Moral distress, ethical climate, and compassion fatigue among oncology nurses: the mediating role of moral distress

In this descriptive-correlational study, 122 nurses working in oncology wards were selected from six hospitals (Imam Khomeini, Imam Hossein, Ayatollah Taleghani, Shariati, Rasoul Akram, and Shohadaye Tajrish) affiliated with Tehran, Shahid Beheshti, and Iran University of Medical Sciences in Tehran, Iran, from October 2023 to March 2024 to clarify the correlation between moral distress, ethical climate, and compassion fatigue in oncology nurses. The convenience sampling method was utilized.

The criteria for inclusion involved nurses who held a bachelor's degree or higher and were actively employed on a full-time basis in a clinical setting. Applicants had to have at least one year of nursing experience in oncology wards (time to gain a sufficient understanding of the workplace environment) and express their readiness to participate in the study. Participants were excluded if they demonstrated a refusal to continue their cooperation in the research or if they left questionnaires incomplete. To mitigate participant dropout, questionnaires were meticulously designed for optimal respondent experience, and participants were encouraged to complete them in their entirety. Following approval from the Ethics Research Center of Shahid Beheshti University of Medical Sciences, the researcher delivered the introduction letter to the hospitals where the study was planned to be implemented. Under hospital management's coordination and consent, the researcher visited nurses during different shifts (morning, evening, and night) and distributed the questionnaire among them. The completed questionnaires were collected in the next shifts. The researcher provided comprehensive explanations regarding questionnaire completion, emphasizing that participants should complete it only once. The questionnaires were distributed in a certain number and then received in the same number.

The study data were collected using a demographic information form, Corley's Moral Distress Questionnaire, Revised Victor and Cullen's Ethical Climate Questionnaire, and the Professional Quality of Life (ProQOL) questionnaire. The demographic information form included age, sex, marital status, work experience, education level, average number of monthly shifts, and financial situation. This form was designed after reviewing the literature and using experts' opinions.

The research objectives were explained to the participants after obtaining approval from the Ethics Research Center of Shahid Beheshti University of Medical Sciences and getting an introduction letter to enter hospitals to make the necessary arrangements. Informed written consent was obtained, and they were also assured of the confidentiality of their information. To maintain confidentiality, participants were told they did not need to reveal personal information like their name, last name, or any other identifying details. Additionally, they were informed about responding to the questionnaires and the voluntary nature of the study. Also, the data will be maintained and stored only by the corresponding author to make it available to researchers for non-commercial use in research work if necessary and upon request.

Descriptive statistics were employed to analyze the data, encompassing the mean and standard deviation for normal quantitative variables and the median and interquartile range for non-normal distributions of quantitative variables. The data normality was assessed through graphical (Q-Q plots, P-P plots, and histograms) and numerical techniques, along with implementing the Kolmogorov-Smirnov test. The report included point estimation and interval estimation at a 95% confidence level to estimate the scores of variables. Spearman's non-parametric test was used because of the lack of normal data distribution. The correlation between primary outcomes and quantitative demographic variables was assessed using the Spearman correlation test. The relationship with two-mode qualitative demographic variables was examined using the Mann-Whitney U test, while the Kruskal-Wallis H test was employed for multi-mode qualitative variables. The analysis mentioned above was executed using SPSS version 27, with a significance level of 5% being considered.

The data was also analyzed using path analysis. The research focused on analyzing and reporting the associations between the dimensions of ethical climate, moral distress, and compassion fatigue while also accounting for the adjustment of age and gender. The path analyses were conducted using Stata 17 software. Model fit was evaluated using the comparative fit index (CFI), Tucker–Lewis's index (TLI), Standardized Root Mean Square Residual (SRMR), and Root Mean Square Error of Approximation (RMSEA). CFI, TLI, and NFI values greater than 0.90 [27] and RMSEA values <0.08 [28] were used as the cutoffs for acceptable fit.

The following hypotheses are proposed based on the arguments and literature search of researchers in this study (Figure 1).

The sample size was determined using information from the pertinent article by Ventovara et al. [3]. The sample size was determined based on ascertaining the moral distress score. To achieve this aim, a sample size of 122 individuals was calculated based on an estimated error of 3% of the expected mean (0.09), a confidence level of 95%, and a standard deviation of 0.5. These calculations were also suitable for other objectives.

One hundred twenty-two questionnaires were distributed to nurses, and all questionnaires were completed and returned. Thus, the study concluded with a final sample size of 122 participants.

The nurses' mean age and work experience were reported as 30.34 ± 4.18 and 4.56 ± 3.51 years, respectively. Most nurses (61.5%) were males, most (63.9%) were single, and 82.8% had a bachelor's degree in nursing. On average, the participants worked 27.20 ± 4.48 monthly shifts (Table 1).

The overall moral distress, compassion fatigue, and EC mean scores were 125.54 ± 37.50, 31.50 ± 9.23, and 49.03 ± 7.94, respectively. The revised EC questionnaire's egoism, benevolence, and principled subscale scores were 15.48 ± 6.34, 17.31 ± 4.12, and 16.22 ± 3.52, respectively (Table 2).

The Kolmogorov–Smirnov test showed that the distribution of the scores of moral distress, ethical climate, and compassion fatigue was not normal (Table 3).

The Spearman correlation test found a significant correlation between nurses' moral distress and compassion fatigue (r=.861, P=0.00). Also, moral distress was correlated to all subscales of the revised EC questionnaire (r=.220, P=0.015).

Despite this, no significant relationship was observed between compassion fatigue and the total EC score (r=.114, P=0.212). There was a significant association between the EC questionnaire's egoism, benevolence, and principled subscales and compassion fatigue (r=-.743, P=0.000, r=.527, P=0.000, r=.336, P=0.000, respectively). The detailed information on the statistical analysis is presented in Table 4.

In the analysis of the theoretical model, an adequate fit was obtained, Chi2/df = 4.07, CD = .74, TLI = .96, CFI = .99, SRMR = .02, and RMSEA = .08 (CI: 0.00–0.193).

The standardized coefficient, also known as the beta coefficient or beta weight, allows for comparison between coefficients. A more significant coefficient, in terms of absolute value, signifies a more significant impact of that variable on the dependent variable than other variables. The sign of the correlation provides the direction of the relationship. The sign indicates whether the two variables are positively or negatively related.

Results showed that when egoistic EC increases by 1 unit, moral distress score increases by 4.22 units. It was statistically significant (p <.001). If Benevolent EC increases by 1 unit, CMD decreases by 2.98 units. This was statistically significant, too (p <.001). The complete results of the analysis are presented in Table 8.

The analysis showed that among the dimensions of EC (including egoism and benevolence and principled climate), egoism directly and indirectly and benevolence indirectly (through moral distress) were significantly related to compassion fatigue. Also, the total effect of egoism and benevolence on compassion fatigue was significant, but the principled EC did not show any direct or indirect impact. The impact of moral distress on compassion fatigue was also substantial, as an increase in the moral distress score decreases the compassion fatigue score by 0.16 points (p-value<.001) (Figure 2).

According to the findings of our study, the moral distress experienced by oncology nurses was at a moderate level. The incidence and severity of moral distress among nurses vary according to the results of studies conducted in different hospital wards and geographical areas. However, moral distress tends to be more prevalent in developing countries than in developed countries [29]. As previously stated, the level of moral distress experienced by nurses varies across different wards of the hospital. Several factors contribute to the complexity of this matter. Moral distress is more severe among intensive care unit nurses and oncology nurses. The primary cause of moral distress experienced by these nurses is the management of issues, particularly those arising during end-of-life care for patients [30, 31]. A systematic review study examining oncology nurses revealed that nurses encountered a low to moderate frequency of moral distress. Of course, the intensity of moral distress was estimated to be high. Our study yielded similar results to this study [32].

The findings from a qualitative research study revealed three primary themes of distressing causes: [1] a sense of powerlessness at various levels (patients/family, team, and organization), [2] concerns related to end-of-life issues, and [3] challenges in team function (including communication and collaboration issues, incompetence among healthcare providers, and inappropriate behavior of colleagues) [33]. All these factors may lead nurses to a position where they believe that "we can't be the nurse we want to be" [34].

Our study observed that variables including level of education, marital status, and financial status significantly influenced the moral distress experienced by oncology nurses. Moreover, it has been determined that there was a notable increase in the severity of moral distress among oncology nurses as they age, gain work experience, and take on more monthly shifts.

A study conducted in Iran revealed that the moral distress experienced by emergency nurses is moderately intense. Higher levels of moral distress were observed in individuals with characteristics such as female gender, less work experience, younger age, and higher education level [35]. Our study yielded similar results to this study regarding moral distress, although there were notable discrepancies in the various factors examined. Our study yielded similar results regarding moral distress levels to this study, although there were notable discrepancies in the multiple factors examined. The results of our research have shown that moral distress increases with increasing work experience. However, the results of the study mentioned above have shown that nurses experience more moral distress with lower work experience. This may pertain to the study domain, as our investigation specifically targeted oncology nurses.

Across all studies, there is a consistent finding of elevated levels of moral distress among oncology nurses. In addition to demographic factors, the lack of nursing staff, nurses with inadequate experience, insufficient support from organizations and colleagues, inadequate education and knowledge among nurses, lack of teamwork, and heavy workload are risk factors for moral distress [36, 37].

Recent studies indicated a correlation between a positive ethical climate and reduced moral distress [38, 39]. Prior research has also demonstrated a correlation between certain dimensions of ethical climate and professional quality of life [40]. Also, our study findings showed that the perceived EC in the oncology wards of the examined society was benevolence. Principled and egoism were the following priorities, respectively.

Considering that EC refers to individuals' overall perception of moral aspects within organizational functions and procedures. Given the variation in individual situations and working experiences, perceptions of EC may differ even within a hospital ward. However, this study presented an overview of the prevailing EC of oncology wards in Tehran.

In an environment characterized by benevolence, it is anticipated that individuals within the unit exhibit a genuine concern for the well-being of their colleagues, both within and beyond the organization. In this particular climate, there is a notable increase in sensitivity and a greater inclination among organizational members to offer support to one another. Consequently, a supportive environment can bolster the perception of organizational assistance, resulting in increased employee commitment [41]. Fortunately, the prevailing EC in the oncology wards of the hospitals being studied was benevolent. It can enhance group benefits, foster friendship and intimacy, strengthen relationships and social responsibility, and improve the performance of nurses [42, 43].

Principled climates encompass enforcing and comprehending regulations, legislations, and expectations within a social unit [41]. The primary focus of egoistic climates is to maximize one's self-interest. The presence of an egoistic climate in organizational environments should be viewed as a warning for officials and planners. This type of climate fosters a culture where personal gains are emphasized, and the primary objective is to achieve maximum personal benefits [44]. Although the hospitals examined did not have a prominent presence of an egoistic climate, it remains essential for politicians involved in hospital systems to continuously consider strategies for mitigating these perceptions within healthcare environments.

Furthermore, our study revealed a significant increase in the perception of an egoistic climate by nurses with advancing age, work experience, and the number of monthly shifts. Conversely, we observed a decline in principled and benevolent climates in older participants with more experience and a greater number of monthly shifts. Two comparable studies conducted in pediatric oncology settings revealed a positive EC perceived by nurses, aligning with the results of our research [3, 45].

Extensive research indicated a strong, positive relationship between compassion fatigue and moral distress [46, 47]. Based on our study's findings, oncology nurses' compassion fatigue was low. According to a systematic review and meta-analysis, the prevalence of compassion fatigue was found to be of moderate extent. Asian regions have experienced higher levels of compassion fatigue than European and American regions. Furthermore, there was a gradual increase in compassion fatigue among nurses from 2010 to 2019, ultimately reaching its peak in 2019. Amongst all nurses, it was observed that those in intensive care units displayed the most prominent symptoms of compassion fatigue, as revealed by a study. Oncology nurses reported less compassion fatigue [48]. Another study showed that the prevalence exhibits considerable variability, and its association with demographic, personal, and professional factors contributes to this variability [49]. As per a systematic review and meta-analysis, the occurrence of compassion fatigue among oncology nurses was recorded at 62.79% [50]. The level of compassion fatigue among oncology nurses was determined to be "moderate," with 22% of them experiencing a "high" risk of compassion fatigue [51]. Various factors would account for the minor disparities between this study and the studies highlighted in recent reviews. The context of the community in which the study was conducted can be a reason. A scoping review found that nurses' personal beliefs about the nursing care provided and the personality traits of psychological inflexibility, passive coping, workplace conflict, and an unhealthy work-life balance can contribute to compassion fatigue. These traits may be different across diverse communities [52].

Our study discovered that, aside from gender, various demographic factors significantly influence the mean score of compassion fatigue among oncology nurses. Age, work experience, and monthly shifts contribute to increased compassion fatigue. According to a recent meta-analysis, demographic variables (such as age, marital status, education history, health status, and gender), work-related variables (including job satisfaction, income satisfaction, years of work experience, professional position, and work environment), as well as other factors such as support, coping strategy, self-compassion, professional recognition, and psychological training, were found to affect nurses' compassion fatigue, significantly [51].

The results of path analysis showed that the predominance of egoistic EC significantly increases moral distress. On the other hand, the benevolence EC was related to reducing moral distress. The effect size demonstrated a considerably larger magnitude regarding egoistic climate. There is no significant correlation between the principled climate and moral distress.

Regarding compassion fatigue, the questionnaire employed reverse scoring, wherein lower scores signified higher levels of compassion fatigue. The study's results regarding direct effect showed that an egoistic EC was associated with a reduction in the average score of compassion fatigue, signifying a substantial increase in compassion fatigue among nurses. Concerning the direct influence of other ethical climate dimensions, no significant effect was observed on compassion fatigue.

As for the indirect path, an egoistic climate significantly increased compassion fatigue with the mediating effect of moral distress. However, the EC of benevolence was indirectly associated with a significant reduction in compassion fatigue. Regarding the magnitude of the impact, the indirect effect of an egoistic EC on compassion fatigue was more than twice that of a benevolent ethical climate.

Unlike the principled climate, the EC of benevolence and egoism had significantly affected compassion fatigue. Examining direct and indirect paths within each dimension of EC has yielded a significant finding. For instance, it has been determined that an egoistic EC substantially influenced nurses' compassion fatigue by intensifying moral distress instead of direct causation.

The study findings showed that enhancing the EC is crucial for mitigating moral distress and decreasing compassion fatigue. Limited interventions have been examined for their efficacy in enhancing ethical climate. A recent study employed a 12-hour workshop intervention to improve ethical decision-making and interpersonal communication skills. The study recommends nursing managers arrange training workshops to foster positive EC among nurses [53]. The incorporation of these programs into both the continuing education of inpatient oncology nurses and the orientation of newly graduated nurses is advisable due to their effectiveness in lessening compassion fatigue through direct attention to ethical climate and moral distress.

In conclusion, clearly defined learning objectives are recommended for all healthcare professions and continuing nursing sciences education to cultivate an ethically responsible and skilled healthcare workforce across all medical specialties, including oncology. It is imperative to acknowledge the substantial impact of religion, spirituality, and cultural frameworks on medical decision-making processes and the provision of long-term patient care [54]. All of these issues should be taken into consideration by policymakers.

It is essential to acknowledge the limitations inherent in this study. Cross-sectional research designs present initial limitations in establishing causal relationships between variables. Self-reported data's susceptibility to subjective biases can negatively influence the accuracy and dependability of the study's conclusions. The study's limited sample size may affect the validity of specific statistical analyses and the generalizability of the results. Therefore, taking necessary precautions when generalizing the data to nurses other than oncology nurses is recommended. A convenience sampling method was used for participant recruitment; this approach is vulnerable to selection bias. Also, The complex interplay of variables underwent analysis, yet the inherent relationships may exhibit further complexities.