Journal of Pediatric Nursing (2014) xx, xxx–xxx
Pediatric Nurses' Grief Experience, Burnout and Job Satisfaction1 Jehad Z. Adwan PhD,RN ⁎ University of Minnesota, School of Nursing, Minneapolis, MN, USA Received 31 December 2013; revised 20 January 2014; accepted 22 January 2014
Key words: Pediatric nurses; Grief; Burnout; Job satisfaction; Intention to leave
Correlations among grief, burnout, and job satisfaction among highly satisfied pediatric nurses were examined using the Revised Grief Experience Inventory (RGEI), Maslach Burnout Inventory (MBI), and Index of Work Satisfaction (IWS). Results showed that grief had significant correlations; positive with burnout, negative with job satisfaction. RN's reported significantly higher emotional exhaustion if their primary patients died and higher guilt if patients died younger. Conclusions suggest a dynamic statistical interaction among nurses' grief, burnout, and job satisfaction representing a pathway to intention to leave their unit, organization, or nursing. Recommendations include implementation and evaluation of grief intervention and education programs. © 2014 Elsevier Inc. All rights reserved.
Background Grief and Nurses COWLES AND RODGERS (1991) defined grief as “dynamic, pervasive, highly individualized process with a strong normative component” (p. 121). Pediatric nurses have a unique caring role because of the vulnerable nature of their patients and the special bonds that can form when caring for patients and their families (Scott, 1994). This relationship can expose nurses to intense and recurrent experiences with unresolved loss and grief when one of their patients suffers from a terminal illness or dies (Couden, 2002; Rashotte, Fothergill-Bourbonnais, & Chamberlain, 1997). Rushton (2004) and Landro (2012) indicated that health professionals may suffer and
Previous presentations: 1. Round table discussion: Center for Children with Special Health Care Needs, University of Minnesota, School of Nursing, March 18th, 2011. 2. Nursing Research Day poster presentation, University of Minnesota, School of Nursing, April 29th, 2011. 1 Funding: This research was supported in part by Sigma Theta Tau, Zeta Chapter at the University of Minnesota. ⁎ Corresponding author: Jehad Z. Adwan, PhD,RN. E-mail address: [email protected]. 0882-5963/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pedn.2014.01.011
experience a variety of emotional and physical symptoms including headaches, fatigue, and depression as they watch the suffering of their patients and families. Others argued that the exposure to patients' suffering and death can render the nurse prone to stress and burnout (Gerow et al., 2009; Kennedy, 2005; Peters, 2012). Maslach, Schaufeli, and Leiter (2001) defined burnout as a syndrome that affects individuals who are in the helping or caring professions, such as nurses, psychologists, and therapists. Burnout has three components: emotional exhaustion (EE), depersonalization (DP), and low personal accomplishment (PA) (Maslach et al., 2001). Staff burnout can trigger turnover challenges for managers and institutions alike. Combined with high emotional exhaustion and depersonalization, low personal accomplishment provide ripe conditions for lower quality nursing care, less consciousness about patient safety and eventually higher rates of turnover (Anderson, 2008). Burnout and job satisfaction have been linked to nurse productivity and retention in the workplace (Anderson, 2008; Erenstein & McCaffrey, 2007; Mrayyan, 2005) as well as the quality of patient care (Billeter-Koponen & Freden, 2005; ToppinenTanner, Ojajarvi, Vaananen, Kalimo, & Jappinen, 2005). The Institute of Medicine (IOM) and American College of Critical Care Medicine (ACCCM) advocated for grief support
2
J.Z. Adwan
programs for staff caring for terminally-ill or dying children (Durall, 2011). Such programs are new and are currently being developed (Altounji, Morgan, Grover, Daldumyan, & Secola, 2013). Pediatric oncology nurses reported a benefit from a grief-support program that was delivered in the form of self-care retreat (Altounji et al., 2013). The program provided “rejuvenation, feelings of appreciation, and a revived passion for the job” (p. 22). In a literature review, (Hildebrandt, 2012) identified several strategies that could be used to resolve grief, and as an added benefit, increase retention of oncology nurses. Some of these strategies included: creating supportive environments, debriefing, providing grief education, and alternating patient assignments (p. 602). Numerous studies have explored nurses' grief experiences but focused only on the description of their feelings (Feldstein & Buschman-Gemma, 1995; Papadatou, Bellali, & Petraki, 2002; Rashotte et al., 1997). On the other hand, ways to deal with grief were explored by others (Ashby, Kosky, Laver, & Sims, 1991; McNeely, 1996; Medland, Howard-Ruben, & Whitaker, 2004; Zander, Hutton, & King, 2013) to help find coping strategies for nurse caregivers to use. Despite the abundance of nursing literature on burnout and job satisfaction, these variables have been addressed either individually or in combination with variables related to some aspects of the clinical environment. No studies to date have explored pediatric nurses' grief experience and its relationship to burnout and job satisfaction. The only published study (Anderson, 2008) that investigated grief experience, burnout and turnover patterns was conducted among certified nursing assistants (CNA's) working at nursing homes. Anderson (2008) concluded that complications from grief could contribute to staff burnout, but the results showed that even though grief had no significant direct impact on turnover patterns, it may have a complex and indirect effect on turnover. Braccia (2005) offered an expert's account and advice on how to avoid burnout as a result of the grief experience. This opinion, however, was based on personal experience rather than evidence. Therefore, the purpose of this study was to explore the relationships between pediatric nurses' self-reported grief experiences resulting from patient deaths with burnout and job satisfaction. The study had the following 2 hypotheses: 1. Grief scores are positively correlated with burnout scores. 2. Grief scores are negatively correlated with job satisfaction scores.
Conceptual Framework The conceptual framework for this study was based on the proposed model for healthcare professionals' grieving process (Papadatou, 2000). The framework highlights the three main study variables: grief, burnout, and job satisfaction. Grief is the central concept of this framework and is explained by antecedents or events that occurred prior to the grief
experience as well as consequences or events that take place as a result of the grief process. The behaviors a person exhibits as a response to the grief experience determine whether an individual engages in effective or ineffective coping in an attempt to resolve the emotional and physical consequences of grief including burnout and job satisfaction as outcomes.
Methods Sample and Sampling A convenience sample of 120 pediatric nurses was recruited from a large midwestern academic medical center's four pediatric patient care units and pediatric float pool RN's. Participants were recruited based on the following criteria: 1. Minimum pediatric experience of 12 months. 2. Worked at 0.5 FTE or more at the time of data collection in pediatric setting in the 12 months before data collection started.
Instrumentation and Measurement The study used four instruments to collect data. The demographic information form (DIF) collected information about the participants and their work environment, while the Revised Grief Experience Inventory (RGEI), the Maslach Burnout Inventory (MBI) and the Index of Work Satisfaction Scale for nurses (IWS) collected data about their respective constructs. Below is a detailed description of the instruments. Demographic Information Form (DIF) The DIF included relevant demographic information about the participants in addition to their nursing experience. Data were also collected about the nurses' specific experiences with dying children in terms of how many deaths they experienced in the past 12 months, and whether or not they were the primary nurse for these patients. The DIF also inquired about major events in the life of the nurse, such as death, divorce, and diagnosis of a serious illness in the direct family. The Revised Grief Experience Inventory (RGEI). The RGEI is a 22-question, multidimensional measure of grief (Lev, Munro, & McCorkle, 1993) which includes four subscales: existential tension (ET) referring to meaning of life and existence, depression (Dep), guilt (Glt), and physical distress (PD). The scoring is a self-report Likert-scale style with five possible responses ranging from strongly disagree (1) to strongly agree (5). The RGEI subscales have reliability alphas of 0.87, 0.80, 0.72, and 0.83 respectively and an overall internal consistency alpha coefficient of 0.93 (Lev et al., 1993). The Maslach Burnout Inventory (MBI) The Maslach Burnout Inventory (MBI) is a 22-job-related item questionnaire designed to measure the concept of
Grief Experience, Burnout and Job Satisfaction burnout among human service and helping professionals (Maslach & Jackson, 1981). The MBI uses seven-point Likert-scale responses to job-related views, where a response of a zero (0) means the feeling never exists and a response of six means that the feeling exists every day. The MBI contains three subscales referring to the three components of burnout; emotional exhaustion (EE), depersonalization (DP), and personal accomplishments (PA). The MBI has shown strong psychometric properties as evidenced by its Cronbach's coefficients alpha for internal consistency of its subscales: 0.89; 0.77; and 0.74, respectively. These three components are assessed independently from each other (no total score), as relationships among them have not been understood fully yet (Maslach, Jackson, & Leiter, 2005). Index of Work Satisfaction for Nurses (IWS) The IWS is a two-part (A and B) instrument. Part A was not used because the researcher's interest was in the total numeric score of the IWS rather than the individual subscale scores. Part A is often used for institutional purposes in evaluating staff job satisfaction through ranking of job satisfaction subscales. Part B is composed of 44 seven-point Likert-scale items that assess the nurse's responses from strong agreement (1) to strong disagreement (7) on job satisfaction among nurses in six different areas: pay, autonomy, task requirements, organizational policies, interaction, and professional status. Overall reliability analysis of the IWS showed a Cronbach's alpha of 0.91.
Design This research used a cross-sectional correlational design to examine potential relationships among the core variables (grief, burnout and job satisfaction scores) as they occurred in the clinical setting without implying a causal relationships among them (Polit & Beck, 2012).
3 scores. Multivariate associations were conducted using multiple regression analyses to find explanations of burnout and job satisfaction by the grief scores (Polit & Beck, 2012).
Results Demographic Description Out of 214 eligible pediatric nurses at a large Midwestern Academic Medical Center, 120 (56%) participated in the study. Paper-and-pencil surveys were returned by 87 participants (72.5%), and 33 surveys (27.5%) were submitted online. Participants' ages ranged from 23–62 years (m = 34, SD = 10). Summary of participants' demographic characteristics are in Tables 1a and 1b.
Nurse Experience With Patient Death Participants reported a range of zero to 40 patient deaths in the previous 12 months (m = 13, SD = 9). Forty seven participants (39%) had served as primary nurses for one or more patients who had died during the prior 12 months. At least 68% (n = 82) of participants reported that they had cared for a patient on or around the day the patient died in the past 12 months. Following their patients' death, 114
Table 1a Demographic descriptions of the participants (nominal variables). Descriptor
Sub-descriptor
Count
%
Gender
Female Male White Other Married Single Separated/Divorced Yes No Alone Spouse/SO, no children Spouse/SO, w/ children W/ children Other Associate Bachelor's Master's Specialized Pulmonary, cardiology, renal Heme/Onc., GI, trauma PICU Blood and marrow transplant Float pool Other
114 6 117 3 69 47 4 50 70 25 31 42 4 18 22 91 7 12 16 29 30 33 11 1
95 5 98 2 58 39 3 42 58 21 26 35 3 15 18 76 6 10 13 24 25 28 9 0.8
Ethnicity Marital Status
Data Collection and Analysis Plan Approvals of the institutional review board (IRB) and the target institution were obtained prior to starting data collection. Recruitment was done by sending email invitations to departmental group emails and posting of recruitment flyers in target pediatric units. Data were collected using identical paper and pencil and online surveys. Upon returning completed surveys, each participant received a $10 gift card as an incentive for participation. Quantitative data were entered, cleaned and analyzed using the Statistical Package for Social Sciences (SPSS) version 17.0. Descriptive statistics were used to describe the nurses' grief scores on the RGEI, burnout scores on the MBI, and job satisfaction scores on the IWS as well as means and standard deviations of participants' demographic characteristics. Correlational statistics were conducted to examine relationships among grief, burnout, and job satisfaction
Have children Living with
Education
Unit type
4
J.Z. Adwan Table 1b
Demographic and work‐related descriptors.
Descriptor
Range
M
SD
Age 23–62 yrs. 34 10 Years of nursing experience 1–40 9 9 Years of pediatric experience 1–36 6.7 6.9 Number of children 0–4 0.89 1.15 Number of negative life events 0–4 0.74 0.98 Number of patients' death in 0–40 13 9 past 12 months Longest time known patient 0–20 yrs. 2.6 yrs. 3.4 before they died Shortest time known patient 0–2 yrs 1.6 yrs 3.5 before they died Youngest patient died 0–18 yrs 1.5 yrs. 3 yrs. Time elapsed since last 0–4 yrs. 0.5 yrs. 0.63 patient death
participants (95%) reported that they had shared their feelings with one or more people around them; 103 (86%) shared feelings with a nursing colleague, 80 (67%) with a spouse or significant other, 74 (63%) with a friend, 22 (18%) with parents/siblings, and only two participants (1.8%) reported sharing their feelings with a therapist or counselor. The majority (n = 96, 80%) shared their feelings with two to three others around them. The majority of those who shared feelings with others (n = 111, 92%) reported that sharing feelings was somewhat or very helpful to them.
Grief Experience The results revealed that nurses do experience grief over their patients' death in similar ways as a patient's family care giver does. Although, the scores of the RGEI subscales were consistent with the tool's validation study, participantsunderstandably-exhibited lesser grief intensity compared to family care givers. Descriptive statistics of the total and subscale scores of the complete grief scores are summarized in Table 1c and are compared to the actual results of the total as well as subscale scores from the validation study (Lev et al., 1993).
Burnout Experience Scoring of the MBI was done by individual subscale without calculating a total burnout score because the full Table 1c
relationships among the three concepts PA, EE, and DP were not fully established (Maslach et al., 2005). The results revealed moderate levels of emotional exhaustion and personal accomplishment in addition to low depersonalization scores among the study particiapants. A summary of the MBI's subscales scores is in Table 2.
Job Satisfaction The results revealed high scores in most job satisfaction subscales. Job satisfaction subscales' descriptive attributes are detailed in Table 3 and are listed for reference. For the purposes of this study, only the total job satisfaction score (range = 120–240, max = 308, m = 190, SD = 24) was used in further analysis, namely correlational and regression analyses.
Adverse Personal Life Events Participants provided information on adverse personal life events that they had experienced during the 12-month period prior to their participation in the study. A death in the immediate family was reported by 16% (n = 19), a diagnosis of major illness in the family by 25% (n = 30), a divorce or separation by 3% (n = 4), and other major loss by 5% (n = 6). The total of instances of adverse personal events (Table 4) was counted for each participant and used as a continuous variable for measurement in relation to other variables in the analysis.
Correlations Among Grief, Burnout, and Job Satisfaction Analysis revealed a statistically significant positive correlation between the RGEI total score and MBI's EE subscale (r = 0.38, p b 0.001), a significant positive correlation with the MBI's DP subscale (r = 0.19, p = 0.04), and a negative correlation with the MBI's PA subscale (r = − 0.244, p = 0.009). Most RGEI subscales are individually correlated to the subscales of the MBI subscales in the same direction of the total RGEI scores. Significant negative correlations were found between total RGEI and total IWS scores (r = − 0.29, p = 0.002). Furthermore, IWS total score had a significant negative correlation with MBI's EE and DP,
Study RGEI compared to normative data (Lev et al., 1993) and its subscales.
Current study (N = 120)
Validation study (N = 418)
Subscales→
Existential tension (30 max)
Depression (30 max)
Guilt (15 max)
Physical distress (42 max)
Total grief score (110 max)
Min–Max x SD Min–Max x SD
6–28 11.13 4.8 – 20.1 8.5
6–29 14.92 7.56 – 23 7
3–14 7.32 2.76 – 10 4.6
7–32 15.09 6.30 – 22.5 9.3
22–94 48.29 17.77 – 75.5 25.7
Grief Experience, Burnout and Job Satisfaction Table 2
5 Table 4
MBI subscales.
Descriptive measures
Emotional Depersonalization Personal accomplishment exhaustion (max: 30) (max: 56) (max: 48)
Min–Max x SD Burnout level
13–48 38 5.7 Moderate
5–45 21 8.58 Moderate
0–19 5.8 4.7 Low
while it had a significant positive correlation with MBI's PA. Table 5 summarizes correlations among the three core variables.
Correlations Between Grief and Demographic Variables Due to non-normal distribution of the following demographic variables, a Spearman's correlation was used and revealed a significant positive correlation between the number of primary patients' deaths and the RGEI's total scores (r = 0.21, p = 0.026). Except for the physical distress subscale, the number of primary patients death was significantly correlated with the RGEI's subscales as follows: existential tension (r = 0.18, p = 0.047), depression (r = 0.24, p = 0.01), and guilt (r = 0.19, p = 0.039). The grief total score was not significantly correlated with the number of patients the nurse cared for on or around the day they had died. The grief's guilt subscale, however, was negatively correlated with the numbers of patients the nurse had cared for around the time of death (r = − 0.19, p = 0.04). Additionally, statistically significant positive correlations were found between the age of the youngest patient who died and the burnout's depersonalization subscale (r = − 0.22, p = 0.03). The average period the nurse knew the patient before they died and grief's guilt scores were also significantly correlated (r = 0.20, p = 0.044). It is important to point out that no statistically significant correlations were identified between the total perceived
Table 3
Descriptive attributes of IWS total and subscale.
IWS scale/subscale
Descriptive measures Max possible Min Max x scores
Pay Autonomy Task requirements Organizational policies Professional status Interactions IWS total
42 56 42 49
6 21 7 6
38 51 37 36
SD 24.67 36.81 20.88 16.7
5.8 6.7 5.2 5.4
Total numbers of adverse life events.
Total adverse personal life events
n
%
0 1 2 3 4
69 20 25 5 1
58 17 21 4 0.8
events event events events events
number of patient's deaths or how recent the latest death was and grief's total scores or its four subscales. The participant's age and gender had no significant correlations with grief, burnout, or their respective subscale scores. Furthermore, the analysis did not show any significant correlations between either participant's length of experience in nursing in general or in pediatric nursing in particular and any of the grief, burnout, or job satisfaction scores, including their respective subscales. Also, the total number adverse personal life events did not reveal any significant correlations with grief, burnout, or job satisfaction total scores or subscales as applicable.
Regression Analysis Multiple regression analysis was used to further examine relationships among variables based on previously established significant correlations. Regression models were created using total grief scores and the four subscales of the RGEI (existential tension, depression, guilt, and physical distress), three MBI subscales (emotional exhaustion, depersonalization, and personal accomplishment), and the total score of the IWS. All variables that are correlated with these subscales were entered into regression models using stepwise process, where each of the subscales was entered as dependent variables and all other correlates with a pvalue b 0.1 were entered as candidate independent variables in the model. Regression analysis showed that the RGEI's guilt subscale score explained about 7% of the variance in the MBI's personal accomplishment score (beta = − 0.56, p = 0.003). The model involving the IWS total scores and the RGEI's existential tension (where the person feels like the purpose of life has been lost) explained 31% of the variance in the MBI's emotional exhaustion score. Conversely, a quarter of grief score variance was predicted by the MBI's
Table 5
Correlations matrix among the RGEI, MBI, & IWS.
Total RGEI 49 70 308
22 49 40.80 4.5 26 68 51.11 7.6 120 240 190.97 24
Total IWS
r p r p
Total RGEI
MBI-PA
MBI-EE
MBI-DP
− 0.29 0.002
− 0.244 0.009 0.238 0.009
0.381 b .001 − 0.464 b .001
0.192 0.042 − 0.358 b .001
6 emotional exhaustion score (beta = 0.89, p b 0.001), the length of time the nurse knew the patient before they had died (beta = 0.12, p = 0.02), and the number of primary patients who died in the previous 12 months (beta = 2.19, p = 0.047).
Discussion It is not surprising that nurses' reactions to the death of their patients are not as intense or overreaching as those of the patient's family members. Consistent with reviewed literature (Brown & Wood, 2009; Feldstein & BuschmanGemma, 1995; Papadatou et al., 2002), this study suggests that nurses experience grief over the death of their patients in ways similar to the patient's family members, yet to lesser extents. This study demonstrated that nurses' grief scores were near-normally distributed indicating a viable range of emotional and physical reactions with significant correlations with all aspects of burnout and further suggesting a statistical explanation to burnout's emotional exhaustion aspect. Even though the current literature does not show the direct relationship between grief and emotional exhaustion, it makes clinical as well as scientific sense to establish that those two concepts are closely related to each other. The available literature, though more anecdotal than empirical, supports that nurses' grief may lead to burnout if not addressed properly (Braccia, 2005; Brown & Wood, 2009; Saunders & Valente, 1994). Nurses felt that sharing feelings about patient death was helpful. Interestingly however; that perception was correlated with emotional exhaustion, which could be interpreted that person-to-person sharing of feelings may actually have undesirable consequences even though nurses engaged in it may see it as a helpful strategy, and that could be the start of the burnout process. Higher levels of grief and its components are associated positively and even suggest explanations to some components of burnout, i.e. emotional exhaustion, which in turn is linked to nurses' intention to leave their unit, hospital, or even nursing as a profession. Grief and its components were correlated negatively with job satisfaction although this cohort of nurses was highly satisfied in most of the work satisfaction subscales. Some of the participants' burnout scores were, though relatively low, associated their perceived numbers of patients' deaths on their unit, their feeling of guilt over these deaths, and their personal existential tension felt as a result of these deaths. Nurses are able to continue their professional growth and have sense of personal accomplishment in an environment where patients' death may bring about grief and burnout. Participants' job satisfaction and grief's existential tension scores suggested a valid explanation of burnout's emotional exhaustion, which could be seen that burnout may be an intermediary force between grief and job satisfaction that could eventually lead the nurse to make a career change. Intention to leave unit was correlated to nurses' grief, burnout and job satisfaction issues which has
J.Z. Adwan potential human resource implications for employers costing money to replace and train new staff.
Recommendations and Conclusions Recommendation for Practice Informal strategies of coping with grief and its consequences such as sharing feelings with peers, loved ones, and relaxation techniques, while seen helpful by some nurses, may not be adequate to overcome the emotional as well as the physical aspects of the grieving process. Worse yet, sharing feelings may actually be working against its perceived benefit. A professional early intervention program for nurses is recommended to be used regularly when a situation calls for grief intervention following a patient's death. There is a need for grief education to help nurses better understand the emotional as well as physical toll of experiencing a patient's death repeatedly over a period of time. Organizational recommendations include reassessment of attrition rates and how grief, burnout, and job satisfaction play a role in the process and the financial impact on the organization.
Recommendations for Research Refining of the RGEI tool is recommended because this is the first known time such an instrument, designed originally for family care givers, to be used in clinical setting to measure staff grief even though it demonstrated strong psychometric properties when compared to the original instrument. It is recommended that the RGEI be examined by a panel of experts to re-formulate its statements to make them suitable to nurse–patient relationship setting rather than a family caregiver setting. Another logical outcome of the current study is to design and evaluate a program of grief education where objectives are established such that therapeutic grief resolution is attained through learning how to cope with patients' deaths as well as organizational early engagement when deaths occur. An evidence-based program that builds on current coping literature among pediatric nurses is needed to evaluate therapeutic strategies among nurses that can be tested for effectiveness and later used to help nurses cope with grief.
Limitations Although the RGEI was based upon the 102-item inventory, which was used on nurses (Sanders, Mauger, & Strong, 1979), the Revised Grief Experience Inventory (RGEI) was tested in its validation study on grieving family caregivers who had experienced the loss of a loved one. Therefore, its statements and questions were geared toward what is expected of a family to experience. Consequently, the
Grief Experience, Burnout and Job Satisfaction responses of the current study participants show lower level of grief which could have affected the results if the instrument were designed for measuring nurses' grief. The Maslach Burnout Inventory is the most widely used burnout measure, which has been used in theory-driven research on the burnout phenomenon. Its multidimensional conceptualization of the concept, although a theoretical strength, represents a limitation at the same time in that these dimensions are understood better when applied to groups of persons. They have not been used, at least for now, to assess burnout among individuals because the instrument does not have well-validated cut-off points for burnout scores (Maslach & Jackson, 1999). The convenient nature of the sample is a limitation. This limitation is shared by non-experimental designs under which this study falls, where a risk for skewed results may occur because of unequal chances of representation of all participants. The nature of the setting where the current study took place and the characteristics of the participants suggest overall equal chances among participants for exposure to patient's death due to the high acuity of the patient population.
Conclusion This study has demonstrated that pediatric nurses are at a real risk of experiencing grief following their patients' death. It is evident that even though the participants of this study were highly satisfied with their jobs as evidenced by their high levels of job satisfaction and personal accomplishments, they still experienced grief in similar ways a family member does-albeit at a lower level. However, once nurses manage to cope with losing one patient, they may have a rerun of the same experience with another patient and the cycle goes on. It is important for nurses, managers, and institutions to understand the real impact of grief on nurses in terms of suffering job satisfaction, higher burnout, which may lead to nurses' thinking about leaving a stressful unit, hospital, or even nursing altogether. This could have significant financial ramifications for institutions. Dealing with grief and its consequences is a shared responsibility of the nurses, managers, and the institutions. Grief interventions where therapeutic coping is learned as a skill could be a useful tool to deal with grief's impact in its early stages before or soon after a patient dies.
References Altounji, D., Morgan, H., Grover, M., Daldumyan, S., & Secola, R. (2013). A self-care retreat for pediatric hematology oncology nurses. US: Sage Publications. Anderson, K. A. (2008). Grief experiences of CNAs: Relationships with burnout and turnover. Journal of Gerontological Nursing, 34, 42–49.
7 Ashby, M. A., Kosky, R. J., Laver, H. T., & Sims, E. B. (1991). An inquiry into death and dying at the Adelaide-Childrens-Hospital - A useful model. Medical Journal of Australia, 154, 165–170. Billeter-Koponen, S., & Freden, L. (2005). Long-term stress, burnout and patient-nurse relations: Qualitative interview study about nurses' experiences. Scandinavian Journal of Caring Sciences, 19, 20–27. Braccia, D. (2005). Follow these tips to avoid burnout from chronic grief and compassion fatigue. ONS News, 20, 6. Brown, C., & Wood, A. (2009). Oncology nurses' grief: A literature review. Clinical Journal of Oncology Nursing, 13, 625–627. Couden, B. (2002). Sometimes I want to run: A nurse reflects on loss in the intensive care unit. Journal of Loss & Trauma, 7, 35–45. Cowles, K., & Rodgers, B. (1991). The concept of grief: A foundation for nursing research and practice. Research in Nursing and Health, 14, 119–127. Durall, A. (2011). Care of the caretaker: Managing the grief process of health care professionals. Pediatric Annals, 40, 266–273. Erenstein, C., & McCaffrey, R. (2007). How healthcare work environments influence nurse retention. Holistic Nursing Practice, 21, 303–307. Feldstein, M., & Buschman-Gemma, P. (1995). Oncology nurses and chronic compounded grief. Cancer Nursing, 18, 228–236. Gerow, L., Conejo, P., Alonzo, A., Davis, N., Rodgers, S., & Domain, E. (2009). A phenomenological study of nurses' experience of grief following patient death. Western Journal of Nursing Research, 31, 1078–1079. Hildebrandt, L. (2012). Providing grief resolution as an oncology nurse retention strategy: A literature review. US: Oncology Nursing Society. Kennedy, B. R. (2005). Stress and burnout of nursing staff working with geriatric clients in long-term care. Journal of Nursing Scholarship, 37, 381–382. Landro, L. (2012). When nurses catch compassion fatigue, patients suffer. Wall Street Journal - Eastern Edition, 259, D1–D3. Lev, E., Munro, B. H., & McCorkle, R. (1993). A shortened version of an instrument measuring bereavement. International Journal of Nursing Studies, 30, 213–226. Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Occupational Behavior, 2, 99–113. Maslach, C., & Jackson, S. E. (1999). Factorial validity of the Maslach Burnout Inventory–Dutch (MBI-NL) among dentists. Journal of Organizational Behavior, 20, 209–217. Maslach, C., Jackson, S. E., & Leiter, M. P. (2005). Maslach Burnout Inventory–general survey–“English”. Behavioral Medicine, 31, 18–27. Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of Psychology, 52, 397–422. McNeely, S. (1996). Stress and coping strategies in nurses from palliative, psychiatric and general nursing areas. Health Manpower Management, 22, 10–12. Medland, J., Howard-Ruben, J., & Whitaker, E. (2004). Fostering psychosocial wellness in oncology nurses: Addressing burnout and social support in the workplace. Oncology Nursing Forum, 31, 47–54. Mrayyan, M. T. (2005). Nurse job satisfaction and retention: Comparing public to private hospitals in Jordan. Journal of Nursing Management, 13, 40–50. Papadatou, D. (2000). A proposed model of health professionals' grieving process. Omega, 41, 59–77. Papadatou, D., Bellali, T., & Petraki, D. (2002). Greek nurse and physician grief as a result of caring for children dying of cancer. Pediatric Nursing, 28, 345–353. Peters, L. (2012). Is work stress in palliative care nurses a cause for concern? A literature review. International Journal of Palliative Nursing, 18, 561–567. Polit, D., & Beck, C. (2012). Resource manual for nursing research: Generating and assessing evidence for nursing practice (9th Edition ed ). Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins.
8 Rashotte, J., Fothergill-Bourbonnais, F., & Chamberlain, M. (1997). Pediatric intensive care nurses and their grief experiences: A phenomenological study. Heart & Lung, 26, 372–386. Rushton, C. (2004). The other side of caring: Caregiver suffering. In B. Carter, & M. Levetown (Eds.), Palliative care for infants, children, and adolescents (pp. 221–243). Baltimore: The John Hopkins University Press. Sanders, C. M., Mauger, P. A., & Strong, P. N. (1979). A manual for the grief experience inventory. Charoltte, NC: Catherine M. Sanders.
J.Z. Adwan Saunders, J., & Valente, S. (1994). Nurses' grief. Cancer Nursing, 17, 318–325. Scott, T. (1994). King Joey. Nursing, 24, 54–55. ToppinenTanner, S., Ojajarvi, A., Vaananen, A., Kalimo, R., & Jappinen, P. D. M. (2005). Burnout as a predictor of medically certified sickleave absences and their diagnosed causes. Behavioral Medicine, 31, 18–27. Zander, M., Hutton, A., & King, L. (2013). Exploring resilience in paediatric oncology nursing staff. Collegian, 20, 17.
Pediatric Nurses' Grief Experience, Burnout and Job Satisfaction1 Jehad Z. Adwan PhD,RN ⁎ University of Minnesota, School of Nursing, Minneapolis, MN, USA Received 31 December 2013; revised 20 January 2014; accepted 22 January 2014
Key words: Pediatric nurses; Grief; Burnout; Job satisfaction; Intention to leave
Correlations among grief, burnout, and job satisfaction among highly satisfied pediatric nurses were examined using the Revised Grief Experience Inventory (RGEI), Maslach Burnout Inventory (MBI), and Index of Work Satisfaction (IWS). Results showed that grief had significant correlations; positive with burnout, negative with job satisfaction. RN's reported significantly higher emotional exhaustion if their primary patients died and higher guilt if patients died younger. Conclusions suggest a dynamic statistical interaction among nurses' grief, burnout, and job satisfaction representing a pathway to intention to leave their unit, organization, or nursing. Recommendations include implementation and evaluation of grief intervention and education programs. © 2014 Elsevier Inc. All rights reserved.
Background Grief and Nurses COWLES AND RODGERS (1991) defined grief as “dynamic, pervasive, highly individualized process with a strong normative component” (p. 121). Pediatric nurses have a unique caring role because of the vulnerable nature of their patients and the special bonds that can form when caring for patients and their families (Scott, 1994). This relationship can expose nurses to intense and recurrent experiences with unresolved loss and grief when one of their patients suffers from a terminal illness or dies (Couden, 2002; Rashotte, Fothergill-Bourbonnais, & Chamberlain, 1997). Rushton (2004) and Landro (2012) indicated that health professionals may suffer and
Previous presentations: 1. Round table discussion: Center for Children with Special Health Care Needs, University of Minnesota, School of Nursing, March 18th, 2011. 2. Nursing Research Day poster presentation, University of Minnesota, School of Nursing, April 29th, 2011. 1 Funding: This research was supported in part by Sigma Theta Tau, Zeta Chapter at the University of Minnesota. ⁎ Corresponding author: Jehad Z. Adwan, PhD,RN. E-mail address: [email protected]. 0882-5963/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pedn.2014.01.011
experience a variety of emotional and physical symptoms including headaches, fatigue, and depression as they watch the suffering of their patients and families. Others argued that the exposure to patients' suffering and death can render the nurse prone to stress and burnout (Gerow et al., 2009; Kennedy, 2005; Peters, 2012). Maslach, Schaufeli, and Leiter (2001) defined burnout as a syndrome that affects individuals who are in the helping or caring professions, such as nurses, psychologists, and therapists. Burnout has three components: emotional exhaustion (EE), depersonalization (DP), and low personal accomplishment (PA) (Maslach et al., 2001). Staff burnout can trigger turnover challenges for managers and institutions alike. Combined with high emotional exhaustion and depersonalization, low personal accomplishment provide ripe conditions for lower quality nursing care, less consciousness about patient safety and eventually higher rates of turnover (Anderson, 2008). Burnout and job satisfaction have been linked to nurse productivity and retention in the workplace (Anderson, 2008; Erenstein & McCaffrey, 2007; Mrayyan, 2005) as well as the quality of patient care (Billeter-Koponen & Freden, 2005; ToppinenTanner, Ojajarvi, Vaananen, Kalimo, & Jappinen, 2005). The Institute of Medicine (IOM) and American College of Critical Care Medicine (ACCCM) advocated for grief support
2
J.Z. Adwan
programs for staff caring for terminally-ill or dying children (Durall, 2011). Such programs are new and are currently being developed (Altounji, Morgan, Grover, Daldumyan, & Secola, 2013). Pediatric oncology nurses reported a benefit from a grief-support program that was delivered in the form of self-care retreat (Altounji et al., 2013). The program provided “rejuvenation, feelings of appreciation, and a revived passion for the job” (p. 22). In a literature review, (Hildebrandt, 2012) identified several strategies that could be used to resolve grief, and as an added benefit, increase retention of oncology nurses. Some of these strategies included: creating supportive environments, debriefing, providing grief education, and alternating patient assignments (p. 602). Numerous studies have explored nurses' grief experiences but focused only on the description of their feelings (Feldstein & Buschman-Gemma, 1995; Papadatou, Bellali, & Petraki, 2002; Rashotte et al., 1997). On the other hand, ways to deal with grief were explored by others (Ashby, Kosky, Laver, & Sims, 1991; McNeely, 1996; Medland, Howard-Ruben, & Whitaker, 2004; Zander, Hutton, & King, 2013) to help find coping strategies for nurse caregivers to use. Despite the abundance of nursing literature on burnout and job satisfaction, these variables have been addressed either individually or in combination with variables related to some aspects of the clinical environment. No studies to date have explored pediatric nurses' grief experience and its relationship to burnout and job satisfaction. The only published study (Anderson, 2008) that investigated grief experience, burnout and turnover patterns was conducted among certified nursing assistants (CNA's) working at nursing homes. Anderson (2008) concluded that complications from grief could contribute to staff burnout, but the results showed that even though grief had no significant direct impact on turnover patterns, it may have a complex and indirect effect on turnover. Braccia (2005) offered an expert's account and advice on how to avoid burnout as a result of the grief experience. This opinion, however, was based on personal experience rather than evidence. Therefore, the purpose of this study was to explore the relationships between pediatric nurses' self-reported grief experiences resulting from patient deaths with burnout and job satisfaction. The study had the following 2 hypotheses: 1. Grief scores are positively correlated with burnout scores. 2. Grief scores are negatively correlated with job satisfaction scores.
Conceptual Framework The conceptual framework for this study was based on the proposed model for healthcare professionals' grieving process (Papadatou, 2000). The framework highlights the three main study variables: grief, burnout, and job satisfaction. Grief is the central concept of this framework and is explained by antecedents or events that occurred prior to the grief
experience as well as consequences or events that take place as a result of the grief process. The behaviors a person exhibits as a response to the grief experience determine whether an individual engages in effective or ineffective coping in an attempt to resolve the emotional and physical consequences of grief including burnout and job satisfaction as outcomes.
Methods Sample and Sampling A convenience sample of 120 pediatric nurses was recruited from a large midwestern academic medical center's four pediatric patient care units and pediatric float pool RN's. Participants were recruited based on the following criteria: 1. Minimum pediatric experience of 12 months. 2. Worked at 0.5 FTE or more at the time of data collection in pediatric setting in the 12 months before data collection started.
Instrumentation and Measurement The study used four instruments to collect data. The demographic information form (DIF) collected information about the participants and their work environment, while the Revised Grief Experience Inventory (RGEI), the Maslach Burnout Inventory (MBI) and the Index of Work Satisfaction Scale for nurses (IWS) collected data about their respective constructs. Below is a detailed description of the instruments. Demographic Information Form (DIF) The DIF included relevant demographic information about the participants in addition to their nursing experience. Data were also collected about the nurses' specific experiences with dying children in terms of how many deaths they experienced in the past 12 months, and whether or not they were the primary nurse for these patients. The DIF also inquired about major events in the life of the nurse, such as death, divorce, and diagnosis of a serious illness in the direct family. The Revised Grief Experience Inventory (RGEI). The RGEI is a 22-question, multidimensional measure of grief (Lev, Munro, & McCorkle, 1993) which includes four subscales: existential tension (ET) referring to meaning of life and existence, depression (Dep), guilt (Glt), and physical distress (PD). The scoring is a self-report Likert-scale style with five possible responses ranging from strongly disagree (1) to strongly agree (5). The RGEI subscales have reliability alphas of 0.87, 0.80, 0.72, and 0.83 respectively and an overall internal consistency alpha coefficient of 0.93 (Lev et al., 1993). The Maslach Burnout Inventory (MBI) The Maslach Burnout Inventory (MBI) is a 22-job-related item questionnaire designed to measure the concept of
Grief Experience, Burnout and Job Satisfaction burnout among human service and helping professionals (Maslach & Jackson, 1981). The MBI uses seven-point Likert-scale responses to job-related views, where a response of a zero (0) means the feeling never exists and a response of six means that the feeling exists every day. The MBI contains three subscales referring to the three components of burnout; emotional exhaustion (EE), depersonalization (DP), and personal accomplishments (PA). The MBI has shown strong psychometric properties as evidenced by its Cronbach's coefficients alpha for internal consistency of its subscales: 0.89; 0.77; and 0.74, respectively. These three components are assessed independently from each other (no total score), as relationships among them have not been understood fully yet (Maslach, Jackson, & Leiter, 2005). Index of Work Satisfaction for Nurses (IWS) The IWS is a two-part (A and B) instrument. Part A was not used because the researcher's interest was in the total numeric score of the IWS rather than the individual subscale scores. Part A is often used for institutional purposes in evaluating staff job satisfaction through ranking of job satisfaction subscales. Part B is composed of 44 seven-point Likert-scale items that assess the nurse's responses from strong agreement (1) to strong disagreement (7) on job satisfaction among nurses in six different areas: pay, autonomy, task requirements, organizational policies, interaction, and professional status. Overall reliability analysis of the IWS showed a Cronbach's alpha of 0.91.
Design This research used a cross-sectional correlational design to examine potential relationships among the core variables (grief, burnout and job satisfaction scores) as they occurred in the clinical setting without implying a causal relationships among them (Polit & Beck, 2012).
3 scores. Multivariate associations were conducted using multiple regression analyses to find explanations of burnout and job satisfaction by the grief scores (Polit & Beck, 2012).
Results Demographic Description Out of 214 eligible pediatric nurses at a large Midwestern Academic Medical Center, 120 (56%) participated in the study. Paper-and-pencil surveys were returned by 87 participants (72.5%), and 33 surveys (27.5%) were submitted online. Participants' ages ranged from 23–62 years (m = 34, SD = 10). Summary of participants' demographic characteristics are in Tables 1a and 1b.
Nurse Experience With Patient Death Participants reported a range of zero to 40 patient deaths in the previous 12 months (m = 13, SD = 9). Forty seven participants (39%) had served as primary nurses for one or more patients who had died during the prior 12 months. At least 68% (n = 82) of participants reported that they had cared for a patient on or around the day the patient died in the past 12 months. Following their patients' death, 114
Table 1a Demographic descriptions of the participants (nominal variables). Descriptor
Sub-descriptor
Count
%
Gender
Female Male White Other Married Single Separated/Divorced Yes No Alone Spouse/SO, no children Spouse/SO, w/ children W/ children Other Associate Bachelor's Master's Specialized Pulmonary, cardiology, renal Heme/Onc., GI, trauma PICU Blood and marrow transplant Float pool Other
114 6 117 3 69 47 4 50 70 25 31 42 4 18 22 91 7 12 16 29 30 33 11 1
95 5 98 2 58 39 3 42 58 21 26 35 3 15 18 76 6 10 13 24 25 28 9 0.8
Ethnicity Marital Status
Data Collection and Analysis Plan Approvals of the institutional review board (IRB) and the target institution were obtained prior to starting data collection. Recruitment was done by sending email invitations to departmental group emails and posting of recruitment flyers in target pediatric units. Data were collected using identical paper and pencil and online surveys. Upon returning completed surveys, each participant received a $10 gift card as an incentive for participation. Quantitative data were entered, cleaned and analyzed using the Statistical Package for Social Sciences (SPSS) version 17.0. Descriptive statistics were used to describe the nurses' grief scores on the RGEI, burnout scores on the MBI, and job satisfaction scores on the IWS as well as means and standard deviations of participants' demographic characteristics. Correlational statistics were conducted to examine relationships among grief, burnout, and job satisfaction
Have children Living with
Education
Unit type
4
J.Z. Adwan Table 1b
Demographic and work‐related descriptors.
Descriptor
Range
M
SD
Age 23–62 yrs. 34 10 Years of nursing experience 1–40 9 9 Years of pediatric experience 1–36 6.7 6.9 Number of children 0–4 0.89 1.15 Number of negative life events 0–4 0.74 0.98 Number of patients' death in 0–40 13 9 past 12 months Longest time known patient 0–20 yrs. 2.6 yrs. 3.4 before they died Shortest time known patient 0–2 yrs 1.6 yrs 3.5 before they died Youngest patient died 0–18 yrs 1.5 yrs. 3 yrs. Time elapsed since last 0–4 yrs. 0.5 yrs. 0.63 patient death
participants (95%) reported that they had shared their feelings with one or more people around them; 103 (86%) shared feelings with a nursing colleague, 80 (67%) with a spouse or significant other, 74 (63%) with a friend, 22 (18%) with parents/siblings, and only two participants (1.8%) reported sharing their feelings with a therapist or counselor. The majority (n = 96, 80%) shared their feelings with two to three others around them. The majority of those who shared feelings with others (n = 111, 92%) reported that sharing feelings was somewhat or very helpful to them.
Grief Experience The results revealed that nurses do experience grief over their patients' death in similar ways as a patient's family care giver does. Although, the scores of the RGEI subscales were consistent with the tool's validation study, participantsunderstandably-exhibited lesser grief intensity compared to family care givers. Descriptive statistics of the total and subscale scores of the complete grief scores are summarized in Table 1c and are compared to the actual results of the total as well as subscale scores from the validation study (Lev et al., 1993).
Burnout Experience Scoring of the MBI was done by individual subscale without calculating a total burnout score because the full Table 1c
relationships among the three concepts PA, EE, and DP were not fully established (Maslach et al., 2005). The results revealed moderate levels of emotional exhaustion and personal accomplishment in addition to low depersonalization scores among the study particiapants. A summary of the MBI's subscales scores is in Table 2.
Job Satisfaction The results revealed high scores in most job satisfaction subscales. Job satisfaction subscales' descriptive attributes are detailed in Table 3 and are listed for reference. For the purposes of this study, only the total job satisfaction score (range = 120–240, max = 308, m = 190, SD = 24) was used in further analysis, namely correlational and regression analyses.
Adverse Personal Life Events Participants provided information on adverse personal life events that they had experienced during the 12-month period prior to their participation in the study. A death in the immediate family was reported by 16% (n = 19), a diagnosis of major illness in the family by 25% (n = 30), a divorce or separation by 3% (n = 4), and other major loss by 5% (n = 6). The total of instances of adverse personal events (Table 4) was counted for each participant and used as a continuous variable for measurement in relation to other variables in the analysis.
Correlations Among Grief, Burnout, and Job Satisfaction Analysis revealed a statistically significant positive correlation between the RGEI total score and MBI's EE subscale (r = 0.38, p b 0.001), a significant positive correlation with the MBI's DP subscale (r = 0.19, p = 0.04), and a negative correlation with the MBI's PA subscale (r = − 0.244, p = 0.009). Most RGEI subscales are individually correlated to the subscales of the MBI subscales in the same direction of the total RGEI scores. Significant negative correlations were found between total RGEI and total IWS scores (r = − 0.29, p = 0.002). Furthermore, IWS total score had a significant negative correlation with MBI's EE and DP,
Study RGEI compared to normative data (Lev et al., 1993) and its subscales.
Current study (N = 120)
Validation study (N = 418)
Subscales→
Existential tension (30 max)
Depression (30 max)
Guilt (15 max)
Physical distress (42 max)
Total grief score (110 max)
Min–Max x SD Min–Max x SD
6–28 11.13 4.8 – 20.1 8.5
6–29 14.92 7.56 – 23 7
3–14 7.32 2.76 – 10 4.6
7–32 15.09 6.30 – 22.5 9.3
22–94 48.29 17.77 – 75.5 25.7
Grief Experience, Burnout and Job Satisfaction Table 2
5 Table 4
MBI subscales.
Descriptive measures
Emotional Depersonalization Personal accomplishment exhaustion (max: 30) (max: 56) (max: 48)
Min–Max x SD Burnout level
13–48 38 5.7 Moderate
5–45 21 8.58 Moderate
0–19 5.8 4.7 Low
while it had a significant positive correlation with MBI's PA. Table 5 summarizes correlations among the three core variables.
Correlations Between Grief and Demographic Variables Due to non-normal distribution of the following demographic variables, a Spearman's correlation was used and revealed a significant positive correlation between the number of primary patients' deaths and the RGEI's total scores (r = 0.21, p = 0.026). Except for the physical distress subscale, the number of primary patients death was significantly correlated with the RGEI's subscales as follows: existential tension (r = 0.18, p = 0.047), depression (r = 0.24, p = 0.01), and guilt (r = 0.19, p = 0.039). The grief total score was not significantly correlated with the number of patients the nurse cared for on or around the day they had died. The grief's guilt subscale, however, was negatively correlated with the numbers of patients the nurse had cared for around the time of death (r = − 0.19, p = 0.04). Additionally, statistically significant positive correlations were found between the age of the youngest patient who died and the burnout's depersonalization subscale (r = − 0.22, p = 0.03). The average period the nurse knew the patient before they died and grief's guilt scores were also significantly correlated (r = 0.20, p = 0.044). It is important to point out that no statistically significant correlations were identified between the total perceived
Table 3
Descriptive attributes of IWS total and subscale.
IWS scale/subscale
Descriptive measures Max possible Min Max x scores
Pay Autonomy Task requirements Organizational policies Professional status Interactions IWS total
42 56 42 49
6 21 7 6
38 51 37 36
SD 24.67 36.81 20.88 16.7
5.8 6.7 5.2 5.4
Total numbers of adverse life events.
Total adverse personal life events
n
%
0 1 2 3 4
69 20 25 5 1
58 17 21 4 0.8
events event events events events
number of patient's deaths or how recent the latest death was and grief's total scores or its four subscales. The participant's age and gender had no significant correlations with grief, burnout, or their respective subscale scores. Furthermore, the analysis did not show any significant correlations between either participant's length of experience in nursing in general or in pediatric nursing in particular and any of the grief, burnout, or job satisfaction scores, including their respective subscales. Also, the total number adverse personal life events did not reveal any significant correlations with grief, burnout, or job satisfaction total scores or subscales as applicable.
Regression Analysis Multiple regression analysis was used to further examine relationships among variables based on previously established significant correlations. Regression models were created using total grief scores and the four subscales of the RGEI (existential tension, depression, guilt, and physical distress), three MBI subscales (emotional exhaustion, depersonalization, and personal accomplishment), and the total score of the IWS. All variables that are correlated with these subscales were entered into regression models using stepwise process, where each of the subscales was entered as dependent variables and all other correlates with a pvalue b 0.1 were entered as candidate independent variables in the model. Regression analysis showed that the RGEI's guilt subscale score explained about 7% of the variance in the MBI's personal accomplishment score (beta = − 0.56, p = 0.003). The model involving the IWS total scores and the RGEI's existential tension (where the person feels like the purpose of life has been lost) explained 31% of the variance in the MBI's emotional exhaustion score. Conversely, a quarter of grief score variance was predicted by the MBI's
Table 5
Correlations matrix among the RGEI, MBI, & IWS.
Total RGEI 49 70 308
22 49 40.80 4.5 26 68 51.11 7.6 120 240 190.97 24
Total IWS
r p r p
Total RGEI
MBI-PA
MBI-EE
MBI-DP
− 0.29 0.002
− 0.244 0.009 0.238 0.009
0.381 b .001 − 0.464 b .001
0.192 0.042 − 0.358 b .001
6 emotional exhaustion score (beta = 0.89, p b 0.001), the length of time the nurse knew the patient before they had died (beta = 0.12, p = 0.02), and the number of primary patients who died in the previous 12 months (beta = 2.19, p = 0.047).
Discussion It is not surprising that nurses' reactions to the death of their patients are not as intense or overreaching as those of the patient's family members. Consistent with reviewed literature (Brown & Wood, 2009; Feldstein & BuschmanGemma, 1995; Papadatou et al., 2002), this study suggests that nurses experience grief over the death of their patients in ways similar to the patient's family members, yet to lesser extents. This study demonstrated that nurses' grief scores were near-normally distributed indicating a viable range of emotional and physical reactions with significant correlations with all aspects of burnout and further suggesting a statistical explanation to burnout's emotional exhaustion aspect. Even though the current literature does not show the direct relationship between grief and emotional exhaustion, it makes clinical as well as scientific sense to establish that those two concepts are closely related to each other. The available literature, though more anecdotal than empirical, supports that nurses' grief may lead to burnout if not addressed properly (Braccia, 2005; Brown & Wood, 2009; Saunders & Valente, 1994). Nurses felt that sharing feelings about patient death was helpful. Interestingly however; that perception was correlated with emotional exhaustion, which could be interpreted that person-to-person sharing of feelings may actually have undesirable consequences even though nurses engaged in it may see it as a helpful strategy, and that could be the start of the burnout process. Higher levels of grief and its components are associated positively and even suggest explanations to some components of burnout, i.e. emotional exhaustion, which in turn is linked to nurses' intention to leave their unit, hospital, or even nursing as a profession. Grief and its components were correlated negatively with job satisfaction although this cohort of nurses was highly satisfied in most of the work satisfaction subscales. Some of the participants' burnout scores were, though relatively low, associated their perceived numbers of patients' deaths on their unit, their feeling of guilt over these deaths, and their personal existential tension felt as a result of these deaths. Nurses are able to continue their professional growth and have sense of personal accomplishment in an environment where patients' death may bring about grief and burnout. Participants' job satisfaction and grief's existential tension scores suggested a valid explanation of burnout's emotional exhaustion, which could be seen that burnout may be an intermediary force between grief and job satisfaction that could eventually lead the nurse to make a career change. Intention to leave unit was correlated to nurses' grief, burnout and job satisfaction issues which has
J.Z. Adwan potential human resource implications for employers costing money to replace and train new staff.
Recommendations and Conclusions Recommendation for Practice Informal strategies of coping with grief and its consequences such as sharing feelings with peers, loved ones, and relaxation techniques, while seen helpful by some nurses, may not be adequate to overcome the emotional as well as the physical aspects of the grieving process. Worse yet, sharing feelings may actually be working against its perceived benefit. A professional early intervention program for nurses is recommended to be used regularly when a situation calls for grief intervention following a patient's death. There is a need for grief education to help nurses better understand the emotional as well as physical toll of experiencing a patient's death repeatedly over a period of time. Organizational recommendations include reassessment of attrition rates and how grief, burnout, and job satisfaction play a role in the process and the financial impact on the organization.
Recommendations for Research Refining of the RGEI tool is recommended because this is the first known time such an instrument, designed originally for family care givers, to be used in clinical setting to measure staff grief even though it demonstrated strong psychometric properties when compared to the original instrument. It is recommended that the RGEI be examined by a panel of experts to re-formulate its statements to make them suitable to nurse–patient relationship setting rather than a family caregiver setting. Another logical outcome of the current study is to design and evaluate a program of grief education where objectives are established such that therapeutic grief resolution is attained through learning how to cope with patients' deaths as well as organizational early engagement when deaths occur. An evidence-based program that builds on current coping literature among pediatric nurses is needed to evaluate therapeutic strategies among nurses that can be tested for effectiveness and later used to help nurses cope with grief.
Limitations Although the RGEI was based upon the 102-item inventory, which was used on nurses (Sanders, Mauger, & Strong, 1979), the Revised Grief Experience Inventory (RGEI) was tested in its validation study on grieving family caregivers who had experienced the loss of a loved one. Therefore, its statements and questions were geared toward what is expected of a family to experience. Consequently, the
Grief Experience, Burnout and Job Satisfaction responses of the current study participants show lower level of grief which could have affected the results if the instrument were designed for measuring nurses' grief. The Maslach Burnout Inventory is the most widely used burnout measure, which has been used in theory-driven research on the burnout phenomenon. Its multidimensional conceptualization of the concept, although a theoretical strength, represents a limitation at the same time in that these dimensions are understood better when applied to groups of persons. They have not been used, at least for now, to assess burnout among individuals because the instrument does not have well-validated cut-off points for burnout scores (Maslach & Jackson, 1999). The convenient nature of the sample is a limitation. This limitation is shared by non-experimental designs under which this study falls, where a risk for skewed results may occur because of unequal chances of representation of all participants. The nature of the setting where the current study took place and the characteristics of the participants suggest overall equal chances among participants for exposure to patient's death due to the high acuity of the patient population.
Conclusion This study has demonstrated that pediatric nurses are at a real risk of experiencing grief following their patients' death. It is evident that even though the participants of this study were highly satisfied with their jobs as evidenced by their high levels of job satisfaction and personal accomplishments, they still experienced grief in similar ways a family member does-albeit at a lower level. However, once nurses manage to cope with losing one patient, they may have a rerun of the same experience with another patient and the cycle goes on. It is important for nurses, managers, and institutions to understand the real impact of grief on nurses in terms of suffering job satisfaction, higher burnout, which may lead to nurses' thinking about leaving a stressful unit, hospital, or even nursing altogether. This could have significant financial ramifications for institutions. Dealing with grief and its consequences is a shared responsibility of the nurses, managers, and the institutions. Grief interventions where therapeutic coping is learned as a skill could be a useful tool to deal with grief's impact in its early stages before or soon after a patient dies.
References Altounji, D., Morgan, H., Grover, M., Daldumyan, S., & Secola, R. (2013). A self-care retreat for pediatric hematology oncology nurses. US: Sage Publications. Anderson, K. A. (2008). Grief experiences of CNAs: Relationships with burnout and turnover. Journal of Gerontological Nursing, 34, 42–49.
7 Ashby, M. A., Kosky, R. J., Laver, H. T., & Sims, E. B. (1991). An inquiry into death and dying at the Adelaide-Childrens-Hospital - A useful model. Medical Journal of Australia, 154, 165–170. Billeter-Koponen, S., & Freden, L. (2005). Long-term stress, burnout and patient-nurse relations: Qualitative interview study about nurses' experiences. Scandinavian Journal of Caring Sciences, 19, 20–27. Braccia, D. (2005). Follow these tips to avoid burnout from chronic grief and compassion fatigue. ONS News, 20, 6. Brown, C., & Wood, A. (2009). Oncology nurses' grief: A literature review. Clinical Journal of Oncology Nursing, 13, 625–627. Couden, B. (2002). Sometimes I want to run: A nurse reflects on loss in the intensive care unit. Journal of Loss & Trauma, 7, 35–45. Cowles, K., & Rodgers, B. (1991). The concept of grief: A foundation for nursing research and practice. Research in Nursing and Health, 14, 119–127. Durall, A. (2011). Care of the caretaker: Managing the grief process of health care professionals. Pediatric Annals, 40, 266–273. Erenstein, C., & McCaffrey, R. (2007). How healthcare work environments influence nurse retention. Holistic Nursing Practice, 21, 303–307. Feldstein, M., & Buschman-Gemma, P. (1995). Oncology nurses and chronic compounded grief. Cancer Nursing, 18, 228–236. Gerow, L., Conejo, P., Alonzo, A., Davis, N., Rodgers, S., & Domain, E. (2009). A phenomenological study of nurses' experience of grief following patient death. Western Journal of Nursing Research, 31, 1078–1079. Hildebrandt, L. (2012). Providing grief resolution as an oncology nurse retention strategy: A literature review. US: Oncology Nursing Society. Kennedy, B. R. (2005). Stress and burnout of nursing staff working with geriatric clients in long-term care. Journal of Nursing Scholarship, 37, 381–382. Landro, L. (2012). When nurses catch compassion fatigue, patients suffer. Wall Street Journal - Eastern Edition, 259, D1–D3. Lev, E., Munro, B. H., & McCorkle, R. (1993). A shortened version of an instrument measuring bereavement. International Journal of Nursing Studies, 30, 213–226. Maslach, C., & Jackson, S. E. (1981). The measurement of experienced burnout. Journal of Occupational Behavior, 2, 99–113. Maslach, C., & Jackson, S. E. (1999). Factorial validity of the Maslach Burnout Inventory–Dutch (MBI-NL) among dentists. Journal of Organizational Behavior, 20, 209–217. Maslach, C., Jackson, S. E., & Leiter, M. P. (2005). Maslach Burnout Inventory–general survey–“English”. Behavioral Medicine, 31, 18–27. Maslach, C., Schaufeli, W. B., & Leiter, M. P. (2001). Job burnout. Annual Review of Psychology, 52, 397–422. McNeely, S. (1996). Stress and coping strategies in nurses from palliative, psychiatric and general nursing areas. Health Manpower Management, 22, 10–12. Medland, J., Howard-Ruben, J., & Whitaker, E. (2004). Fostering psychosocial wellness in oncology nurses: Addressing burnout and social support in the workplace. Oncology Nursing Forum, 31, 47–54. Mrayyan, M. T. (2005). Nurse job satisfaction and retention: Comparing public to private hospitals in Jordan. Journal of Nursing Management, 13, 40–50. Papadatou, D. (2000). A proposed model of health professionals' grieving process. Omega, 41, 59–77. Papadatou, D., Bellali, T., & Petraki, D. (2002). Greek nurse and physician grief as a result of caring for children dying of cancer. Pediatric Nursing, 28, 345–353. Peters, L. (2012). Is work stress in palliative care nurses a cause for concern? A literature review. International Journal of Palliative Nursing, 18, 561–567. Polit, D., & Beck, C. (2012). Resource manual for nursing research: Generating and assessing evidence for nursing practice (9th Edition ed ). Philadelphia, PA: Wolters Kluwer Health/Lippincott Williams & Wilkins.
8 Rashotte, J., Fothergill-Bourbonnais, F., & Chamberlain, M. (1997). Pediatric intensive care nurses and their grief experiences: A phenomenological study. Heart & Lung, 26, 372–386. Rushton, C. (2004). The other side of caring: Caregiver suffering. In B. Carter, & M. Levetown (Eds.), Palliative care for infants, children, and adolescents (pp. 221–243). Baltimore: The John Hopkins University Press. Sanders, C. M., Mauger, P. A., & Strong, P. N. (1979). A manual for the grief experience inventory. Charoltte, NC: Catherine M. Sanders.
J.Z. Adwan Saunders, J., & Valente, S. (1994). Nurses' grief. Cancer Nursing, 17, 318–325. Scott, T. (1994). King Joey. Nursing, 24, 54–55. ToppinenTanner, S., Ojajarvi, A., Vaananen, A., Kalimo, R., & Jappinen, P. D. M. (2005). Burnout as a predictor of medically certified sickleave absences and their diagnosed causes. Behavioral Medicine, 31, 18–27. Zander, M., Hutton, A., & King, L. (2013). Exploring resilience in paediatric oncology nursing staff. Collegian, 20, 17.
