Burnout and Turnover Intention Among Nurses in Governmental Hospitals in Makkah, Saudi Arabia: A Cross-Sectional Correlational Study

1. Introduction

Nurse turnover is a persistent workforce concern because it can disrupt staffing continuity, increase organizational costs, intensify the workload of remaining staff, and threaten care quality. Turnover intention—the conscious willingness to leave a current position—is an important precursor of actual departure and is therefore a practical indicator for workforce planning. International evidence identifies burnout, unfavorable work environments, inadequate organizational support, and reduced job satisfaction as central correlates of nurses’ intentions to leave [1,2,3,4].

Burnout is an occupational phenomenon characterized by emotional exhaustion (EE), depersonalization (DP), and reduced personal accomplishment (PA). These dimensions should be interpreted separately because they reflect distinct aspects of occupational strain [1]. Burnout has been associated with adverse nurse outcomes, including lower job satisfaction, diminished perceived quality of care, and greater organizational or position turnover [2,3]. Recent network and meta-analytic evidence further supports a meaningful relationship between burnout, organizational support, and turnover intention [4,5].

The Saudi nursing workforce operates within a rapidly developing health system that depends on both Saudi and expatriate nurses. Earlier Saudi studies linked quality of work life and work-environment characteristics with emotional exhaustion and intention to leave [6,7]. Research among Saudi critical care nurses also identified substantial burnout and emphasized the roles of workload, staffing, and managerial support [8,9]. In Jeddah, burnout varied by nationality and clinical unit, suggesting that workforce and organizational context may shape nurses’ experiences [10].

Despite this evidence, data from governmental hospitals in Makkah remain limited. Makkah’s hospitals serve a diverse population and a multinational nursing workforce in a context that may include fluctuating service demand. Local evidence is therefore needed to clarify the levels of burnout and turnover intention and to identify groups that may benefit from organizational support. This study aimed to (1) describe nurses’ turnover intention and burnout; (2) examine associations of demographic and work-related characteristics with turnover intention and burnout; and (3) determine the relationships between burnout dimensions and turnover intention among nurses in governmental hospitals in Makkah.

2. Materials and Methods

2.1. Study Design and Setting

A quantitative cross-sectional correlational design was used. The study was conducted during the first quarter of 2024 in governmental hospitals within the Makkah Health Cluster, Saudi Arabia. Reporting was organized in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) recommendations [11].

2.2. Participants and Sampling

At the time of the study, the nine hospitals in the cluster employed approximately 7000 nurses. OpenEpi was used to estimate the sample required for a descriptive study using an anticipated frequency of 50%, 5% absolute precision, and a 95% confidence level; the resulting target was 365 participants. Snowball sampling was used: participating nurses were invited to share the survey with potentially eligible colleagues. Because recruitment relied on peer referral and an open electronic link, the number of nurses who received the invitation could not be established and a response rate could not be calculated.

Eligible participants were registered nurses who had worked in one of the targeted hospitals for at least six months and were willing to participate. Nurses with less than six months of employment or who did not meet the other eligibility criteria were excluded.

2.3. Measures

The questionnaire had three parts. The first collected age, sex, nationality, marital status, educational level, years of experience, hospital, and department.

Turnover intention was assessed using the 12-item Anticipated Turnover Scale (ATS), developed by Hinshaw and Atwood [12]. Items are rated from 1 (strongly disagree) to 7 (strongly agree); six items are reverse-scored, and total scores range from 12 to 84, with higher scores indicating stronger turnover intention. The ATS has demonstrated acceptable reliability and validity across nursing studies [13]. For the present analysis, scores were classified as very low (12–22.29), moderately low (22.30–32.59), low (32.60–42.89), moderate (42.90–53.19), high (53.20–63.49), moderately high (63.50–73.79), or very high (73.80–84.00).

Burnout was measured with the 22-item Maslach Burnout Inventory–Human Services Survey, comprising EE (nine items), DP (five items), and PA (eight items) [14]. Items are scored from 0 (never) to 6 (every day). The operational cutoffs used in the source analysis were EE ≤17, 18–29, and ≥30; DP ≤5, 6–11, and ≥12; and PA ≤33, 34–39, and ≥40. Higher EE and DP and lower PA represent greater burnout. The source analysis also reported a composite burnout classification (≤55 low, 56–79 moderate, and ≥80 high); because the MBI is fundamentally multidimensional, the three dimensions remain the primary outcomes in this manuscript.

2.4. Data Collection

Following ethical approval, an anonymous self-administered questionnaire was distributed electronically to eligible nurses through institutional email and peer referral. Participation was voluntary. No direct personal identifiers were collected. Responses were exported to an electronic dataset for analysis and stored on a password-protected device accessible to the research team.

2.5. Statistical Analysis

Categorical characteristics are presented as frequencies and percentages, and scale scores as means. Chi-square tests were used to examine associations between participant characteristics and categorized turnover-intention and burnout outcomes. Cramer’s V was reported for significant associations as an estimate of association strength. Pearson’s correlation coefficients quantified relationships between continuous burnout and turnover-intention scores. Statistical significance was set at p < 0.05 (two-sided). Values originally reported as p = 0.000 are presented as p < 0.001. Because numerous bivariate tests were conducted without multiplicity correction and several categories contained few observations, subgroup findings are interpreted as exploratory.

2.6. Ethical Considerations

Ethical permission was granted by the Department of Nursing’s Ethical Review Board before recruitment. Electronic informed consent was obtained within the questionnaire. Participation was voluntary, no incentive was provided, and participants could complete the questionnaire privately. Data were anonymized and reported only in aggregate.

3. Results

3.1. Participant Characteristics

The study included 365 nurses. Most were women (83.8%), aged 31–40 years (56.7%), married (60.5%), and educated to bachelor’s level (66.0%). Saudi and non-Saudi nurses were represented almost equally. The largest experience group had 11–15 years of service (28.8%), and the most frequently reported departments were outpatient (15.3%), intensive care (14.2%), and emergency care (11.2%) (Table 1). Hospital information was available for 355 participants; it was not reported for 10.

3.2. Turnover Intention and Burnout

The mean ATS score was 51.92, corresponding to moderate turnover intention. Overall, 41.64% of nurses were in the moderate category, 27.67% in the high category, and 12.88% in the moderately high category (Table 2).

Mean EE was 17.78 (low), mean DP was 27.84 (high), and mean PA was 6.57 (low). Based on the study’s composite classification, mean overall burnout was 52.19 (low). Nevertheless, the dimensional profile showed substantial occupational strain: 85.75% had high DP, and all participants were classified as having low PA. The overall-burnout category was available for 364 participants (Table 3).

3.3. Associations with Demographic and Work-Related Characteristics

None of the examined demographic or work-related characteristics was significantly associated with categorized turnover intention (all p > 0.05; Table 4).

Educational level, years of experience, and department were significantly associated with EE and DP in the source analysis. Years of experience, hospital, and department were associated with PA, while nationality was associated with the reported overall-burnout category (Cramer’s V = 0.598). Table 5 also contained an EE-by-hospital result with p < 0.001 but no corresponding effect size; this result is retained as reported but should be verified against the analysis output. Given the number of comparisons and small cell counts for several hospitals and departments, all subgroup associations should be considered exploratory.

3.4. Relationships between Burnout and Turnover Intention

Turnover intention was positively correlated with EE (r = 0.127, p = 0.016), DP (r = 0.283, p < 0.001), and the reported overall-burnout score (r = 0.235, p < 0.001). These relationships were small. PA was not significantly correlated with turnover intention (r = 0.083, p = 0.113) (Table 6).

4. Discussion

This study found moderate turnover intention among nurses in governmental hospitals in Makkah. Although the study’s composite burnout score was categorized as low, the dimensional results were concerning: DP was high and PA was low. Turnover intention was positively related to EE, DP, and the composite burnout score, while PA was not significantly related to turnover intention. Demographic and work-related characteristics did not differentiate turnover intention, but several were associated with burnout outcomes in exploratory analyses.

The positive association between burnout and turnover intention is consistent with international evidence. Kelly et al. linked nurse burnout with subsequent organizational and position turnover [2], while a recent network analysis identified close interconnections between burnout symptoms and turnover intention [4]. Studies from Oman and other settings similarly reported that adverse work environments and burnout accompany poorer retention-related outcomes [3,15]. The present correlations were small, indicating that burnout is relevant but does not alone explain nurses’ intentions to leave. Compensation, scheduling, leadership, professional development, family considerations, and perceived organizational support may also contribute [5,6,7].

DP showed the strongest correlation with turnover intention. DP reflects psychological distancing and an impersonal response to recipients of care; persistent distancing may weaken professional identification and attachment to the organization. Comparable studies have linked burnout dimensions with turnover intention among clinical nurses [15,16,17,18]. However, the exceptionally high DP and uniformly low PA classifications in this sample warrant caution. Differences in scoring, cultural interpretation, survey administration, or data processing may contribute to this pattern. Verification of item coding—particularly reverse direction and the interpretation of PA—should be completed before submission.

The findings should also be understood in the Saudi context. Studies of Saudi nurses have linked staffing and resource adequacy, participation in hospital affairs, leadership support, and quality of work life with emotional exhaustion or intention to leave [6,7,8,9]. Burnout among nurses in Jeddah also differed by nationality and work unit [10]. In the present study, nationality was strongly associated with the composite burnout category. Expatriate nurses may face additional adaptation, communication, family-separation, or employment-related pressures, but these mechanisms were not measured and should not be assumed from nationality alone.

No demographic or work-related factor was associated with turnover intention. This finding suggests that retention risk may extend across workforce groups rather than being concentrated in a single demographic category. It also reinforces the value of examining modifiable organizational conditions rather than relying only on demographic profiling. Meta-analytic evidence indicates that organizational support is inversely related to turnover intention, highlighting recognition, development opportunities, fair management, and supportive working conditions as actionable priorities [5].

The subgroup associations involving education, experience, hospital, and department may help generate hypotheses for local workforce planning, but they are not definitive. Some hospitals and departments contained very few participants, numerous tests were conducted without adjustment, and the analysis did not control for confounding or clustering within hospitals. Future studies should use probability or stratified sampling, document response rates, confirm the measurement structure and reliability of both instruments in the study population, and apply multivariable models with hospital-level clustering. Longitudinal designs are needed to determine whether burnout predicts subsequent turnover.

4.1. Implications for Nursing Management

Nursing leaders should monitor EE, DP, PA, and turnover intention as distinct workforce indicators. Interventions should prioritize adequate staffing, manageable workload, psychologically safe leadership, recognition, opportunities for professional development, and accessible well-being support. Because turnover intention was moderate across demographic groups, organization-wide prevention is appropriate, supplemented by targeted assessment of departments with high demands. Repeated measurement after implementation would help determine whether interventions improve burnout dimensions and retention intentions.

4.2. Strengths and Limitations

The study included nurses from multiple governmental facilities and examined both burnout dimensions and turnover intention using established instruments. Nevertheless, the cross-sectional design prevents causal inference, and snowball sampling increases selection bias and limits representativeness. The recruitment denominator and response rate were unavailable. Self-reported data may be affected by recall, social-desirability, and common-method bias. Current-sample reliability coefficients and evidence of linguistic or cultural validation were not reported. Several hospital and department categories were very small, potentially violating chi-square assumptions. Multiple comparisons were not adjusted, no multivariable analysis addressed confounding, and hospital clustering was not modeled. The source tables also contained missing or inconsistent entries, including 10 unreported hospital values, one missing composite-burnout classification, and a significant EE-by-hospital p value without an effect size. Finally, interpretation of the composite burnout score is limited because the MBI is designed primarily for dimensional interpretation.

5. Conclusions

Nurses in governmental hospitals in Makkah reported moderate turnover intention. EE, DP, and the reported composite burnout score were positively but weakly correlated with turnover intention, with DP showing the strongest relationship. The high DP and low PA profile deserves organizational attention even though the composite burnout category was low. These findings support system-level strategies that strengthen staffing, supervisory support, professional recognition, and staff well-being. Because of the cross-sectional nonprobability design and limitations in the subgroup analyses, the results should be interpreted as associative and exploratory rather than causal.