Analyzing the effect of theoretical mindfulness-nursing-education-programme on students’ stress, resilience, and heart rate variability: a quasi-experimental study

Abstract

Background

Stress adversely affects nursing students’ well-being and clinical communication. This study examined the effects of a newly developed mindfulness-based nursing education programme (MINDNUEDU) on perceived stress, resilience, and heart rate variability (HRV) among nursing students.

Methods

A quasi-experimental, two-arm pre–post design was conducted at a university in Taiwan. Seventy nursing students were allocated to either an eight-week mindfulness intervention group (n = 36) or a standard curriculum control group (n = 34) based on academic scheduling constraints. Primary outcomes were perceived stress, measured using the Perceived Stress Scale, and resilience, measured using the Connor–Davidson Resilience Scale. Secondary outcomes included HRV indices (SDNN, RMSSD, and HF power), assessed within the intervention group. Adjusted between-group differences over time were analyzed using Generalized Estimating Equations, controlling for baseline values and year of study.

Results

After adjustment for baseline differences, significant Group × Time interactions were observed for perceived stress (p = .004) and resilience (p < .001), indicating greater improvements in the intervention group compared with controls. The mean increase in resilience exceeded the minimal clinically important difference. A significant dose–response relationship was also identified, with higher attendance linearly associated with greater reductions in stress (p = .003) and greater gains in resilience (p = .004). Within the intervention group, session-level analyses demonstrated acute post-session increases in parasympathetic-related HRV indices across several weeks.

Conclusions

Participation in the MINDNUEDU programme was associated with reduced perceived stress and increased resilience among nursing students. While physiological findings reflected short-term autonomic changes during sessions rather than confirmed long-term trait adaptation, the results support the feasibility of integrating structured mindfulness training into nursing education. Randomized controlled trials with longitudinal follow-up are needed to confirm sustainability and causal effects.

Background

Academic stress is a widespread concern among university students, with evidence suggesting comparatively higher stress levels in some Asian regions [1]. Persistent stress during higher education has been associated with psychological distress, reduced academic engagement, and increased risk of attrition. In addition to psychological effects, stress has been linked to alterations in autonomic nervous system functioning, which may be reflected in reduced vagal tone and lower heart rate variability (HRV), indicators of diminished physiological adaptability.

Within this broader context, nursing students may be particularly vulnerable. Alongside academic demands, they undertake intensive clinical practicums that involve exposure to human suffering, death, ethical dilemmas, and the potential for medical errors [2]. These experiences require substantial emotional regulation and professional responsibility during training. Such stressors have been associated with anxiety, reduced learning motivation, and academic disengagement, with potential implications for clinical performance and patient safety [3].

Strengthening adaptive coping strategies is therefore an important priority in nursing education. Resilience commonly defined in this context as the capacity to adapt positively to academic and clinical stressors has been identified as a protective factor associated with lower stress and better psychological well-being among university students [4, 5]. However, resilience may not develop automatically in high-stress environments and may require structured educational support. Existing approaches in nursing education often focus on stress awareness, while fewer provide systematic, theory-informed methods aimed at enhancing students’ regulatory capacities.

Mindfulness-based interventions have been increasingly explored as one potential strategy. Studies involving nursing students and healthcare trainees suggest that mindfulness training may reduce perceived stress and enhance resilience [6, 7]. Some evidence also indicates that mindfulness practices are associated with changes in HRV, including indices such as the standard deviation of normal-to-normal intervals (SDNN), which is commonly used as a non-invasive marker of autonomic flexibility [8]. However, much of the existing literature relies primarily on self-reported outcomes, and physiological indicators are less consistently examined. In addition, recent meta-analyses have noted considerable heterogeneity in intervention structure and intensity, highlighting the need for clearer theoretical frameworks to guide program design and interpretation of outcomes [9, 10].

To address these gaps, the present study developed a theory-informed mindfulness program grounded in Monitoring and Acceptance Theory (MAT) [11]. MAT proposes that mindfulness comprises both monitoring (heightened present-moment awareness) and acceptance (a non-reactive stance toward experience) and suggests that the combination of these components may support adaptive responses to stress. In the MINDNUEDU intervention, monitoring was operationalized through body scan and mindful breathing practices to enhance interoceptive awareness, while acceptance was incorporated through cognitive reappraisal exercises aimed at promoting more adaptive processing of clinical challenges. Rather than assuming direct physiological effects, this study examines whether participation in a theory-based program is associated with changes in both psychological outcomes (stress and resilience) and a physiological indicator (SDNN). Accordingly, this study aimed to develop and evaluate the effects of a theory-based Mindfulness Nurse Education programme (MINDNUEDU) on stress, resilience, and SDNN among nursing students.

Methods

Research design and setting

This study employed a quasi-experimental, two-arm parallel pre- and post-test design. A randomized controlled trial was not feasible due to the rigid structure of the nursing curriculum and mandatory clinical placements. To avoid interference with required coursework and ensure consistent attendance throughout the 8-week program, participants were allocated based on academic schedule compatibility rather than random assignment. This study was conducted at a private university of science and technology in Southern Taiwan. The intervention was held in a dedicated, quiet multi-purpose classroom specifically arranged to provide a private and distraction-free environment, which is essential for conducting mindfulness meditation and group reflections.

Sample

The sample size was estimated based on the effect sizes reported by Lever Taylor, Strauss [12], which ranged from d = 0.22 to 1.07 for mindfulness interventions. To ensure adequate sensitivity, a medium effect size (Cohen’s d = 0.65) was targeted. Using G*Power 3.1, with a statistical power of 0.80 and a significance level (α) of 0.05, the required minimum sample was estimated to be 76 participants. This calculation included a projected attrition rate of 12% [13] to ensure sufficient power to detect significant changes in perceived stress.

The inclusion criteria were full-time university nursing students who were able to communicate in Chinese and had access to mobile devices. Students were excluded if they had previously participated in similar mindfulness activities, were currently receiving psychotherapy, or had a diagnosis of severe mental illness. Participants assigned to the control group received the existing curriculum only, whereas those in the intervention group received the MINDNUEDU program in addition to the existing curriculum.

Recruiting

The study was conducted across the fist semesters 2021. Participants were recruited during the first week of the first semester using a multi-channel strategy to ensure broad outreach Initial outreach involved displaying promotional posters on primary campus bulletin boards and within academic buildings. Furthermore, faculty mentors assisted by announcing the study during class sessions and department meetings. Interested students-initiated contact by scanning a QR code on the recruitment materials or reaching out to a research assistant. Following the initial contact, potential participants were screened according to the inclusion criteria to ensure study eligibility.

Screening and ethical procedures

On the first day of the study, the Principal Investigator (PI) provided a comprehensive briefing covering the study’s objectives, potential benefits and risks, and the participants’ rights and obligations. Written informed consent was obtained from all eligible participants prior to the commencement of pre-test data collection (T1).

Allocation and blinding

Due to the rigid nature of the nursing curriculum and to ensure intervention feasibility, a quasi-experimental design was employed instead of randomized allocation. Group assignment was determined by academic schedule compatibility: students whose timetables permitted consistent attendance throughout the 8-week MINDNUEDU program were assigned to the intervention group, while those with scheduling conflicts from the same academic year served as the control group. To account for the non-randomized group assignment and potential baseline differences, Generalized Estimating Equations (GEE) was utilized. To ensure objectivity, data collection at T1 and T2 (PSS14-C and CDRS-C questionnaires) was managed by research assistants blinded to the study protocol. A separate assistant was responsible for recording SDNN data before and after each intervention session. The participant flow, including screening, allocation, and attrition, is illustrated in Fig. 1.

Fig. 1.

Flowchart of participant recruitment and allocation

MINDNUEDU Intervention

The MINDNUEDU programme was guided by the Monitoring and Acceptance Theory (MAT) [11], targeting its two core mechanisms: Attention Monitoring and Acceptance. Attention Monitoring aimed to cultivate heightened awareness of present-moment sensory and cognitive experiences, participants practiced techniques such as the body scan, mindful walking, and mindful stretching. These were further integrated into daily routines through mindful eating and hand-washing, facilitating the transfer of attentional skills into everyday contexts. Acceptance focused on regulating affective reactivity by fostering a non-judgmental and receptive attitude toward internal and external stressors. Participants were trained to approach thoughts, emotions, and bodily sensations with openness rather than avoidance or suppression. Specific practices, such as “allowing/letting be” and “recognizing the territory of aversion,” were used to help participants remain present with difficult experiences and reduce impulsive or habitual reactions.

The intervention was delivered as an eight-week program adapted from Williams and Penman [14]. The program incorporated structured meditation and mindfulness practices designed to systematically cultivate participants’ awareness of their present-moment experiences and daily lived experiences. The program was implemented through weekly sessions, with a consistent seven-day interval between each session. Each face-to-face group session included guided practice, experiential exercises, and group discussion, and lasted approximately 2 to 3 h. The curriculum was divided into three modules: Module 1 comprised two sessions: “awareness and automatic pilot” and “living with our heads”. Model II consisted of two other sessions: “gathering scattered minds” and “recognising the territory of aversion”. Module III consisted of Sessions 5, 6, 7, and 8: “allowing/letting be,” “thoughts are not facts”, “how can I best take care of myself?” and “maintaining and extending new learning”. See Table 1 for a summary of the MINDNUEDU.

Table 1.

Summary MINDNUEDU program (adapted from Hölzel et al. (2011) and Kuyken and Evans (2014)

Goals	Description	Session	Practices
Attention to regulation	Sustaining attention on the chosen object; whenever distracted, returning attention to the object	Session 1: Awareness and Automatic Pilot. Session 4: Recognize the Territory of Aversion	Session 1: Body scan and raisin exercise using five senses (visual, listening, smelling, touching and tasting) Session 4:Mindfulness of breath, body, sounds to deepen awareness
Body awareness	Focus is usually an object of internal experience: sensory experiences of breathing, emotions, or other body sensations	Session 1: Awareness and Automatic Pilot Session 4: Recognize the Territory of Aversion Session 2: Living in Our Heads Session 3: Gathering Scattered Minds	Session 1: as above Session 4: as above Session 2: Working with difficulties without struggle Session 3: Being and Doing Model practice: body scan and short mindfulness of breath; 3-Minute breathing space; mindful movement
Emotion regulation: reappraisal	Approaching ongoing emotional reactions in a different way (nonjudgmentally, with acceptance)	Session 4: Recognize the Territory of Aversion Session 5: Allowing /Letting Be Session 6: Thoughts Are Not Facts	Session4: as above Session 5: Exposure to difficult thoughts and feelings to transform them Session 6: Purposefully exposing to strong thoughts and feelings -Techniques for working wisely with difficult thoughts
Emotion regulation: exposure, extinction, and reconsolidation	Exposing oneself to whatever is present in the field of awareness; letting oneself be affected by it; refraining from internal reactivity	Session 3: Gathering the Scattered Mind Session 4: Recognize the Territory of Aversion Session 5: Allowing /Letting Be Session 7: How Can I Best Take Care of Myself? Session 8: Maintaining and Extending New Learning	Session 3: as above Session 4: as above Session 5: as above Sesson7: Develop self-care action plan into individual daily activities Session8: Personalized mindfulness practices for individual clients; maintain momentum of positive gains
Change in perspective on the self	Detachment from identification with a static sense of self	Sessions 1–8	Cultivating awareness with equanimity by frequently home practice

To ensure programme fidelity, the intervention was delivered by a certified instructor from the Oxford Mindfulness Centre with over 6 years of clinical experience, strictly adhering to the standardized protocol. To address quality control and individual attention within the group, participants were encouraged to maintain weekly written logs of their formal and informal practices. These logs served as a primary tool for individual monitoring and feedback. Following, each session’s practice, the instructor inquired about participants’ meditation experiences and discussed their homework logs. A similar structure was used to deliver the cognitive-behavioural components, namely, noticing direct experience, exploring the effects of bringing awareness to direct experience, and personal reflections on the relevance of managing daily life events, such as emotions and relationships with others. The control group continued their standard nursing curriculum, which did not include any formal mindfulness, meditation, or stress management training, ensuring no overlap with the intervention components.

Measures

The Chinese versions of the Perceived Stress Scale (PSS14-C) [15] and Connor–Davidson Resilience Scale (CDRS-C) [16] were administered at baseline (T1) and at week eight (T2) to assess perceived stress and resilience, respectively. The PSS14-C measured the degree to which individuals perceive their lives as unpredictable, uncontrollable, and overloaded during the previous month. The scale consisted of 14 items rated on a five-point Likert scale (0 = never to 4 = very often). Items 4, 5, 6, 7, 9, 10, and 13 were reverse-scored before calculating the total score. Higher total scores indicated greater perceived stress. The original Chinese version of the PSS14-C has demonstrated good internal consistency, with Cronbach’s alpha ranging from 0.84 to 0.86. In the present study, the PSS14-C showed excellent internal reliability, with Cronbach’s alpha coefficients of 0.89 at T1 and 0.89 at T2, indicating good reliability.

The CDRS-C [16] was used to assess resilience, defined as the capacity to withstand and recover from adversity. The scale consists of 22 items rated on a 5-point Likert scale (0 = not at all to 4 = almost always). Higher total scores indicate greater resilience. The CDRS-C has demonstrated strong internal consistency, with Cronbach’s alpha coefficients of 0.94 at both T1 (baseline) and T2 in current study. In addition to the self-reported psychological measures, participants’ physiological responses in the intervention group were assessed using SDNN as an objective indicator of autonomic nervous system activity.

To provide a more comprehensive assessment of autonomic function, additional HRV parameters were analyzed using the Eureka HRV device. The root mean square of successive differences (RMSSD) and high-frequency power (HF) were used as indicators of parasympathetic activity and vagal tone, reflecting the body’s recovery capacity. The Stress Index was also calculated to quantify cardiovascular tension and sympathetic–parasympathetic balance [17, 18]. The reliability of these HRV metrics was supported by the high precision of the device, which demonstrated over 99% accuracy in RR interval measurement compared with medical-grade systems (Biograph Infinity, TFDA No. 011374) and 99.99% accuracy in simulator testing. All HRV parameters were analyzed in accordance with the 1996 international guidelines for heart rate variability measurement [19].

Ethical considerations

This study was approved by the university’s Institutional Review Board (anonymised for review). Participation was voluntary and the participants provided written informed consent to participate. This study was conducted in compliance with the principles of the Declaration of Helsinki. Eligible students were approached by emailed advertisements inviting them to attend the research briefing. An overview of the study aims, methods, participants’ inclusion criteria, and duration of the MINDNUEDU was presented to potential participants. If potential participants met the study criteria and agreed to participate, they were required to return the consent form before completing the questionnaire at T1.

Data collection

Data collection was conducted across the 2021 academic semesters. To minimize bias, research assistants who were blinded to the study protocol were responsible for obtaining written informed consent and administering the psychometric instruments (PSS14-C and CDRS-C) at both T1 (pre-test) and T2 (post-test).

For physiological assessment, heart rate variability (HRV) was measured using the Eureka TW system, following a standardized short-term recording protocol. A separate research assistant recorded the standard deviation of normal-to-normal intervals (SDNN) immediately before and after each MINDNUEDU session. SDNN was selected as the primary physiological outcome because it is a validated index of overall autonomic nervous system (ANS) activity and autonomic flexibility. All data collection procedures were standardized to ensure consistency throughout the study period.

To enhance data reliability, HRV measurements were conducted in a quiet classroom with Tatami flooring to minimize environmental noise and external interference. All sessions were scheduled between 15:30 and 18:00 to control for potential circadian variations in ANS activity. Prior to recording, participants rested in a relaxed supine position for a 5-minute stabilization period, followed by a 5-minute recording period. To maintain signal integrity and reduce respiration-related artifacts, participants were instructed to remain silent and breathe spontaneously during the recording. HRV signals were monitored in real time by the research assistant, who ensured successful data transmission and verified technical integrity throughout the procedure [20].

Data analyses

Statistical analyses were conducted using IBM SPSS Statistics (version 29.0). Descriptive statistics were used to summarize participant characteristics and outcome variables. Baseline homogeneity between the intervention and control groups was assessed using chi-square or Fisher’s exact tests for categorical variables and independent t-tests for continuous variables.

Intervention effects on perceived stress and resilience were analyzed using Generalized Estimating Equations (GEE), including group, time, and Group × Time interaction terms, with study year entered as a covariate where appropriate. GEE accounts for within-subject correlations in repeated measures and is consistent with the Intention-to-Treat (ITT) principle. The Group × Time interaction term was considered the primary indicator of intervention effect. Effect sizes (Cohen’s d) were calculated by dividing the interaction coefficient by the pooled baseline standard deviation. For physiological outcomes, session-by-session heart rate variability (HRV) indices (SDNN, HF, and LF) were analyzed using paired t-tests to evaluate acute pre–post changes within the intervention group. Statistical significance was set at p < .05.

Results

Sample characteristics

Seventy participants completed the study (intervention group, n = 36; control group, n = 34). The intervention group included three males (8.3%) and 33 females (91.67%). The control group included three males (8.8%) and 31 females (91.2%). By the study year, there were 11 (32.4%), 3 (8.8%), 16 (47.1%), and 4 (11.8%) participants in the 1st, 2nd, 3rd, and 4th study year, respectively. Fisher’s exact test for sex between the two groups showed non-significant difference (p = .64). However, Fisher’s exact test indicated that there was a significant difference in the year of study between the two groups (χ² = 7.99, p = .04). Moreover, post hoc analysis indicated that second-year students had significantly higher perceived stress than first-year students, whereas no significant differences were found among the other grade levels.

Relationships between samples’ characteristics and perceived stress and resilience

An independent t-test was conducted to examine whether the participants’ sex was an impact factor for perceived stress and resilience. The results indicated that there were no significant differences for perceived stress (p = .77) and resilience (p = .66) between men and women. The ANOVA revealed significant differences in perceived stress (F [3,66] = 3.77, p = .015) among different study years. A Bonferroni post-hoc test indicated a significant difference in mean perceived stress between second- and first-year participants. The other study years did not differ significantly between the groups in terms of perceived stress. See Table 2.

Table 2.

Comparison of mean differences in perceived stress and resilience between the control and intervention groups (n = 70)

Group Outcomes	Control (n = 34)	Intervention (n = 36)	95% CI	t-test	Cohen’s d
Perceived stress				t (68) = 2.25 (p = .49),	0.478
Mean (M) differences	-4.71(6.64)	-8.61(7.79)	(0.44,7.36)
M(SD) at T1	30.42 (6.55)	30.42 (6.61)
M(SD) at T2	25.75(8.86)	21.81 (7.52)
Resilience				t (68) = 3.48 (p < .001)	0.83
Mean differences	-2.24(11.06)	-11.97(12.30)	(-4.15,-15.33)
M(SD) at T1	57.68 (15.12)	54.00(14.06)
M(SD) at T2	59.91(15.31)	65.97(13.69)

The relationship between participants’ perceived stress and resilience scores was examined using Spearman’s correlation. The results showed a significant correlation (r ₌ -.61, p < .001), indicating that participants with high resilience also had lower levels of perceived stress.

Perceived stress

At the baseline of the study (T1), an independent samples t-test confirmed that there was no significant difference in perceived stress levels between the intervention group (M = 30.42, SD = 6.61) and the control group (M = 30.44, SD = 6.55), t(68) = 0.016, p = .988, demonstrating comparable initial stress levels. Although a baseline imbalance was observed in the “year of study” (p = .04), the intervention group reported significantly lower perceived stress (M = 21.81, SD = 7.52) compared with the control group (M = 25.74, SD = 8.86), t(68) = 2.004, p = .049.

The GEE analysis further demonstrated a significant Group × Time interaction for perceived stress (β = −5.070, Wald χ² = 8.145, p = .004), indicating that the intervention group experienced a significantly greater reduction in stress over time compared with the control group. The magnitude of this effect was in the medium-to-large range. Year of study was not a significant predictor in the adjusted model (p = .614), suggesting that the intervention effect remained robust after accounting for baseline differences (see Table 3).

Table 3.

GEE parameter estimates for the effects of MINDNUEDU on perceive stress and resilience (n = 70)

Dependent variable	Effect	B	SE	95% CI	Wald χ2	p
Perceive stress	(Intercept)	36.98	3.99	[29.19, 44.80]	85.65	< 0.001
	Time	-8.87	2.45	[-13.63, -4.02]	12.96	< 0.001
	Group (Exp.)	4.40	2.92	[-1.33, 10.12]	2.26	0.133
	Year of study	-0.77	1.51	[-3.73, 2.20]	0.26	0.614
	Time × Group	-5.07	1.78	[-8.55, -1.58]	8.15	0.004
Resilience	(Intercept)	62.06	4.93	[52.40, 71.72]	158.62	< 0.001
	Time (Post)	2.24	1.87	[-1.43, 5.90]	1.43	0.232
	Group (Exp.)	-2.44	3.49	[-9.23, 4.41]	0.49	0.486
	Year of study	-1.84	1.61	[-4.99, 1.31]	1.31	0.253
	Time × Group	9.74	2.75	[4.34, 15.13]	12.51	< 0.001

Resilience

The intervention group showed a substantial increase in resilience scores from T1 (M = 54.00) to T2 (M = 65.97), whereas the control group’s scores remained relatively stable (M = 57.68 at T1, M = 59.91 at T2). An analysis of change scores revealed a highly significant difference between groups (t(68) = -3.476, p < .001), with the intervention group demonstrating a mean improvement of 11.97 points compared to 2.24 points in the control group.

The GEE model for resilience corroborated the initial findings by demonstrating a highly significant Group × Time interaction (Wald X² = 12.512, p < .001), which indicates that the observed improvements in resilience were uniquely attributable to the MINDNUEDU intervention. Furthermore, after adjusting for the “year of study” as a covariate, the intervention group exhibited a significantly higher resilience score at T2 relative to the baseline (β = 9.737, p < .001), indicating a large effect size, confirming the specific efficacy of the training program (See Table 3).

Dose-response effects on psychological outcomes

To evaluate the cumulative impact of intervention intensity, an Analysis of Covariance (ANCOVA) was conducted to examine the dose-response relationship across four exposure levels: control (0 sessions), low-dose (1–3 sessions), moderate-dose (4–6 sessions), and high-dose (7–8 sessions).

Impact on stress reduction

A significant dose–response relationship was observed for perceived stress (F = 5.238, p = .003, partial η² = 0.202). As participation frequency increased, the magnitude of stress reduction progressively improved, with mean decreases of − 4.16, − 6.42, − 8.19, and − 10.81 across increasing levels of session attendance, respectively. In addition, a significant interaction between dose level and year of study was identified (p = .025, partial η² = 0.139). This finding indicates that although the intervention reduced perceived stress overall, the strength of the dose–response effect varied according to students’ academic stage.

Impact on resilience

A significant main effect of intervention dose was also found for resilience (F = 4.821, p = .004, partial η² = 0.182). Post hoc pairwise comparisons indicated that participants in the high-dose group (7–8 sessions) demonstrated significantly greater improvements in resilience compared with the control group (0 sessions), with a mean difference of 12.77 points (p = .007). No significant interaction between intervention dose and year of study was observed (F = 1.421, p = .245), suggesting that the resilience-enhancing effects of the MINDNUEDU program were consistent across academic cohorts.

Heart rate variability (HRV)

To measure the immediate effects of the MINDNUEDU intervention, the participants’ HRV was examined before and after each weekly session. To better differentiate between sympathetic and parasympathetic activity, this study expanded its physiological analysis to include HF power and RMSSD, both of which primarily reflect vagal tone. The concurrent increase in HF, RMSSD, and SDNN, alongside the reduction in the Stress Index, provides robust evidence that the intervention effectively enhanced parasympathetic activation rather than merely altering overall heart rate variability. The intervention’s physiological impact was assessed through four key metrics: Stress Index, SDNN, RMSSD, and HF power. Time-domain analysis showed significant increases in SDNN and RMSSD, indicating enhanced overall autonomic flexibility and parasympathetic tone. Concurrently, frequency-domain analysis revealed a significant rise in HF power, reflecting an immediate relaxation response through vagal activation. These improvements were further corroborated by a significant reduction in the instrument-measured Stress Index, providing robust objective evidence for the efficacy of the MINDNUEDU intervention.

To further investigate these responses across the 8-week program, Wilcoxon Signed-Rank Tests were conducted for each session. As detailed in Table 4, the physiological effects were most robust during the early and middle stages of the intervention (e.g., Week 1 and Week 4), where participant attendance was at its peak (N = 29 and N = 22, respectively). In Week 1, SDNN significantly increased from a median of 51.90 ms to 68.85 ms, while HF power rose from 1092.0 ms² to 1612.5 ms² (p < .001), signaling strong immediate vagal activation.

Table 4.

Multidimensional HRV analysis and stress index results

Session	Parameter	Pre-test Median (IQR)	Post -test Median (IQR)	Z	P
Week 1 n = 29	SDNN	51.9(26.98)	68.85 (38.58)	-3.34	< 0.001
	HF	1092(1324.75)	1612.5 (2978.5)	-3.58	< 0.001
	RMSSD	40(29.98)	51.5 (42)	-2.65	0.008
	Stress Index	66(63.25)	41 (37)	-3.32	< 0.001
WEEK 2 n = 15	SDNN	50.6(41.4)	51.2(48.2)	-1.81	0.069
	HF	473(2085)	1092 (2272)	-2.50	0.012
	RMSSD	30.9(40.6)	44.2(45.2)	-2.56	0.011
	Stress Index	68(106)	70(80)	− 0.483	0.629
WEEK3 n = 27	SDNN	54.6(43.9)	76.9(44.83)	-2.43	0.015
	HF	958(1943)	1926(3017.25)	-2.4	0.016
	RMSSD	40.1(42.12)	70.5(42.7)	-2.94	0.003
	Stress Index	73.5(70.25)	34(46)	-3.16	0.002
WEEK 4 n = 22	SDNN	58.6(25)	71.8(29.4)	-3.59	< 0.001
	HF	1044(1633)	2090(3298)	-3.56	< 0.001
	RMSSD	49.5(25.9)	61.3(43.7)	-3.36	< 0.001
	Stress Index	64(50.5)	37(22.5)	-3.825	< 0.001
WEEK 5 n = 18	SDNN	54.85(34.95)	70.2(31.25)	-2.98	0.003
	HF	1044.5(1852)	1549.5(1292.25)	-1.42	0.157
	RMSSD	46.7(30.08)	57.38(39.63)		0.014
	Stress Index	71(34.95)	42(38)	-2.66	0.008
WEEK 6 n = 12	SDNN	56.3(48.3)	76.55(51.85)	-1.71	0.087
	HF	941(1918.5)	1963(2997.5)	-1.08	0.279
	RMSSD	44.2(45.5)	58.85(51.98)	-1.01	0.311
	Stress Index	72.5((106.5)	36(90.5)	-1.92	0.055
WEEK 7 n = 17	SDNN	60.8(17.45)	79.05(40.58)	-2.20	0.028
	HF	1436(1745)	2122(3093.5)	-1.302	0.193
	RMSSD	47.2(37.6)	1436(56.33)	-1.88	0.059
	Stress Index	58.5(48.25)	34(56)	-2.08	0.037
WEEK 8 n = 18	SDNN	59.85(23.05)	72.7(16.15)	-2.50	0.012
	HF	1464.5(1671.25)	1947(2003.75)	− 0.68	0.500
	RMSSD	53.08(31.4)	61.55(21.13)	-2.07	0.039

Participant numbers (n) fluctuated weekly due to non-negotiable university activities and academic scheduling conflicts. Wilcoxon signed-rank tests were performed on a per-session basis, including only participants with complete pre- and post-test data for each respective week. The consistent trend in HRV improvements across sessions supports the acute physiological benefits of the MINDNUEDU practice”

In the latter half of the program, despite a reduction in sample size (reaching a minimum of N = 12 in Week 6), the trend of physiological improvement remained consistent. Although the increase in SDNN showed only marginal significance in Week 6 (p = .087) due to reduced statistical power, significant enhancements in SDNN (p = .012) and RMSSD (p = .039) were sustained through Week 8. Furthermore, the Stress Index consistently showed significant post-session reductions in most weeks (e.g., Week 7, p = .037), confirming that the MINDNUEDU practice effectively mitigated physiological stress throughout the intervention period (See Table 4).

Discussion

The theory-based MINDNUEDU program effectively reduced perceived stress and enhanced resilience in nursing students compared to the control group, directly addressing our primary research question. These psychological improvements were corroborated by significant gains in HRV parameters (SDNN, HF, RMSSD) and a decreased Stress Index. These findings indicate a pivotal shift from sympathetic-dominant ‘fight-or-flight’ arousal to parasympathetic-mediated ‘rest-and-digest’ stability. Such autonomic flexibility facilitates more adaptive cognitive processing and reduced stress appraisal, providing a physiological foundation for the observed increase in self-regulation and resilience.

The proficiency of the instructor, particularly one with a clinical psychology background, was pivotal in fostering participant engagement and home-based practice. The MINDNUEDU sessions followed a structured pedagogical cycle: thematic instruction, guided practice, and group reflection. By maintaining an inquisitive stance of curiosity and acceptance, the instructor facilitated the exploration of perceptions during practice, subsequently linking individual experiences to broader personal and professional contexts. This process enabled students to transform immediate sensations into generalizable coping strategies, encouraging the practical application of mindfulness to manage stress in diverse clinical and academic situations.

This study contributes to the field by providing novel evidence through a multi-method assessment of a theory-driven intervention specifically tailored for nursing students. Unlike previous research, our study distinguishes itself by integrating the Monitoring and Acceptance Theory (MAT) with high-fidelity physiological monitoring to evaluate feasibility in an understudied population of Asian nursing students. The theory-based MINDNUEDU program effectively reduced perceived stress and enhanced resilience among nursing students, directly addressing the primary research question. These improvements were corroborated by significant gains in HRV parameters (SDNN, HF, RMSSD) and a decreased Stress Index, indicating a shift from sympathetic-dominant ‘fight-or-flight’ arousal to parasympathetic-mediated ‘rest-and-digest’ stability [18].

Through the lens of Monitoring and Acceptance Theory (MAT), MINDNUEDU likely enhanced students’ sensory monitoring while fostering non-judgmental acceptance. These psychological processes translated into measurable autonomic regulation, where increased vagal activity—reflected by high-fidelity HRV data—facilitated a shift in cognitive processing, reducing negative appraisals of stressors [21, 22]. This dual-assessment approach—combining psychological scales with objective HRV data—addresses a critical gap in existing nursing education literature. By utilizing the Eureka device (offering 99.99% accuracy), we ensure that these findings are not merely subjective perceptions but are grounded in measurable autonomic nervous system regulation. This synergy between physiological flexibility and psychological self-regulation empowers students to perceive stressors as ‘manageable challenges’ rather than overwhelming threats, strengthening their internal psychological buffer [23].

Contextual analysis revealed that the observed HRV variability during Weeks 5 and 6 coincided with the university’s midterm examination period. This peak in academic demand and elevated baseline sympathetic activation likely masked immediate physiological improvements, explaining the lack of statistical significance in HRV parameters during those specific weeks. Furthermore, the nature of the MINDNUEDU curriculum transitioned over the 8-week period. The somatic-focused practices in the first half (e.g., Body Scan) appeared more potent in eliciting immediate parasympathetic responses (HF and RMSSD). In contrast, the second half shifted toward cognitive and social integration (e.g., Mindful Listening), where the benefits appeared to transition from direct physiological relaxation to broader psychological stress management and self-regulation.

While the 55% attendance rate is a noted limitation, the significant physiological and psychological gains achieved despite this variability underscore the intervention’s high potency for this population. This suggests that even inconsistent exposure to structured mindfulness can yield meaningful shifts in autonomic balance. The reliability of these captured improvements is further reinforced by the technical precision of the Eureka HRV device, which maintains 99.99% simulator accuracy and complies with 1996 international HRV standards. Such high measurement validity ensures that the observed session-by-session enhancements reflect genuine physiological recovery rather than measurement noise or instrumentation error.

Ultimately, these findings highlight MINDNUEDU’s capacity to expand students’ “physiological reservoir.” By strengthening this physiological foundation, the intervention enables nursing students to recover more rapidly from the sympathetic arousal triggered by intense clinical or academic stressors. This prevents the cumulative psychological burnout common in nursing education, effectively transforming what were previously perceived as overwhelming threats into manageable professional challenges.

Beyond immediate stress reduction, the MINDNUEDU program significantly enhanced participants’ resilience, a finding consistent with meta-analyses in university settings [5]. A primary explanation for this improvement lies in the dual cultivation of physiological and cognitive flexibility. From a physiological perspective, the session-by-session increases in SDNN and RMSSD represent an expansion of the ‘physiological reservoir’ within the autonomic nervous system. This strengthened foundation enables students to recover more rapidly from sympathetic arousal triggered by academic or clinical stressors, thereby preventing cumulative burnout.

Furthermore, the program fosters resilience by replacing automatic ‘fight-or-flight’ reactions with ‘thoughtful reactions. By cultivating conscious awareness and non-judgmental acceptance, participants transform overwhelming threats into ‘manageable challenges’ [23]. This internal shift is confirmed by the significant inverse relationship between resilience and perceived stress: as physiological stability and resilience increased, subjective stress decreased. In the high-pressure context of nursing education, mindfulness acts as an internal psychological buffer, allowing students to process difficult emotions without becoming overwhelmed.

The practical importance of these findings is underscored by the Minimal Clinically Important Difference (MCID). While a 5-point change on the CD-RISC is considered clinically meaningful [24], participants in the MINDNUEDU group achieved a mean increase of 11.97 points. This substantial gain, more than double the MCID threshold, indicates that the intervention provided a perceptibly beneficial boost to students’ psychological resources rather than a mere statistical fluctuation. The attainment of an 11.97-point gain—nearly double the MCID—serves as a ‘unique selling point of the MINDNUEDU model, proving its clinical utility in enhancing the psychological resources of a high-pressure student population. While the present study did not directly test mediating neurophysiological mechanisms, the high-fidelity HRV data obtained via the Eureka device (99.99% accuracy) provides a robust physiological basis for these resilient outcomes.

Regarding baseline differences, sophomore students (Year 2) exhibited significantly higher perceived stress than freshmen (Year 1), a finding that aligns with the structural transition in the nursing curriculum. While first-year students focus on foundational general education, the sophomore year marks a critical juncture where the workload intensifies with core professional courses, such as Fundamental and Medical-Surgical Nursing, alongside initial clinical practicum requirements. By pinpointing this specific transitional stress peak in the sophomore year, our study offers a unique pedagogical insight: mindfulness interventions are most critical when students first bridge the gap between theoretical foundational courses and intense professional clinical training.

The shift from theoretical education to rigorous professional training and clinical entry imposes a substantial cognitive and emotional burden. This transitional stress explains the elevated baseline scores among sophomores, as they must rapidly adapt to specialized professional skills and clinical environments that differ fundamentally from their first-year experience.

The average weekly attendance was 55%, with only 19.44% of participants achieving full attendance. Compared to previous studies reaching 80% [25], this rate reflects the significant real-world challenges of implementing extracurricular interventions within a rigorous nursing curriculum. However, the fact that significant improvements in SDNN, RMSSD, and the Stress Index were captured despite this 55% attendance rate suggests the intervention possesses high potency; even partial exposure provided a measurable physiological buffer.

Regarding resilience gains, the univariate analysis demonstrated a significant dose-response effect (F = 4.821, p = .004), with an effect size of partial η² = 0.182. Post-hoc pairwise comparisons showed that the high-dosage group (4 sessions) yielded significantly higher resilience improvements compared to the low-dosage group (1 session), with a mean difference of 12.77 points (p = .007). Unlike the results for stress reduction, the interaction between dose and grade level was not statistically significant (F = 1.421, p = .245), suggesting that the positive impact of increased intervention frequency on resilience was consistent across all students, regardless of their year of study.

While the lack of full compliance may have masked the true magnitude of the program’s impact, these findings provide a strong foundation for future implementation. To enhance feasibility, we propose transitioning from a standalone model to a modularized, credit-bearing curriculum or a hybrid delivery format (combining in-person and asynchronous digital sessions). Integrating mindfulness as a formal component of professional development would reduce scheduling conflicts and prevent it from becoming an additional burden during peak examination periods [26].

Limitations

Several limitations should be acknowledged. First, the quasi-experimental design may introduce selection bias. Specifically, there is the potential for self-selection bias, as students who volunteered for the program may have had higher motivation for stress management compared to the general student population. This factor limits the generalizability of the findings, and future research should seek to include more diverse samples with varied levels of motivation to provide a broader understanding of the intervention’s effects.

Additionally, the lack of longitudinal follow-up represents the sustainability of these benefits though exceeding the MCID for resilience, remains unknown. While the intervention demonstrated substantial improvements in resilience (exceeding the MCID) and HRV, the sustainability of these benefits over time remains uncertain. Given the fluctuating demands of the nursing curriculum, including high-stress periods such as midterms and clinical rotations, it is essential for future studies to incorporate long-term follow-up assessments to assess whether the observed improvements in autonomic flexibility and psychological resilience are maintained during subsequent high-pressure periods.

Another limitation relates to the variability in intervention exposure. The 55% average attendance rate largely attributable to unavoidable academic scheduling conflicts and clinical requirements, limits conclusions regarding the minimum effective dose of the intervention program and may have attenuated the observed effects. Variability in session attendance also restricts the ability to fully examine does-effects between intervention exposure and psychological or physical outcomes. Future research should prioritize integrating such training into the formal academic schedule or adopting hybrid delivery formats to enhance compliance and clarify the relationship between intervention exposure and physiological outcomes. Overall, these limitations suggest a cautious interpretation while providing clear directions for strengthening methodological rigor in future nursing education research.

Conclusion

This study advances the field by implementing MINDNUEDU, a theory-driven mindfulness program specifically tailored for nursing students. By integrating psychological scales with objective HRV assessments, it provides unique insights into the transitional stress patterns and implementation challenges faced by Asian nursing students, a population underrepresented in mindfulness research. The program demonstrated significant efficacy in enhancing both physiological and psychological resilience. Session-by-session monitoring revealed a shift toward parasympathetic dominance, providing an objective buffer against stress that remained evident even during peak examination periods. Notably, the 11.97-point gain in resilience, more than double the MCID threshold, underscores the intervention’s practical impact despite a 55% average attendance rate. Given the pronounced stress observed during the sophomore transition to clinical courses, integrating such training into the formal nursing curriculum via modularized or hybrid formats is recommended. These findings provide a robust foundation for developing self-regulation tools to support a resilient future nursing workforce, though further research is needed to confirm long-term sustainability and optimal implementation strategies.

Authors’ contributions

Concept development: T-J T, C-JW. Data analysis: T-J T, C-JW. Manuscript writing & editing: T-J T, C-JW. Final appraisal: T-J T, C-JW.

Funding

This study was supported by Ministry of Science and Technology (MOST 110-2511-H-255-009), Taiwan.

Data availability

Research data are not shared.

Declarations

Ethics approval and consent to participate

All procedures were performed per the relevant guidelines (for example. Declaration of Helsinki).

The authors confirm that all methods were performed per the relevant guidelines and regulations.

This study was approved by the Chang Gung Medical Foundation Institutional Review Board and Human Research Ethics Committee (reference number: 202100833A3D001) before commencing the study.

Informed consent: Informed consent was obtained from all participants involved in the study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.