Development and validation of a nomogram for predicting the risk of mental health problems of factory workers and miners

OBJECTIVE: A nomogram for predicting the risk of mental health problems was established in a population of factory workers and miners, in order to quickly calculate the probability of a worker suffering from mental health problems. METHODS: A cross-sectional survey of 7500 factory workers and miners in Urumqi was conducted by means of an electronic questionnaire using cluster sampling method. Participants were randomly assigned to the training group (70%) and the validation group (30%). Questionnaire-based survey was conducted to collect information. A least absolute shrinkage and selection operator (LASSO) regression model was used to screen the predictors related to the risk of mental health problems of the training group. Multivariate logistic regression analysis was applied to construct the prediction model. Calibration plots and receiver operating characteristic-derived area under the curve (AUC) were used for model validation. Decision curve analysis was applied to calculate the net benefit of the screening model. RESULTS: A total of 7118 participants met the inclusion criteria and the data were randomly divided into a training group (n=4955) and a validation group (n=2163) in a ratio of 3:1. A total of 23 characteristics were included in this study and LASSO regression selected 12 characteristics such as education, professional title, age, Chinese Maslach Burnout Inventory, effort–reward imbalance, asbestos dust, hypertension, diabetes, working hours per day, working years, marital status and work schedule as predictors for the construction of the nomogram. In the validation group, the Brier score was 0.176, the calibration slope was 0.970 and the calibration curve of nomogram showed a good fit. The AUC of training group and verification group were 0.785 and 0.784, respectively. CONCLUSION: The nomogram combining these 12 characteristics can be used to predict the risk of suffering mental health problems, providing a useful tool for quickly and accurately screening the risk of mental health problems.