Impact of Exposure to Patient Death or Near Death on Compassion Fatigue in Pediatric Intensive Care Nurses

Kyle S. Richardson, BSc, RN; Meredith MacKenzie Greenle, PhD, RN, CRNP, CNE


Am J Crit Care. 2020;29(4):285-291. 

In This Article


The goal of this study was to determine whether a relationship exists between exposure to patient death or near death and the outcomes of compassion satisfaction, burnout, and secondary traumatic stress among nurses working in pediatric intensive care. We were surprised to find no relationship between nurse-reported exposure to patient death or near death and the outcomes of interest. Other key findings included associations of compassion satisfaction and burnout with educational attainment, experience, and the nurse age-experience interaction term.

The lack of a significant relationship between exposure to patient death or near death and the outcomes of compassion satisfaction, burnout, and secondary traumatic stress conflicts with the findings of both Adwan[11] and Watson and Feld.[13] In those studies, death of a pediatric patient and a nurse's grief were significantly associated with burnout. Both studies used the Maslach Burnout Inventory rather than the ProQOL. Although these tools have a similar construct, in previous studies using both tools, participants score differently on each tool. In 1 study,[14] nurses who completed both the Maslach Burnout Inventory and the ProQOL scored below the high-risk threshold on the Maslach Burnout Inventory but had high scores for burnout and secondary traumatic stress on the ProQOL.

The findings may be explained by our sample, as none of the participants reported low compassion satisfaction or high burnout or secondary traumatic stress. In assessing for traumatic events, we did not separate events resulting in patient death from events the patient survived. In addition, nurses with more than 2 years of experience composed 89% of our sample. Such nurses may have developed appropriate mechanisms to cope with traumatic events. Further investigation is needed to elucidate the relation between exposure to patient death and psychological outcomes in nurses.

In this sample, nurses with a master's or higher degree had higher compassion satisfaction and higher burnout scores than did nurses with an associate's or bachelor's degree. Previously reported results regarding the association between educational attainment and compassion satisfaction are conflicting. In a meta-analysis examining compassion satisfaction and burnout among nurses globally, Zhang et al[15] found that having a bachelor's or a master's degree was associated with higher compassion satisfaction and lower burnout scores than was having a diploma or an associate's degree. They did not, however, compare the results of those with a master's or higher degree with the results from those with a bachelor's degree. In contrast, Sacco et al[16] found that critical care nurses with an associate's or master's degree had higher compassion satisfaction than did nurses with a bachelor's degree. In another study of pediatric nurses, Roney and Acri[12] found no relationship between educational attainment and compassion satisfaction scores.

We did not explore the role(s) that nurses with a master's or a doctoral degree performed, and nurses who are more removed from providing direct patient care (eg, nurse managers, nurse practitioners, clinical care specialists) may have different risk factors for burnout than nurses who work primarily at the bedside. The impact of educational attainment was small in our sample and is unlikely to be clinically relevant. However, little research has been performed on this population, and studies are warranted to investigate the impact of educational attainment and specific nursing role on psychological outcomes.

We found that an interaction between age and experience was associated with compassion satisfaction and burnout. Age increased the protective effect of experience on compassion satisfaction scores; that is, if experience was held constant, older nurses would be expected to have higher compassion satisfaction scores than younger nurses. But age also increased the risk for burnout associated with experience: again, if experience was held constant, older nurses would be expected to have higher burnout scores than younger nurses. Both Berger et al[17] and Sacco et al[16] found an association between age and compassion satisfaction. Berger et al reported that younger nurses had lower compassion satisfaction scores than older nurses, and Sacco et al reported that nurses who were older than 50 years had higher compassion satisfaction scores than nurses younger than age 40 years. Both of these studies suggest that age contributes to compassion satisfaction.

But experience is also important; Berger et al[17] found that nurses with 20 or more years of experience had higher compassion satisfaction than did nurses with less experience. Roney and Acri[12] found high compassion satisfaction and low burnout within their sample, which heavily comprised second-career nurses (who were starting out in nursing at an older age). Zhang et al[15] hypothesized that less experienced nurses may not have had the repetitive exposure to trauma that drives secondary traumatic stress, nor to the cumulative exhaustion that leads to burnout, whereas older and more experienced nurses have developed the coping mechanisms required to maintain compassion satisfaction and avoid burnout. Our research findings suggest that neither age nor experience alone tells the whole story; instead, age and experience interact.

Younger, newer nurses may be less likely to have experienced personal or professional loss before starting their nursing careers. Research shows that new-graduate nurses do not have resources available with which to cope with patient death, and when available resources were used, they were not adequate.[18] Second-career nurses may have developed during their lifetime coping mechanisms that enable them to resist burnout and retain compassion satisfaction. Second-career nurses may, however, be more likely to have external stressors that contribute to burnout; Meyer et al[19] discovered that personal stressors, and life stress at baseline, are predictors of compassion fatigue and burnout. Further work is needed to examine specifically whether a difference exists between nurses for whom nursing is a first career and those for whom nursing is a second career, and how best to support each group. We suggest that health care organizations recognize that stressors may differ on the basis of age and experience, and tailor interventions to different groups. Examples of this include a workshop focused on pediatric nurses who are new mothers and may now begin overidentifying with patients, or a residency program adapted to fit the different needs of first-career and second-career nurses.


Because of the survey nature of this study, we relied on nurse self-report. Although we noted in the consent form that survey responses would be kept confidential, nurses—especially those who received the survey through their work e-mail—may have felt pressured to answer questions with a bias toward positive feelings. Also, those with lower burnout and secondary traumatic stress or higher compassion satisfaction may have been more likely to complete the survey.

We would have liked to obtain more granular detail for several key variables. We did not ask participants to identify whether they worked in a pediatric or a neonatal intensive care setting, to specify their role in their unit (eg, bedside nurse, nurse manager, nurse educator), or to identify the outcome of resuscitation events (eg, whether a code ended with the patient dying or surviving). Other unexplored potential confounders include socioeconomic status, time spent away from the bedside, and policies regarding grief counseling after a patient dies.

Another limitation is the small sample size. We did not analyze power before the study, as this was intended as a pilot study given the limited literature on the topic. A follow-up study is warranted with a larger, more nationally representative sample. Finally, only 1 participant had an associate's degree, so our findings cannot be generalized to nurses who hold such a degree.