LASSO regression-based nomogram for distinguishing nontuberculous mycobacterial pulmonary disease from pulmonary tuberculosis: a clinical risk prediction model.

This study aims to identify predictive factors that can effectively distinguish between non-tuberculous mycobacterial pulmonary disease (NTM-PD) and pulmonary tuberculosis (PTB) by comparing their clinical characteristics and laboratory indicators, and to construct a risk prediction model based on these factors to improve the accuracy of clinical diagnosis, optimize treatment strategies, and reduce the risks of misdiagnosis and drug resistance. Data from 150 hospitalized patients treated at The Second People's Hospital of Fuyang City between January 2021 and December 2022 were collected, including 50 cases with NTM-PD and 100 cases with PTB. By gathering clinical data and laboratory inflammation markers of the patients, key predictive factors were identified using LASSO regression. These factors were further analyzed through univariate and multivariate logistic regression analysis to determine the influencing factors of NTM-PD, thereby constructing a nomogram prediction model. The accuracy of the model was evaluated through calibration curves, ROC curves, and Hosmer-Lemeshow goodness-of-fit tests, while its clinical utility was assessed using decision curve analysis and clinical impact curves.The two groups showed significant differences in age, BMI, bronchiectasis, and lung cavitation (P < 0.05). LASSO regression analysis identified age, BMI, bronchiectasis, and lung cavitation as four key variables. Multivariate logistic regression analysis revealed that old age, bronchiectasis, and lung cavitation were risk factors for NTM-PD, while low BMI acted as a protective factor. The nomogram model constructed based on these four variables demonstrated excellent predictive performance. The Hosmer-Lemeshow goodness-of-fit test indicated good model fit (P = 0.448), with an area under the ROC curve (AUC) of 0.861 (95% CI: 0.798-0.923). Decision curve and clinical impact curve analyses suggested that the model has the potential to optimize clinical decision-making within a threshold probability range of 0.05 to 0.75. By analyzing the critical differential characteristics between NTM-PD and PTB, we successfully developed a prediction model based on age, BMI, bronchiectasis, and lung cavitation, which effectively assesses the risk of NTM-PD occurrence, providing clinicians with a practical diagnostic aid.