Establishment of a nomogram-based risk prediction model for chronic obstructive pulmonary disease in residents aged 40 years and over in Shandong province

  Objective  To establish a nomogram-based risk prediction model for chronic obstructive pulmonary disease (COPD) among individuals aged 40 years and over for improving early diagnosis and treatment of COPD in the population.

  Methods  The data on COPD surveillance in 2019 among 4 558 residents at ages of 40 years and above in Shandong province were collected. Significant impact factors of COPD were screened out with unconditional logistic regression analysis to establish a nomogram model for predicting COPD risk. Receiver operating characteristic (ROC) curve and calibration curve were used to test the fitting of the established model.

  Results  Logistic regression analysis revealed following significant impact factors of COPD: 9 risk factors consisting of at ages of 50 and above, disease history of chronic bronchitis, disease history of tuberculosis, parental disease history of chronic bronchitis, with wheezing while suffering from cold, history of recurrent wheezing, ever smoking, current smoking, frequent exposure to secondhand smoke before the age of 14, disease history of pneumonia or bronchitis before the age of 14 and 3 protective factors including female gender, being obesity, and having ventilation while cooking. Using the 12 influencing factors identified, a well-fitting nomogram model for COPD risk prediction was established, with the area under the ROC curve (AUC) of 0.786 (95% confidence interval: 0.766 – 0.806) and the concordance index (C-index) of 0.786. There was no significant difference between the model-predicted data and actual data (χ² = 0.40, P = 0.818).

  Conclusion  The established nomogram model is of good effect for predicting COPD risk in community residents aged 40 years and above and could be adopted in primary screening of the disease in the population.