns as intrusive images, decreased empathetic attunement, sleep difficulties, and avoidance behaviors (Stamm, 2010). Elevated scores on this subscale represent greater incidence of compassion fatigue. The Compassion Fatigue/Secondary Trauma (CF) su bscale includes ProQOL - V5 items 2, 5, 7, 9, 11, 13, 14, 23, 15, and 28. Independent Variables Professional Stress Independent variables of interest were gleaned from the literature review and from practice experience of the researcher and include d those variables considered to be relevant to the experience of compassion fatigue by hospice social work clinicians. Professional stress variables of interest include d exposure to death, work overload, number of years working in the hospice setting, and employ ment status. These variables are operationalized as follows. Exposure to D eath Exposure to death wa s measured in this survey with one item inquiring about the number of deaths the respondent ha d been exposed to over the past 30 days. This question wa s included as various researchers have noted that professional caregivers who experience repeated exposure to patient deaths in a short period of time may be at greater risk for stress and compassion fatigue (Abendroth & Flannery, 2006; Melvin, 2012; Suppl e - Diaz & Mattison, 1992). Respondents we re able to enter any number up to three digits, providing a ratio level measurement. 38 Work O verload Work overload w a s measured using five questions from the Lait and Wallace (2002) study exploring how certain wo addition to work overload, Lait and Wallace (2002) examined the extent to which factors such as work autonomy, collegiality, role conflict, the routine nature of the work, and unmet expectations affect ed with human services workers as well as previous studies relating to job stress to develop a 75 - item survey instrument for their study. Zero - order correlations were comple ted with no evidence of collinearity (Lait & Wallace, 2002). The Lait and Wallace (2002) study instrument included f ive items specific to work overload. T wo of these five items were constructed specifically for the Lait and Wallace (2002) study and consider ed the number of meetings and the amount of paperwork required of respondents. The final three Lait and Wallace (2002) study items were adapted from an earlier study (Caplan, Cobb, & French, 1975) on work stress and consider ed the pace of work, th e heaviness of workload, and lack of time to complete necessary tasks. All five of these Lait and Wallace (2002) items measuring work overload were included in the present study. These five items were measured on a five point Likert scale consisting of Strongly Disagree (1), Disagree (2), Neither Agree Nor Disagree (3), Agree (4), and Strongly Agree (5). Higher scores on this subscale we five - item measure of work overload was .78. Work overloa d wa s included as a variable in this inquiry as the literature has suggested that end - of - life caregivers with consistently high workloads may be at greater risk for professional stress and fatigue (Abendroth & Flannery, 006; Supple - Diaz & Mattison, 1995). 39 Number of Years Working in Hospice Setting Th is professional stress variable solicit ed the length of time, in years, that respondents have been employed as a social worker within any and all hospice settings. Responses beg a n with Less Than One Year, and continue d with One to Two Years, Two to Five years, Five to Ten Years, and More Than Ten Years. This variable wa s included as previous research has suggested that lack of experience in providing end - of - life care creates greater stress and anxiety (Harper , 1977). In addition, research has also suggested that mental health professionals with several years of work experience with traumatized individuals are at higher risk for negative quality of life impacts (Ghahramanlou & Brodbeck, 2000; Harr et al, 2014; Moulden & Firestone, 2007). Employment S tatus Employment status solicit ed whether respondents we re employed as full time, part time, or as needed (PRN ) hospice social workers. This variable was included in this study due to the this question we re measured on the nominal level as Full Time, Part Time, or PRN. Worker - Related Variables Worker - related variables of interest include d the experience of the death of a family member or close friend within the past year, and a current diagnosis or diagnosis within the past 12 months of clinical depression. These variables we re operationalized as follows. 40 Personal Experience of the De ath of a Family Membe r or Friend This worker - related variable was included as a variable in this inquiry as previous research has suggested that personal history of loss is correlated with higher burnout and compassion fatigue scores (Simon, Pryce, Roff & Klemmack, 2005; Supple - Diaz & Mattison, 1992). This question was asked in a nominal level Yes or No manner. Current D iagnosis or D iagnosis within the past Year of C linical D epression This variable wa s asked in a nominal level Yes or No manner as well. This personal variable was included as prior studies have suggested a correlation between a diagnosis of clinical depression in end - of - life care workers and compassion fatigue (Abendroth & Flannery, 2006; Lawson, 2008; Whitebird et al., 2013). Empat hy Personal Distress subscales of the Interpersonal Reactivity Index (IRI), Version 2 (Davis, 1983). Research has suggested that individuals who are exceptionally empathetic may be more at risk for the development of compassion fatigue (Abendroth & Flannery, 2006; Badger et al., 2008; Slocum - Gori et al., 2011). Consequently, empathy was included as an independent variable in this study. The IRI is a 28 - item, self - administer ed questionnaire designed to measure differences in cognitive and emotional components of empathy. The instrument was developed for use in social psychological research and has been widely used to assess empathy of psychotherapists (Hassenstaub, Dziobek, Rogers, Wolf, & Convit, 2007; Hatcher, Favorite, Hardy, Goode, Deshetler, & Thomas, 2005; Wilbertz, Brakemeier, Zobel, Harter, &Schramm, 2009) physicians 41 (Yarnold, Bryant, Nightengale, & Martin, 1996), and medical students (Evans, Coman, & Stanley, 1988). The IRI contains four discrete, seven - item subscales measuring various separate but related facets of empathy. The Perspective Taking (PT) subscale measures the reported tendency to spontaneously adopt the perspectives of others and to perceive exp eriences from identify with characters in movies, novels, and other fictional situations. The final two subscales, Empathetic Concern (EC) and Personal Distress (P positive and negative emotional reactions to the difficult experiences of others. The Empathetic Concern subscale inquires about personal feelings of warmth, compassion, and concern for others, while the Personal Distress subscale measures personal feelings of anxiety and Concern and the Personal Distress subscales explicitly assess individual differences in emotional responses to observed emotionality in others. Consequently, these subscales appear ed most appropriate to the intent of this inquiry and were the only two included in the study survey . The use of individual subscales of the IRI is endorsed by the author (Davis, 1983), who suggests that each subscale separately measures various aspects of empathy. All items on the IRI are scored on a five point Likert scale ranging from Does No t Describe Me At All (0), to Describes Me Very Well (4). The instrument yields a score for each individual subscale ranging from zero to 28 , with higher scores indicating higher levels of the a spect of empathy being measured by each subscale . The IRI ha s excellent psychometric properties (Davis, 1983). Internal reliability coefficients using standardized alphas range from 42 .70 to .78 for the four subscales and t est - retest reliability correlations range from .61 to .79 for females and from .62 to .8 for males (Davis, 1983). Moderating Variables Professional Supervision The researcher conducted an exhaustive search of various databases including Health and Psychosocial Instruments (HaPI), PsychINFO, AgeLine, MedLine, Social Sciences Abstracts, EBSCOHost , and ProQuest in order to locate a standardized instrument to measure both scope and perceived helpfulness of professional social work supervision in the workplace. Several instruments were considered including the M anchester Clinical Supervision Scale - 26 [MCSS] (Winstanley & White, 2002), the Supervisory Styles Inventory (Friedlander & Ward, 1984), the Supervisory Working Alliance Inventory (Efstation, Patton, & Kardash, 1990) and the Supervisory Questionnaire (Worthington & Roehlke, 1979). While these instruments afford ed benefits such as ease of administration and scoring and strong reliability and validity, none adequately met the desired requirements of instrument brevity and relevance to scope and/or perceived helpfulness of social work supervision . Consequently, survey questions regarding scope and helpfulness of professional social work supervision in the workplace were generated by the researcher and informed by Sue and Ritter (2012) and Dillman, Smyth, and Christian (2009) methods for writing valid web - based survey questions. These methods consider addressing respondent - centered threats to validity such as social desirability, inaccurate estimates, and non - attitudes as well as guidelines for choosing words and forming questions such as applica bility to respondents, technical accuracy, simple and familiar wording, and brevity (Dillman et al., 2009; Sue & Ritter, 2012). 43 Information solicited regarding extent and helpfulness of professional social work supervision in the hospice setting included eleven questions regarding the professional discipline d routinely scheduled meetings with their supervisor, the number of contacts per week with the supervisor, the number of minutes per w eek of contact with the supervisor, the type of contact with the supervisor, the perceived helpfulness of the supervisor, the perceived supportiveness of the supervisor, and whether or not the supervisor assist ed with processing work stress. These supe rvision - related questions we re operationalized as follows. Professional discipline wa s measured in a nominal manner Social Work, Degree, No College Degree, Unknown, and Other. This question and these categories wer e based on research that has suggested that, despite recommendations by the N ASW (2005; 20 13), social workers in hospice settings are frequently supervised by disciplines other than social work, most often nursing (Reese & Raymer, 2004). The next four questions were intended to explore the extent and type of contact respondents ha d with t heir supervisor. Whether or not respondents ha d routinely scheduled supervisory meetings with their supervisor and whether or not respondents ha d unscheduled meetings with their supervisor were solicited using a simple nominal measure of Yes and No. Numb er of contacts per week with supervisor denote d the average number of times respondents ha d substantial interactions, both scheduled and unscheduled, with their supervisor related to work with patients and families and/or reactions to that work. This ques tion was measured using five exclusive categories beginning with Zero and progressing through One to Two, Three to 44 Four, Five to Six, and More than Seven . An additional question soliciting information regarding the total number of minutes per week of con tact with supervisor wa s included in order to provide greater insight and understanding concerning the extent of contact with the supervisor. This question wa s also measured using five exclusive categories beginning with Less than Fifteen Minutes, and con tinuing with 16 to 30 Minutes, 31 to 60 Minutes, 61 to 90 Minutes, More Than 90 Minutes. The type of contact with the supervisor denote d the manner in which respondents ha d scheduled and/or unscheduled substantial interactions with their supervisor regarding work with patients and families and/or reactions to that work. Responses represent ed a nominal measure using distinct categories. These responses include One on One, G roup, Telephone, supervisors assist them with processing the stress and emotionally challenging aspects of their work was solicited using a simple Nominal Yes or No response. The subsequent four questions consider perceived helpfulness and supportiveness of the supervisor. The perceived helpfulness of the supervisor indicate d ass isting them in coping with the demands of that work. Two questions regarding supervisor helpfulness we re asked in order to differentiate between helpfulness related to the work itself and helpfulness related to processing emotional reaction to the work. In order to maintain consistency among all survey questions, responses to both questions were on a five point Likert scale of Unhelpful (1), Somewhat Unhelpful (2), Neither Helpful Nor Unhelpful (3), Somewhat Helpful (4), Helpful (5). The perceived supp ortiveness of the supervisor denote d 45 individuals and their families and as well as with their emotional reactions to their work. Supervisor supportiveness was solicited using two questions that differentiated between supportiveness related to the work and supportiveness related to emotional reactions to the work. Responses were on a five point Likert scale of Unsupportive (1), Some what Unsupportive (2), Neither Supportive Nor Unsupportive (3), Somewhat Supportive (4), Supportive (5). Compassion Satisfaction Compassion satisfaction was measured using the Compassion Satisfaction (CS) subscale of the P roQOL - V 5 (Stamm, 2010), discussed previously. The Compassion Satisfaction subscale measures the pleasure or gratification individuals experience as they carry out their professional ssion fatigue subscales, scores range from 10 to 50 with elevated scores suggestive of higher rates of satisfaction within the professional caregiving role. ProQOL - V items 3, 6, 12, 16, 18, 20, 22, 24, 27, and 30 comprise the Compassion Satisfaction subsca le (Stamm, 2010). Demographic Variables Socio - demographic variables solicited for this inquiry include d gender, race, age, marital status, and highest social work degree obtained. In addition, respondents were asked to rate the importance of religion/ spirituality in their daily life. Options for this question include d Not Important, Somewhat Important, Important, and Very Important. The researcher included this question as prior research has suggested an inverse relationship between spirituality and compassion fatigue among medical personnel (Huggard, 2003 ) as well as a positive impact of religion and spirituality on mental and physical health ( Helm , Hays, Flint, Koenig, & Blazer , 2000; Koenig , George, & Peterson, 1998; Plante & Sherman, 2001). Fina lly , respondents we re 46 asked to indicate the size of the hospice, based on average daily census, at which they we re employed. The survey questions, design, and process were pre - tested by ten social work colleagues representing both practicing social work clinicians and research authorities as recommended by Dillman and associates (2009) and Sue and Ritter (2012). The pre - test proved helpful in identifying minor concerns regarding question clarity and understanding, ease of use, and survey completion time and prompted minor changes to the survey w ording. Data Analysis At the conclusion of data collection, data were downloaded from the Survey Monkey website to Statistical Package for the Social Sciences (SPSS 22.0) software. Data were cleaned to identify e xtreme and missing values. No extreme values or outliers were noted. A small number of participants (n = 2) had missing values on one or more variables. List wise deletion was utilized in these cases (deVaus, 2001). No participants were dropped from the dataset due to missing values. Following data cleaning, data were coded appropriately using numeric codes and reverse coding as needed. A survey codebook was developed delineating the structure, contents, and layout of the data file (Sue & Ritter, 2009). Descriptive statistics explored sample characteristics utilizing frequencies and percentages. Overall mean ratings were identified for the Compassion Fatigue and Compassion Satisfaction subscales of the ProQOL - V5 (Stamm, 2010) and these s cores were compared with benchmark data available for the ProQOL instrument. Overall mean ratings were identified for the Empathetic Concern and the Personal Distress subscales of the IRI (Davis, 1983) and for all other independent and moderating variable s. 47 Bivariate analyses were completed to explore whether significant associations existed between the outcome variable (compassion fatigue) and the various predictor variables. One way analyses of variance (ANOVA), Pearson correlations and independent samp le t tests were utilized to test hypothesis one regarding the association between professional stress and compassion fatigue and hypothesis two regarding the association between worker - related factors and compassion fatigue. These relationships were inves tigated using a probability level of .05 to determine statistical significance. A priori power analys e s using SPSS SamplePower were completed to determine necessary sample size to test study hypotheses. For hypotheses one and two testing relationships between compassion fatigue and the various professional stress and worker - related variables, power analysis suggested that a sample size of 47 was required to detect a moderate effect ( r =. 30) with 80 percen t power to correctly reject the null hypothesis for correlations. For ANOVA tests, power analysis indicated a sample size of 24 would provide 80 percent power to detect a plus or minus one unit mean difference. For independent t tests related to hypothes es one and two, power analysis suggest ed that a sample size of five provided 80 percent power to detect a significant mean difference of two using a probability level of .05 . For hypotheses three, four, five, and six regarding potential moderating influence of compassion satisfaction and professional supervision, a priori power a nalysis suggested that a sample size of 85 was required to detect a moderate effect size ( r = . 3 0) w ith 80 percent power to = 55) did not meet the sample size requirement . Consequently, the interaction models testing hypotheses three, four, five, and six w ere not able to be completed. 48 CHAPTER FOUR: RESULTS This chapter presents information regarding study response rates and descriptive statistics for all study variables. In addition, this chapter discusses the outcomes of statistical analyses and presents results using various tables. Response Rate Init ially, 35 of the ninety invited Michigan hospice organizations (3 8.89 %) agreed to participate in this study. However, one hospice that initially agreed to participate later emailed the researcher and stated that the survey link had not been forwarded to t heir social workers due to lack of time related to an overwhelming amount of work. Consequently, a final total of 34 hospice organizations of eligible Michigan hospice organizations participated in the study (3 7.78 %) . Fifty - five social workers from these 34 hospices completed the study survey. Sample Characteristics Survey participants included 51 women (92.7%) and four men (7.3%). Twelve participants (21.8%) identified their marital status as single, 40 (72.7%) reported being married or partnered, an d three (5.5%) reported being divorced. Forty - nine respondents identified as Caucasian (89.1%), two as African American (3.6%), two as Hispanic, and two as Other (3.6%). Participants ranged in age from 25 years to 68 years with a mean age of 42 years (SD 11.5). Religion/Spirituality was very important to them. Ten respond ents (18.2%) indicated that Religion/Spirituality was important, 11 reported (20%) it was somewhat important, while seven (12.7%) respondents indicated that Religion/Spirituality was not important to them. Table One provides information regarding sample characteristics. 49 Table 1 Sample Characteristics (N = 55) Variable n Percent M SD Gender Female Male 51 92.7% 4 7.3% Age Marital Status 42.1 11.5 Single 12 21.8% Married or Partnered 40 72.7% Divorced 3 5.5% Race/Ethnicity White or Caucasian 49 89.1% Black or African American 2 3.6% Hispanic or Latino 2 3.6% Other Race 2 3.6% Highest Social Work Degree MSW BSW Importance of Religion/Spirituality 50 90.9% 5 9.1% Not Important Somewhat Important Important Very Important 7 12.7% 11 20.0% 10 18.2% 27 49.1% The size of the participating hospices varied widely , with 10 (18.2%) respondents indicating that they worked for a hospice with an average daily census (ADC) of zero to 50, 11 respondents (20%) reported working for a hospice with an ADC of 51 to 100, 12 respondents (21.8%) worked for a hospice with an ADC o f 101 to 150, two (4%) for a hospice with an ADC of 151 to 200 ADC, and 17 (30.9%) participants indicated working for a hospice with a 200+ ADC. Table Two provides information regarding ADC of hospic es represented in this study. 50 Table 2 Average Daily Census of Represented Hospices (N = 55) ______________________________________________________________________________ Census n Percent 0 to 50 10 18.2% 51 to 100 11 20.0% 101 to 150 12 21.8% 151 to 200 2 3.6% 200+ 17 30.9% Missing 3 5.4% ______________________________________________________________________________ Descriptive Statistics Compassion Fatigue and Compassion Satisfaction The C ompassion F atigue and C ompassion S atisfaction subscales of the ProQOL (Stamm, 2010) were used in this study to measure major variables of interest. Reliability was calculated for each subs cale and found to be moderate to high (Enge l & Schutt, 2009) . The Compassion were consistent with F atigue subscale (0.81) and the Compassion Satisfaction subscale (0.88). The mean score on the Compassion Fatigue subscale of the ProQ O L (Stamm, 2010) was fixed to 50 with a standard deviation of 10 according to scoring instr uctions provided by the author . Compassion fatigue scores below 43 indicate d low compassion fatigue, scores between 44 and 56 designate d moderate compassion fatigue, and scores over 57 were suggestive of high compassion fatigue (Stamm, 2010). Compassion fatigue scores for respondents in this study ranged from 33.3 to 76.4. Twelve respondents (21.8%) scored in the low range of compassion 5 1 fatigue, 31 (56%) scored in the moderate range, and 12 respondents (21.8%) scored in the high range of compassion fatigue. These results are noted in Table Three. Table 3 Compassion Fatigue Scores (N = 55) Compassion Fatigue Score n Percent Low (<43) 12 21.8% Moderate (44 - 56) 31 56.4% High (>57) 12 21.8% The mean score for the Compassion Satisfaction subscale was fixed to 50 with a standard deviation of 10 according to scoring instructions provided by the author (Stamm, 2010). Compassion satisfaction scores below 43 indicate d low compassion satisfaction, scores between 44 and 56 designate d moderate compassion satisfaction, and scores over 57 we re suggestive of high compassion satisfaction (Stamm, 2010). Compassion satisfaction scores ranged from 31.3 to 63.5 in this study . Eleven respondents (20%) scored in the low range for compassion satisfaction, 24 (43.6%) scored in the moderate range, and 20 respondents (36.4%) scored in the high range for compassion satisfaction. Compassion s atisfaction scores are summarized in Table Four. Table 4 Compassion Satisfaction Scores (N = 55) ______________________________________________________________________________ Compassion Satisfaction Score n Percent Low (<43) 11 20% Moderate (44 - 56) 24 43.6% High (>57) 20 36.4% _______________________________________________ _______________________________ 52 Compassion fatigue and compassion satisfaction wer e demonstrated to be inversely correlated in this study ( r = - .376, p < .05) ; as compassion satisfaction for study respondents increase d , their compassion fatigue decrease d . Professional Stress Regarding e xposure to death as a professional stress, the number of deaths experienced by study respondents in the p ast 30 days ranged from two to 40, with an average of 14.1 caseload deaths in the past 30 days (SD 10.5). Work overload as a professional stressor was measured using five questions from the Lait and Wallace (2002) study exploring how certain work conditio Reliability was calculated for these questions and found to be high (Enge l Possible combined s cores for these questions ranged from zero to 20 with higher s cores suggestive of higher workload. Mean score for respondents in this study was 10.9 (SD 4.2), indicating moderate work overload among study participants. The final two aspects of professional stress included employment status and number of years w orking in hospice care. Forty - two respondents ( 76.4 %) indicated that they were employed as full time hospice social workers, four (7. 2 %) reported being employed part time, and six (1 0 . 9 %) respondents identified their employment status as PRN ( a s n eeded). The number of years respondents reported being employed by any hospice organization varied widely. Six (1 0 . 9 %) indicated being employed with hospice for less than one year, four (7. 2 %) for one to two years, 17 (3 0 . 9 %) for two to five years, 11 (2 0 . 0 %) f or five to 10 years, and 13 (2 3 . 6 %) respondents reported being employed by hospice for 10 years for more. Table Five outlines descriptive information regarding professional stress. 53 Table 5 Professional Stress (N = 55) Variable n Percent M SD Caseload Deaths in Past 30 Day 14.1 10.5 Work Overload (0 20) 10.9 4.2 Employment Status Full Time 42 76.4 % Part Time 4 7. 3 % PRN 6 1 0 . 9 % Missing 3 5.4% Years Employed With Hospice Less than 1 Year 6 1 0 . 9 % 1 to 2 Years 4 7. 3 % 2 to 5 Years 17 3 0 . 9 % 5 to 10 Years 11 2 0 . 0 % 10 Years+ 13 2 3 . 6 % Missing 4 7.3% Worker - Related Factors Twenty - one respondents (38.2%) reported having experienced the death of a family member or close friend within the past 12 months. Twelve respondents (21.8%) indicated having been diagnosed with depression within the past twelve months, while 25 participa nts (45.5%) indicated having been diagnosed with depression in their lifetime. Respondent empathy was measured using the Empathetic Concern and the Personal Distress subscales of the Interpersonal Reactivity Index [IRI] (Davis, 1983). Reliabil ity was calculated for each subscale and found to be acceptable (Enge l & Schutt, 2009) . The Empathetic 0 .717 in this study. These s cores we re consistent with 54 subscale (0.73) and for the Personal Distress subscale (0.75). Scores for both the Personal Distress an Empathetic Concern subscales of the IRI range from zero to 28 with higher scores indicative of higher levels of empathy . Mean score s for the Personal Distress and Empathetic Concern subscales for women suggested by the author (Davis, 1983) are 12.28 (SD 5.01) and 21.67 (SD 3.83) respectivel y . Personal Distress mean score in this study was 19.36 (SD 2.88) , while the mean score for the Empathetic Concern subscale was 16.21 (SD 3.12) , suggesting moderate levels of empathy among study respondents. The Personal Distress subscale mean score in this study was slightly higher than the mean score of suggested by the author (Davis, 1983) , while the Empathetic Concern subscale score in this study was somewhat lower than the mean score provided by the author. The instrument author (Davis, 1983) does not provide cut scores for any of the IRI subscales, but indicates that the mean scores are as above. Table Six provides descriptive statistics for worker - related factors. Table 6 Worker - Related Factors (N = 55) Variable n Percent M SD Death of Family Member/Friend Past 12 Months 21 38.2% Depression Diagnosis P ast 12 Months 12 21.8% Depression Diagnosis Lifetime 25 45.5% Total Worker Empathy Personal Distress (Range: 0 - 28) 19.36 2.88 Empathetic Concern (Range: 0 - 28) 16.21 3.12 55 Supe rvision supervision in their work with hospice patients. Twenty - seven respondents (49.1%) reported that their supervisor held a degree in nursing while 17 (30.9%) indicated that their supervisor held a (9.1%) reported that their supervisor held a college degree other than Nursing or Social Work and one (1.8%) indicated that their supervisor did not have a college degree. Twenty respondents (36.4%) indicated that they had routinely scheduled meetings with their supervisor while 43 (78.2%) reported that they had unscheduled meetings with their supervisor. Twenty - nine respondents (52.7%) indicated having one to tw o contacts with their supervisor each week, 11 (20%) reported having three to four, three (5.5%) indicated having five to seven, and three (5.5%) reported having more than seven contacts. Seven (12.7%) participants indicated having no contact at all wit h their supervisor each week. Length of contact with supervisors varied widely. Twenty - one respondents (38.2%) indicated having less than 15 minutes of contact with their supervisor each week, 16 (29.1%) indicated having 16 to 30 minutes of weekly cont act, nine (16.4%) reported 31 to 60 minutes, one (1.8%) reported 61 to 90 minutes, and six (10.9%) reported having more than 90 minutes of weekly contact with their supervisor. One - on - one contact (17 or 30.9%) and group meetings (16 or 29.1%) were the mo st frequently cited types of contact with supervisors. In addition, 12 respondents (21.8%) identified email as their most frequent type of contact with their supervisor while seven respondents (12.7%) identified telephone. Twenty - four respondents (43.6%) responded affirmatively to whether or not their supervisor assisted them with processing the stress and emotionally challenging aspects of their 56 work. In regards to supervisor helpfulness, scores ranged from zero to eight w ith a mean score of 5.4 (SD 2.6). Supervisor supportiveness scores ranged from zero to eight with a mean score of 5.8 (SD 2.5). Supervis ion characteristics are presented in T able Seven . Table 7 Supervision Characteristics (N = 55) Variable n Percent M SD Supervisor Degree Nursing 27 49.1% Social Work 17 30.9% Other 5 9.1% None 1 1.8% Missing 5 9.1% Routinely scheduled meetings Yes 20 3 6 . 4 % No 33 6 0 . 0 % Missing 2 3.6% Unscheduled meetings Yes 43 78 . 2% No 10 1 8 . 2 % Missing 2 3.6% Number of Contacts Each Week None 7 12.7% 1 to 2 29 52.7% 3 to 4 11 20.0% 5 to 7 3 5.5% 7+ 3 5.5% Missing 2 3.6% Length of Contact Less than 15 Minutes 21 38.2% 16 to 30 Minutes 16 29.1% 31 to 60 Minutes 9 16.4% 61 to 90 Minutes 1 1.8% 90 Minutes + 6 10.9% Missing 2 3.6% 57 Type of Contact One on One 17 30.9% Group 16 29.1% Telephone 7 12.7% Email 12 21.8% Missing 3 5.5% Helpfulness of Supervisor 5.4 2.6 Supportiveness of Supervisor 5.8 2.5 Bivariate Analysis Bivariate analys i s was completed to explore whether significant associations existed between compassion fatigue and professional stress variables and between compassion fatigue and worker - related variables . One way analyses of variance (ANOVA), Pearson correlations and independent sample t tests were utilized to test hypothesis one regarding the association between professional str ess and compassion fatigue and hypothesis two regarding the association between worker - related factors and compassion fatigue. These relationships were investigated using a probability level of .05 to determine statistical significance. Results of these tests i ndicated that work overload was significantly associated with compassion fatigue ( r = .380, p , suggesting that as work overload increased, compassion fatigue increased. In addition, diagnosis of clinical depression within the past 12 m onths was significantly associated with compassion fatigue ( t = - 4.11, p < .05 ) . Compassion fatigue was higher in those with a depression diagnosis in last 12 months (59.42 , SD 9.54) compared to those not diagnosed with depression in past 12 months (47.3 , SD 8.79) . N o statistically significant relationship was found between compassion fatigue and exposure to death, employment status, number of years employed with hospice , death of a family member or friend 58 within the past 12 months, worker empathy, or li fetime diagnosis of depression. R esults of bivariate analysis are outlined in Table Eight . T able 8 Bivariate Analysis of Compassion Fatigue and Professional Stress and Compassion Fatigue and Worker - Related Factors (N = 55) Compassion Fat igue r or t Professional Stress Caseload Deaths in Past 30 Days - 0 .044 Work Overload 0 .380* Number of Years Employed - 0 .652 Employment Status - 1.512 Worker Related Factors Death of family Member or Friend Past 12 Months - 0 . 906 Depression Diagnosis Past 12 Months - 4.11* Lifetime Depression Diagnosis - 1.99 Worker Empathy 0 .123 * p<.05 Compassion fatigue and compassion satisfaction mean scores were calculated comparing various demographic, supervisory, and hospice agency characteristics. No significant differences were noted in compassion fatigue mean scores based on respondents age, marital status, social work degree, number of years e mployed with hospice, employment status, ADC of the hospice, supervisor degree, frequent scheduled or unscheduled meetings with supervisor, or number of contacts per week with supervisor. Compassion fatigue mean scores were found to be 59 significantly diffe rent ( p < .05) based on the importance of religion/spirituality and type of contact with supervisor. No significant differences were noted in compassion satisfaction mean scores based on respondent age, marital status, importance of religion/spiritualit y, social work degree, number of years employed with hospice, employment status, supervisor degree, frequent scheduled or unscheduled visits with supervisor, number of contacts per week with supervisor, or type of contact with supervisor. Compassion satis faction scores were found to be significantly different ( p < .05) based on ADC of hospice that employed respondents. Results of compassion fatigue and compassion satisfaction mean score comparisons are presented in Table Nine. Table 9 Comparative Compassion Fatigue and Compassion Satisfaction Scores (N = 55) Variable Compassion Fatigue Compassion Satisfaction M SD M SD ______________________________________________________________________________ Age < 35 Years 51.04 10.91 49.43 10.18 35 to 50 Years 48.97 9.09 50.92 8.79 >50 Years 50.27 10.68 49.34 11.91 Marital Status Married/Partnered 49.84 9.64 49.93 10.55 Single/Divorced 48.80 9.69 49.37 9.53 Soc ial Work Degree MSW 49.39 8.13 43.76 16.08 BSW 50.06 10.24 50.62 9.20 Importance of Religion Very Important/Important 51.99 9.81 * 50.91 9.34 Somewhat/Not Important 45.92 9.37 * 48.12 11.29 60 ADC of Hospice 0 to 100 48.87 10.16 49.70 10.14 101 to 200 50.79 13.03 56.14 6.11 * 200+ 51.04 8 .10 46.83 9.85 * Employment Status Full Time 54.45 5.97 46.76 11.21 Part Time/PRN 49.06 10.84 51.39 9.21 Number of Years Employed w/ Hospice <5 Years 49.18 11.00 51.66 8.63 >5 Years 51.09 9.74 49.91 10.45 Supervisor Degree Social Work 51.58 10.47 48.86 10.01 Nursing 50.12 10.04 49.75 10.48 Scheduled Meeting s w/Supervisor Yes 49.97 10.08 49.26 10.80 No 50.34 10.57 51.59 8.39 Unscheduled Meetings w/Supervisor Yes 45.24 9.26 48.26 13.28 No 51.25 10.13 50.57 9.14 Number of Contacts Per Week None 54.57 13.65 56.14 5.09 1 to 2 48.63 10.17 47.15 9.89 3 or More 50.41 8.10 52.09 10.34 Type of Contact One to One 47.81 10.49 * 49.24 9.83 Group 47.25 9.08 * 49.43 12.34 Telephone/Email 54.74 9.83 * 51.19 8.24 ______________________________________________________________________________ * p < .05 61 CHAPTER FIVE: DISCUSSION T his study sought to explore the prevalence of compassion fatigue among hospice social w orkers and to investigate potential moderating factors in the experience of compassion fatigue by these s ocial workers. not all study hypotheses could be tested. However, the data was able to yield notable and helpful information regarding co mpassion fatigue among hospice social workers. This chapter offers detailed discussion of study results and recommendations , outlines study limitations , and suggests directions for future research. Major Findings Results of this study suggest ed that compassion fatigue is indeed a genuine concern among hospice social workers . P revious studies have noted the prevalence of compassion fatigue among end - of - life care nurses ( Abendroth & Flannery , 2005 ; Melvin, 2012; Potter et al., 2010 ), hospice caregivers (Thompson, 2007), and health care professionals in hospitals, care homes, and community - based health services (Slocum - Gori et al., 2011) . H owever this study is among the first to substantiate the reality of compassion fatigue specifi cally among social w orkers in the hospice setting. This finding, considered in combination with other study results, offers a necessary first step that may be helpful in identifying appropriate interventions to counter the negative consequences of compassion fatigue. One such intervention may be a focus on compassion satisfaction. While hospice social workers in this study exhibited moderate to high levels of compassion fatigue, they were also found to have moderate to high levels of compassion satisfaction. In addition, a moderate negative correlation was noted between compassion fatigue 62 and compassion satisfaction ( r = - .376, p < .05) , suggesting that as compassion satisfaction increase d , compassion fatigue decrease d . Th ese result s, which were consistent with previous studies t hat ha ve found similar negative correlations between compassion fatigue and compassion satisfaction among hospice workers (Alkema et al., 2008), palliativ e care workers (Slocum - Gori et al., 2011) and social work students ( Harr et al., 2014) , appear to be a promising option for assisting hospice social workers to cope with this professional hazard. Th e negative correlation of compassion fatigue and comp assion satisfaction highlights the importance of finding ways to assist hospice social workers to recognize and connect to the positive and affirming aspects of their work. T his may potentially be achieved through a variety of workplace activities. S uperv isors and co - workers alike may encourag e and recogniz e individual and collective hospice social worker strengths , knowledge, and abilities . Particularly helpful and successful interventions with patients and families may be routinely and publicly recognized by supervisors and the hospice interprofessional team. Social work or interprofessional staff meetings may consistently offer opportunity for social workers to share meaningful or significant encounters with patie nts and families as a means to build positive feelings about their work. And finally, s upervisors may encourage personal reflection and journaling of meaningful interactions with patients and families as well as important lessons An additional study finding that may offer insight in to potential protective mechanisms against compassion fatigue wa s the significant correlation found between w ork overload and compass ion fatigue. This finding was consiste nt with previous studies that suggested that high caseloads (Abendroth & Flannery, 2006; Supple - Diaz & Mattison, 1995) high work volumes (DiTullio & MacDonald, 1999), and the progressive constriction of time available for attention 63 to patient needs and emo tional processing (DiTullio & MacDonald, 1999) were primary sources of stress for hospice workers. Work overload as a correlate to compassion fatigue noted in this and other stud ies is suggestive of the need f or hospice administrators and social work supervisors alike to maintain vigilance regarding work responsibilities that may contribute to work overload for their social workers. These may include high number of workplace meetings, large amounts of requisite paperwork , and high caseload volume, acuity, and/or turnover. Monitoring of and adjustment to these potential work overload issues may be helpful in lowering risk for compassion fatigue among hospice social workers . This correlation between work overload and comp assion fatigue may also have implications for hospice agency policy regarding social worker and other interprofessional staff patient visits per day or per wee k would be helpful. Perhaps hospice agency policies that cap the number of times and the number of hours social workers and other interprofessional staff are on - call would be beneficial in reducing the risk of work overload in the hospice setting. Or p er haps policies that offer generous earned time off packages and requirements that hospice social workers and interprofessional staff utilize their earned time off may assist hospice staff to feel less overwhelmed by their work. Depression diagnosis within the past 12 months was found to be positively correlated with compassion fatigue in this study as well . This correlation, which was consistent with previous research studies that found self - identified depressed respondents demonstrated higher level s of compassion fatigue than those who did not report a diagnosis of depression (Abendroth & Flannery, 2006; Lawson, 2008 ; Whitebird et al., 2013 ) , raises compelling considerations 64 regarding compassion fatigue and mental health. One may consider the manne r in which depression and compassion fatigue are related. Does depression make one more susceptible to compassion fatigue or might the experience of compassion fatigue create conditions ripe for the development of depression? Do the two develop independe nt of or in conjunction with one another? It is interesting to note that while the majority of respondents in this study exhibited moderate to high compassion fatigue, just 12 respondents reported a depression diagn osis within the past 12 months. This ma y be an indicat ion that compassion fatigue is not necessarily a pivotal factor in the development of depression or similar mental health concerns. This finding also suggests the need for supervisors to be sufficiently connected to their social workers and other professional staff in order to monitor for symptoms of depression or other mental health concerns. While supervisors cannot and should not act as therapists to their s ocial workers, they certainly can and should be alert to symptoms of depression or other mental health concerns in their staff . In noting these symptoms, supervisors may facilitate referrals to employee assistance program s or to other mental health profes sionals to ensure that their staff receive the resources and support necessary to successfully cope with these issues . In addition, hospice organizations would appear to benefit from proactive policies and activities aimed at promoting mental health and preventing mental disorders in the hospice setting . These may include in - service trainings by mental health professionals regarding signs and symptoms of depression , anxiety or other mental disorders, periodic depression and compassion fatigue screenings, facilitation of team building and peer bonding activities to build community and reduce isolation , and provision of space and opportunity for group support and debriefing. Failure to work to prevent mental health issues or intervene when these become apparent may have grave consequences on several levels. I ndividual social worker s may become 65 apathetic toward their work and their patients. The H ospice interprofessional team may not fu nction optimally with a social worker who is struggling with the ramifications of depression and compassion fatigue. Hospice agencies may experience the fiscal implications of decreased productivity and increased sick days associated with depressed social workers. And most importantly, vulnerable hospice patients and families may receive less than the highest quality, most compassionate care from a social worker besieged by depression and/or compassion fatigue . Finally, the co mbination of depression and compassion fatigue would certainly appear to put hospice social workers at risk for leaving hospice work for a less taxing or stressful field of practice. This is a critical issue as hospice s attempt to decrease social work attrition as well as attract n ew social workers in order t o meet the growing demand for hospice services (NHPCO, 2012). It was interesting to note that while t correlated with decreased compassion fatigue in this study , mean compassion fatigue scores were found to be higher among those respondents who identified the type of contact they most often have with their supervisor as email/telephone than among those who identified either one to one or group contact . This finding appears to sug either in individual or group meetings, may be an important consideration in assisting hospice social workers to cope with the stress of their work. While email and telephone communication offer conven ient and efficient means for social workers and their supervisors to address the logistical and practical issues of hospice social work , this finding may suggest that face to face contact with a caring supervisor is helpful and necessary in processing the stress and emotionally Face to face communication, either individually 66 or in group format , offer s benefit s that communication via technology does not. For example, face to face communication affords the opp ortunity to read non - verbal communication and body language and to give and receive non - verbal support through eye contact and physical presence (Fussell, 2002) . Perhaps these benefits provide enhanced opportunity for hospice social workers to process and cope with the stress of their work, resulting in lower compassion fatigue scores. An additional consideration related to individual or group contact versus email/telephone communication and compassion fatigue may be related to the nature of hospice work . All members of the hospice interprofessional team , including social workers carry out their work wherever patients reside. This can include private homes, hospice inpatient facilities, hospitals, skilled nursing facilities, and assisted living facilities. With the exception of hospice inpatient facilities, hospice social workers and other interprofessional team members carry out their work primarily in isolation from their co - workers ; they are alone with dying patients and their families withou t the support of other interprofessional team members . It may be that this isolation contributes to their compassion fatigue as they have little opportunity to interact with and support one another in their day - to - day work. Perhaps contact with superviso rs on an individual basis and with supervisors and co - workers in a group setting offers opportunity for processing work stress in a manner that that is missing from their day to day work lives. C ompassion satisfaction mean scores were significantly lower for respondents who reported working for a hospice organization with an ADC of 200 or more than they were for respondents from smaller hospice organizations. T h is result raise s several compelling questions regarding the size of hospice organization s . What is the relationship between the ADC of a patients and their families but also for its professional staff? Do larger hospice organizations 67 require their social workers to manage higher caseloads and if so, does this detract from opportunities for social workers to connect to the more meaningful and satisfying aspects of their work? Is it possible that large hospice organizations wi th a gr eat number of social workers struggle to stay sufficiently connected to these social workers to m onitor their w ork stress and provide support and assistance when and where needed? Since there are no studies investigating the relationship between ADC and c ompassion fatigue in the hospice setting, e mpirical evidence answering these and similar questions may prove helpful in understanding the nuances of compassion fatigue and compassion satisfaction and how hospice administrators can provide the best environm ent and most helpful resources for keeping their staff emotionally and physically healthy for long - term hospice work. Additional Findings A number of hypothesized correlates were found to have no statistically significant relationship with compassion fatigue in the present study . However these results provided helpful information as well. No correlation was noted between caseload deaths and compassion fatigue. This was a surprising result as a number of p rior research studies found that frequent and ongoing exposure to death by end - of - life caregivers was a significant correlate in the development of compassion fatigue (Abendroth & Flannery, 2006; Melvin, 2012; Supple - Diaz & Mattison, 1995). A potential explanation for this inconsistency ma y have been the number of caseload deaths experienced by respondents in the present study . While previous empirical studies do not provide details regarding precisely how many patient deaths study participants were exposed to that resulted in correlat ions with compassion fatigue, perhaps these numbers were significantly higher than the mean exposure to death (14.1, SD: 10.5) in the present study. 68 An additional consideration that may speak to this inconsistency is that prior research considered this correlation from the perspective of nurses rather than social workers. Perhaps the physical care that nurses provide to hospice patients impacts them in a different manner than does the emotional care and support that social workers provide. Perhaps hosp smaller caseloads (NHPCO, 2014) allow opportunity for them to develop more intimate relationships with their patients and families that results in more intense emotional reactions when those patients die. Perhaps social workers training and education exposes them to loss and grief and related coping mechanisms that are helpful to them in dealing with patient deaths. Professional stress variables of n umber of years employed with hospice and employment status were not found to be correlated with compassion fatigue in this study. While there was no prior empirical evidence for such correlation s , the researcher found these results to be unexpected as one may surmise that daily and long term exposure to dying and death through ful l time or longstanding hospice employment may create increased opportunity for the development of compassion fatigue. However, i t may be that full time or longstanding e mployment in hospice care forces social workers to develop and utilize resources and supports that allow them to successfully cope with their day to day exposure to death and dying. Full time and long term hospice social workers have likely had increased opportunity for on - the - job training and education regarding death and dying , professional grief, and coping mechanisms . They may also have had increased opportunity to develop strong and enduring bonds with peers who offer a safe space to debrief about the challenges and pressures of hospice work. Fina lly, f ull time and long term hospice social workers may also have had more abundant occasions to have meaningful interactions with their dying patients and their family members that sustain them in their work , despite the stress of their ongoing exposure t o death and dying. 69 Death of family member or friend was also not found to be correlated with compassion fatigue in this study. This finding differs from personal experiences of loss were correlated with higher burnout scores among oncology social workers and from Supple - that suggested that personal l demands of their work. This is an interesting finding and one that prompts the researcher question if perhaps this result may be indicative of positive connections between these two variabl es. Does the experience of personal loss for hospice social workers provide opportunity for them to more genuinely understand and empathize with their patients and families in a manner that they may have been unable to without the experience of personal l oss ? Does significant personal loss necessitate the development of coping skills to more capably manage the str ? Answers to these questions may provide some insight in to this inconsistency with prior research. An additional c onsideration may be the very small number of studies that have considered death of a family member or friend as a correlate with compassion fatigue. Clearly, additional studies are necessary to substantiate the correlation of these two variables. Surprisingly, w orker empathy was not found to be correlated with compassion fatigue in this study. This result is at odds with prior studies suggesting positive correlations between empathetic attunement and compassion fatigue in hospice and pallia tive care workers ( Slocum - Gori et al., 2011) , hospice nurses ( Abendroth & Flannery , 2006) , and hospital social workers ( Badger et al., 2008) . It also differs from theoretical suggestions that the risk for compassion fatigue increases relative to empathetic aptitude (Adams et al., 2006; Figley, 2002). While this difference cannot be fully understood or explained, it must be noted that, with the exception of 70 one study with social workers, prior research regarding the relationship of these two variables has been carried out with hospice or other end - of - life care nurses . This raises questions regarding the potential differences between nurses and s ocial workers education and training . Social work education , particularly at the graduate level, includes information and training about transference and counter - transference issues as well as the importance of boundaries in professional relationships (C SWE, 2008). One must consider if th is training enhances social - empathizing with their patients and families and to maintain appropriate emotional boundaries. Study Limitations As with any study, t his study is not without its shortcomings. A significant activities in effort to increase the sample size. These included both email and telephone contacts direc tly to Michigan hospice agencies requesting participation as well as attempts to access Michigan hospice social workers through the Hospice and Palliative Care Association of Michigan (HPCAM). Initially, the researcher received an affirmative response from HPCAM administration regarding their willingness to forward the study link to their social work members. However, for unknown reasons, the survey link was not forwarded to HPCAM social work members and requests for information about this from HPCAM a dministration went unanswered. While the exact reasons why Michigan hospice organizations or indiv idual hospice social workers chose not to participate in this study cannot be known, the researcher has posited possible explanations. Perhaps the d ecision not to participate was indicative of work overload and time constraints for hospice organizations a nd individual hospice social workers. Previous 71 research has suggested that high caseloads (Abendroth & Flannery, 2006; Supple - Diaz & (DiTullio & MacDonald, 1999) are issues faced by many hospice workers and work overload was one of the significant correlations with compassion fatigue noted in this study. raises additional questions regarding the sample itself. One must consider if perhaps both the hospice organizations that agreed to participate as well as the individua l respondents were those who were feeling less overwhelmed with their work and , consequently , found adequate time to participate. In this case, one might surmise that the prevalence of compassion fatigue within hospice social work may be under - represented in this study , while compassion satisfaction may be over - represented . An additional potential explanation for lack of participation in this study may be the content of the study itself. Perhaps hospice administrators and supervisors who received the requ est to participate felt uneasy about the investigation of compassion fatigue among their social workers. Perhaps inquiry in to if and how professional supervision may be helpful in moderating compassion fatigue was information hospice administrators and s upervisors did not want to have due to its potential organizational and financial implications. This potential explanation causes the r esearcher to question if she may have had greater response from hospice social workers if she had solicited their part icipation directly rather than through their agency or supervisor. survey itself . Since the request to participate in this study came through their supervisor or their hospice administrator, it is certainly possible that participants responded to survey questions in a 72 more positive manner than they would have if the request had come directly from the researcher. Perhaps they may have been concerned that honest response s regarding their feelings of s tress about their work would have negative implications for them with their supervisor or their agency, despite reassurances from the researcher regarding the anonymous and confidential nature of the survey . In this case, co mpassion fatigue may under - represented in this study while compassion satisfaction may be over - represented. A final explanation for lack of participation in this study and the resulting small sample may perhaps have been related to negative experiences with previous research studies. One invited hospice organization declined to participate in this study citing adverse experiences with previous researchers and research studies and a second hospice organization requested the ion due to prior negative research experiences. One may surmise that there were other hospice organizations that did not respond likewise but may have had similar negative experiences that influenced their decision not to participate. Regardles ( N = 55) created a situation in which a number of study hypotheses could not be analy zed and gleaned from the study. In addition to this significant limitation, other limitations were noted as well. A quantitative, cross - sectional design was chosen for this study as it offered an ideal method to conduct the study within a restricted time period and with minimal resources. H owever, the cross - sion fatigue at just one point in time . This did not allow for a more comprehensive understanding of if and how perceptions 73 and experiences of compassion fatigue may have change d over time due to shifting individual and organizational circumstances. In a ddition, t he quantitative nature of the study did not offer opportunity for the researcher to follow up with participants regarding their responses or to probe for deeper and richer understanding regarding both their positive and negative t houghts and feel ings about their work with dying individuals. T he web - based nature of the study was helpful in addressing potential issues of sample attrition and subject fatigue (Engel & Schutt, 2009) as well as eliminating costs associated with postage, paper, and dat a entry . It also significantly shorten ed the time required to gather data (Dillman et al., 2009; Sue & Ritter, 2012). However, this manner of collecting data also created risk of coverage bias due to the exclusion of those hospices and social workers w ith limited access to the internet . An additional limitation of this study wa s its focus on a specific geographical region. B oth the hospice organizations located within the state of Michigan a s well as the individuals employed therein may have possess ed geographic and regional idiosyncrasies that may potentially have impacted study results and generated concern regarding generalizability . However, all hospice organizations nationwide are subject to the same stringent federal guidelines and conditions of participation for staffing and the provision of care to patients and families (NHPCO, 2014). Consequently, hospice organizations located within Michigan were likely to be at least partially representative of hospice organizations nationwide in terms of staffing levels and care provided. Implications for the Social Work Profession This exploratory study was intended to contribute to the slight but growing body of knowledge regarding the issue of professional stress and compassion fatigue within the hos pice 74 social work setting. In addition, this study sought to provide preliminary information regardi ng potential interventions and resources that might serve as protective mechanisms against compassi on fatigue such as consistent access to and use of professional social work supervision and the promotion of compassion satisfaction . While the study was limited by the aforementioned issues, it did provide valuable implications for hospice social workers specifically as well as the broader social work pro fession. Social work practice with the terminally ill is a highly specialized field requiring p ractitioners who are comfortable with and capable of providing compassionate and competent psychosocial care to individuals in their last days and weeks of l ife. The findings noted in this and other studies substantiate the reality of compassion fatigue for end - of - life social workers (Abendroth & Flannery, 2006; Supple - Diaz & Mattison, 1992; Thompson, 2007) . The relationships found in this study between comp assion fatigue and work overload a s well as worker diagnosis of depression along with factors s uggest ed by previous research (Abendroth & Flannery, 2005; Lawson, 2008; Melvin, 2012; Simon et al., 2005; Slocum - Gori et al., 2001; Supple - Diaz & Mattison, 1992 ) suggest to this researcher that enhanced efforts and additional resources are necessary to ensure that hospice social workers are sufficiently equipped for and supported in their work. This may be accomplished through multiple avenues. Social workers themselves must learn to identify signs and symptoms of compassion fatigue and to understand that these symptoms are not representative of personal flaw s or weakness. Social workers must then be encouraged to seek guidance and assistance from trusted co - w orkers and supervisors. Schools of social work must consider expanding their curricula to include courses intended to prepare social work students for end - of - life caregiving. Courses that include information 75 regarding psychosocial dynamics of death and dying, potential ethical dilemmas faced by dying patients and their families as well as the practitioners who work with and care for them, the grief process, and best - practice end - of - life interventions would appear to be compulsory in this preparation. In addition, schools of social work would do well to provide education and information to social work students regarding the risks for and presentation of compassion - being . In ad dition, information regarding the concept of compassion satisfaction and the importance of connecting would be helpful . It is this researcher bear a responsibility for preparing students to recognize and manage compassion fatigue as a potential professional hazard. Hospice agencies, too, must implement policies and practices that encourage, support, and assist social workers and other inter professional team members in identifying, normalizing, and coping with the demands of their work. Such policies and practices may include caps on caseloads and monitoring of caseload acuity, required attendance at inservices and staff meetings, and genero us earned - time - off packages to allow for adequate time away from work to renew and reenergize. In addition, hospice administrators would do well to direct resources toward developing and implementing trainings regarding the unique stress of hospice work, compassion fatigue as a very real risk, practical ways to cope, and how compassion satisfaction may serve to protect hospice workers from this professional risk. Finally, hospice agency administrators could ensure that professional supervision is availabl e for social workers, either face - to face or in a more fiscally conservative manner such as via telephone, skype, and email. It is this researcher most beneficial in this regard : the administrative function to monitor caseload size and makeup; 76 the educational function to provide orientation and ongoing instruction regarding best practice skills in end - of - life social work; and the supportive function to assist social workers in r ecognizing and managing the stress of their work , to monitor for symptoms of depression and/or compassion fatigue, and to increase positive feelings about their work and their professional identity (Kadushin, 1992b; Kadushin & Harkness, 2002). The broade r hospice community would do well to address the issue of compassion fatigue in the hospice setting as well. The National Hospice and Palliative Care Organization (NHPCO) along with state hospice organizations such as Hospice and Palliative Care of Michig an (HPCAM) may advocate for and support policies and activities within individual hospice organizations that seek to address this issue. They may ensure that information and trainings are routinely available at their annual conferences and on their websit es. Finally, NHPCO, HPCAM and other state level hospice organizations may encourage and provide funding for ongoing research regarding compassion fatigue and best practices to assist hospice staff in coping with the stress of their work. Preparing social workers for and normalizing the experience of compassion fatigue through S chool of Social W ork curricula and hospice agency policies and practices appears imperative to precluding social workers from interpreting the difficult emotional an d psychological effects of their work as evidence of their personal or professional incompetence and from potentially leaving their positions as hospice social workers or departing the social work profession altogether. In addition, these suggested change s appear necessary in order to ensure that hospice social workers are able to practice in a manner that allows for dying individuals and their families to receive the highest quality, most compassionate care and that allows practitioners to thrive both per sonally and professionally. 77 Finally, social work practitioners and scholars alike must engage in broad and ongoing empirical research in this area of study. Such research is necessary to provide solid and unassailable evidence of the reality of compassion fatigue in hospice social work, the associated emotional, psychological, and fiscal costs of this issue, and the use of professional supervision , a focus on compassion satisfaction, and other interventions as means to mitigate this professional risk. It is this researcher acquired and conveyed, that suggested additions to school of social work curricula and introduction of hospice agency policies and practices intended to addre ss th ese issue will be considered and implemented. Recommendations for Future Research Clearly, additional inquiry into the risks, presentation, and mitigation of compassion fatigue is imperative if hospice social workers are to thrive both professio nally and personally and if dying individuals and their families are to receive the best possible care in the final days of their lives. It is evident from the dearth of current research that is specifically related to hospice social work and compassion f atigue that additional empirical studies are necessary to provide enhanced verification of the prevalence of compassion fatigue among hospice social workers and improved understanding of associated risk factors. In addition, future studies must explore what types of support and resources may be most beneficial in the prevention and mitigation of compassion fatigue. Social workers are a heterogeneous group with varying needs and preferred ways of coping. Consequently, a variety of resources and support s must be empirically explored including professional supervision, a focus on compassion satisfaction, informal support and collegiality, self - care activities, in - service education, and routine acknowledgement and normalization of patient losses and associa ted professional grief. 78 In order to be most helpful and meaningful, both qualitative and quantitative studies must be utilized. Quantitative research studies would offer additional insight in to the prevalence of compassion fatigue in hospice social wo rk and the various correlates associated this professional risk among this population. Qualitative research studies including focus groups and intensive interviewing (Engle & Schutt, 2009) with hospice social workers would afford further understanding of the experiences of hospice social workers from their perspective and provide information about this issue that may not be captured in researcher - determined categories and measurements of quantitative methods. The limitations of this study and the consist ent limitations noted in previous studies available in this area of inquiry must be addressed in future research as well. Studies with larger samples and greater geographic representation are necessary to improve representativeness and generalizability (T hyer, 2001). In addition, future studies with adequate sample sizes to investigate potential moderating factors of compassion fatigue, such as professional supervision and compassion satisfaction, are necessary. F uture studies must also address the issu e of self - selected and purposive sampling methods in order to address concerns about selection bias. Finally, the use of longitudinal study designs rather than, or in combination with, cross sectional designs would be helpful in understanding the developm ent and presentation of compassion fatigue over time (Eng e l & Schutt, 2009). Finally, it must be noted that with the exception of nurses, none of the current research considers if or how various disciplines providing end - of - life care may differ in the type or extent of professional stress and compassion fatigue they experience. Each discipline involved in caring for the dying interacts with and experiences these patients and their dying process in a di stinct manner and context. Consequently, i t would appear that each discipline is likely to 79 perceive and experience the stress and impact of their work in very different ways. Additional research regarding how each hospice discipline perceives and copes with their work would be helpful in providin g a richer and more complete understanding of compassion fatigue and its impact on hospice workers. Conclusion T his study is among the first to substantiate the reality of compassion fatigue among hospice social workers and to suggest compassion satisfaction as a potential protective mechanism. Despite its limitations, this study provides a solid base for fu ture re search regarding compassion fatigue among social workers in the hospice setting. Clearly, additional inquiry in to the risks, presentation, and mitigation of this professional hazard is imperative if hospice social workers are to thrive both personally an d professionally and if dying individuals and their families are to receive the best possible care in the final days of their lives. 80 APPENDICES 81 APPENDIX A R equest for Participation Email Communication Hello - I am a doctoral student at Michigan State University. I am currently conducting a dissertation study regarding professional stress in hospice social work via a brief, online survey. Might you be willing to connect me with the individual in your agency who super vises your hospice social workers so that I might discuss with her or him having (Individual Hospice Name) social workers participate in this brief online survey? I am hopeful that their thoughts and opinions will contribute to improved understanding of t he gifts and stressors of hospice social work and to recognizing what might help hospice social workers to thrive in their work. Thank you for your time and assistance. Please feel free to contact me if you have questions or concerns. Sally Pelon, LMS W Doctoral Candidate Michigan State University School of Social Work 616 - 403 - 3253 pelonsal@msu.edu 82 APPENDIX B I nstitutional Review Board Approval December 12, 2014 To: Anne Hughes 240 Baker Hall East Lansing MI 48824 Re: IRB# x14 - 1249e Category: Exempt 2 Approval Date: December 11, 2014 Title: Compassion fatigue in Hospice social work: Potential moderating factors The Institutional Review Board has completed their review of your project. I am pleased to advise you that your project ha s been deemed as exempt in accordance with federal regulations. The IRB has found that your research project meets the criteria for exempt status and the criteria for the protection of human subjects in exempt research. Under our exempt policy the Princip al Investigator assumes the responsibilities for the protection of human subjects in this project as outlined in the assurance letter and exempt educational material. The IRB office has received your signed assurance for exempt research. A copy of this sig ned agreement is appended for your information and records. Renewals : Exempt protocols do not need to be renewed. If the project is completed, please submit an Application for Permanent Closure . Revisions : Exempt protocols do not require revisions. Howev er, if changes are made to a protocol that may no longer meet the exempt criteria, a new initial application will be required. Problems : If issues should arise during the conduct of the research, such as unanticipated problems, adverse events, or any prob lem that may increase the risk to the human subjects and change the category of review, notify the IRB office promptly. Any complaints from participants regarding the risk and benefits of the project must be reported to the IRB. Follow - up : If your exempt project is not completed and closed after three years, the IRB office will contact you regarding the status of the project and to verify that no changes have occurred that may affect exempt status. Please use the IRB number listed above on any forms submi tted which relate to this project, or on any correspondence with the IRB office. Good luck in your research. If we can be of further assistance, please contact us at 517 - 355 - 2180 or via email at IRB@msu.edu. Thank you for your cooperation. 83 Sincerely, Harry McGee, MPH SIRB Chair cc: Sally Pelon 84 APPENDIX C E mail Contacts with Participating Hospices Initial Email (Supervisors/Administrators Forwarded to Social Workers) Hello - I am inviting you to participate in a research study regarding professional stress among Hospice social workers. The purpose of this study is to gather information about the experience of compassion fatigue in hospice social work as well as to understand what things might serve to protect social workers from experiencing compassion fatigue in their work. I am asking you to participate because you are currently employed as a social worker with a you a link to the online survey and I would very much appreciate your taking about 15 minutes of your time to complete the survey, anonymously and confidentially. Participation in this study is completely voluntary. Thank you in advance for your part icipation in this study. Your thoughts and opinions will contribute to improved understanding of the gifts and stressors of hospice social work and to recognizing what might help hospice social workers to thrive in their work. Please feel free to contact me with any questions or concerns. Sally Pelon, LMSW Doctoral Candidate Michigan State University 616 - 331 - 6588 pelonsal@msu.edu Second Email Hello - A few days ago you received an email requesting your participation in my research study regarding compassion fatigue in hospice social work. Below you will find a link to the survey. The survey will take about 15 minutes to complete and is completely vol untary, anonymous and confidential. The link to the study that may be copied and pasted in to your browser is https://www.surveymonkey.com/s/3G7SH2S and the password is 'hospice' Than k you in advance for your participation in this study. Your thoughts and opinions will contribute to better understanding the gifts and stressors of hospice social work and to recognizing what might help hospice social workers to thrive in their work. Pl ease feel free to contact me with any questions or concerns. 85 Sally Pelon, LMSW Doctoral Candidate Michigan State University 616 - 331 - 6588 pelonsal@msu.edu Final Email Hello - Last week I sent you an email with a l ink to my research study regarding compassion fatigue in our understanding of the benefits and risks of hospice social work and to recognize what might a ssist hospice social workers to thrive in their work. Below is the link to the online survey. If you have already completed the survey, thank you! I appreciate your time and your contribution to understanding this important issue. Please feel free to contact me with any questions or concerns. Sally Pelon, LMSW Doctoral Candidate Michigan State University 616 - 331 - 6588 pelonsal@msu.edu 86 APPENDIX D Study Survey Hello - You are being asked to participate in a research study regarding professional stress among Hospice social workers. The purpose of this study is to gather information about the experience of compassion fatigue in hospice social wor k as well as to understand what things might serve to protec t social workers from the experience of compassion fatigue in their wo rk. You have been selected as a research participant because you a re currently employed as a social worker with a hospice organization in the state of Michigan. Your participa tion will i nvolve completing an online survey through SurveyMonkey; the survey should take about 15 minutes to complete. Participation in this research is completely voluntary. You have the right to decline to participate an d you may also change your mind about part icipating a t any time during the study and withdraw. You may also choose to not answer sp ecific questions. If you choose not to participate, there will be no negative impacts on your work or on your standing as a hospic e social worker in the state of Michi gan. Neither you as an individual participant or the hospice which you represent are identifiable from participation in the study. The collected results will be presented in combined form as part of a PhD dissertation, in publications in peer - reviewed jou rnals, and in presentations at professional conferences. There are no costs to you as a participant other than the time it takes for you to complete the survey. If you choose to parti cipate, you may also choose to submit your email address for a chance to win a drawing for one of ten $20 Amazon.com electronic gift cards. I f you have concerns or questions about this research, please contact the researchers: Sally Pelon, LMSW 401 W. Fulton Avenue Grand Rapids, MI 49504 616 - 331 - 6588 pelonsal@msu.edu Anne, K. Hughes, PhD, LMSW Baker Hall, 655 Auditorium Road, Room 240 East Lansing, MI 48824 517 - 432 - 6181 hughesa@msu.edu If you have questions or concerns about your role and rights as a research participant, would like to obtain information or to offer input, or to register a complaint about this study, you may contact, anonymously if you wish, the Michigan State University Human Research Protection 87 program at 517 - 355 - 2180, fax 517 - 432 - 4502, or email irb@msu.edu or mail 207 Olds Hall, MSU, East Lansing, MI 4882 4. By clicking on the link below, you voluntarily agree to participate in this online study. You may exit the survey at any time by clicking the 'exitsurvey' tab located at the top right corner of each page. Thank you for your time! 1. What is your gend er? Female Male Transgender Declined 2. What is your marital status? Single Married Partnered Divorced Separated Widowed Declined 3. What is your race? American Indian or Alaskan Native Asian Black or African American Native Hawaiian or Other Pacific Islander White Hispanic/Latino Multiracial: Asian/Black or African American Multiracial: Asian/White Multiracial: Black or African American/White Multiracial: Other Combination Declined 4. What is your age in years? 5. How important is religion/spiritual ity in your life? Not important Somewhat Important Important Very Important Declined 88 6. What is the highest social work degree you have achieved? BSW MSW DSW/PhD Declined When you help people, you have direct contact with their lives. As you may have found, your compassio n for those you help can affect you in positive and negative ways. Below are some questions about your experiences, both posi tive and negative, as a helper. Consider each of the following questions about you and your current work situa tion. 7. I am preoccupied with more than one person I help Never Rarely Sometimes Often Very Often 8. I get satisfaction from being able to help people. Never Rarely Sometimes Often Very Often 9. I jump or am easily startled by unexpected sounds. Never Rarely Sometimes Often Very Often 10. I feel invigorated after working with those I help. Never Rarely Sometimes Often Very Often 11. I find it difficult to separate my personal life from my life as a helper. Never Rarely Sometimes Often Very Often 12. I think I might have been affected by the traumatic stress of those I help. Never Rarely Sometimes Often Very Often 13. Because of my helping, I have felt "on edge" about various things. Never Rarely Sometimes Often Very Often 14. I like my work as a hel per. Never Rarely Sometimes Often Very Often 15. I feel depressed because of the traumatic experiences of the people I help. Never Rarely Sometimes Often Very Often 16. I feel as though I am experiencing the trauma of someone I have helped. Never Rarely Sometimes Often Very Often 17. I am pleased with how I am able to keep up with helping techniques and protocols. Never Rarely Sometimes Often Very Often 18. My work makes me feel satisfied. Never Rarely Sometimes Often Very Often 89 19. I have happy thought s and feelings about those I help and how I could help them. Never Rarely Sometimes Often Very Often 20. I believe I can make a difference through my work. Never Rarely Sometimes Often Very Often 21. I avoid certain activities or situations because they remind me of frighten ing experiences of the people I help. Never Rarely Sometimes Often Very Often 22. I am proud of what I can do to help. Never Rarely Sometimes Often Very Often 23. As a result of my helping, I have intrusive, frightening thoughts. Nev er Rarely Sometimes Often Very Often 24. I have thoughts that I am a "success" as a helper. Never Rarely Sometimes Often Very Often 25. I can't recall important parts of my work with trauma victims. Never Rarely Sometimes Often Very Often 26. I am happy that I chose to do this work. Never Rarely Sometimes Often Very Often Please use the scale to rate the degree to which each statement applies to you in your current work situation. 27. I have to attend too many meetings in this job. Strongly Disagree Di sagree Neither Agree Nor Disagree Agree Strongly Agree 28. My job involves a lot of paperwork. Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 29. I have to work very fast to get everything done in my job. Strongly Disagree Dis agree Neither Agree Nor Disagree Agree Strongly Agree 30. My workload is too heavy in my job. Strongly Disagree Disagree Neither Agree Nor Disagree Agree Strongly Agree 31. I do not have enough time to get everything done in my job. Strongly Disagree Dis agree Neither Agree Nor Disagree Agree Strongly Agree The following statements inquire about your thoughts and feelings in a variety of situations. For each item, indicate how well it describes you by choosing the appropriate number on the scale. Read eac h item carefully before responding. Answer as honestly as you can. 90 32. I often have tender, concerned feelings for people less fortunate than me. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 33. Sometimes I don't feel very sorry for people when they are having problems. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 34. In emergency situations, I feel apprehensive and ill - at - ease. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 35. When I see someone being taken advantage of, I feel kind of protective towards them. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 36. I sometimes feel helpless when I am in the middle of a very emotional situation. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Wel l 37. When I see someone get hurt, I tend to remain calm. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 38. Other people's misfortunes do not usually disturb me a great deal. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 39. Be ing in a tense, emotional situation scares me. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 40. When I see someone being treated unfairly, I sometimes don't feel very much pity for them. Does Not Describe Me At All 0 1 2 3 4 Describes Me Ver y Well 41. I am usually pretty effective in dealing with emergencies. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 42. I would describe myself as a pretty soft - hearted person. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 43. I tend to lose control during emergencies. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well 44. When I see someone who badly needs help in an emergency, I go to pieces. Does Not Describe Me Well 0 1 2 3 4 Describes Me Very Well The next set of questions is related to your current supervisor and the supervision you receive in your workplace. Please read the following questions carefully and respond with the answer that best reflects your current supervisory situation. 91 45. What professional de gree is held by your current supervisor? MSW BSW Masters Degree in Nursing (MSN) Bachelors Degree in Nursing (BSN) No College Degree Unknown Other (please specify) 46. Do you have routinely scheduled meetings with your current supervisor? Yes No 47. Do y ou have unscheduled meetings with your supervisor? Yes No 48. How many contacts do you typically have with your supervisor i n an average week, scheduled or unscheduled, related to your work with patients and families and/or your reactions to that work? Zero One to Two Three to Four Five to Six Seven or More 49. What is the average number of minutes you spend with your supervisor i n an average week, scheduled or unscheduled, related to your work with patients and families and/or your reactions to that work? Less Than 15 Minutes 16 to 30 Minutes 31 to 60 Minutes 61 to 90 Minutes More Than 90 Minutes 50. What type of contacts do you most often have with your supervisor. Please check only one. One on One Group Telephone Email Other (please specify) 51. D oes your supervisor assist you in processing the stress and emotionally c hallenging aspects of your role as a hospice social worker? Yes No 92 52. When you meet with your supervisor to discuss your work with patients and families, how helpful do you find your supervisor to be? Not Helpful Somewhat Unhelpful Neither Helpful Nor Unhelpful Somewhat Helpful Helpful N/A 53. When you meet with your supervisor to discuss your emotional reactions to your work, how helpful do you find your supervisor to be? Not Helpful Somewhat Unhelpful Neither Helpful Nor Unhelpful Somewhat Helpful Helpful N/A 54. When you meet with your supervisor to discuss your work with patients and families, how supportive do you find your supervisor to be? Not Supportive Somewhat Un supportive Neither Supportive Nor Unsupportive Somewhat Supportive Supportive N/A 55. When you meet with your supervisor to discuss your emotional reactions to your work, how supportive do you find your supervisor to be? Not Supportive Somewhat Unsupp ortive Neither Supportive Nor Unsupportive Somewhat Supportive Supportive N/A 56. What is the average daily census of the hospice where you are currently employed? Zero to 50 51 to100 101 to 150 151 to 200 More Than 200 57. Approximately how many patients on your caseload have died in the past 30 days? 93 58. What is your current employment status? Full Time Part Time PRN (Please indicate how many hours you typically work per week) 59. How many years total have you been employed with any hospice organization? Less Than One Year One to Two Years Two to Five Years Five to Ten Years More Than Ten Years 60. Have you experienced the death of a family member or close friend in the p ast 12 months? Yes No 61. Do you currently or have you had a diagnosis of depression in the past 12 months? Yes No 62. Have you had a diagnosis of depression in your lifetime? Yes No Thank you for your time! If you would like to be entered in to a drawing for one of ten $20 Amazon electronic gift cards, please send an email to pelonsal@msu.edu and simply include 'Survey Drawing' as your subject - no need to include any other text. Winners will be notified after the Survey closes . 94 REFERENCES 95 R EFERENCES Abendroth, M., & Flannery, J. (2006) Predicting the risk of compassion fatigue: A study of hospice nurses. Journal of Hospice and Palliative Nursing, 8 (6), 346 - 356. Adams, R.E., Boscarino, J.A., & Figley, C.R. (2006) Compassion fatigue and psychological distress among social workers: A validation study. American Journal of Orthopsychiatry, 76 (1), 103 - 108. Ainsworth, M., Blehar, M., Waters, E., & Wall, S. (1978) Patterns of attachment. Hillsdale , NJ: Earlbaum Publishing Alkema, K., Linton, J.M., & Davies, R. (2008) A study of the relationship between self - care, compassion satisfaction compassion fatigue and burnout among hospice professionals. Journal of Social Work in End - of - Life and Palliat ive Care, 4 (2), 101 - 119. American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5 th ed.). Washington, DC. Badger, K., Royse, D., & Craig, C. (2008). Hospital social workers and indirect trauma: An explo ratory study of contributing factors. Health and Social Work, 33 (1), 63 - 71. Baird, K., & Kracen, A.C. (2006). Vicarious traumatization and secondary traumatic stress: A research synthesis. Counselling Psychology Quarterly, 19 (2), 181 - 188. Barker, R.L. (2003). The social work dictionary (5 th Ed.). Washington, DC: National Association of Social Workers Press. Bell, H., Kulkarni, S., & Dalton, L. (2003) Organizational prevention of vicarious trauma. Families in Society, 84(4), 463 - 470. Bennett, C.S. (2008) Attachment - informed supervision for social work field education. Clinical Social Work, 36, 97 - 107. Bern - Klug, M., Gessert, C, & Forbes, S. (2001). The need to revise assumptions about the end - of - life: Implications for social work practice. Health & Social Work, 26 (1), 38 - 48. Bowlby, J. (1969) Attachment and loss: Attachment (Vol.1, 2 nd ed). New York, NY: Basic Books. Caplan, R.D., Cobb, S., & French, J.P.R. (1975). Relationship of cessation of smoking with job stress, p ersonality, and social support. Journal of Applied Psychology, (60), 211 - 219. 96 Centers for Disease Control and Prevention (2014). Atlanta, GA. Retrieved from http://www.cdc.gov/ Centers for Medicare and Medicaid Services (2013). Baltimore, MD. Retrieved from http://www.cms.gov/ Christ, G.H., & Blacker, S. (2005) Series introduction: The profession of social work in end - of - life and palliative care. Journal of Palliative Medi cine, 8 (2), 415 - 417. Cicirelli, V.G. (2006). Fear of death in mid - old age. Journal of Gerontology, 61B (2), 75 - 81. Clark, E.J. (2011). The challenge of sadness in social work. Retrieved from NASW News: http://www.socialworkers.org/pubs/news/2011/05/sadness - social - work.asp Clark, E.J. (1989). Offsetting burnout in the thanatologic setting: Recognition and emphasis of vention. Loss, Grief, and Care, 3 (1 - 2), 115 - 123. Clark, L., Leedy, S., McDonald, L., Muller, B., Lamb, C, Mendez, T., Kim S., & Schonwetter, R. (2007) Spirituality and job satisfaction among hospice interdisciplinary team members. Journal of Palliat ive Medicine, 10 (6), 1321 - 1328. Conner, S.R. (2007) Development of hospice and palliative care in the United States. Omega, 56 (1), 89 - 99. Corcoran, K.J. (1989). Interpersonal stress and burnout: Unraveling the role of empathy. Journal of Social Behavior and Personality, 4 (1), 141 - 144. Council on Social Work Education (2008) Educational Policy and Accreditation Standards. Alexandria, VA. Retrieved from: http://www.cswe.org Dane, B., & Chachkes, E . (2001) The cost of caring for patients with illness: Contagion to the social worker. Social Work in Health Care, 33 (2), 31 - 51. Davidson, K.W. (1985). Social work with cancer patients: Stresses and coping patterns. Social Work in Health Care, 10(4), 73 - 82. Davidson, K.W., & Foster, Z. (1995) Social work with dying and bereaved clients: Helping the workers. Social Work in Health Care, 21 (4), 1 - 15. Davis, M.H. (1980). A multidimensional approach to individual differences in empathy. Journa l of Social and Administrative Sciences: Catalog of Selected Documents in Psychology, 10, 1 - 17. 97 Davis, M.H. (1983). Measuring individual differences in empathy: Evidence for a multidimensional approach. Journal of Personality and Social Psychology , 44 (1), 113 - 126. Davis, M.H., Soderlund, T., Cole, J., Gadol, E., Kute, M., & Myers, M. (2004). Cognitions associated with attempts to empathize: How do we imagine the perspective of another? Personality and Social Psychology Bulletin, 30, 1625 - 16 35. DeLoach, R. (2003). Job satisfaction among hospice interdisciplinary team members. American Journal of Hospice and Palliative Care, 20 (6), 434 - 440. deVaus, D. (2002). Analyzing social science data: 50 key problems in data analysis. Thousand Oaks, CA: Sage Publications. Dillman, D.A., Smyth, J.D., & Christian, L.M. (2009) Internet, e mail, and mixed - mode surveys: The tailored design method (3 rd ed). Hoboken, NJ: Wiley Publications. DiTullio, M., & MacDonald, D. (1999) The struggle for the soul of hospice: Stress, coping and change among hospice workers. American Journal of Hospice and Palliative Care, 16 (5), 641 - 655. Engel, R.J., & Schutt, R. K. (2009) The practice of research in social work (2 nd ed.). Thousand Oa ks, CA: Sage Publications. Efstation, J.F., Patton, M.J., & Kardash, C.M. (1990). Supervisory Working Alliance Inventory. Retrieved from PsycTESTS. doi: 10.1037/t02667 - 000 Evans, B.J., Coman, J., & Stanley, R.O., (1988). Scores on the profile of nonver bal sensitivity: A sample of Australian medical students. Psychological Reports, 62, 903 - 906. Feifel, H. (1977). Death and dying in modern America. Death Education, 1 (1), 5 - 14. Figley, C.R. (1988) Victimization, trauma, and traumatic stress. Th e Counseling Psychologist, 16 (4), 635 - 641. Figley, C.R. (1995) Compassion fatigue: Toward a new understanding of the costs of caring. In Stamm, B.H. (Ed.), Secondary traumatic stress: Self - care issues for clinicians, researchers and educators (pp. 3 - 29). Baltimore, MD: Sidran Press. - care. Journal of Clinical Psychology, 58 (11), 1433 - 1441. Fonagy, P. (2001) Attachment theory and psychoanalysis. New York, NY: Other Press. Friedlander, M.L., & Ward, L.G. (1984). Development and validation of the supervisory styles inventory. Journal of Counseling Psychology, 31 (4), 541 - 557. 98 Freudenberger, H.J. (1974) Staff burnout. Journal of Social Issues, 30 (1 ), 159 - 165. Fussell, S. R. (2002). The verbal communication of emotions: Interdisciplinary perspectives . Mahwah, NJ : L. Erlbaum Associates Gallavan, D.B., & Newman, J.L. (2013). Predictors of burnout among correctional mental health professionals. Psychological Services, 10 (1), 115 - 122. Ghrahmanlou, M., & Brodbeck, C. (2000). Predictors of secondary trauma in sexual assault trauma counselors. International Journal of Emergency Mental Health, 2 (4), 229 - 400. Grasso, A. J. (1994). Management style, jobs satisfaction, and service effectiveness. Administration in Social Work, 18 (4), 89 - 105. Greene, R.R. (1991) Supervision in social work with the aged and their families. Journal of Gerontological Social Work, 17 (1/2), 139 - 144. Harkness, D. (1995) The art of helping in supervised practice: Skills, relationships, and outcomes. The Clinical Supervisor, 9 (2), 31 - 42. Harkness, D., & Hensley, H. (1991) Changing the focus of social work supervision: Effects on client satisfactio n and generalized contentment. Social Work, 36 (6), 506 - 512. Harper, B.C. (1977) Death: The coping mechanism of the health professional . Greenville, SC: Southeastern University Press. Harr, C.R., Brice, T.S., Riley, K., & Moore, B. (2014). The i mpact of compassion fatigue and compassion satisfaction on social work students. Journal of the Society for Social Work and Research, 5 (2), 233 - 251. Harrison, R.L. & Westwood, M.J. (2009). Preventing vicarious traumatization of mental health therap ists: Identifying protective practices. Psychotherapy Theory, Research, Practice, and Training, 46 (2), 203 - 219. Hassenstab, J., Dziobek, I., Rogers, K., Wolf, O.T., & Convit, A. (2007). Knowing what others know , feeling what others feel: A controlled study of empathy in psychotherapists. Journal of Nervous and Mental Disorders, 195 (4), 277 - 281. Hatcher, S.L., Favorite, T.K., Hardy, E.A., Goode, R.L., Deshetler, L.A., & Thomas, R.M. (2005). An analogue stud y of therapist empathetic process: Working with difference. Psychotherapy: Theory, Research, Practice, Training, 42 (2), 198 - 210. Helm, PL, Hays, J. C, Flint, E., Koenig, H. G., & Blazer, D. G. (2000). Does private religious activity prolong surviv al? A six - year follow - up study of 3,851 older adults. Journal of Gerontology , 55A, M400 - M405. 99 Hernandez, P., Gangsei, D., & Engstrom, D. (2007). Vicarious resilien ce: A new concept in work with those who survive trauma. Family Process, 46 (2), 229 - 241. Hooper, C., Craig, J., Janvrin, D.R., Wetsel, M.A., & Reimels, E. (2010). Compassion satisfaction, burnout, and compassion fatigue among emergency nurses compared with nurses in other selected inpatient specialties. Journal of Emerge ncy Nursing, 38 (5), 420 - 427. Hospice Foundation of America. (2013). Washington DC. Retrieved from http://www.hospicefoundation.org/ Hospice of Michigan. (2014). Detroit, MI. Retrieved from http://www.hom.org/ Huggard, P. (2003). Compassion fatigue: How much can I give? Medical Education, 37 (1), 163 - 164. Joinson, C. (1992) Coping with compassion fatigue. Nursing, 22, 116 - 121. Joubert, D., Hocking, A., & Hampson, R. (2013). Social work in oncology: Managing vicarious trauma - The impact of professional supervision. Social Work in Health Care, 52, 296 - 210. Kadushin, A. (1992) Social work supervision: An updated survey. The Clinical Supervisor, 10 (2), 10 - 27. Kadushin, A. (1992) Supervision in social work (3 rd Ed). New York, NY: Columbia University Press. Kadushin, A., & Harkness, D. (2002) Supervision in social work ( 4 th Ed.) New York, NY: Columbia University Press. Keidel, G.C. (200 2) Burnout and compassion fatigue among hospice caregivers. American Journal of Hospice and Palliative Care, 19 (3), 200 - 205. Kim, H., & Lee, S.Y. (2009). Supervisory communication, burnout, and turnover intention among social workers in health care settings. Social Work in Health Care, 48, 364 - 385. Koenig, H. G., George, L. K., & Peterson, B. L. (1998). Religiosity and remission from depression in medically older patients. American Journal of Psychiatry, 155, 536 - 542. Kroenk e, K., Strine , T.W., Sptizer, R.L., Williams, J.B., Berry, J.T., & Mokdad, A.H. (2009). The PHQ - 8 as a measure of current depression in the general population. Journal of Affect Disorders, 114, 163 - 173. 100 Lait, J., & Wallace, J.E. (2002). Stress at work: A study of organizational - professional conflict and unmet expectations. Relations Industrielles, 57 (3), 463 - 487. Larsen, D. & Stamm, B.H. (2008). Professional quality of life and trauma therapists. In Joseph, S.A. & Linley, P.A. (eds.). Trauma, recover y, and growth: Positive psychological perspectives on post - traumatic stress. 275 - 293. Hoboken, NJ: Wiley Press. Lawson, S.K. (2008) Hospice social workers: The incidence of compassion fatigue. (Doctoral Dissertation). University Microform Internati onal 3331396. Lawson, G., & Myers, J.E. (2011). Wellness, professional quality of life, and career - sustaining behaviors: What keeps us well? Journal of Counseling and Development, 89 (2), 163 - 171. Leiter, M.P., & Maslach, C. (2004). Areas of work life: A structured approach to organizational predictors of job burnout. Emotional and Physiological Processes and Positive Intervention Strategies Research in Occupational Stress and well Being, 3, 91 - 134. Leon, A.M., Altholz, J.A.S ., & Dziegielewski, S.F. (1999) Compassion fatigue: Considerations for working with the elderly. Journal of Gerontological Social Work, 32 (1), 43 - 63. - being. Journal of Social and Clinical Psychology, 26 (3), 385 - 403. Lloyd, C., King, R., & Chenoweth, L. (2002) Social work, stress, and burnout: A review. Journal of Mental Health, 11 (3), 255 - 265. Maslach, C. (1976) Burnout. Human Behavior, 5, 16 - 22. Maslach, C. (1978) Job burnout: How people cope. Public Welfare, 36 (2), 56 - 58. Maslach, C., & Jackson, S.E. (1981). The measurement of experienced burnout. Journal of Occupational Behavior, 2, 99 - 113. McCann, I.L., & Pearlman, L.A. (199 0) Vicarious traumatization: A framework for understanding the psychological effects of working with victims. Journal of Traumatic Stress, 3 (1), 131 - 149. Melvin, C.S. (2012) Professional compassion fatigue: What is the true cost of nurses caring for the dying? International Journal of Palliative Nursing, 18 (12), 606 - 611. Mena, K.C., & Bailey, J.D. (2007). The effects of the supervisory working alliance on worker outcomes. Journal of Social Service Research, 34 (1), 55 - 65. 101 Mor, V., & Laliberte, L. (1984). Burnout among hospice staff. Health and Social Work , 9 (4), 274 - 283. Moulden, H.M., & Firestone, P. (2007). Vicarious traumatization: The impact on therapists who work with sexual offenders. Trauma, Violence, and Abuse, 8 ( 1), 67 - 83. Munson, C.E. (2002) Handbook of clinical social work supervision (3 rd Ed). Binghamton, NY: Haworth Press Najjar, N., Davis, L.W., Beck - Coon, K., & Doebbeling, C.C. (2009). Compassion fatigue: A review of the research to date and relevance to cancer care providers. Journal of health Psychology, 14 (2), 267 - 277. National Association of Social Workers (2013). Best practice standards in social work supervision. Washington, DC. Retrieved from http://www.socialworkers.org/practice/naswstandards/supervisionstandards2013.pdf National Association of Social Workers (2008) Code of Ethics for Social Workers. Wa shington, DC. Retrieved from: http://www.socialworkers.org/pubs/code/code.asp National Association of Social Workers (2005) Standards for social work practice in health care settings. Washington DC. Retrieved from http://www.naswdc.org/practice/standards/naswhealthcarestandards.pdf National Hospice and Palliative Care Organization (2013). Alexandria, VA. http://www.nhpco.org/about/hospice - and - palliative - care National Hospice and Palliative Care Facts and Figures: Hospice Care in America. (2013) Alexandria, VA. Newman, B.M. and Newman, P.R. (2009) Development through life: A psychosocial approach. Belmont, California: Wadsworth Cengage Learning. Newswald - McCalip, R. (2001). Development of the secure counselor: Examples supporting Pistole and Watkins (1995) d iscussion of attachment theory in counseling supervision. Counselor Education and Supervision, 41 (1), 18 - 27. Noe, K., Smith, P.C., & Younis, M. (2012) Calls for reform to the US Hospice system. Aging International, 37, 228 - 237. Nulty, D.D. (2008). T he adequacy of response rates to online and paper surveys: What can be done? Assessment and Evaluation in Higher Education, 33 (3), 301 - 314. Pearlman, L.D., & MacIan, P.S. (1995) Vicarious traumatization: An empirical study of the effects of trauma w ork on trauma therapists. Professional Psychology and Research Practice, 26, 558 - 565. 102 Pearlman, L.D., & Saakvitne, K.W. (1995) Treating therapists with vicarious traumatization and secondary traumatic stress disorders. In Figley, C.R. (Ed.) Compassion fatigue: Coping with secondary traumatic stress disorder in those who treat the traumatized. New York , NY : Bruner/Mazel Publishing. Pilsecker, C. (1979) Terminal cancer: A challenge for social work. Social Work in Health Care, 4 (4), 369 - 379. Pistole, M.C., & Watkins, C.E. (1995) Attachment theory, counseling process, and supervision. The Counseling Psychologist , 23 (3), 457 - 476. Plante, T. G., & Sherman, A. S. (Eds.). (2001). Faith and health: Psychological perspectives. New York, NY: Guilford Press. Potter, P., Deshields, T., Davanbeigi, J., Berger, J., Cipriano, D., Norris, L., & Olsen, S. (2010). Compassion fatigue and burnout: Prevalence among oncology nurses. Clinical Journal of Oncology Nursing. 14 (5), 56 - 62. Poulin , J.E., & Walter, C.A. (1993). Burnout in gerontology social work. Social Work, 38 (3), 305 - 310. Radey, M., & Figley, C.R. (2007) The social psychology of compassion. Clinical Social Work Journal, 35, 207 - 214. Reese, D.J. (2011) Interdisciplinar y perceptions of the social work role in hospice: Building upon the classic Kulys and Davis study. Journal of Social Work in End - of - Life and Palliative Care, 7, 383 - 406. Reese, D.J, & Raymer, M. (2004) Relationships between social work involvement and hospice outcomes: Results from a national hospice social work survey. Social Work, 49 (3), 415 - 422. caseworkers. Social Work, 39 , 75 - 80. Rynde, T. (1997). Stress: Creating an environment to prevent burnout. Health Care Forum Journal, 40 (4), 54 - 57. Sabo, B. (2011) Reflecting on the concept of compassion fatigue . The Online Journal of Issues in Nursing, 16 (1), 1 - 18. Saleebey, (D). (1996). The strengths perspective in social work practice: Extensions and cautions. Social Work, 41 (3), 296 - 305. 103 S amantrai, K. (1992). Factors in the decision to leave: Retaining social workers with MSWs in public child welfare. Social Work, 37 (5), 454 - 45 8. Saunders, D.C. (2001) Social work and palliative care: The early history. British Journal of Social Work, 31, 791 - 799. Scannapieco, M., & Connell - Carrick, K. (2003). Do collaborations with schools of social work make a difference for the field of child welfare?: Practice, retention and curriculum. Journal of Human Behavior in the Social Environment, 7 (1/2), 35 - 52. Simon, C.E., Pryce, J.G., Roff, L.L., & Klemmack, D. (2005) Secondary traumatic stress and oncology social work: Protecting compassion from fatigue and compromising the Journal of Psychosocial Oncology, 23 (4), 1 - 14. Slatten, L.A., Carson, K.D., & Carson, P.P. (2011) Compassion fatigue and burnout: What managers should know. The Health Care Manager, 30 (4), 325 - 333. Slocum - Gori, S., Hemsworth, D., Chan, W., Carson, A., & Kazanjian, A. (2011) Understanding compassion satisfaction, compassion fatigue, and burnout: A survey of the hospice palliative care workforce. Palliative Med icine, 27 (2), 172 - 178. Smart, D, English, A., James. J., Wilson, M., Daratha, K.B., Childres, B., & Magera, C. (2014). Compassion fatigue and satisfaction: A cross - sectional survey among US health care workers. Nursing and Health Sciences, 16 (1), 3 10. Stamm, B. H. (2002). Measuring compassion satisfaction as well as fatigue: Developmental history of the compassion satisfaction and fatigue test. In Figley, C. R (ed.). Treating compassion fatigue , (pp 107 119) . New York , N Y : Brunner Routl edge. Stamm, B. H. (2008). The ProQOL Test Manual (2 nd ed.) Baltimore: Sidran Press and ProQOL.org. Stamm, B.H. (2010). The concise ProQOL manual (2 nd ed.). Pocatello, ID: ProQOL.org. Sue, V.M., & Ritter, L.A. (2012) Conducting online surveys (2 nd ed.) Los Angeles, CA: Sage Publications. Sudeck, E.R. (2012). Exploring compassion fatigue among members of the hospice Supple - Diaz, L., & Mattison, D. (19 92) Factors affecting survival and satisfaction: navigating a career in oncology social work. Journal f Psychosocial Oncology, 10 (1), 111 - 131. 104 Swetz, K.M., Harrington, S.E., Matsuyama, R.K., Shanafelt, T.D., & Lyckholm, L.J. (2009) Strategies for avoi ding burnout in hospice and palliative medicine: Peer advice for physicians on achieving longevity and fulfillment. Journal of Palliative Medicine, 12 (9), 773 - 777. Sze, W.C., & Ivker, B. (1986) Stress in social workers: The impact of setting and role . Social Casework, 67 (3), 141 - 148. Tehrani, N. (2007). The cost of caring: The impact of secondary trauma on assumptions, values, and beliefs. Counseling Psychology Quarterly, 20 (4), 325 - 339. Thompson, G. (2007). Exploring the relationship betwee n faith development and compassion fatigue in hospice practitioners. Chaplaincy Today, 23 (2), 25 - 31. Thyer, B.A. (Ed.) (2001) The handbook of social work research methods. Thousand Oaks, CA: Sage Publications. Trippany, R.L., White Kress, V.E., & Wilcoxon, S.A. (2004) Preventing vicarious trauma: What counselors should know when working with trauma survivors. Journal of Counseling and Development, 82 (1), 31 - 37. Tsui, M.S. & Ho, W. (1997) In search of a comprehensive model of social work s upervision. The Clinical Supervisor, 16 (2), 181 - 205. Vachon, M.L.S. (1995) Staff stress in hospice/palliative care: A review. Palliative Medicine, 9, 91 - 122. Whitebird, R.R., Asche, S.E., Thompson, S.L., Rossom, R., & Heinrich, R. (2013). Stress, burnout, compassion fatigue, and mental health in hospice workers in Minnesota. Journal of Palliative Medicine, 16 (12), 1534 - 1539. Wilberts, G., Brakemeier, E., Zobel, I., Harter, M., & Schramm, E. (2010). Exploring preoperational features in chronic depression. Journal of Affective Disorders, 124 (3), 262 - 269. Winstanley, J., & White, E. (2002). Clinical supervision: Models, measures, and best practice. Nurse Researcher, 10 (4), 7 - 38. Worthington, E.L., & Roehlke, H.J. (1979). Effective supervision as perceived by beginning counselor in training. Journal of Counseling Psychology, 26, 54 - 73. Yarnold, P.R., Bryant, F.B., Nightingale, S.D., & Martin, G.J. (1996).Assessing physician empathy using the interpersonal re activity index: A measurement model and cross - sectional analysis. Psychology, Health, and Medicine, 1 (2), 207 - 221. 105 Young, J.L., Derr, D.M., Cichillo, V.J., & Bressler, S. (2011). Compassion satisfaction, burnout, and secondary traumatic stress in he art and vascular nurses. Critical Care Nursing, 34 (3), 227 - 134.