BackgroundDespite the increasing prevalence of poor sleep quality among medical students, only few studies have identified the factors associated with it sing methods from epidemiological surveys. Predicting poor sleep quality is critical for ensuring medical Students’ good physical and mental health. The aim of this study was to develop a comprehensive visual predictive nomogram for predicting the risk of poor sleep quality in medical students.MethodsWe investigated medical Students’ association with poor sleep quality at JiTang College of North China University of Science and Technology through a cross-sectional study. A total of 5,140 medical students were randomized into a training cohort (75%) and a validation cohort (25%). Univariate and multivariate logistic regression models were used to explore the factors associated with poor sleep quality. A nomogram was constructed to predict the individual risk of poor sleep quality among the medical students studied.Results31.9% of medical students in the study reported poor sleep quality. We performed multivariate logistic analysis and obtained the final model, which confirmed the risk and protective factors of poor sleep quality (p < 0.05). Protective factors included the absence of physical discomfort (OR = 0.638, 95% CI: 0.546–0.745). Risk factors included current drinking (OR = 0.638, 95% CI: 0.546∼0.745), heavy study stress (OR = 2.753, 95% CI: 1.456∼5.631), very heavy study stress (OR = 3.182, 95% CI: 1.606∼6.760), depressive symptoms (OR = 4.305, 95% CI: 3.581∼5.180), and anxiety symptoms (OR = 1.808, 95% CI: 1.497∼2.183). The area under the ROC curve for the training set is 0.776 and the area under the ROC curve for the validation set is 0.770, which indicates that our model has good stability and prediction accuracy. Decision curve analysis and calibration curves demonstrate the clinical usefulness of the predictive nomograms.ConclusionOur nomogram helps predict the risk of poor sleep quality among medical students. The nomogram used includes the five factors of drinking, study stress, recent physical discomfort, depressive symptoms, and anxiety symptoms. The model has good performance and can be used for further research on and the management of the sleep quality of medical students.