Abstract: Objective To understand the mental health status of the occupational population in Urumqi,construct a risk prediction column chart model,and provide reference for the adoption of targeted measures to promote the mental health of the occupational population. Methods The occupational population of Urumqi in Xinjiang was sampled from April to December 2021 using cluster sampling as the study population. The questionnaires were administered using the general information questionnaire,the effort-reward imbalance（ERI）,the Pittsburgh sleep quality index（PSQI）,and the symptom checklist-90（SCL-90）. The Pearson analysis was used to analyze the correlations between ERI and PSQI scale scores,and the univariate logistic regression and multivariate logistic regression were used to screen out the risk factors related to mental health in the occupational population. A column chart was established based on multivariate logistic regression results. The predictive ability was assessed using the area under the curve（AUC） of the subjects' work characteristics（ROC）,calibration curves and decision curve analysis（DCA）. A total of 2 000 questionnaires were distributed,and 1 863 questionnaires were validly recovered,with a response rate of 93.15%. Results Among 1 863 occupational population of Urumqi,790 were positive for mental health,with a prevalence rate of 42.4%. Except for the sleep efficiency dimension and the return dimension,the total scores and the remaining dimensions of ERI and PSQI showed positive correlations（r=0.125-0.350,all P<0.01）. Based on the univariate logistic regression and multivariate logistic regression showed that marital status（OR=2.455,95%CI=0.960-6.276）,engaged in occupation（OR=1.915,95%CI=1.096-3.346）,monthly income（OR=5.127,95%CI=2.211-11.887）,smoking（OR=1.497,95%CI=1.097-2.041）,work stress（OR=1.897,95%CI=1.538-2.340） and sleep quality problems（OR=4.119,95%CI=3.332-5.092） were the independent risk factors for mental health in the occupational population（all P<0.05）. The analysis of the ROC curve showed that the AUC was 0.737（95%CI=0.654-0.702）. The model was validated using the Bootstrap method with 1 000 self-samples and a corrected C-index index of 0.9. The results showed that this column chart model predicted that the incidence of mental health status in occupational populations was basically the same as the actual incidence,and it could be used for clinical prediction. The DCA decision curve showed that the model has good clinical applicability. Conclusions The mental health status of occupational groups is mainly affected by factors such as marital status,occupation,monthly income,smoking,work pressure,and sleep quality problems. The column chart model constructed in this study has high accuracy and discriminative power in predicting the mental health status of occupational groups.

Key words: Urumqi, Occupational population, Mental health, Column charts