Self-compassion, palliative care self-efficacy, and spiritual care preparedness among nursing students: a cross-sectional study

Abstract

Preparing nursing students to deliver holistic palliative care requires both clinical competence and readiness to address patients’ spiritual needs. Spiritual care preparedness may be associated with psychological resources and educational experiences; however, these relationships remain insufficiently understood, particularly in culturally specific contexts such as Saudi Arabia. To examine the associations among self-compassion, palliative care self-efficacy, and spiritual care preparedness among nursing students, and to explore whether palliative care self-efficacy demonstrates a cross-sectional indirect association between self-compassion and spiritual care preparedness. A single-center cross-sectional study was conducted between March and April 2026 among 284 third- and fourth-year nursing students at Northern Border University, Saudi Arabia. Data were collected using self-reported questionnaires, including the Self-Compassion Scale, the Palliative Care Self-Efficacy Scale, and the Student Survey of Spiritual Care. Data were analyzed using heteroscedasticity-robust regression models and bootstrap-based indirect effect analysis, adjusting for demographic and educational variables. Participants reported modest palliative care self-efficacy (M = 2.39, SD = 0.75), moderate spiritual care preparedness (M = 4.10, SD = 0.95), and moderate self-compassion (M = 3.19, SD = 0.78). In adjusted analyses, spiritual care preparedness was positively associated with higher self-compassion (B = 0.352, p < .001) and higher palliative care self-efficacy (B = 0.328, p < .001). Bootstrap analysis indicated a significant indirect association of self-compassion with spiritual care preparedness through palliative care self-efficacy (indirect effect B = 0.063; 95% BCa CI: 0.020–0.128). Self-compassion and palliative care self-efficacy were positively associated with spiritual care preparedness among nursing students. The findings also suggest a cross-sectional indirect association involving palliative care self-efficacy; however, the study design does not permit conclusions regarding causal pathways or mechanisms. These results support further investigation of confidence-building and psychological skill development in palliative care education through longitudinal and interventional research.

Introduction

Providing high-quality palliative care extends beyond technical symptom management; it necessitates a deep capacity to integrate psychological support and spiritual sensitivity into the care of patients facing serious, life-limiting illnesses. Within this clinical landscape, spiritual care preparedness, defined as a nurse’s readiness to perceive, honor, and respond to a patient’s inner spiritual needs, stands as a cornerstone of holistic practice. For undergraduate nursing students, however, developing this readiness is an uneven process shaped by the complex interplay of formal educational exposure and individual emotional resources. Uncovering the modifiable factors that nurture spiritual care preparedness during these formative training years is a vital priority for nursing educators and healthcare systems dedicated to compassionating end-of-life services^1–3.

Social Cognitive Theory provides a useful framework for understanding factors associated with the development of spiritual care preparedness among nursing students. As established by⁴, self-efficacy dictates whether an individual will initiate, persist through, and ultimately excel at emotionally challenging tasks. In nursing education, palliative care self-efficacy reflects a student’s deeply felt confidence in their ability to deliver clinical interventions, manage painful communications, and offer psychological comfort^5–7. When professional confidence is low, students often feel vulnerable or hesitant when facing dying patients, inadvertently restricting their capacity to offer comprehensive care, particularly in sensitive domains like spiritual support. Empirically, intentional palliative training and meaningful clinical exposure have consistently been shown to elevate this confidence, demonstrating that educational environments can reshape a student’s professional trajectory^8–10.

However, technical training alone cannot fully prepare a student for the emotional weight of terminal care. Individual emotional capacities heavily influence how a student absorbs and responds to clinical suffering. Self-compassion, the practice of extending kindness to oneself, recognizing suffering as a shared human experience, and holding one’s own distress in balanced, non-judgmental awareness, acts as a vital emotional buffer. It protects students from the paralyzing effects of self-criticism when confronted with complex care demands. While the broader literature linking a clinician’s internal self-compassion to outward behavioral competence continues to evolve, growing evidence within nursing emphasizes that self-compassion directly enriches core caregiving attributes, including caring behaviors and clinical competence^11–13. Theoretically, by nurturing their own emotional well-being, students may preserve the emotional resources associated with applying classroom knowledge in clinical practice^14,15.

While these internal psychological resources provide a necessary shield against burnout, a critical empirical question remains: how exactly does a student’s internal self-kindness translate into visible, outward clinical readiness? To date, the precise relationships linking a nursing student’s emotional resilience with their reported readiness to address a patient’s spiritual vulnerability have not been fully charted^16,17. Grounded in Social Cognitive Theory, psychological safety and emotional balance do not alter complex clinical behaviors in a vacuum; instead, they function by strengthening task-specific confidence⁴. In palliative care, this confidence is uniquely embodied as palliative care self-efficacy¹⁸. Higher levels of self-compassion may help students manage the emotional demands associated with caring for seriously ill and dying patients, they may preserve the cognitive and emotional bandwidth necessary to build true professional confidence during demanding clinical rotations¹⁹. From a Social Cognitive Theory perspective, palliative care self-efficacy may represent a potential explanatory pathway linking self-compassion with spiritual care preparedness.

Building upon this integrated framework, the present study examines a model where palliative care self-efficacy acts as the core mechanism through which self-compassion ultimately contributes to spiritual care preparedness. Opting for a theory-informed, cross-sectional indirect effect analysis allows us to move past surface-level correlations and instead examine patterns of association among theoretically related constructs. This direction aligns closely with recent nursing literature championing the role of professional self-efficacy in developing spiritual care preparedness among trainees^20,21, alongside a growing global call for structured educational strategies that honor the spiritual dimensions of healing^22,23.

In Saudi Arabia, this pathway carries deep cultural resonance. Nursing curricula are intentionally crafted to balance international palliative standards with culturally grounded expectations of care, which place a profound emphasis on the spiritual aspects of patient support rooted in Islamic values. Yet, despite an inspiring national drive to expand palliative access and strengthen the healthcare workforce under Saudi Arabia’s Vision 2030, undergraduate preparation for end-of-life care varies widely depending on the institution and region^24–26. Localized research is therefore essential to design supportive student initiatives and meaningful curricula.

This need is acutely felt in the northern border region of Saudi Arabia, where undergraduate nursing students navigate high-stakes environments requiring the seamless convergence of complex medical actions and deeply rooted spiritual values. Investigating these integrated pathways within this specific cohort shifts the educational lens from merely testing textbook knowledge toward identifying educational and psychological factors associated with a student nurse’s readiness to provide culturally aligned care. Ultimately, understanding how self-compassion and palliative care self-efficacy interact to shape spiritual care preparedness can guide training methods that are as clinically excellent as they are culturally congruent.

Although previous research has examined self-compassion, palliative care self-efficacy, and spiritual care preparedness as separate or related constructs, limited evidence is available regarding the mechanisms linking these variables among undergraduate nursing students. In particular, the potential role of palliative care self-efficacy in explaining the association between self-compassion and spiritual care preparedness has received little empirical attention. Accordingly, this study aimed to examine the relationships among self-compassion, palliative care self-efficacy, and spiritual care preparedness among nursing students in Northern Saudi Arabia. In addition, the study evaluated whether palliative care self-efficacy functions as an indirect pathway linking self-compassion with spiritual care preparedness. By examining self-efficacy as a potential intermediary factor, the study seeks to enhance understanding of how personal psychological resources may be associated with greater readiness to provide spiritually sensitive care and to inform future research and educational strategies within palliative care nursing education^8,20.

Objectives

1. To assess the levels of self-compassion, palliative care self-efficacy, and spiritual care preparedness among nursing students.

2. To examine the associations among self-compassion, palliative care self-efficacy, spiritual care preparedness, and relevant demographic and educational characteristics.

3. To evaluate whether palliative care self-efficacy demonstrates a cross-sectional indirect association between self-compassion and spiritual care preparedness.

Research questions

1. What are the levels of self-compassion, palliative care self-efficacy, and spiritual care preparedness among nursing students?

2. How are self-compassion, palliative care self-efficacy, spiritual care preparedness, and demographic and educational characteristics associated with one another?

3. Does palliative care self-efficacy demonstrate a cross-sectional indirect association between self-compassion and spiritual care preparedness?

Hypotheses

H1: Self-compassion is positively associated with both palliative care self-efficacy and spiritual care preparedness among nursing students.

H2: The association between self-compassion and spiritual care preparedness is indirectly associated with palliative care self-efficacy in a cross-sectional framework.

Methods section

Study design

This single-center, cross-sectional study examined associations among self-compassion, palliative care self-efficacy, and spiritual care preparedness among undergraduate nursing students at the Faculty of Nursing, Northern Border University, Arar, Saudi Arabia. Data were collected between March and April 2026. The study also evaluated whether palliative care self-efficacy demonstrated a cross-sectional indirect effect in the association between self-compassion and spiritual care preparedness. Given the cross-sectional design, all findings were interpreted as associative rather than causal and do not establish temporal ordering, mediation, or underlying mechanisms. This study was conducted as part of a larger research project approved under a single ethics protocol. Although the participant cohort, setting, and data collection period overlapped with a previously published study (Baqeas et al., 2026), the present analysis addressed different research questions and examined self-compassion and palliative care self-efficacy, variables that were not analyzed in the previous publication.

Setting

The research was conducted at the Faculty of Nursing, Northern Border University, located in Arar, Saudi Arabia. This institution provides a relevant academic environment for examining palliative care competencies, as nursing education in Saudi Arabia integrates clinical training with ethical and spiritual dimensions of care deeply influenced by Islamic values.

Sampling and participants

Eligible participants were third- and fourth-year undergraduate nursing students because these cohorts had progressed beyond foundational coursework and had opportunities for clinical exposure relevant to palliative and end-of-life care. Students were eligible if they were actively enrolled during the study period. Clinical exposure was not assumed solely based on academic year; rather, prior experience caring for a dying patient was assessed directly through a study questionnaire and analyzed as a separate variable. Students who declined participation or submitted questionnaires with more than 10% missing responses on the study scales were excluded from the final analysis.

A stratified random sampling strategy was employed to ensure representation across academic year (third and fourth year) and sex. Eligible students were first grouped into strata based on these characteristics. Within each stratum, participants were selected using a computer-generated random number sequence applied to the official university enrollment roster, ensuring that all eligible students within a given stratum had an equal probability of selection. Students who were absent during data collection or declined participation were replaced by other students randomly selected from the same stratum using the same procedure. This approach maintained the planned sample size and proportional representation across academic-year and sex strata. A total of 340 students were invited to participate. Of these, 302 returned questionnaires, corresponding to a response rate of 88.8%. Following data screening, 18 questionnaires were excluded because of excessive missing data, resulting in a final analytic sample of 284 students. The final sample comprised 97 third-year students (34.2%) and 187 fourth-year students (65.8%). The participant recruitment, replacement, exclusion, and analysis procedures are summarized in Fig. 1.

Fig. 1.

Participant flow diagram.

Sample size adequacy was determined as a priori using G*Power software. Assuming a medium effect size (f² = 0.15), an alpha level of 0.05, and statistical power of 0.95 for multiple regression analysis, the minimum required sample size was 200 participants. The final sample of 284 exceeded this requirement, indicating adequate statistical power for the planned analyses. Participants’ ages ranged from 19 to 23 years (M = 21.40, SD = 1.31; median = 21.00, IQR = 2.00), which is consistent with the expected age range of undergraduate nursing students at these academic levels. During data screening, three implausible age values (> 50 years) were identified as data-entry errors. Following identification and treatment of three implausible age values during data cleaning, the final analytic dataset showed an age range of 19–23 years. No age outliers or influential observations were identified in the final analytic dataset.

Instruments and measurement

Data were collected using a demographic questionnaire and three standardized instruments. The demographic questionnaire obtained information on participants’ age, sex, academic year, formal palliative care training, and prior experience caring for a dying patient. Formal palliative care training was assessed by asking participants whether they had previously attended any extracurricular educational activity, such as workshops, seminars, short courses, or training programs outside the standard undergraduate nursing curriculum (yes/no). Information regarding the duration, frequency, and timing of such training was not collected. Prior experience caring for a dying patient was assessed by asking participants whether they had directly cared for dying patients during their clinical training placements (yes/no). Information regarding the frequency, duration, and intensity of these experiences was not collected. To definitively resolve prior inconsistencies regarding the operationalization of the compassion construct, the Self-Compassion Scale (SCS) was the sole instrument used to measure this psychological resource; the Compassion Competence Scale was not administered in this study.

Palliative Care Self-Efficacy Scale (PCSS). Palliative care self-efficacy was assessed using the 12-item PCSS²⁷. Items are rated on a 4-point Likert scale ranging from 1 (not at all confident) to 4 (very confident). Total scores range from 12 to 48, with higher scores indicating greater confidence in delivering palliative care. For descriptive purposes, mean item scores are reported to facilitate interpretation. The Arabic version of the PCSS has previously undergone translation, cultural adaptation, and psychometric validation among Arabic-speaking nursing populations and demonstrated satisfactory validity and reliability. In the present study, internal consistency was excellent (Cronbach’s α = 0.942).

Student Survey of Spiritual Care (SSSC). Spiritual care preparedness was assessed using the 9-item Student Survey of Spiritual Care (SSSC). The instrument was originally developed to evaluate students’ perceived preparedness, confidence, and readiness to provide spiritual care in clinical practice^28,29. Responses are recorded on a 6-point Likert scale ranging from 1 (strongly disagree) to 6 (strongly agree). Total scores range from 9 to 54, with higher scores indicating greater spiritual care preparedness. For descriptive purposes, mean item scores are reported. Previous studies have demonstrated acceptable validity and reliability of the SSSC among nursing students. In the present study, internal consistency was good (Cronbach’s α = 0.874).

Self-Compassion Scale (SCS). Self-compassion was assessed using the validated Arabic version of Neff’s 26-item Self-Compassion Scale (SCS)¹⁶. The scale consists of six subdimensions: self-kindness, self-judgment, common humanity, isolation, mindfulness, and overidentification. Items are rated on a 5-point Likert scale ranging from 1 (almost never) to 5 (almost always). Consistent with the original scoring guidelines, negatively worded items were reverse scored before calculating the overall self-compassion score, with higher scores indicating greater self-compassion. The Arabic version has demonstrated satisfactory psychometric properties in previous studies. In the current sample, internal consistency reliability was excellent (Cronbach’s α = 0.947).

Data collection procedures

Data were collected over two months (March–April 2026) using a structured online questionnaire administered through Google Forms. Surveys were completed during designated sessions in a university computer laboratory; however, participation was not part of any mandatory academic activity. The questionnaire required approximately 10–15 min to complete. To minimize duplicate submissions, the survey was configured to accept only one response per participant and responses were reviewed for potential duplication before analysis. To minimize social desirability bias and perceived coercion, faculty members were explicitly absent from the room. A trained research assistant, who had no teaching, supervisory, evaluative, or administrative relationship with the participants, provided introductory instructions and remained available only for technical support while maintaining distance from respondents. Before beginning the survey, students were informed that participation was entirely voluntary, that declining participation would have no effect on their grades, academic standing, or relationship with the university, and that they could withdraw from the study at any time without penalty. No personally identifying information was collected, and survey data were stored in password-protected files accessible only to the research team to ensure confidentiality and data security.

Ethical considerations

The study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Institutional Review Board of Northern Border University (HAP 09 A 043; Decision No. 15/26/H). Participation was entirely voluntary, and electronic informed consent was obtained before survey initiation. Recruitment was conducted through official university communication channels. Participation was not associated with any course requirement, academic activity, or incentive. Students were explicitly informed that declining participation or withdrawing from the study would have no effect on their academic grades, academic standing, or relationships with faculty members. Anonymity was ensured using an online survey that did not collect names, student identification numbers, e-mail addresses, IP addresses, or other directly identifying information. Participants first completed an electronic consent form; those selecting “agree” proceeded to the questionnaire, whereas those selecting “decline” were automatically exited from the survey. Survey responses were stored anonymously and analyzed only in aggregate form. Furthermore, the research assistant who facilitated data collection had no teaching, supervisory, evaluative, or administrative relationships with the participants.

Statistical analysis

Missing Data. Missing data were minimal. During data screening, three implausible age values (> 50 years) were identified as likely data-entry errors. Because the original values could not be verified from source records, these observations were treated as missing and replaced using median imputation. To evaluate the potential impact of this decision, a sensitivity analysis excluding these three cases was conducted. The pattern and statistical significance of the main findings remained unchanged, indicating that the study conclusions were robust to the handling of these observations. No other variables contained missing data, and all scale items were complete. Common Method Bias. Because all variables were collected using self-report measures, Harman’s single-factor test was conducted. No single factor accounted for more than 50% of the total variance, providing preliminary evidence against substantial common method bias. As a sensitivity analysis, the main analyses were repeated using an alternative unreversed SCS scoring specification.

The direction, magnitude, and statistical significance of the findings were substantively unchanged. Therefore, only results based on the conventional reverse-scored SCS are presented as the primary analyses. Cross-Sectional Indirect Effect Analysis. Consistent with the cross-sectional study design, the analysis was framed as an indirect effect analysis rather than a test of mediation or causal mechanisms. PROCESS Model 4 with 5,000 bootstrap resamples was used to estimate indirect effects and generate bias-corrected and accelerated (BCa) 95% confidence intervals. Age, sex, academic year, formal palliative care training, and prior experience caring for a dying patient were included as covariates. Heteroscedasticity-consistent standard errors were estimated using the HC3 covariance estimator. Residual plots, standardized residuals, leverage values, and Cook’s distance statistics were examined to assess model assumptions and identify influential observations. No influential cases exceeding conventional thresholds were identified. Median regression was conducted as a sensitivity analysis, and the direction and statistical significance of the findings were consistent with those obtained from the primary regression models. Findings were interpreted as associative rather than causal, and no conclusions regarding temporal ordering or mediation processes were inferred.

Results section

Table 1 summarizes the demographic characteristics and descriptive statistics of the study sample (N = 284). More than half of the participants were female (58.5%), and approximately two-thirds were fourth-year nursing students (65.8%). Only 33.8% of participants reported having received formal palliative care training, while 29.6% reported prior experience caring for a dying patient during clinical training. Participants had a mean age of 21.40 years (SD = 1.31), with ages ranging from 19 to 23 years. The mean palliative care self-efficacy score was 2.39 (SD = 0.75) on a 1–4 scale, indicating a modest level of confidence in providing palliative care. The mean spiritual care preparedness score was 4.10 (SD = 0.95) on a 1–6 scale, suggesting a moderate level of preparedness to address patients’ spiritual needs. Similarly, the mean self-compassion score was 3.19 (SD = 0.78) on a 1–5 scale, reflecting a moderate level of self-compassion among the participants. The study instruments demonstrated good to excellent internal consistency reliability, with Cronbach’s alpha coefficients of 0.942 for the Palliative Care Self-Efficacy Scale, 0.874 for the Student Survey of Spiritual Care, and 0.947 for the Self-Compassion Scale. Overall, the findings indicate moderate levels of self-compassion and spiritual care preparedness, relatively modest palliative care self-efficacy, and limited exposure to formal palliative care training among the nursing students.

Table 1.

Sample characteristics and descriptive statistics for demographic and study variables.

Variable	n	Mean / %	SD	Median	IQR	Range	Cronbach’s α
Female sex	166	58.50%	—	—	—	—	—
Fourth academic year	187	65.80%	—	—	—	—	—
Formal palliative care training	96	33.80%	—	—	—	—	—
Prior experience caring for a dying patient	84	29.60%	—	—	—	—	—
Age (years)	284	21.40	1.31	21.00	2.00	19–23	—
Palliative care self-efficacy	284	2.39	0.75	2.33	1.00	1.00–4.00	0.942
Spiritual care preparedness	284	4.10	0.95	4.00	1.22	1.00–6.00	0.874

Table 2 presents the bivariate associations between demographic and psychosocial factors and the study outcomes, namely palliative care self-efficacy (PCSE) and spiritual care preparedness (SCP). As shown in Panel A, age demonstrated weak but significant positive correlations with both PCSE (ρ = 0.187, p = .002) and SCP (ρ = 0.242, p < .001), indicating that older students tended to report higher levels of self-efficacy and preparedness. No significant association was found between age and self-compassion (ρ = 0.029, p = .629). Significant positive correlations were observed between PCSE and SCP (ρ = 0.329, p < .001), PCSE and self-compassion (ρ = 0.224, p < .001), and SCP and self-compassion (ρ = 0.331, p < .001), suggesting that students with higher self-compassion and greater palliative care self-efficacy were more likely to report higher spiritual care preparedness.

Table 2.

Bivariate associations of demographic and psychosocial factors with PCSE and SCP.

Panel A. Spearman correlations
Variable pair	ρ	p
Age and palliative care self-efficacy	0.187	0.002
Age and spiritual care preparedness	0.242	< 0.001
Age and self-compassion	0.029	0.629
Palliative care, self-efficacy and spiritual care preparedness	0.329	< 0.001
Palliative care, self-efficacy and self-compassion	0.224	< 0.001
Spiritual care preparedness and self-compassion	0.331	< 0.001

Panel B. Mann-Whitney U comparisons
Predictor	Outcome	Group 0 median (IQR)	Group 1 median (IQR)	U	p	FDR-adjusted p
Sex (male vs. female)	Palliative care self-efficacy	2.29 (0.90)	2.33 (1.08)	9402.5	0.566	0.566
Sex (male vs. female)	Spiritual care preparedness	4.00 (1.08)	4.11 (1.22)	8739.0	0.121	0.139
Academic year (third vs. fourth)	Palliative care self-efficacy	2.00 (0.83)	2.50 (1.00)	5700.0	< 0.001	< 0.001
Academic year (third vs. fourth)	Spiritual care preparedness	3.78 (0.89)	4.11 (1.17)	6926.0	0.001	0.002
Formal palliative care training (no vs. yes)	Palliative care self-efficacy	2.17 (0.83)	2.75 (1.04)	5793.5	< 0.001	< 0.001
Formal palliative care training (no vs. yes)	Spiritual care preparedness	4.00 (0.89)	4.44 (1.22)	6598.5	< 0.001	< 0.001
Prior experience caring for a dying patient (no vs. yes)	Palliative care self-efficacy	2.17 (0.96)	2.67 (0.75)	5566.5	< 0.001	< 0.001
Prior experience caring for a dying patient (no vs. yes)	Spiritual care preparedness	4.00 (0.89)	4.33 (1.36)	6949.0	0.021	0.029

Note. Nonparametric tests were used due to non-normality. Group 1 corresponds to the more exposed category. FDR adjustment was applied to control the false discovery rate.

Panel B summarizes the Mann–Whitney U test results comparing PCSE and SCP across demographic and educational groups. No significant differences were observed between male and female students in either PCSE (p = .566) or SCP (p = .121), and these findings remained non-significant after false discovery rate (FDR) adjustment. In contrast, fourth-year students reported significantly higher PCSE and SCP scores than third-year students (both FDR-adjusted p ≤ .002). Students who had received formal palliative care training also demonstrated significantly higher levels of PCSE and SCP compared with those without training (both FDR-adjusted p < .001). Likewise, participants with prior experience caring for a dying patient reported significantly greater PCSE (FDR-adjusted p < .001) and SCP (FDR-adjusted p = .029) than those without such experience.

Overall, the bivariate analyses indicate that academic progression, formal palliative care training, and prior experience caring for dying patients are associated with higher palliative care self-efficacy and spiritual care preparedness. Furthermore, self-compassion was positively associated with both outcomes, supporting its potential role as an important psychosocial factor in preparing nursing students for palliative and spiritual care practice.

Figure 2 visually demonstrates a consistent trend toward higher palliative care self-efficacy and spiritual care preparedness among students with greater educational and clinical exposure. Across the six panels, the distributions for fourth-year students, those who received formal palliative care training, and those with prior experience caring for dying patients generally show higher central values than their comparison groups. Although considerable overlap is evident between groups, the overall shifts in the distributions support the positive patterns observed in the bivariate analyses and highlight the variability in individual responses within each subgroup.

Fig. 2.

Distribution of palliative care self-efficacy and spiritual care preparedness across key educational and clinical exposure groups. Caption: Group-wise distributions of palliative care self-efficacy and spiritual care preparedness. Legend: Multi-panel box-and-jitter plots should display palliative care self-efficacy and spiritual care preparedness by academic year, formal palliative care training, and prior experience caring for a dying patient. Median lines, interquartile boxes, whiskers, and overlaid observations should be shown. This figure visually complements the Mann-Whitney comparisons reported in Table 2.

Table 3 presents the heteroscedasticity-robust regression model examining factors associated with palliative care self-efficacy. The model explained 21.8% of the variance in palliative care self-efficacy (R² = 0.218) among the participants (N = 284). Being in the fourth academic year was significantly associated with higher palliative care self-efficacy (B = 0.380, SE = 0.092, 95% CI = 0.200–0.559, p < .001). Similarly, students who had received formal palliative care training demonstrated significantly greater self-efficacy compared with those without such training (B = 0.311, SE = 0.101, 95% CI = 0.113–0.509, p = .002). Prior experience caring for a dying patient was also positively associated with palliative care self-efficacy (B = 0.229, SE = 0.101, 95% CI = 0.031–0.427, p = .023). In addition, self-compassion emerged as a significant positive predictor of palliative care self-efficacy (B = 0.193, SE = 0.061, 95% CI = 0.075–0.312, p = .001). Conversely, age and sex were not significantly associated with palliative care self-efficacy (both p > .05). All VIF values were below 2.0, indicating the absence of problematic multicollinearity.

Table 3.

Heteroscedasticity-robust regression model for palliative care self-efficacy.

Predictor	B	SE	95% CI	p
Intercept	1.379	0.248	0.894 to 1.864	< 0.001
Age	0.000	0.006	-0.012 to 0.011	0.939
Female sex	-0.022	0.082	-0.183 to 0.139	0.789
Fourth academic year	0.380	0.092	0.200 to 0.559	< 0.001
Formal palliative care training	0.311	0.101	0.113 to 0.509	0.002
Prior experience caring for a dying patient	0.229	0.101	0.031 to 0.427	0.023
Self-compassion	0.193	0.061	0.075 to 0.312	0.001

Model statistics: R² = 0.218; N = 284; all VIFs < 2.0.

substantial variance in self-efficacy remains attributable to unmeasured determinants.

Table 4 presents the heteroscedasticity-robust regression model examining factors associated with spiritual care preparedness. The model explained 26.7% of the variance in spiritual care preparedness (R² = 0.267) among the participants (N = 284). Age was a significant positive predictor of spiritual care preparedness (B = 0.021, SE = 0.008, 95% CI = 0.005–0.037, p = .009), indicating that older students tended to report higher levels of preparedness. Self-compassion (B = 0.352, SE = 0.092, 95% CI = 0.172–0.533, p < .001) and palliative care self-efficacy (B = 0.328, SE = 0.084, 95% CI = 0.163–0.493, p < .001) were also significant positive predictors, suggesting that students with higher self-compassion and greater confidence in providing palliative care were more prepared to deliver spiritual care. In contrast, sex, academic year, formal palliative care training, and prior experience caring for a dying patient were not significantly associated with spiritual care preparedness (all p > .05). Additionally, all VIF values were below 2.0, indicating no evidence of problematic multicollinearity among the predictor variables. Overall, the findings highlight age, self-compassion, and palliative care self-efficacy as significant independent predictors of spiritual care preparedness.

Table 4.

Heteroscedasticity-robust regression model for spiritual care preparedness.

Predictor	B	SE	95% CI	p
Intercept	1.545	0.364	0.831 to 2.258	< 0.001
Age	0.021	0.008	0.005 to 0.037	0.009
Female sex	-0.009	0.112	-0.229 to 0.211	0.935
Fourth academic year	0.048	0.125	-0.198 to 0.294	0.700
Formal palliative care training	0.208	0.123	-0.033 to 0.449	0.091
Prior experience caring for a dying patient	-0.084	0.129	-0.336 to 0.168	0.512
Self-compassion	0.352	0.092	0.172 to 0.533	< 0.001
Palliative care self-efficacy	0.328	0.084	0.163 to 0.493	< 0.001

Model statistics: R² = 0.267; N = 284; all VIFs < 2.0.

Table 5 presents the total, direct, and indirect associations between self-compassion and spiritual care preparedness. Self-compassion showed a significant positive total association with spiritual care preparedness (B = 0.415, SE = 0.094, 95% CI [0.232, 0.599], p < .001), indicating that higher levels of self-compassion were associated with greater preparedness to provide spiritual care. After accounting for palliative care self-efficacy, the direct association between self-compassion and spiritual care preparedness remained statistically significant (B = 0.352, SE = 0.092, 95% CI [0.172, 0.533], p < .001). This finding suggests that self-compassion independently contributes to spiritual care preparedness. Furthermore, a significant indirect association was observed through palliative care self-efficacy (B = 0.063, 95% BCa CI [0.020, 0.128]). Because the bootstrap confidence interval did not include zero, the indirect association was considered statistically significant. The indirect effect accounted for approximately 15.2% of the total association (0.063/0.415 × 100), indicating that palliative care self-efficacy accounted for a significant indirect association between self-compassion and spiritual care preparedness. Overall, these findings support the presence of a significant cross-sectional indirect association involving palliative care self-efficacy.

Table 5.

Total, direct, and indirect associations between self-compassion and spiritual care preparedness.

Effect	B	SE	95% CI	p
Total association (c path)	0.415	0.094	0.232 to 0.599	< 0.001
Direct association (c′ path)	0.352	0.092	0.172 to 0.533	< 0.001
Indirect association through palliative care self-efficacy (a × b)	0.063	—	0.020 to 0.128*	—

*Bias-corrected and accelerated bootstrap confidence interval based on 5,000 resamples.

Figure 3 illustrates cross-sectional indirect association between self-compassion and spiritual care preparedness through palliative care self-efficacy. Higher self-compassion was associated with greater palliative care self-efficacy, which in turn was positively associated with spiritual care preparedness. The significant indirect effect (B = 0.063, 95% BCa CI [0.020, 0.128]) indicates that palliative care self-efficacy demonstrated a significant indirect association between self-compassion and spiritual care preparedness. Because the direct effect of self-compassion on spiritual care preparedness remained significant after accounting for the mediator (B = 0.352, p < .001), the findings support the proposed cross-sectional indirect effect model, suggesting that self-compassion contributes to spiritual care preparedness both directly and indirectly through enhanced palliative care self-efficacy.

Fig. 3.

Cross-sectional indirect association between self-compassion and spiritual care preparedness through palliative care self-efficacy. Caption: Adjusted cross-sectional indirect-effect model. Legend: The path diagram should present the association between self-compassion and spiritual care preparedness, with palliative care self-efficacy specified as the intervening variable. The model should be adjusted for age, sex, academic year, formal palliative care training, and prior experience caring for a dying patient. Use the non-causal title “cross-sectional indirect effect” rather than “mediation,” and report the verified coefficients shown in Table 5.

Discussion

Among 284 third- and fourth-year nursing students at a university in northern Saudi Arabia, palliative care self-efficacy was modest, falling just below the midpoint of its response scale, whereas spiritual care preparedness and self-compassion were at moderate levels. The three constructs were positively and significantly intercorrelated, and both palliative care self-efficacy and spiritual care preparedness increased with age. After adjustment for covariates, higher academic seniority, formal palliative care training, prior experience caring for a dying patient, and greater self-compassion were independently associated with stronger palliative care self-efficacy. Notably, only a minority of participants reported previous formal palliative care training, which in this study referred to extracurricular workshops, seminars, and training activities rather than formal curriculum-based instruction. This limited exposure to structured palliative care education may partly explain the observed variability in self-efficacy and preparedness and highlights the potential value of expanding educational opportunities in this area. Spiritual care preparedness was independently associated with older age, greater self-compassion, and higher palliative care self-efficacy, whereas the associations with academic year and training were attenuated after adjustment. Furthermore, palliative care self-efficacy demonstrated a significant indirect association in the relationship between self-compassion and spiritual care preparedness, underscoring its role as an important mechanism linking personal psychological resources with readiness to provide holistic, spiritually sensitive care.

The modest level of palliative care self-efficacy reported by our participants is consistent with evidence that nurses and nursing students frequently feel underconfident in end-of-life care and that this confidence is closely tied to targeted preparation^8,27. It also aligns with documented gaps in undergraduate palliative care preparedness in the wider region^30,31 and with descriptions of palliative care in Saudi Arabia as a developing field that is expanding under national health reform but remains unevenly embedded in professional training²⁶. Our finding that fourth-year status, formal training, and prior experience caring for a dying patient were each independently associated with higher self-efficacy agrees with intervention and review evidence indicating that structured education and clinical exposure are among the most reliable correlates of palliative care confidence^8,9. The positive correlations of age with both self-efficacy and spiritual care preparedness are also broadly compatible with this body of work, in which longer exposure to clinical practice tends to accompany greater confidence^8,9. Although the age effect was statistically significant, the magnitude of the association was modest (B = 0.021), suggesting a relatively small increase in spiritual care preparedness across the observed age range. These results therefore add regional, student-level data to a literature that has been dominated by studies of practicing nurses, and they extend it by situating self-compassion alongside the more commonly examined educational and experiential correlates.

The positive associations of self-compassion with both self-efficacy and spiritual care preparedness partially agree with the theoretical proposition that a self-compassionate stance supports compassionate engagement with others¹³. This interpretation should be made cautiously, because a systematic review concluded that the empirical basis linking provider self-compassion to patient-directed care remains limited and that no included study measured patient-reported outcomes¹³. Our results are more directly consistent with a multi-university Saudi study in which self-compassion was associated with caring behaviour and compassion competence among nursing students¹¹, and with reviews reporting that self-compassion interventions can strengthen self-compassion and related caregiving capacities in nurses¹².

The independent association between palliative care self-efficacy and spiritual care preparedness is in line with recent work identifying professional self-efficacy as the strongest predictor of spiritual care preparedness among nursing and midwifery students²⁰. The moderate level of spiritual care preparedness we observed echoes systematic review findings that nurses and students tend to report moderate spiritual care competence, with formal preparation often limited^22&23. We found no significant sex differences in either outcome; regional evidence on sex-related differences in these constructs remains limited. Considered as a whole, the present associations are broadly consistent with the existing literature on the separate constructs, while the explicit modeling of self-efficacy as an intermediary between self-compassion and spiritual care preparedness addresses an area where integrated, regional evidence has been limited.

One possible interpretation of the observed indirect association is offered by self-efficacy theory, which holds that confidence in performing a task develops through mastery experiences, vicarious learning, and supportive feedback, and in turn shapes whether individuals attempt and persist with demanding care activities⁴. Within this framework, self-compassion may function as an internal resource that helps students approach the emotional demands of dying patients without excessive self-criticism, which may be associated with perceptions of clinical encounters as more manageable and with greater reported confidence. The partial mediation we observed could suggest that self-compassion relates to spiritual care preparedness both directly, perhaps through a generally compassionate orientation, and indirectly, through its association with higher palliative care self-efficacy, which is in turn associated with greater spiritual care preparedness.

The attenuation of the academic-year and training associations with spiritual care preparedness after adjustment suggests that these variables share variance with other factors included in the model, particularly age, self-compassion, and palliative care self-efficacy. However, because of the cross-sectional design, the present data cannot determine whether any of these factors explain, mediate, or account for the observed associations. The association of greater age with both outcomes may reflect a range of correlated influences, including accumulated life experiences, clinical exposure, or other unmeasured characteristics. Similarly, although self-compassion showed independent associations with both palliative care self-efficacy and spiritual care preparedness, the mechanisms underlying these relationships remain uncertain and require investigation in longitudinal studies. This would be consistent with reports that spiritual care content is unevenly represented in nursing curricula and that structured educational strategies for spiritual care are still being developed and tested^9,22. The association of greater age with both outcomes may reflect accumulated life and clinical experience rather than age itself. The cultural and religious context of Saudi Arabia may also be relevant to understanding spiritual care preparedness. For example, it is possible that cultural norms or religious values could influence how students perceive and respond to spiritual care concepts. However, cultural and religious factors were not directly measured in the present study; therefore, any such explanation should be regarded as speculative and as a hypothesis for future research rather than a conclusion supported by the current data.

Taken together, the findings suggest several preliminary implications that should be interpreted cautiously given the cross-sectional design. Because palliative care self-efficacy was modest yet closely linked to training and clinical exposure, undergraduate curricula could benefit from explicit, sequenced palliative care content paired with supervised clinical placements that allow graded mastery experiences with seriously ill patients^8,9. Given the observed positive association between palliative care self-efficacy and spiritual care preparedness, educational approaches that support students’ confidence alongside knowledge and skills development may warrant further investigation. However, the present findings do not establish whether improving self-efficacy would lead to improved spiritual care preparedness, and intervention studies are needed to evaluate this possibility (20, 22). The independent contribution of self-compassion raises the possibility that brief, evidence-informed self-compassion or well-being components could serve a dual purpose, supporting student wellbeing while potentially reinforcing the personal resources associated with compassionate, holistic care^11,12. At the health-system level, strengthening the palliative care preparedness of the future nursing workforce is congruent with national efforts to expand palliative and end-of-life services under Saudi health reform²⁶. Such curricular and faculty-development investments would require local evaluation, but they align the education of nursing students with the holistic, culturally grounded care that palliative contexts in the region increasingly require. Investing in faculty capacity to teach and role-model spiritual and end-of-life care may be especially important, since the limited integration of this content in curricula appears to constrain how far formal coursework alone can advance preparedness.

Strengths, limitations and future research directions

This study has several strengths, including an adequately powered sample, the use of Arabic-language instruments with prior validation in regional nursing populations^16,27, high internal consistency across all scales, and an analytic approach that combined heteroscedasticity-robust regression, median regression as a sensitivity analysis, and bootstrapped mediation, which together support the stability of the observed patterns. Several limitations should temper interpretation. The cross-sectional design precludes causal inference, and the mediation model, although theoretically motivated, describes associations measured concurrently rather than effects unfolding over time. All measures were self-reported, creating the potential for social desirability bias, common-method variance, and reporting inaccuracies, particularly for spirituality- and compassion-related constructs. In addition, because participation was voluntary, nonresponse bias cannot be excluded, and students who chose to participate may have differed systematically from those who did not. Although a stratified random sampling approach was used, the study was conducted at a single nursing faculty, which may limit the generalizability of the findings to students from other universities or regions. Differences in institutional characteristics, curricular content, clinical training opportunities, and cultural contexts may influence palliative care self-efficacy and spiritual care preparedness. Therefore, the findings should be generalized with caution, and multicenter studies are needed to confirm the results in more diverse nursing education settings. Formal palliative care training and prior experience caring for dying patients were assessed using dichotomous self-report measures. Information regarding the duration, frequency, quality, and recency of these experiences was not collected; therefore, potential differences in the intensity of educational and clinical exposure could not be examined. Although observations were analyzed at the individual level, students were nested within academic classes, and potential class-level clustering effects were not explicitly modeled. Three implausible or nonnumeric age entries were recoded as missing and imputed to the sample median, a pragmatic decision that preserved the full sample but reflects a minor data-quality limitation. The self-compassion score also warrants caution in interpretation because uncertainty may arise from the modified scoring approach; although sensitivity analyses using an alternative specification produced similar findings, comparisons with studies using different scoring methods should be made carefully. Finally, the models explained only a minority of the variance in each outcome, indicating that additional unmeasured factors likely contribute to the outcomes studied.

Future research should prioritize longitudinal and multicentre studies across Saudi regions to clarify the temporal ordering suggested by the present associations and to improve generalizability, alongside intervention studies that test whether enhancing self-compassion and palliative care self-efficacy translates into measurable gains in spiritual care preparedness, ideally incorporating patient-reported outcomes given the noted scarcity of such evidence¹³. Curriculum evaluation and qualitative studies exploring how students experience spiritual and end-of-life care would help explain the modest variance observed here, and rigorous validation of educational interventions would strengthen the evidence base for curricular reform.

Conclusion & recommendations

This single-center cross-sectional study found that nursing students reported modest palliative care self-efficacy and moderate levels of self-compassion and spiritual care preparedness. Self-compassion and palliative care self-efficacy were positively associated with spiritual care preparedness, and palliative care self-efficacy demonstrated a significant cross-sectional indirect effect in the association between self-compassion and spiritual care preparedness. Given the cross-sectional design, these findings should be interpreted as associative rather than causal and do not establish temporal ordering or underlying mechanisms.

The findings suggest that self-compassion, palliative care self-efficacy, and spiritual care preparedness may be relevant and interrelated constructs within nursing education. However, any educational or practice implications should be considered preliminary. While pilot interventions aimed at enhancing self-compassion or palliative care self-efficacy may warrant exploration, their effectiveness should be evaluated using rigorous longitudinal, quasi-experimental, or randomized controlled study designs before wider curricular or institutional implementation is considered. Future multicenter studies employing prospective and experimental methodologies are needed to confirm the observed associations, clarify potential causal pathways, and determine whether targeted educational interventions can improve spiritual care preparedness among nursing students.

Implications for nursing education

The findings suggest that integrating psychological and competency-based perspectives within nursing education may be relevant for promoting spiritual care preparedness. Given the observed associations between palliative care self-efficacy, self-compassion, and spiritual care preparedness, nursing curricula may benefit from providing opportunities for students to develop competencies related to palliative and spiritual care. Approaches such as active learning, simulation, reflective practice, and supervised clinical experiences may represent promising educational strategies; however, their effectiveness was not directly evaluated in the present study. Similarly, incorporating culturally sensitive spiritual care content and supporting faculty development may help create learning environments that foster holistic care competencies. Future longitudinal and intervention studies are needed to determine whether educational strategies designed to enhance self-compassion and palliative care self-efficacy lead to improvements in spiritual care preparedness. Therefore, the present findings should be interpreted as providing a rationale for further educational research rather than evidence of the effectiveness of specific educational interventions.

Author contributions

FAM (Fathia Ahmed Mersal) was responsible for data analysis and contributed to the interpretation of statistical findings. MOA (Muamar Odeh Aldalaeen) led the development of the discussion section. ONA (Ohoud Naif Aldughmi), FAE (Fatma Abdou Eltaib), AGA (Abdulhamid Gharib Alrwili) and LMMAN (Lobna Mohamed Mohamed Abu Negm) contributed to writing the introduction, conclusion, and implications of the study. AAE (Amal Ahmed Elbilgahy) was responsible for the methodology design, while MHB (Manal Hassan Baqeas) led the data collection with the support of all authors.The primary draft of the manuscript was prepared with substantial contributions and support from all authors. All authors participated in revising the manuscript critically for important intellectual content, approved the final version, and agreed to be accountable for all aspects of the work.

Funding

The authors gratefully acknowledge the approval and the support of this research study by the grant No (NBU-FFR-2026-112-06) from the Deanship of Scientific Research at Northern Border University, Arar, Saudi Arabia.

Data availability

All the data generated was included in this article.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki. Ethical approval was granted by the Institutional Review Board of Northern Border University (HAP-09-A-043; Decision No. 15/26/H). Written informed consent was obtained electronically from all participants before data collection.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.