Other Titles
Interpretable Psychosocial Pathways to Poor Sleep in a High-Stress Workforce: A Decision-Tree Analysis of Military Personnel [Title Slide]
Abstract
Purpose: Sleep disturbances are common in high-stress occupational environments and strongly influence psychological and functional outcomes. This study identified hierarchical psychosocial and occupational predictors of poor sleep among active-duty military personnel using an interpretable machine learning model.
Methods: A cross-sectional study of 1,386 Taiwanese army personnel was conducted between 2020 and 2021. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI ≥ 5 as poor sleep). Depressive symptoms, perceived stress, and resilience were evaluated using validated scales. A chi-squared automatic interaction detection (CHAID) decision-tree algorithm was applied to identify hierarchical predictors and risk pathways in training (70%) and testing (30%) datasets. Model performance was assessed using area under the curve (AUC), F1 score, and classification accuracy.
Results: Overall, 86.4% of participants met the criteria for poor sleep quality. Depressive symptoms were the strongest predictor, followed by perceived stress and resilience. Among individuals with minimal depressive symptoms, low resilience markedly increased poor-sleep risk (83.0% vs. 43.5%). Perceived stress further differentiated sleep outcomes among those with mild depressive symptoms (80.5% vs. 50.0%). Brigade type moderated risk among participants with moderate depressive symptoms, with those in combined-arms or reserve brigades showing higher prevalence (94.4%) than those in special forces (79.5%). The CHAID model achieved AUCs of 0.746 (training) and 0.712 (testing), with F1 scores exceeding 0.90, confirming stable and interpretable predictive performance.
Conclusion and Implications: Depressive symptoms, perceived stress, and resilience jointly shape hierarchical pathways to poor sleep in high-stress environments. Resilience serves as a protective factor, whereas occupational context functions as a structural determinant of sleep vulnerability, an occupational “sleep desert.” These findings inform nursing and occupational health practice by highlighting the need for integrated screening, resilience-building, and stress management programs to promote sleep health across military and civilian high-demand professions.
Notes
Presenter notes available in attached slide deck. To see the notes in Adobe Acrobat, go to Tools > Comment or look at your Layers Panel. If the notes were saved as comments or layers, you can toggle them visible.
References:
Erez, D., Lieberman, H. R., Rafael, N., Ketko, I., & Moran, D. S. (2025). The impact of a 20-h rotating watch schedule on cognitive and mood states in submarine operations. Journal of Sleep Research, 34(4), e14400. https://doi.org/10.1111/jsr.14400
Johnson, D. A., Attarian, H., Hale, L., & Knutson, K. (2025). Sleep deserts: From observational to interventional environmental patterns of sleep health and sleep disorders. Sleep Health. https://doi.org/10.1016/j.sleh.2025.06.008
Lotti, S., Moretton, M., Bulgari, M., Costantini, L., Dall'Asta, M., De Amicis, R., Esposito, S., Ferraris, C., Fiorini, S., Formisano, E., Giustozzi, D., Guglielmetti, M., Membrino, V., Moroni, A., Napoletano, A., Perone, N., Proietti, E., Tristan Asensi, M., Vici, G.,... Dinu, M. (2025). Association between shift work and eating behaviours, sleep quality, and mental health among Italian workers. European Journal of Nutrition, 64(2), 97. https://doi.org/10.1007/s00394-025-03600-5
Norful, A. A., Albloushi, M., Zhao, J., Gao, Y., Castro, J., Palaganas, E., Magsingit, N. S., Molo, J., Alenazy, B. A., & Rivera, R. (2024). Modifiable work stress factors and psychological health risk among nurses working within 13 countries. Journal of Nursing Scholarship, 56(5), 742-751. https://doi.org/10.1111/jnu.12994
Pasyar, N., Rambod, M., Abbasi, A., & Salmanpour, M. (2025). Social relational quality and ethical climate as the predictors of sleep quality in employees of the operating room: A hierarchical linear regression analysis. BMC Health Services Research, 25(1), 718. https://doi.org/10.1186/s12913-025-12903-6
Sigma Membership
Non-member
Type
Presentation
Format Type
Text-based Document
Study Design/Type
Cross-Sectional
Research Approach
Quantitative Research
Keywords:
Stress and Coping, Workforce, Sub-Acute Care, Sleep Quality, Resilience, Hardiness, Stress, Psychological Stress, Military Personnel, Decision Tree, Decision Trees, Taiwan
Recommended Citation
Li, Shih-Ting; Hong, Yu-Chia; Chang, Yun-Cune; and Chiang, Hui-Hsun, "Psychosocial Pathways to Poor Sleep Among High-Stress Military Workforce: A Decision-Tree Analysis" (2026). International Nursing Research Congress (INRC). 70.
https://www.sigmarepository.org/inrc/2026/presentations_2026/70
Conference Name
37th International Nursing Research Congress
Conference Host
Sigma Theta Tau International
Conference Location
Toronto, Ontario, Canada
Conference Year
2026
Rights Holder
All rights reserved by the author(s) and/or publisher(s) listed in this item record unless relinquished in whole or part by a rights notation or a Creative Commons License present in this item record. All permission requests should be directed accordingly and not to the Sigma Repository. All submitting authors or publishers have affirmed that when using material in their work where they do not own copyright, they have obtained permission of the copyright holder prior to submission and the rights holder has been acknowledged as necessary.
Review Type
Abstract Review Only: Reviewed by Event Host
Acquisition
Proxy-submission
Date of Issue
2026-07-27
Funder
Ministry of National Defense, Medical Affairs Bureau, Taiwan
Psychosocial Pathways to Poor Sleep Among High-Stress Military Workforce: A Decision-Tree Analysis
Toronto, Ontario, Canada
Purpose: Sleep disturbances are common in high-stress occupational environments and strongly influence psychological and functional outcomes. This study identified hierarchical psychosocial and occupational predictors of poor sleep among active-duty military personnel using an interpretable machine learning model.
Methods: A cross-sectional study of 1,386 Taiwanese army personnel was conducted between 2020 and 2021. Sleep quality was assessed using the Pittsburgh Sleep Quality Index (PSQI ≥ 5 as poor sleep). Depressive symptoms, perceived stress, and resilience were evaluated using validated scales. A chi-squared automatic interaction detection (CHAID) decision-tree algorithm was applied to identify hierarchical predictors and risk pathways in training (70%) and testing (30%) datasets. Model performance was assessed using area under the curve (AUC), F1 score, and classification accuracy.
Results: Overall, 86.4% of participants met the criteria for poor sleep quality. Depressive symptoms were the strongest predictor, followed by perceived stress and resilience. Among individuals with minimal depressive symptoms, low resilience markedly increased poor-sleep risk (83.0% vs. 43.5%). Perceived stress further differentiated sleep outcomes among those with mild depressive symptoms (80.5% vs. 50.0%). Brigade type moderated risk among participants with moderate depressive symptoms, with those in combined-arms or reserve brigades showing higher prevalence (94.4%) than those in special forces (79.5%). The CHAID model achieved AUCs of 0.746 (training) and 0.712 (testing), with F1 scores exceeding 0.90, confirming stable and interpretable predictive performance.
Conclusion and Implications: Depressive symptoms, perceived stress, and resilience jointly shape hierarchical pathways to poor sleep in high-stress environments. Resilience serves as a protective factor, whereas occupational context functions as a structural determinant of sleep vulnerability, an occupational “sleep desert.” These findings inform nursing and occupational health practice by highlighting the need for integrated screening, resilience-building, and stress management programs to promote sleep health across military and civilian high-demand professions.
Description
This session presents an interpretable machine learning approach to uncover psychosocial and occupational pathways leading to poor sleep among military personnel. Participants will learn how hierarchical predictors, such as depressive symptoms, perceived stress, resilience, and occupational context, interact to shape sleep vulnerability and inform data-driven strategies for resilience and sleep health promotion.