Relationships Among Palliative Care, Ethical Climate, Empowerment, and Moral Distress in Intensive Care Unit Nurses

The study hypotheses were as follows: (1a) Increased levels of moral distress are associated with a poor ethical climate. (1b) Access to palliative care affects the relationship between ethical climate and moral distress. (2a) Increased levels of moral distress are associated with decreased empowerment. (2b) Access to palliative care affects the relationship between moral distress and empowerment. (2c) A curvilinear relationship exists between moral distress and empowerment. (3) Access to a full specialist palliative care team is associated with lower levels of moral distress.

We conducted a nationwide web-based survey via the American Association of Critical-Care Nurses (AACN) e-newsletter (>10 000 recipients, untracked for receipt of email). Potential participants included critical care nurses who receive e-newsletters or who access AACN’s social media (an unknown number of viewers). Criteria for participation included the following: current employment as an ICU nurse, provision of direct patient care in an adult ICU, and having provided care for a dying patient within the past 6 months. A total of 288 potential qualified participants opened the survey and agreed to the informed consent; 235 participants completed the survey, yielding an 82% completion rate. Three other participants completed most of the survey, including key variables. These participants were included, yielding a total sample size of 238. We identified no significant differences in demographics between those who did and did not complete the survey. Participant demographics did not significantly differ from AACN member demographics. A power analysis indicated that the needed sample size was at least 222 participants, and the recruitment goal was met in 1 week in March 2015. The University of California, San Francisco, granted ethical approval for human participants.

The survey included 3 standardized instruments and 1 questionnaire developed specifically for this study. The Moral Distress Scale–Revised (MDS-R) includes 21 items describing specific scenarios, and participants are asked to rate each item for frequency and intensity of moral distress.¹⁵  Calculated composite scores have a possible range of 0 to 336, with higher scores indicating higher levels of distress.

The Hospital Ethical Climate Survey (HECS) is a 26-item scale with 5 domains relating to nurses’ relationships with patients, peers, physicians, managers, and the hospital.²⁴  Each item is rated from 1 to 5, and calculated composite mean scores range from 1.00 to 5.00. A score of greater than 3.50 is interpreted as a positive ethical climate.

The Psychological Empowerment Instrument (PEI), a measure of perceived work environment, includes 12 statements that describe the participant’s self-perception.²⁵  Calculated composite mean scores range from 1.00 to 7.00, with higher scores reflecting greater perceived empowerment.

We developed a questionnaire to assess nurses’ perceptions of access to, use of, and barriers to palliative care in the ICU. We created questions with input from previous studies, palliative care researchers, ICU nurses, and palliative care clinicians and pilot tested the questionnaire with a small group of ICU nurses. The questionnaire included triggers for consultation, methods of palliative care inclusion on a patient’s care team, and the level of nurse education on palliative care topics. This study focused only on the questions related to palliative care access.

We collected data with an online survey tool (Qualtrics) administered through the University of California, San Francisco. We reviewed the data and then imported them into statistics software (SPSS version 22.0, IBM) for analysis. We used the Cronbach α to analyze instrument scores for reliability (scores: MDS-R = 0.97, HECS = 0.93, and PEI = 0.89). We assessed frequencies, descriptive statistics, and distributions and evaluated correlations among primary independent variables, demographic characteristics, and moral distress. We included variables found to be significantly associated with moral distress in a multiple regression model. Categorical variables were dummy-coded as dichotomous variables, with a designated reference variable for analysis in the regression model.

Previous study findings are equivocal regarding ICU experience as a significant factor for the explanation of moral distress.^2,11,26,27  Therefore, we included ICU experience in the model as a potentially meaningful covariate in the overall regression analysis.

Study participants were primarily white and female, and most held a bachelor’s degree in nursing (Table 1). A broad range of facility types and sizes were represented. Seventy-nine percent of participants (185 of 235) reported that they felt confident to describe palliative care in the ICU setting, suggesting that they had knowledge and understanding of the definition of palliative care. Seventy-four percent of participants (175 of 235) reported having routine interdisciplinary patient rounds, and for 92% of these (161 of 175), rounds included the bedside nurse. Twenty-nine percent of nurses (67 of 235) could request palliative care consultation without a direct physician order.

MDS-R scores were from 0 to 225, with a mean (SD) score of 96.5 (55.8); higher scores reflect greater distress. The HECS scores were from 1.96 to 5, with a mean (SD) score of 3.9 (0.5); higher scores indicate a positive ethical climate. PEI scores were from 1 to 7 (possible 1–7), with a mean (SD) score of 5.3 (0.8); higher scores indicate greater perceived empowerment. Access to a full palliative care team was defined as specialist palliative care clinicians, including physicians, nurses, social workers, and spiritual care practitioners. Seventy-three percent of participants (171 of 235) claimed access to a full palliative care team.

Moral distress was negatively correlated with empowerment (Table 2). Analysis of a curvilinear relationship between empowerment and moral distress was statistically significant (R² change = 0.02; F = 4.866; P = .03) in regression analysis. Participants scoring lowest on the PEI scale had lower MDS-R scores, participants scoring highest on the PEI scale had lower MDS-R scores, and participants scoring moderately high on the PEI had higher MDS-R scores. All associations were small but statistically significant. We analyzed MDS-R subscales for moral distress frequency and intensity in relationship to the PEI score. We found a negative correlation (r = −0.188; P = .005) between moral distress frequency and PEI score. Participants with higher empowerment scores had lower moral distress frequency scores. We found no significant relationship between moral distress intensity and PEI score. These findings support our study hypotheses. Contrary to the study hypothesis, we found no moderating effect of access to palliative care in the relationship between empowerment and moral distress.

Moral distress and ethical climate were negatively correlated, with moral distress levels lower in a positive ethical climate, thereby supporting the study hypothesis. The association between ethical climate and moral distress was not influenced by access to palliative care.

We found a weak but significant correlation between access to palliative care and the experience of moral distress. Nurses reporting access to a full palliative care team reported higher levels of moral distress. The mean (SD) MDS-R score for participants with access to a full palliative care team was 102.3 (57.2), which was 22.5 points higher than the mean MDS-R score of those who did not have access to a full palliative care team (mean [SD], 79.8 [49.4]). This finding was contrary to the study hypothesis.

Multiple regression findings show that 218 participants were included after listwise deletion (Table 3). Factors that significantly contributed to explaining variance in moral distress were ethical climate, access to a full palliative care team, number of beds in the unit, ethnicity, and educational degree. Ethical climate was the largest unique contributor to the variance in levels of moral distress, with more positive ethical climates associated with lower levels of moral distress. Participants working in units with more beds reported higher levels of moral distress than did those in units with fewer beds. African American participants reported lower levels of moral distress than did participants of other races. Those with nursing diplomas reported less moral distress than did those with higher levels of education. Empowerment and facility type did not make significant unique contributions to the explanation of moral distress variation in the presence of the other factors. We found a positive correlation between HECS and PEI (r = 0.354, P < .001).

Novel findings of this study are that moral distress has an inverse relationship with ethical climate, a curvilinear relationship with empowerment, and a positive correlation with access to palliative care. The strongest relationship was between moral distress and ethical climate, corroborating prior study findings. Each relationship contributes to the larger picture of the moral distress experience of ICU nurses. Education level, ethnicity, and number of ICU beds contributed to nurses’ experience of moral distress, although ethical climate was the most important contributor.

Our study findings validate and support data reflecting an inverse relationship in which participants who work in an ethically supportive environment experience lower levels of moral distress.^5,16,28  The finding of a small but significant correlation between levels of moral distress and the inclusion of nurses in daily patient rounds (r = 0.2, P = .004) supports the proposition that positive ethical climates and collaborative work environments reduce moral distress.^2,9,16,28  Recent data suggest that structured communication processes that include the full team in unit-based ethics conversations and goals-of-care rounds²⁹  or that offer ethics or moral distress consultation services³⁰  have a positive impact on factors contributing to moral distress. Inclusion of nurses in interdisciplinary rounds may help increase collaboration and communication about goals of care and ethical issues as they arise and may subsequently diminish moral distress triggers.

A key component of the operational definition of moral distress is that the individual is prevented from fulfilling a perceived obligation.⁷  Results from this study support those of previous studies that found an inverse relationship between empowerment and moral distress frequency.¹²  Individual nurse empowerment has been a target of educational interventions, particularly related to end-of-life care and team building, yet without success in lowering moral distress levels.^3,17  Findings from this study suggest that increased knowledge and perceived empowerment are not sufficient to attenuate moral distress.

The curvilinear relationship between moral distress and empowerment suggests that participants who perceive very low empowerment and very high empowerment have lower moral distress scores than nurses who perceive moderately high empowerment. One possible interpretation is that nurses with very low perceived empowerment scores may also experience a lesser sense of moral obligation and in turn experience less moral distress, whereas those with very high empowerment scores may be able to deliver their desired level of care despite work environment obstacles. These findings support AACN recommendations that moral distress must be addressed at both individual and organizational levels.³¹ 

Empowerment lost its significant unique contribution for explaining moral distress in the presence of the ethical climate variable, in part because of the correlation between ethical climate and empowerment. Our findings suggest that nurses who are moderately empowered may have an increased sense of obligation related to ethical delivery of patient care. However, in the face of less supportive ethical environments and if not fully empowered to act ethically, nurses may experience higher levels of moral distress.

Our findings did not support the study hypothesis that nurses with access to palliative care resources would experience less moral distress. In fact, nurses who reported having access to a palliative care team experienced higher levels of moral distress. Additional factors that need to be considered are the level of integration of the palliative care teams into the ICU team and clinicians’ beliefs and attitudes about palliative care. Collaboration and integration of care delivery teams affect moral distress.^2,11,16  Although the stated goal of palliative care includes interdisciplinary collaboration,^23,32  insufficient team integration may invoke conflict rather than support. The results of previous studies reflect the inconsistent presence of palliative care in the ICU.²¹  In this study, 74% of participants reported participation in routine daily interdisciplinary patient care rounds, and of those, just 30% (52 of 175) included a palliative care team member. This may reflect an opportunity for further integration of palliative care in ICUs.

An alternative interpretation of the inverse correlation between moral distress and access to palliative care may be that the existence of a palliative care team elevates expectations of care for dying patients, but if the palliative care team is inactive in the unit or inaccessible, these elevated expectations may be unmet. This explanation is consistent with previous findings related to differences in care teams’ understanding of obligations and beliefs of best practice for quality care delivery.⁵  Another possible explanation lies in gaps in clinicians’ understanding of the goal of palliative care and agreement about whether it is appropriate in the ICU. Study findings revealed an opportunity for education: more than 20% of ICU nurses did not feel confident to describe palliative care.

Most ICU palliative care services are provided via consultation.²¹  In a consultation model, there will inevitably be differing perspectives on whether or not to include palliative care services. Without established, standardized consult triggers, inconsistent referral patterns allow varied expectations among individual clinicians that can foster conflict among team members. To facilitate clear expectations and establish a more integrated team approach, institutions should develop a standardized approach to palliative care referral and utilization.

A study limitation is the inability to link participants to specific facilities, limiting comparability of organizations other than within broad categories. Self-selection bias of participants who are more knowledgeable about or interested in palliative care may be present, limiting generalizability to the overall population of ICU nurses. Additionally, we did not collect information on beliefs about palliative care, and the study included only nurses.

A strength of this study is the broad sample of ICU nurses in the AACN nationwide database, which provides a range of perspectives across geographical and practice settings. The anonymity of a web-based survey may have encouraged more honest responses to sensitive questions about work environment and experiences regarding moral distress. Using updated, validated instruments advances the growing knowledge about moral distress and related factors. The novel findings of this study contribute to the understanding of nurse empowerment, experience of moral distress, and integration of palliative care into ICU care.

Discussions about moral distress continue to underscore the need for well-integrated teams to respectfully collaborate and communicate goals and strategies for individualized patient care. There remains a knowledge gap for nurses regarding the role and potential benefit of palliative care for patients in the ICU. Individual empowerment is a meaningful concept for ICU nurses, but it is not more important than the nurse’s relationship to the team and unit. Our study findings support the incorporation of organizational support and team collaboration, including leadership, staff, and clinicians, for any attempted changes in a care delivery process when planning interventions to improve ICU end-of-life care or the ethical climate in the workplace. Education about palliative care and ethics³³  may contribute to each individual’s ability to work within upgraded expectations, but all organizational change must be actively supported by administrative and clinical teams.

We recommend further studies of the collaboration and integration of palliative care teams into ICUs. Qualitative studies may provide more insight into the role of palliative care in the ICU and relationships with nurses’ moral distress. Also, exploring other components of access to and use of palliative care services from the perspective of primary and specialist palliative care providers may give insight into the support available for nurses delivering end-of-life care in the ICU.