, 김은혜1
, 서수량1
, 이선영1
, 안소라1
, 김혜진1
, 김예슬1
, 김유진1
, 박미정1
, 공현중2,3,4
, 이사람2,4
, 홍영주2
, 정민호1
, Eunhye Kim1
, Suryang Seo1
, Sunyoung Lee1
, Sora An1
, Hyejin Kim1
, Ye Seul Kim1
, Youjin Kim1
, Mijeong Park1
, Hyoun-Joong Kong2,3,4
, Saram Lee2,4
, Yeong Joo Hong2
, Minho Jung1
1서울대학교병원 간호부문
2서울대학교병원 융합의학과
3서울대학교병원 융합의학기술원
4서울대학교 의과대학 의학과
1Department of Nursing, Seoul National University Hospital, Seoul, South Korea
2Department of Transdisciplinary Medicine, Seoul National University Hospital, Seoul, South Korea
3Institute of Convergence Medicine with Innovative Technology, Seoul National University Hospital, Seoul, South Korea
4Department of Medicine, Seoul National University College of Medicine, Seoul, South Korea
© 2026 Korean Society of Nursing Science
This is an Open Access article distributed under the terms of the Creative Commons Attribution NoDerivs License (http://creativecommons.org/licenses/by-nd/4.0) If the original work is properly cited and retained without any modification or reproduction, it can be used and re-distributed in any format and medium.
Conflicts of Interest
No potential conflict of interest relevant to this article was reported.
Acknowledgements
None.
Funding
This study was supported by the Seoul National University Hospital Research Fund in 2025.
Data Sharing Statement
Please contact the corresponding author for data availability.
Author Contributions
Conceptualization: EJ, EK, SA, YK, MJ. Methodology: EJ, EK, SA, YK, MJ. Software: MJ. Validation: EK, SA, YK, MJ. Formal analysis: EJ, EK, MJ. Investigation: EK, Sunyoung L, SA, HK, YSK, YK. Resources: MJ. Data curation: EJ, EK, MJ. Visualization: MJ. Supervision: EJ, EK. Project administration: EJ, EK. Funding acquisition: EJ. Writing–original draft: SS, Sunyoung L, SA. Writing–review & editing: EJ, EK, HK, YSK, YK, MP, HJK, Saram L, YJH, MJ. Final approval of the manuscript: all authors.
| Characteristic | Total (N=120) | Intervention group (n=70) | Control group (n=50) | p | Effect size |
|---|---|---|---|---|---|
| Age (yr) | 28.0 (26.0–32.5) | 28.0 (26.0–32.0) | 28.5 (25.0–34.0) | .491a) | 0.17b) |
| Gender | .172c) | .12d) | |||
| Women | 118 (98.3) | 70 (100.0) | 48 (96.0) | ||
| Men | 2 (1.7) | 0 (0.0) | 2 (4.0) | ||
| Marital status | .254e) | .05d) | |||
| Single | 87 (72.5) | 48 (68.6) | 39 (78.0) | ||
| Married | 33 (27.5) | 22 (31.4) | 11 (22.0) | ||
| Education | >.999c) | .00d) | |||
| Bachelor’s degree | 115 (95.8) | 67 (95.7) | 48 (96.0) | ||
| Master’s degree or higher | 5 (4.2) | 3 (4.3) | 2 (4.0) | ||
| Clinical career (mo) | 58.0 (23.5–110.0) | 55.5 (23.0–93.0) | 61.5 (24.0–120.0) | .243a) | 0.18b) |
The average posterior probabilities for each model were as follows: Model 2 (0.958, 0.972), Model 3 (0.937, 0.928, 0.914), Model 4 (0.960, 0.955, 0.953, 0.897), and Model 5 (0.925, 0.945, 0.959, 0.919, 0.956).
AIC, Akaike information criterion; AWE, approximate weight of evidence; BIC, Bayesian information criterion; BLRT, bootstrap likelihood ratio test; SABIC, sample-size adjusted Bayesian information criterion.
Values are age-adjusted least-squares means with 95% confidence intervals. p-values represent comparisons between the intervention and control groups within each latent profile based on the age-adjusted ANCOVA model. Bonferroni-adjusted post hoc comparisons for the profile main effect were as follows: a=Healthy-Recovery (n=25), b=Moderate (n=50), and c=Burnout-Vulnerable (n=45). Overall effects were obtained from the two-way ANCOVA adjusted for age, with partial η² presented in parentheses. CI, confidence interval; LS mean, least-squares mean; P×I, profile-by-intervention interaction; partial η², partial eta squared.
| Characteristic | Total (N=120) | Intervention group (n=70) | Control group (n=50) | p | Effect size |
|---|---|---|---|---|---|
| Age (yr) | 28.0 (26.0–32.5) | 28.0 (26.0–32.0) | 28.5 (25.0–34.0) | .491 |
0.17 |
| Gender | .172 |
.12 |
|||
| Women | 118 (98.3) | 70 (100.0) | 48 (96.0) | ||
| Men | 2 (1.7) | 0 (0.0) | 2 (4.0) | ||
| Marital status | .254 |
.05 |
|||
| Single | 87 (72.5) | 48 (68.6) | 39 (78.0) | ||
| Married | 33 (27.5) | 22 (31.4) | 11 (22.0) | ||
| Education | >.999 |
.00 |
|||
| Bachelor’s degree | 115 (95.8) | 67 (95.7) | 48 (96.0) | ||
| Master’s degree or higher | 5 (4.2) | 3 (4.3) | 2 (4.0) | ||
| Clinical career (mo) | 58.0 (23.5–110.0) | 55.5 (23.0–93.0) | 61.5 (24.0–120.0) | .243 |
0.18 |
| Model | Information criteria | Accuracy of classification | Fit differences | Class proportions (%) | |||
|---|---|---|---|---|---|---|---|
| AIC | AWE | BIC | SABIC | Entropy | BLRT (p) | ||
| 1 | 5,169.22 | 5,272.97 | 5,197.09 | 5,165.48 | - | - | - |
| 2 | 4,917.33 | 5,084.75 | 4,961.93 | 4,911.34 | .890 | <.001 | 37.5, 62.5 |
| 3 | 4,848.74 | 5,079.72 | 4,910.07 | 4,840.51 | .837 | <.001 | 20.8, 37.5, 41.7 |
| 4 | 4,804.08 | 5,098.43 | 4,882.13 | 4,793.61 | .874 | <.001 | 8.3, 28.3, 25.0, 38.3 |
| 5 | 4,805.35 | 5,163.13 | 4,900.12 | 4,792.63 | .886 | .141 | 26.7, 28.3, 3.3, 36.7, 5.0 |
| Classification | Healthy-Recovery (n=25)a | Moderate (n=50)b | Burnout-Vulnerable (n=45)c | F | p | η2 | Post hoc test (Bonferroni) |
|---|---|---|---|---|---|---|---|
| Perceived health status | |||||||
| PCS | 89.67±6.04 | 76.01±8.88 | 60.32±10.03 | 93.39 | <.001 | .615 | a>b>c |
| MCS | 80.15±9.35 | 70.30±7.04 | 51.14±8.01 | 125.71 | <.001 | .682 | a>b>c |
| Occupational fatigue and recovery | |||||||
| Chronic fatigue | 49.33±17.19 | 71.73±15.95 | 91.04±13.42 | 60.69 | <.001 | .509 | c>b>a |
| Acute fatigue | 53.20±15.20 | 77.67±10.13 | 94.81±10.11 | 108.61 | <.001 | .650 | c>b>a |
| Recovery | 86.00±12.55 | 56.40±14.33 | 39.70±12.97 | 94.89 | <.001 | .619 | a>b>c |
| Outcome variables | |||||||
| Sleep quality (PSQI) | 4.68±2.08 | 7.60±2.82 | 10.51±2.99 | 37.37 | <.001 | .390 | c>b>a |
| Work-life balance | 107.72±17.09 | 86.82±14.58 | 70.36±14.07 | 50.86 | <.001 | .465 | a>b>c |
| Shift work job satisfaction | 32.60±6.44 | 26.96±5.66 | 25.53±5.69 | 12.29 | <.001 | .174 | a>b,c |
| Turnover intention | 6.72±2.72 | 8.38±2.89 | 10.07±2.68 | 12.11 | <.001 | .171 | c>a,b |
| Outcome | Latent profile | Control LS mean (95% CI) | Intervention LS mean (95% CI) | p | Profile post hoc | Overall ANCOVA effects, p (partial η²) |
|---|---|---|---|---|---|---|
| Sleep quality | Healthy-Recoverya | 5.00 (2.54–7.46) | 4.56 (3.31–5.81) | .754 | c>b>a | Profile: <.001 (.342); Intervention: .252 (.012); P×I: .855 (.003) |
| Moderateb | 7.85 (6.79–8.91) | 7.32 (6.17–8.47) | .502 | |||
| Burnout-Vulnerablec | 11.19 (9.89–12.49) | 10.08 (9.02–11.13) | .191 | |||
| Work-life balance | Healthy-Recoverya | 109.41 (96.04–122.77) | 106.50 (99.68–113.31) | .702 | a>b>c | Profile: <.001 (.396); Intervention: .489 (.004); P×I: .992 (.000) |
| Moderateb | 87.78 (82.03–93.54) | 85.96 (79.71–92.21) | .672 | |||
| Burnout-Vulnerablec | 71.83 (64.74–78.93) | 69.73 (63.98–75.48) | .648 | |||
| Shift work job satisfaction | Healthy-Recoverya | 26.40 (21.40–31.39) | 34.53 (31.98–37.07) | .005 | a>c | Profile: .017 (.069); Intervention: .047 (.034); P×I: .010 (.078) |
| Moderateb | 26.31 (24.16–28.46) | 27.59 (25.26–29.93) | .425 | |||
| Burnout-Vulnerablec | 26.65 (24.00–29.31) | 24.62 (22.47–26.77) | .240 | |||
| Turnover intention | Healthy-Recoverya | 7.60 (5.25–9.96) | 6.05 (4.85–7.25) | .246 | c>a,b | Profile: <.001 (.159); Intervention: .363 (.007); P×I: .545 (.011) |
| Moderateb | 8.39 (7.38–9.40) | 8.52 (7.42–9.62) | .862 | |||
| Burnout-Vulnerablec | 10.28 (9.03–11.53) | 10.12 (9.11–11.14) | .847 |
Values are presented as median (Q1–Q3) for continuous variables and number (%) for categorical variables. By Wilcoxon rank sum test. Cohen’s d. By Fisher’s exact test. Cramer’s V. By Pearson’s chi-square test.
The average posterior probabilities for each model were as follows: Model 2 (0.958, 0.972), Model 3 (0.937, 0.928, 0.914), Model 4 (0.960, 0.955, 0.953, 0.897), and Model 5 (0.925, 0.945, 0.959, 0.919, 0.956). AIC, Akaike information criterion; AWE, approximate weight of evidence; BIC, Bayesian information criterion; BLRT, bootstrap likelihood ratio test; SABIC, sample-size adjusted Bayesian information criterion.
Values are presented as mean±standard deviation unless otherwise stated. MCS, mental component summary; PCS, physical component summary; PSQI, Pittsburgh Sleep Quality Index; η2, eta squared.
Values are age-adjusted least-squares means with 95% confidence intervals. p-values represent comparisons between the intervention and control groups within each latent profile based on the age-adjusted ANCOVA model. Bonferroni-adjusted post hoc comparisons for the profile main effect were as follows: a=Healthy-Recovery (n=25), b=Moderate (n=50), and c=Burnout-Vulnerable (n=45). Overall effects were obtained from the two-way ANCOVA adjusted for age, with partial η² presented in parentheses. CI, confidence interval; LS mean, least-squares mean; P×I, profile-by-intervention interaction; partial η², partial eta squared.
