Balancing mental health through predictive modeling for healthcare workers during public health crises
Abstract During public health emergencies such as SARS, Ebola, and COVID-19, healthcare workers (HCWs) are often on the front lines, placing them at increased risk for adverse mental health outcomes, particularly depression and anxiety. Despite this risk, there remains a scarcity of research focused...
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-08-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-14403-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849235496619737088 |
|---|---|
| author | Jiana Wang Lin Feng Nana Meng Cong Yang Fanfan Cai Xin Huang Yihang Sun Kristin K. Sznajder Lu Zhang Pin Yao |
| author_facet | Jiana Wang Lin Feng Nana Meng Cong Yang Fanfan Cai Xin Huang Yihang Sun Kristin K. Sznajder Lu Zhang Pin Yao |
| author_sort | Jiana Wang |
| collection | DOAJ |
| description | Abstract During public health emergencies such as SARS, Ebola, and COVID-19, healthcare workers (HCWs) are often on the front lines, placing them at increased risk for adverse mental health outcomes, particularly depression and anxiety. Despite this risk, there remains a scarcity of research focused on developing predictive models to forecast the depression and anxiety levels of healthcare workers under challenging conditions. A total of 349 HCWs were selected from a Tertiary Grade-A hospital in the city of Shenyang, Liaoning Province in China. Depression and anxiety were assessed using the Patient Health Questionnaire (PHQ-9) and the Generalized Anxiety Disorder (GAD-7) scale, respectively. This study employed a random forest classifier (RFC) to predict depression and anxiety levels of HCWs from three perspectives: individual, interpersonal, and institutional with SHAP values to assess the contribution of factors. The Synthetic Minority Over-sampling Technique (SMOTE) was employed to address the issue of imbalanced data distribution. The prevalence of depression and anxiety among HCWs was 28.37% and 33.52%, respectively. The prediction model was developed using a training dataset (70%) and a test dataset (30%). The area under the curve (AUC) for depression and anxiety was 0.88 and 0.72, respectively. Additionally, the mean values of the 10-fold cross-validation results were 0.77 for the depression prediction model and 0.79 for the anxiety prediction model. For the depression prediction model, the top ten most significant predictive factors were: burnout, resilience, emotional labor, adaptability, working experience (< 1 year), being a physician, social support, average work time last week (9–11 h), age (28–30 years), age (31–35 years). For the anxiety prediction model, the top ten most significant predictive factors were: burnout, adaptability, emotional labor, age (31–35 years), average work time last week (9–11 h), resilience, being a physician, social support, working experience (< 1 year), and being female. It is essential to develop interventions that provide support both before and after a public health emergency, aiming at mitigating symptoms of depression and anxiety. The machine learning models in this study, using innovative SMOTE methodology to balance datasets with smaller sample sizes, identified key leverage points to prevent depression and anxiety among frontline HCWs, including mitigating burnout among HCWs, bolstering their resilience and adaptability, and ensuring reasonable work hours. |
| format | Article |
| id | doaj-art-e574c59f0eee496498701794615640d2 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-e574c59f0eee496498701794615640d22025-08-20T04:02:45ZengNature PortfolioScientific Reports2045-23222025-08-0115111010.1038/s41598-025-14403-3Balancing mental health through predictive modeling for healthcare workers during public health crisesJiana Wang0Lin Feng1Nana Meng2Cong Yang3Fanfan Cai4Xin Huang5Yihang Sun6Kristin K. Sznajder7Lu Zhang8Pin Yao9School of Public Health, Health Science Center, Ningbo UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Social Medicine, School of Public Health, China Medical UniversityDepartment of Public Health Sciences, Pennsylvania State University College of MedicineDepartment of Medical Humanities, School of Public Health, China Medical UniversityDepartment of Health Management, Shenyang Women’s and Children’s HospitalAbstract During public health emergencies such as SARS, Ebola, and COVID-19, healthcare workers (HCWs) are often on the front lines, placing them at increased risk for adverse mental health outcomes, particularly depression and anxiety. Despite this risk, there remains a scarcity of research focused on developing predictive models to forecast the depression and anxiety levels of healthcare workers under challenging conditions. A total of 349 HCWs were selected from a Tertiary Grade-A hospital in the city of Shenyang, Liaoning Province in China. Depression and anxiety were assessed using the Patient Health Questionnaire (PHQ-9) and the Generalized Anxiety Disorder (GAD-7) scale, respectively. This study employed a random forest classifier (RFC) to predict depression and anxiety levels of HCWs from three perspectives: individual, interpersonal, and institutional with SHAP values to assess the contribution of factors. The Synthetic Minority Over-sampling Technique (SMOTE) was employed to address the issue of imbalanced data distribution. The prevalence of depression and anxiety among HCWs was 28.37% and 33.52%, respectively. The prediction model was developed using a training dataset (70%) and a test dataset (30%). The area under the curve (AUC) for depression and anxiety was 0.88 and 0.72, respectively. Additionally, the mean values of the 10-fold cross-validation results were 0.77 for the depression prediction model and 0.79 for the anxiety prediction model. For the depression prediction model, the top ten most significant predictive factors were: burnout, resilience, emotional labor, adaptability, working experience (< 1 year), being a physician, social support, average work time last week (9–11 h), age (28–30 years), age (31–35 years). For the anxiety prediction model, the top ten most significant predictive factors were: burnout, adaptability, emotional labor, age (31–35 years), average work time last week (9–11 h), resilience, being a physician, social support, working experience (< 1 year), and being female. It is essential to develop interventions that provide support both before and after a public health emergency, aiming at mitigating symptoms of depression and anxiety. The machine learning models in this study, using innovative SMOTE methodology to balance datasets with smaller sample sizes, identified key leverage points to prevent depression and anxiety among frontline HCWs, including mitigating burnout among HCWs, bolstering their resilience and adaptability, and ensuring reasonable work hours.https://doi.org/10.1038/s41598-025-14403-3Mental healthHealthcare workersPredictive modelingPublic health crises |
| spellingShingle | Jiana Wang Lin Feng Nana Meng Cong Yang Fanfan Cai Xin Huang Yihang Sun Kristin K. Sznajder Lu Zhang Pin Yao Balancing mental health through predictive modeling for healthcare workers during public health crises Scientific Reports Mental health Healthcare workers Predictive modeling Public health crises |
| title | Balancing mental health through predictive modeling for healthcare workers during public health crises |
| title_full | Balancing mental health through predictive modeling for healthcare workers during public health crises |
| title_fullStr | Balancing mental health through predictive modeling for healthcare workers during public health crises |
| title_full_unstemmed | Balancing mental health through predictive modeling for healthcare workers during public health crises |
| title_short | Balancing mental health through predictive modeling for healthcare workers during public health crises |
| title_sort | balancing mental health through predictive modeling for healthcare workers during public health crises |
| topic | Mental health Healthcare workers Predictive modeling Public health crises |
| url | https://doi.org/10.1038/s41598-025-14403-3 |
| work_keys_str_mv | AT jianawang balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT linfeng balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT nanameng balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT congyang balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT fanfancai balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT xinhuang balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT yihangsun balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT kristinksznajder balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT luzhang balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises AT pinyao balancingmentalhealththroughpredictivemodelingforhealthcareworkersduringpublichealthcrises |