1. Introduction
Nurses constitute the largest segment of the healthcare workforce and are central to healthcare delivery, yet their psychological wellbeing remains an important occupational-health concern. Emergency departments are particularly demanding because nurses provide rapid, technically complex and emotionally intensive care while facing unpredictable patient flow, trauma, critical illness and death. Repeated exposure to suffering may contribute to secondary traumatic stress and compassion fatigue, while chronic workload and organisational stressors may contribute to burnout [17, 15, 12]. Reviews of emergency medicine further identify repeated exposure to patient suffering, traumatic events and demanding clinical environments as important contributors to compassion fatigue [3]. At the same time, nurses may retain compassion satisfaction, the fulfilment and professional meaning derived from helping others, even when negative work experiences are present [13, 14].
In sub-Saharan Africa, workforce shortages, high caseloads and constrained resources may intensify these pressures. A regional systematic review and meta-analysis identified a substantial burden of compassion fatigue among healthcare professionals, with nurses among the most affected groups [5]. A broader scoping review similarly identified compassion fatigue as an occupational-health concern across healthcare professions and highlighted individual-, organisational- and system-level antecedents [2]. In Zambia, Kanyanta et al. [6] reported high secondary traumatic stress in 23% of nurses, high burnout in 26% and high compassion satisfaction in 32.4% among nurses in selected teaching hospitals in Lusaka. However, evidence focused specifically on emergency department nurses remains limited.
Understanding professional quality of life in emergency nursing is important for workforce retention, staff wellbeing and the sustainability of high-quality emergency care. Evidence from emergency and critical care nursing indicates that resilience is shaped by working pressure, organisational conditions, social support and opportunities for personal growth [10]. Burnout has been associated internationally with organisational and position turnover, poorer safety climate, adverse events, lower patient satisfaction and lower nurse-assessed quality of care [7, 9]. Context-specific evidence is therefore needed to inform preventive and supportive interventions in Zambian referral hospitals.
This study determined the prevalence of compassion fatigue, burnout and compassion satisfaction among emergency department nurses in three selected Level 3 hospitals in Lusaka, Zambia, and examined socio-demographic and work-related factors associated with these outcomes.
2. Materials and Methods
2.1. Study design and setting
A quantitative cross-sectional study was conducted in the emergency departments of University Teaching Hospitals–Adult Hospital, University Teaching Hospitals–Children’s Hospital and Levy Mwanawasa University Teaching Hospital in Lusaka, Zambia. These are major referral and teaching hospitals providing emergency care to trauma, acute medical and critically ill patients.
2.2. Population, sample and sampling
The target population comprised 281 registered nurses working in the emergency departments of the three hospitals. Eligible participants were registered nurses providing direct patient care who had worked in the emergency department for at least six months and provided informed consent. Nurses with less than six months of emergency experience, those on leave, and staff working solely in administrative or managerial roles were excluded. A sample of 165 was determined using Yamane’s finite-population formula with a 5% margin of error [16]. Proportionate stratified random sampling allocated 104 participants to UTH–Adult Hospital, 19 to UTH–Children’s Hospital and 42 to LMUTH.
2.3. Data collection and measures
Data were collected using a structured self-administered questionnaire. Socio-demographic variables included age, sex, marital status, education and hospital. Work-related variables included nursing experience, emergency department experience, shift pattern, patient load, overtime and consideration of leaving emergency nursing because of stress. Professional quality of life was measured using the 30-item Professional Quality of Life Scale, Version 5 (ProQOL-5) [13]. A pilot study was conducted among 10 nurses from Cancer Diseases Hospital, which was not included in the main study.
2.4. ProQOL-5 scoring
Each ProQOL-5 item was scored from 1 (never) to 5 (very often). Ten items each measured compassion satisfaction, burnout and secondary traumatic stress; compassion fatigue was operationalised using the secondary traumatic stress subscale. Positively worded burnout items were reverse-scored before summation. Each subscale ranged from 10 to 50: scores ≤22 were classified as low, 23–41 as moderate and ≥42 as high. For regression analysis, compassion fatigue and burnout were dichotomised as low versus moderate/high; compassion satisfaction was analysed as moderate versus high because no participant recorded a low score.
2.5. Statistical analysis
Data were entered, cleaned and analysed using Stata version 14. Frequencies, percentages, means and standard deviations summarised participant characteristics and ProQOL-5 outcomes. Pearson’s chi-square test or Fisher’s exact test assessed bivariate associations. Separate binary logistic regression models were fitted for elevated compassion fatigue, elevated burnout and high compassion satisfaction. Variables with bivariate p<0.20 were considered for inclusion, while age and sex were retained as potential confounders. Crude and adjusted odds ratios with 95% confidence intervals were reported. Statistical significance was set at p<0.05. Model fit was assessed using likelihood-ratio and Hosmer–Lemeshow tests, and multicollinearity using variance inflation factors.
2.6. Validity and reliability
The standardised ProQOL-5 was reviewed for contextual relevance. Internal consistency was assessed using Cronbach’s alpha, with coefficients ≥0.70 considered acceptable. In the study sample, alpha coefficients were 0.81 for compassion satisfaction, 0.71 for burnout and 0.76 for compassion fatigue/secondary traumatic stress.
2.7. Ethical considerations
Ethical approval was obtained from the University of Lusaka Research Ethics Committee and the National Health Research Authority of Zambia before data collection. Administrative permission was obtained from the participating hospitals. Participation was voluntary, informed consent was obtained, no identifying information was recorded on the questionnaire, and study data were stored securely for research purposes.
3. Results
All 165 planned questionnaires were complete and included in the analysis. UTH–Adult Hospital contributed 104 respondents (63.0%), LMUTH 42 (25.5%) and UTH–Children’s Hospital 19 (11.5%). Most participants were female (78.2%), aged 20–29 years (52.1%), married (65.5%) and diploma-qualified (83.6%). All were registered nurses.
Regarding work characteristics, 43.6% had 2–5 years of overall nursing experience and 39.4% had 6–10 years. Most (59.4%) had 2–5 years of emergency department experience, 66.1% worked rotational shifts and 54.5% cared for more than 15 patients per shift. All respondents reported working overtime, and 43.0% had considered leaving emergency nursing because of work-related stress.
Mean ProQOL-5 scores were 38.24 (SD 6.70) for compassion satisfaction, 25.70 (SD 5.42) for burnout and 28.67 (SD 6.70) for compassion fatigue. Internal consistency was acceptable to good across the three subscales (Cronbach’s alpha 0.71–0.81).
Compassion fatigue was low in 17.6%, moderate in 74.5% and high in 7.9% of respondents. Burnout was low in 27.9%, moderate in 71.5% and high in 0.6%. Compassion satisfaction was moderate in 61.2% and high in 38.8%; no participant had low compassion satisfaction. For inferential analysis, 82.4% were classified as having elevated compassion fatigue and 72.1% as having elevated burnout.
No socio-demographic or work-related factor was significantly associated with elevated compassion fatigue at the bivariate level. Elevated compassion fatigue was more frequent among nurses with less than two years of emergency experience than among those with at least two years (89.1% vs 78.2%), but the association was not statistically significant (p=0.075). No examined factor was significantly associated with elevated burnout; burnout was more frequent among women than men (75.2% vs 61.1%), but this difference was not significant (p=0.096).
Marital status and education were associated with high compassion satisfaction in bivariate analysis. High compassion satisfaction was reported by 50.9% of single respondents compared with 32.4% of married respondents (p=0.021), and by 59.3% of bachelor’s degree holders compared with 34.8% of diploma holders (p=0.017).
In adjusted models, no factor was independently associated with elevated compassion fatigue or elevated burnout. Nurses with at least two years of emergency experience had lower estimated odds of elevated compassion fatigue, although the confidence interval included the null (AOR 0.42, 95% CI 0.16–1.07; p=0.068). Female nurses had higher estimated odds of elevated burnout than male nurses, but the association was not statistically significant (AOR 1.94, 95% CI 0.89–4.23; p=0.097). For compassion satisfaction, married respondents had lower odds of high compassion satisfaction than single respondents (AOR 0.42, 95% CI 0.21–0.85; p=0.016), while bachelor’s degree holders had higher odds than diploma-qualified respondents (AOR 2.74, 95% CI 1.13–6.62; p=0.026).
Table 1. Distribution of Professional Quality of Life outcomes (N=165)
| Outcome | Low n (%) | Moderate n (%) | High n (%) |
|---|---|---|---|
| Compassion fatigue | 29 (17.6) | 123 (74.5) | 13 (7.9) |
| Burnout | 46 (27.9) | 118 (71.5) | 1 (0.6) |
| Compassion satisfaction | 0 (0.0) | 101 (61.2) | 64 (38.8) |
Source: Study data.
Table 2. ProQOL-5 descriptive statistics and internal consistency
| Outcome | Mean (SD) | Range | Cronbach’s α |
|---|---|---|---|
| Compassion satisfaction | 38.24 (6.70) | 25–48 | 0.81 |
| Burnout | 25.70 (5.42) | 15–43 | 0.71 |
| Compassion fatigue | 28.67 (6.70) | 14–44 | 0.76 |
Source: Study data.
Table 3. Multivariable logistic regression results
| Outcome / factor | AOR | 95% CI | p-value |
|---|---|---|---|
| Elevated compassion fatigue: age ≥30 years | 1.59 | 0.69–3.68 | 0.280 |
| Elevated compassion fatigue: female sex | 1.33 | 0.50–3.54 | 0.568 |
| Elevated compassion fatigue: ED experience ≥2 years | 0.42 | 0.16–1.07 | 0.068 |
| Elevated compassion fatigue: >15 patients/shift | 0.61 | 0.26–1.44 | 0.259 |
| Elevated burnout: age ≥30 years | 0.78 | 0.39–1.55 | 0.482 |
| Elevated burnout: female sex | 1.94 | 0.89–4.23 | 0.097 |
| High compassion satisfaction: age ≥30 years | 1.47 | 0.75–2.88 | 0.259 |
| High compassion satisfaction: female sex | 0.81 | 0.36–1.80 | 0.608 |
| High compassion satisfaction: married | 0.42 | 0.21–0.85 | 0.016 |
| High compassion satisfaction: bachelor’s degree | 2.74 | 1.13–6.62 | 0.026 |
| High compassion satisfaction: >15 patients/shift | 1.78 | 0.90–3.51 | 0.095 |
Source: Study data.
4. Discussion
The principal finding was the coexistence of substantial moderate occupational strain with preserved professional fulfilment. Nearly three-quarters of participants had moderate compassion fatigue and moderate burnout, while all participants reported moderate or high compassion satisfaction. This pattern accords with the ProQOL model, which recognises that positive and negative effects of helping may occur simultaneously rather than as mutually exclusive states [13].
The predominance of moderate compassion fatigue is consistent with broader nursing evidence. Yi et al. [17] reported that registered nurses and nursing students commonly experienced moderate compassion fatigue, while Pan et al. [12] identified emergency nurses as particularly vulnerable because of repeated trauma exposure and stressful clinical conditions. This is consistent with emergency-medicine evidence describing compassion fatigue as arising in settings characterised by repeated exposure to suffering, secondary traumatic stress and occupational strain [3]. In the present study, high compassion fatigue (7.9%) was lower than the 23% high secondary traumatic stress reported by Kanyanta et al. [6] among nurses in selected Lusaka teaching hospitals. Differences in clinical units, participant composition, timing and score distributions may partly account for this variation. Importantly, the relatively small high category should not obscure the 74.5% in the moderate category, who may benefit from preventive support.
High burnout was uncommon (0.6%), but 71.5% of respondents had moderate burnout. The proportion with high burnout was considerably lower than the 26% reported by Kanyanta et al. [6]. Because this study was cross-sectional and relied on self-report, it cannot determine why severe burnout was rare. Nonetheless, widespread moderate burnout remains relevant. A meta-analysis found an inverse association between resilience and burnout among nurses, supporting resilience as a potentially important protective resource [1]. International evidence links nurse burnout with poorer safety climate, more adverse events, lower patient satisfaction and lower nurse-assessed quality of care [9], although patient outcomes were not measured in the present study.
Compassion satisfaction remained comparatively strong: 38.8% had high compassion satisfaction and 61.2% had moderate levels. This high category was slightly greater than the 32.4% reported by Kanyanta et al. [6]. The finding suggests that emergency nurses may continue to derive meaning, competence and fulfilment from patient care despite emotional strain and heavy work demands. Preserving these positive dimensions is important because resilience and constructive professional engagement have been associated with compassion satisfaction and may buffer adverse work outcomes [8, 14]. Qualitative synthesis among emergency and critical care nurses also identifies social support, active learning, occupational benefit and self-regulation as factors relevant to resilience and professional growth [10].
Age and sex were not independently associated with any of the three outcomes. This is broadly consistent with the Zambian findings of Kanyanta et al. [6] for age and for secondary traumatic stress and burnout by sex. Reviews also indicate that demographic predictors are inconsistent across settings and should be interpreted alongside work environment and available support [11].
Marital status was independently associated with compassion satisfaction: married respondents had lower odds of high compassion satisfaction than single respondents. The study did not measure household responsibilities, family support or work-family conflict, so the finding should not be interpreted as evidence that marriage itself reduces compassion satisfaction. Rather, it identifies a subgroup difference that warrants further investigation and supports equitable, non-stigmatising work-life and psychosocial support. This broader focus is consistent with evidence that compassion fatigue reflects interacting individual, organisational and system-level influences rather than a single personal characteristic [2].
Education was also independently associated with compassion satisfaction. Bachelor’s degree holders had approximately 2.7 times the odds of high compassion satisfaction compared with diploma-qualified nurses. Degree-level preparation may be associated with greater professional confidence, autonomy or opportunities for role development, but causality cannot be inferred. Only 27 respondents held bachelor’s degrees, and the confidence interval was relatively wide. Nevertheless, the finding is compatible with evidence that higher professional qualifications may contribute positively to compassion satisfaction in high-acuity healthcare settings [14].
No measured work-related factor independently predicted the three outcomes. This absence of statistical significance should not be interpreted as evidence that working conditions are unimportant. Overtime was universal and therefore could not be tested; rotational shifts and high patient loads were also common, reducing between-participant contrast. In addition, dichotomising ProQOL outcomes may have reduced information, and the sample size limited power for some categories. The descriptive finding that 43.0% had considered leaving emergency nursing because of work-related stress remains important for workforce planning and retention. Research on emergency and critical care nurses likewise highlights high job pressure, burnout and turnover as reasons to strengthen resilience and organisational support [10].
From a practice perspective, the findings support preventive rather than crisis-only approaches. Confidential wellbeing or ProQOL screening, referral pathways, counselling access, peer support following distressing cases, supportive supervision and mentorship for nurses newer to emergency care could be considered and evaluated. Hospital managers may also review patient allocation, overtime, rest periods and rotational scheduling. Mindfulness-based interventions may also be considered as one component of staff support: a systematic review of emergency and critical care nurses found that most included interventions improved psychological wellbeing, compassion or resilience [4]. These actions should be treated as strategies to test rather than as causal remedies established by this cross-sectional study.
Educational progression and continuing professional development may help strengthen the rewarding aspects of emergency nursing. At the same time, workforce support should recognise the coexistence of occupational strain and professional fulfilment. Interventions that focus only on reducing burnout may miss opportunities to reinforce professional meaning, recognition, teamwork and career development.
5. Strengths and Limitations
The study achieved the planned sample of 165 nurses from three major emergency-care sites using proportionate allocation. It used the standardised ProQOL-5 to assess positive and negative dimensions of professional quality of life simultaneously, achieved complete data, demonstrated acceptable-to-good internal consistency (α=0.71–0.81), and used separate multivariable models with checks for model fit and multicollinearity.
Limitations include the cross-sectional design, which precludes temporal or causal inference; reliance on self-report, with possible social-desirability and response-style bias; restriction to three Level 3 hospitals in Lusaka; limited variability in some exposures, including universal overtime; small numbers in some categories; and dichotomisation of outcomes for logistic regression. The questionnaire did not measure potentially relevant factors such as resilience, coping style, psychological support or work-family conflict.
6. Conclusion
Moderate compassion fatigue and burnout were widespread among emergency department nurses in the three selected Level 3 hospitals in Lusaka, although severe levels were uncommon. Compassion satisfaction remained moderate to high in all respondents. Married nurses had lower odds of high compassion satisfaction, whereas bachelor’s degree holders had higher odds than diploma-qualified nurses. No measured factor independently predicted elevated compassion fatigue or burnout. The findings support preventive staff-wellbeing measures, workload and scheduling review, retention-focused support and equitable professional-development opportunities, alongside further longitudinal and mixed-methods research in Zambian emergency-care settings. Future intervention research should also evaluate resilience-building, mindfulness-based approaches and organisational support in this context [10, 4].
7. Acknowledgements
The author acknowledges the University of Lusaka, the management of University Teaching Hospitals–Adult Hospital, University Teaching Hospitals–Children’s Hospital and Levy Mwanawasa University Teaching Hospital, and the emergency department nurses who participated in the research.
8. AI-Assisted Manuscript Preparation Disclosure
No AI tool was not used to generate or alter study data or statistical results. The author remains responsible for verifying the manuscript, references, interpretations and final submitted version.