Predictors of anxiety-depression comorbidity in Chinese medical staff: A Random Forest analysis based on the Health Ecology Model

Background Anxiety-depression comorbidity is a serious issue among medical staff, yet few studies have analyzed its influencing factors using a comprehensive approach. This study aims to explore the prevalence and key predictors of this comorbidity among Chinese medical staff based on the health ecology model. Methods A multi-center cross-sectional study was conducted with 1,468 medical staff from 6 hospitals in China. Self-measures including Generalized Anxiety Disorder 7-item (GAD-7), Patient Health Questionnaire-9 (PHQ-9), Self-rated Health Scale, Physical Exercise Questionnaire, Pittsburgh Sleeping Quality Index (PSQI), Night Shift frequency Questionnaire and the Chinese Version Perceived Stress Scale were used to evaluate participants' prevalence of anxiety-depression comorbidity and its influencing factors. The chi-squared test and binary logistic regression were used to analyze the influencing factors of anxiety-depression comorbidity, the random forest model (RFM) was used to determine the importance of the predictors. Results The prevalence of anxiety-depression comorbidity was 34.60%. The individual (self-rated health), behavioral (physical exercise, sleep disorder), and living and working (night shift frequency, perceived stress) factors were significantly associated with anxiety-depression comorbidity ( P  < 0.05). Based on the RFM, the top five predictors were higher perceived stress, sleep disorder, poor self-rated health, high night shift frequency and less physical exercise in sequence. Conclusions Anxiety-depression comorbidity is highly prevalent among Chinese medical staff. Perceived stress acts as the strongest predictor. Effective interventions should prioritize stress management, sleep improvement, reasonable shift scheduling, and physical exercise to mitigate this mental health burden.