Objective To explore the risk factors of impaired physical function (IPF) in patients with rheumatoid arthritis (RA), and to construct a nomogram model for predicting the risk of IPF in RA patients. Methods A total of 628 hospitalized RA patients from 9 3A hospitals in Hubei province were selected by convenience sampling. The general information questionnaire, the Health Assessment Questionnaire Disability Index (HAQ-DI) and the Hospital Anxiety and Depression Scale (HADS) were used for cross-sectional survey. Logistic regression model was used to analyze the risk factors of IPF in the RA patients. R software was used to construct a nomogram model, which was later validated in another sample of 88 RA patients. Results A total of 508 RA patients had IPF. Multivariate logistic regression analysis showed that ethnicity, alcohol drinking, disease activity score, age, hospitalization times and anxiety score were risk factors of IPF. The area under the ROC curve of the nomogram model was 0.811; the calibration curve was close to the ideal curve. External validation results showed that the area under ROC curve was 0.862. Conclusion The risk prediction model of IPF of RA patients constructed has good discrimination and calibration, which can provide reference for clinical individualized prediction of IPF risk of RA patients, and is helpful for intervention of RA patients.