What question did this study set out to answer?

Identify factors linked to densitometric osteoporosis in COPD patients and create a predictive nomogram.

March 26, 2026Open Access

Development of a Clinical Nomogram for Predicting Densitometric Osteoporosis in Patients with Chronic Obstructive Pulmonary Disease Based on NHANES Data

Key Points

Identify factors linked to densitometric osteoporosis in COPD patients and create a predictive nomogram.
Retrospective analysis using NHANES data for COPD patients.
Defined osteoporosis with a T-score of ≤ -2.5.
Used Least Absolute Shrinkage and Selection Operator and Boruta for feature selection.
Conducted ROC analysis to evaluate predictive models.
Included data from 1351 COPD patients.
Key factors for osteoporosis: age, weight, height, OP history of self-report, hemoglobin, HDL cholesterol.
Optimal predictive model identified with four variables: age, weight, height, OP history of self-report.
Nomogram showed good calibration and predictive performance with AUCs of 0.841 and 0.833 in train and test sets respectively.

Abstract

Objective: This study aimed to identify factors associated with densitometric osteoporosis (OP) in patients with Chronic Obstructive Pulmonary Disease (COPD) and develop a predictive model for OP. Methods: Retrospective analysis was conducted using data from COPD patients in the National Health and Nutrition Examination Survey database. OP was defined with a T-score of ≤ − 2.5 at the lumbar spine or femoral neck. Variables with more than 25% missing values were excluded, while others were imputed. Collinearity analysis was performed, and data were randomly divided into a 7:3 train-test set ratio. The Least Absolute Shrinkage and Selection Operator and Boruta algorithms were used for feature selection. The selected variables were evaluated with Receiver Operating Characteristic (ROC) analysis to identify the optimal predictive model, which was then used to construct a nomogram. The nomogram’s efficacy and stability were validated in the test set. Results: Data from 1351 COPD patients were included. Age, weight, height, OP history of self-report, hemoglobin, and high-density lipoprotein cholesterol were identified as significant factors for OP. Those variables were combined into 7 different models. The ROC analysis results revealed that among the seven models, the model containing four variables was the optimal model. The nomogram, based on four variables (age, weight, height, and OP history of self-report), demonstrated good calibration and predictive performance (areas under curve: 0.841 in the train set, 0.833 in the test set). The nomogram showed high clinical net benefit and stability in subgroup analysis. Conclusion: Age, weight, height, and OP history of self-report are significantly associated with densitometric OP in COPD patients. The constructed nomogram, based on these factors, provides an effective and stable tool for early identification of high-risk OP patients in clinical practice. Keywords: chronic obstructive pulmonary disease, osteoporosis, nomogram, Boruta, comorbidity

Bookmark

View Full Paper

Bookmark

View Full Paper

Development of a Clinical Nomogram for Predicting Densitometric Osteoporosis in Patients with Chronic Obstructive Pulmonary Disease Based on NHANES Data

Key Points

Abstract

Cite This Study