Machine learning driven-tracing the risk predictors for Chronic Obstructive Pulmonary Disease patients from a retrospective study

Machine learning driven-tracing the risk predictors for Chronic Obstructive Pulmonary Disease patients from a retrospective study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Machine learning driven-tracing the risk predictors for Chronic Obstructive Pulmonary Disease patients from a retrospective study Tianyi Yang, Qiang Li, Yuemu Li, Songhao Du, Baihua Jiang, Hanwei Zhao, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8574459/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract Background: Chronic Obstructive Pulmonary Disease (COPD) is a major respiratory disorder, and identifying its predictive factors is essential for early intervention to improve clinical outcomes. Methods: This study analyzed 2023 data from the National Health Interview Survey (NHIS), including 27,388 eligible participants (1,499 with COPD, 25,889 without). Data were split into a 70% training set and a 30% validation set. LASSO regression, univariate, and multivariate logistic regression were applied to screen predictors. A nomogram was constructed, and its performance was assessed using ROC curves, calibration curves, and Decision Curve Analysis (DCA). Results: Eleven independent predictors of COPD were identified in this study. Based on these predictors, a nomogram was constructed to assess the risk of COPD. Results showed good agreement between the predicted probabilities of COPD and the actual incidence rates, and DCA confirmed the clinical utility of this nomogram. Conclusion: The nomogram based on these 11 predictors can assist clinicians in estimating COPD risk for early intervention. However, it may not fully capture all complex biological processes and interactions related to COPD. Chronic Obstructive Pulmonary Disease a retrospective study Machine learning Risk predictors Risk populations Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Chronic Obstructive Pulmonary Disease (COPD) is a progressive and irreversible respiratory disease that ranks among the leading causes of morbidity and mortality worldwide, posing substantial challenges to public health systems and individual health [ 1 – 3 ]. The Global Burden of Disease Study 2019 estimated that in 2019, there were over 251 million COPD patients globally, with the disease causing approximately 3.23 million deaths annually—accounting for 5.6% of total global deaths[ 1 , 3 ]. The World Health Organization (WHO) further emphasizes that the public health threat of COPD is constantly escalating, noting that driven by population aging and continued tobacco use, COPD is projected to become one of the leading causes of death by 2030[ 1 ]. In the United States, COPD was the sixth leading cause of death in 2021, with approximately 14.2 million people living with the disease. Notably, the prevalence of COPD is significantly higher in rural areas and among low-income populations than in urban areas and among high-income groups, revealing significant health disparities[ 4 , 5 ]. COPD is characterized by persistent airflow limitation, chronic symptoms (e.g., dyspnea, sputum production), and impaired quality of life. It is also frequently accompanied by comorbidities, which exacerbate disease progression and increase the demand for healthcare services [ 6 , 7 ]. Studies have shown that over 97% of COPD patients have at least one comorbidity, and approximately 53% are diagnosed with four or more comorbidities [ 7 ]. Common comorbidities include ischemic heart disease, heart failure, osteoporosis, anemia, lung cancer, and depression. These comorbidities not only complicate COPD management and increase healthcare costs but also elevate patient mortality[ 7 – 10 ]。 Early identification of high-risk populations is crucial for mitigating the impact of COPD, as timely interventions—including smoking cessation, pulmonary rehabilitation, and targeted pharmacotherapy in accordance with the Global Initiative for Chronic Obstructive Lung Disease guidelines—have been proven to reduce COPD-related mortality[ 10 ]. Specifically, smoking cessation can slow the annual decline in lung function among COPD patients from 42–82 mL/year to 0–49 mL/year[ 11 ]. For patients with recent acute exacerbations, pulmonary rehabilitation (including aerobic exercise and breathing training) can reduce readmission rates and mortality while significantly improving quality of life[ 10 , 12 ]. However, the implementation of these interventions requires a prior diagnosis of COPD. Unfortunately, existing diagnostic tests have limited sensitivity for early-stage patients with no obvious symptoms; many individuals are only diagnosed with COPD after disease progression, resulting in missed opportunities for early intervention[ 13 , 14 ]. Therefore, the development of new diagnostic methods and early risk assessment tools is essential for improving patient prognosis and alleviating the overall disease burden. Based on 2023 National Health Interview Survey (NHIS) data, this study screened predictors using The Least Absolute Shrinkage and Selection Operator (LASSO) regression and univariate/multivariate logistic regression, and constructed and validated a nomogram. The aim of this study is to provide a reliable tool for the clinical early identification of high-risk populations for COPD. The flow chart of this study is shown in Fig. 1 . 2. Material and methods 2.1 Research Participants and Research Design All data and materials are publicly available from the Centers for Disease Control and Prevention ( https://www.cdc.gov/nchs/nhis/documentation/index.html ) and can be accessed online. The NHIS study was approved by the Disclosure Review Board of the National Center for Health Statistics, and all participants provided written informed consent. This study analyzed all subjects in 2023. The data exclusion criteria were as follows: 1. Subjects with missing covariate information were excluded; 2. Subjects with unclear COPD diagnosis were excluded. A total of 27,388 subjects were finally recruited. 2.2 Variable Definition The definition of COPD was determined based on participants' responses to the variable corresponding to Question ID (CON.0010.00.1): "Have you EVER been told by a doctor or other health professional that you had Chronic Obstructive Pulmonary Disease, C.O.P.D.?" Participants were categorized into two groups: the COPD group (n = 1,499) and the non-COPD group (n = 25,889). 2.3 Covariates Covariates of COPD were categorized into four major groups: sociodemographic factors, health status, comorbid conditions, and healthcare data, as shown in Table 1 . Table 1 Table of Variables and Their Definitions Variables Variable Definitions Age (AGE_P) 18–30, 31–45, 46–65, > 65 Sex (SEX) Male, Female Race (HISDETP_A) Hispanic (Mexican/Mexican American), Hispanic (all other groups), Not Hispanic Educational Attainment (EDUCP_A) Below high school education, High School Graduate, Above high school education Marital Status (R_MARITL) Married, Unmarried Cohabitation, Neither of the above Economic Status (RATCAT_A) poor, Relatively poor, Not poor Smoking Status (SMKCIGST_A) Smokes daily, Occasionally smokes, Previously smoked, Never smoked Region (REGION) Northeast, Midwest, South, West Body Mass Index (BMICAT_A) Underweight, Healthy weight, Overweight, Obesity High Cholesterol Status (CHLEV_A) Yes, No Diabetes Mellitus (PREDIB_A) Yes, No Hypertension Status (HYPEV_A) Yes, No 2.4 Statistical Analysis A total of 27,388 eligible participants were included in this study based on predefined inclusion and exclusion criteria. The data were randomly split into a training set and a validation set in a 7:3 ratio. All statistical analyses were performed using R software. Normally distributed data are presented as mean ± standard deviation (mean ± SD), non-normally distributed data as median (interquartile range, IQR), and categorical variables as counts (percentages). For statistical comparisons, the t-test was used for normally distributed continuous variables, the Wilcoxon rank-sum test for non-normally distributed continuous variables, and the chi-square test for categorical variables. LASSO regression model was applied to the training set data. A 10-fold cross-validation was performed to determine the optimal lambda (λ) value for selecting risk factors for COPD. The selected variables were then incorporated into univariate and multivariate logistic regression analyses for further variable refinement. Risk factors with a P-value < 0.05 were included in the nomogram model. After constructing the nomogram, the Receiver Operating Characteristic (ROC) curve and calibration curve were used to evaluate its predictive performance, and Decision Curve Analysis (DCA) was employed to assess its clinical utility. 3. Result 3.1 Baseline Characteristics of Study Participants A total of 27,388 eligible participants were included in this study based on predefined inclusion and exclusion criteria. To assess whether there were significant differences in 12 baseline characteristics (covariates, see Table 1 ) between participants with and without COPD, the included participants were categorized into two groups: the COPD group and the non-COPD group. The baseline table was generated using the tableone R package [ 15 ]. Differences in various baseline characteristics between the two groups were analyzed using the t-test or chi-square test, with a significance level set at P < 0.05 (Table 2 ). The results indicated that all 12 characteristics were significantly associated with the grouping (COPD vs. non-COPD). Table 2 Analysis of Participants' Baseline Characteristics Level NO COPD COPD p-value n 25889 1499 AGEP_A (%) 18–30 1536 (5.9) 6 (0.4) 65 8753 (33.8) 976 (65.1) SEX_A (%) Male 12033 (46.5) 620 (41.4) < 0.001 Female 13856 (53.5) 879 (58.6) HISDETP_A (%) Hispanic (Mexican/Mexican American) 2187 (8.4) 38 (2.5) < 0.001 Hispanic (all other groups) 1737 (6.7) 45 (3.0) Not Hispanic 21965 (84.8) 1416 (94.5) EDUCP_A (%) Below high school education 2053 (7.9) 239 (15.9) < 0.001 High School Graduate 6421 (24.8) 504 (33.6) Above high school education 17415 (67.3) 756 (50.4) MARITAL_A (%) Married 12080 (46.7) 499 (33.3) < 0.001 Unmarried Cohabitation 1760 (6.8) 70 (4.7) Neither of the above 12049 (46.5) 930 (62.0) RATCAT_A (%) poor 2471 (9.5) 298 (19.9) < 0.001 Relatively poor 4526 (17.5) 451 (30.1) Not poor 18892 (73.0) 750 (50.0) SMKCIGST_A (%) Smokes daily 1859 (7.2) 387 (25.8) < 0.001 Occasionally smokes 686 (2.6) 65 (4.3) Previously smoked 6304 (24.4) 696 (46.4) Never smoked 17040 (65.8) 351 (23.4) REGION (%) Northeast 3983 (15.4) 221 (14.7) < 0.001 Midwest 5685 (22.0) 366 (24.4) South 9570 (37.0) 605 (40.4) West 6651 (25.7) 307 (20.5) CHLEV_A (%) Yes 8151 (31.5) 855 (57.0) < 0.001 No 17738 (68.5) 644 (43.0) BMICAT_A (%) Underweight 370 (1.4) 55 (3.7) < 0.001 Healthy weight 8001 (30.9) 369 (24.6) Overweight 8977 (34.7) 453 (30.2) Obesity 8541 (33.0) 622 (41.5) PREDIB_A (%) Yes 4387 (16.9) 488 (32.6) < 0.001 No 21502 (83.1) 1011 (67.4) HYPEV_A (%) Yes 9373 (36.2) 976 (65.1) < 0.001 No 16516 (63.8) 523 (34.9) 3.2 Screening of Risk Factors for COPD Symptoms A total of 27,388 participants were randomly divided into a training set (n = 19,171) and a validation set (n = 8,217) at a ratio of 7:3 for subsequent analyses. To screen for risk factors associated with COPD, the LASSO regression model was applied to all samples in the training set. Based on the aforementioned covariates, the LASSO regression model was constructed using the glmnet R package[ 16 ], with 10-fold cross-validation performed. Using the optimal lambda value of 0.00007, 24 dummy variables (representing 12 candidate risk factors) were identified (Fig. 2 ). Results of the LASSO regression indicated that all 12 candidate risk factors were important features, which were thus designated as candidate features. In Fig. 2 , LASSO coefficient path plot (top): This plot illustrates the changes in feature coefficients under different values of the regularization parameter (λ). The x-axis represents the values of the regularization parameter, while the y-axis denotes the absolute values or scaled values of the feature coefficients. LASSO regularization path plot (bottom): This plot demonstrates the model fitting performance corresponding to different values of the regularization parameter (λ) in the LASSO algorithm. The x-axis indicates the values of the regularization parameter, and the y-axis represents the model performance metrics. To further identify COPD-related risk factors, univariate logistic regression analysis was applied to all samples in the training set to screen the candidate features derived from the LASSO regression. Odds ratios (OR) and 95% confidence intervals (95% CI) were calculated, and a forest plot was generated to display the 12 candidate features (Fig. 3 , P < 0.05 ). Results of the univariate logistic regression showed that never smoking (SMKCIGST_A4) significantly reduced the likelihood of COPD, whereas increasing age significantly increased the likelihood of COPD. In Fig. 3 , The x-axis represents the value of OR (Odds Ratio). The interpretation of OR values is as follows: 1) When OR = 1: The predictor variable has no effect on the probability of event occurrence (neutral effect); 2) When OR > 1: The predictor variable is positively associated with the event occurrence; specifically, an increase in or the presence of the variable increases the likelihood of the event happening; 3) When OR < 1: The predictor variable is negatively associated with the event occurrence; specifically, an increase in or the presence of the variable decreases the likelihood of the event happening. Multivariate logistic regression analysis was then used to screen the 12 candidate features derived from the univariate logistic regression analysis, with OR and 95% CI calculated. This yielded 11 predictors associated with COPD occurrence (excluding REGION [region]): AGE_P (age), SEX (gender), race (HISDETP_A), education level (EDUCP_A), HYPEV_A (presence of hypertension), marital status (R_MARITL), BMI (BMICAT_A), SMKSTAT2 (smoking status), presence of high cholesterol (CHLEV_A), presence of diabetes (PREDIB_A), and RATCAT (economic status) ( Figure 4 , P<0.05 ). 3.3 Nomogram Construction To further evaluate the overall predictive ability of the combined predictors, a nomogram was constructed using the rms R package based on the predictors identified from the LASSO regression model and univariate/multivariate logistic regression analyses, with all samples in the training set included ( Figure 5 ). Each predictor was assigned a corresponding score (Point) via the nomogram. The sum of scores for all risk factors yields the Total Point, which enables the estimation of COPD incidence. A higher total score indicates a greater likelihood of developing COPD. In Figure 5 , predictors are displayed on the left side. On the right side, scales are marked on the line corresponding to each predictor, representing the range of possible values for that predictor. The length of the line segment reflects the magnitude of the predictor's contribution to the disease. The β(X−m) terms in the plot denote the individual scores corresponding to each predictor at different values. "Total score" represents the cumulative score obtained by summing the individual scores corresponding to the values of all predictors. "Pr" denotes the risk of developing the disease. 3.4 Evaluation of the Nomogram To assess the predictive performance of the nomogram, its validity and discriminative ability were evaluated using various metrics in both the training and validation sets. Calibration curves for the training and validation sets were generated using the calibrate R package[17] to evaluate the predictive accuracy of the nomogram ( Figure 6 ). A calibration curve closer to the diagonal line indicates better predictive accuracy of the model. The high degree of overlap between the predicted probabilities of the nomogram and the reference line demonstrated that the nomogram had high predictive accuracy. In Figure 6 , the x-axis represents the predicted event rate, and the y-axis represents the observed actual event rate, both ranging from 0 to 1 (which can be interpreted as event rates, expressed as percentages). The dashed diagonal line serves as the reference line, representing the scenario where predicted values are equal to actual values. The "Ideal" line denotes the scenario where the model’s predictions are perfectly consistent with actual outcomes under ideal conditions. The "Bias-corrected" line represents the performance of the model trained after repeated bootstrap sampling of the samples. The "Apparent" line represents the performance of the dataset. Decision curves for the training and validation sets, generated using the rmda R package, showed that the net benefit of the nomogram was higher than both the "all" line and "none" line, indicating the model has practical value ( Figure 7 ). Additionally, ROC curves were plotted using the pROC R package[18]. The results revealed that the nomogram model exhibited excellent predictive performance in both the training set (the area under the curve, AUC = 0.84) and the validation set (AUC = 0.84) ( Figure 8 ). In Figure 7 , the x-axis represents the threshold probability (Pt). In a risk assessment tool, the probability that Patient i is diagnosed with the disease is denoted as Pi. When Pi reaches a certain threshold (denoted as Pt), the patient is defined as positive, and treatment is administered. At this point, there are benefits of treating patients (considered as "benefits"), as well as harms of treating non-patients and losses resulting from failing to treat patients (considered as "harms"). The y-axis represents the net benefit (NB), calculated as benefits minus harms. Different curved lines in the plot represent different clinical diagnostic models (see the legend for identification). Among these lines, two represent extreme scenarios: 1) The "None" curve indicates that all samples are classified as negative (Pi < Pt), meaning no one receives treatment, and the net benefit is 0; 2)The "All" curve indicates that all samples are classified as positive, meaning everyone receives treatment, and the net benefit is represented by a line with a negative slope. In Figure 8 , the x-axis represents the false positive rate (1 - specificity), and the y-axis represents the true positive rate (sensitivity). Each point on the curve corresponds to the sensitivity and specificity of a screening test under a specific positive cutoff value.The AUC reflects the value of the diagnostic test: the larger the AUC (i.e., the closer it is to 1.0), the higher the diagnostic accuracy; conversely, the closer the AUC is to 0.5, the lower the diagnostic accuracy. 4. Discussion Smoking status was the dominant predictive factor in this study, which aligns with evidence from decades of epidemiological and mechanistic research. Such studies have confirmed that tobacco smoke is the single most significant modifiable cause of COPD[ 10 , 19 – 21 ]. Mechanistically, cigarette smoke induces irreversible damage to airway epithelial cells, activates oxidative stress pathways, and triggers chronic inflammation via pro-inflammatory cytokines—all of which are core processes in the pathogenesis of COPD[ 22 – 25 ]. Our findings further underscore the critical role of smoking cessation interventions, which remain one of the core components of the 2023 Global Initiative for Chronic Obstructive Lung Disease guidelines [ 10 ]. Notably, research indicates that even smoking cessation in the advanced stages of COPD can reduce the rate of acute disease exacerbations[ 26 – 29 ], a conclusion that further validates the practical value of this nomogram in identifying smoking populations for targeted interventions. Age (AGE_P) emerged as a strong independent predictive factor for COPD, reflecting the cumulative impact of aging on lung structure and function. Studies have shown that the prevalence of COPD is common among individuals aged 40 years and older and increases with advancing age[ 30 , 31 ]. This is consistent with our analytical results, highlighting the value of this nomogram in prioritizing screening for middle-aged and elderly populations. Implementing early detection in this group can slow the rapid progression of the disease. In this study, hypertension (HYPEV_A), diabetes mellitus (PREDIB_A), and hypercholesterolemia (CHLEV_A) were identified as independent predictive factors, supporting the notion that comorbidity status in COPD is a key driver of adverse outcomes[ 32 , 33 ]. Furthermore, chronic airway inflammation in COPD can spread to the systemic circulation, leading to endothelial dysfunction, insulin resistance, and dyslipidemia. These abnormalities not only exacerbate COPD but also worsen comorbid conditions [ 34 – 38 ], further underscoring the clinical relevance of this nomogram. The study also confirms that socioeconomic factors, including educational level (EDUCP_A), economic status (RATCAT), and marital status (R_MARITL), are key predictive factors for COPD. Previous research has demonstrated that lower educational levels are associated with insufficient health literacy, which contributes to delayed diagnosis of COPD and poor adherence to inhaled therapy [ 39 , 40 ]. Additionally, individuals with lower incomes face a higher risk of developing COPD[ 5 , 40 – 42 ], a finding corroborated by our analysis. Marital status has received relatively little attention as a predictive factor in COPD research; however, this study found an association between marital status and COPD risk, which may be linked to social support. Spousal support, for instance, can improve medication adherence[ 43 – 45 ]. Conversely, single or divorced participants experience higher levels of psychological stress—a known trigger for acute COPD exacerbations. This emphasizes the need to address social isolation in COPD risk assessment. Body Mass Index (BMI, BMICAT_A) was significantly associated with COPD, consistent with previous studies [ 46 – 48 ]. Moreover, the observed gender (SEX)-related differences in COPD risk align with a growing body of evidence regarding gender-specific susceptibility to lung injury[ 49 – 51 ]. Additionally, race (HISDETP_A) was included as a predictive factor in our analysis. This study offers several methodological and clinical strengths: 1) It utilized data from the 2023 NHIS, which features a large, representative sample—minimizing selection bias to the greatest extent possible; 2) The statistical methods employed were rigorous; 3) Comprehensive validation was conducted; 4) The predictive factors are easily accessible. However, the study also has limitations: 1) The NHIS is a cross-sectional survey, precluding causal inference; 2) COPD diagnosis was based on self-report, which may lead to underdiagnosis of early asymptomatic COPD or overdiagnosis due to recall bias; 3) Confounding factors such as air pollution, occupational exposure, and genetic factors were not measured; 4) Participants with missing covariate data were excluded, which may have introduced selection bias. 5. Conclusion The aim of this study was to identify predictive factors for the development of COPD. Through baseline analysis, data were divided into training and validation sets, followed by the identification of predictive factors for COPD development using LASSO regression, univariate, and multivariate logistic regression. A nomogram model was constructed to predict the risk of disease occurrence, and its predictive value was validated in both the training and validation sets using calibration curves, DCA curves, and ROC curves. In summary, multiple statistical methods were used to analyze the impact of various covariates on the disease, ultimately confirming that AGE_P (age), SEX (gender), race (HISDETP_A), education level (EDUCP_A), HYPEV_A (presence of hypertension), marital status (R_MARITL), BMI (BMICAT_A), SMKSTAT2 (smoking status), presence of high cholesterol (CHLEV_A), presence of diabetes (PREDIB_A), and RATCAT (economic status) can serve as predictors of COPD development. A nomogram was constructed based on these 11 predictors to predict the risk of disease occurrence. The predictive model we developed holds promise as a potential tool to guide clinicians in predicting the probability of COPD development, which may facilitate the implementation of early intervention measures and improve clinical outcomes. However, biological systems are highly dynamic and complex, and this model may not fully capture all relevant biological processes and interactions. Declarations 6. Ethics approval and consent to participate This study analyzed publicly available data from Centers for Disease Control and Prevention (https://www.cdc.gov/nchs/nhis/documentation/index.html). As this study involved the secondary analysis of de-identified public data, further ethical approval and consent to participate were not required. 7. Consent for publication Not applicable. 8. Availability of data and materials The datasets generated and/or analyzed during the current study are available in the Centers for Disease Control and Prevention (https://www.cdc.gov/nchs/nhis/documentation/index.html). 9.Competing interests The authors declare that they have no competing interests. 10.Funding The authors declare that no funding was received for this study. References Wang, H., et al., Global, regional, and national burden of chronic obstructive pulmonary disease from 1990 to 2019. Front Physiol, 2022. 13 : p. 925132. Boers, E., et al., Global Burden of Chronic Obstructive Pulmonary Disease Through 2050. JAMA Netw Open, 2023. 6 (12): p. e2346598. Li, H.Y., et al., Global, regional and national burden of chronic obstructive pulmonary disease over a 30-year period: Estimates from the 1990 to 2019 Global Burden of Disease Study. Respirology, 2023. 28 (1): p. 29-36. Liu, Y., et al., Trends in the Prevalence of Chronic Obstructive Pulmonary Disease Among Adults Aged ≥18 Years - United States, 2011-2021. MMWR Morb Mortal Wkly Rep, 2023. 72 (46): p. 1250-1256. Raju, S., et al., Rural Residence and Poverty Are Independent Risk Factors for Chronic Obstructive Pulmonary Disease in the United States. Am J Respir Crit Care Med, 2019. 199 (8): p. 961-969. Negewo, N.A., P.G. Gibson, and V.M. McDonald, COPD and its comorbidities: Impact, measurement and mechanisms. Respirology, 2015. 20 (8): p. 1160-71. Smith, M.C. and J.P. Wrobel, Epidemiology and clinical impact of major comorbidities in patients with COPD. Int J Chron Obstruct Pulmon Dis, 2014. 9 : p. 871-88. Divo, M. and B.R. Celli, Multimorbidity in Patients with Chronic Obstructive Pulmonary Disease. Clin Chest Med, 2020. 41 (3): p. 405-419. Bazell, C., et al., Variation in Prevalence and Burden of Chronic Obstructive Pulmonary Disease by State and Insurance Type in the United States. Chronic Obstr Pulm Dis, 2025. 12 (2): p. 158-174. Agustí, A., et al., Global Initiative for Chronic Obstructive Lung Disease 2023 Report: GOLD Executive Summary. Eur Respir J, 2023. 61 (4). van Eerd, E.A., et al., Smoking cessation for people with chronic obstructive pulmonary disease. Cochrane Database Syst Rev, 2016. 2016 (8): p. Cd010744. Greening, N.J., et al., An early rehabilitation intervention to enhance recovery during hospital admission for an exacerbation of chronic respiratory disease: randomised controlled trial. Bmj, 2014. 349 : p. g4315. Stolz, D., et al., Towards the elimination of chronic obstructive pulmonary disease: a Lancet Commission. Lancet, 2022. 400 (10356): p. 921-972. Wang, L., et al., Construction and validation of a risk prediction model for chronic obstructive pulmonary disease (COPD): a cross-sectional study based on the NHANES database from 2009 to 2018. BMC Pulm Med, 2025. 25 (1): p. 317. Panos, A. and D. Mavridis, TableOne: an online web application and R package for summarising and visualising data. Evid Based Ment Health, 2020. 23 (3): p. 127-130. Friedman, J., T. Hastie, and R. Tibshirani, Regularization Paths for Generalized Linear Models via Coordinate Descent. J Stat Softw, 2010. 33 (1): p. 1-22. Graffelman, J. and F. van Eeuwijk, Calibration of multivariate scatter plots for exploratory analysis of relations within and between sets of variables in genomic research. Biom J, 2005. 47 (6): p. 863-79. Robin, X., et al., pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics, 2011. 12 : p. 77. Gan, H., et al., Smoking: a leading factor for the death of chronic respiratory diseases derived from Global Burden of Disease Study 2019. BMC Pulm Med, 2022. 22 (1): p. 149. Regan, E.A., et al., Genetic epidemiology of COPD (COPDGene) study design. Copd, 2010. 7 (1): p. 32-43. Muro, S., [Cigarette smoking is the most important causal factor for developing chronic obstructive pulmonary disease (COPD)]. Nihon Rinsho, 2011. 69 (10): p. 1735-40. Schumacher, F.R., et al., RNAi Screen for NRF2 Inducers Identifies Targets That Rescue Primary Lung Epithelial Cells from Cigarette Smoke Induced Radical Stress. PLoS One, 2016. 11 (11): p. e0166352. Zhang, Y., et al., Nrf2/HO-1 signaling activation alleviates cigarette smoke-induced inflammation in chronic obstructive pulmonary disease by suppressing NLRP3-mediated pyroptosis. J Cardiothorac Surg, 2024. 19 (1): p. 58. Pera, T., et al., Role for TAK1 in cigarette smoke-induced proinflammatory signaling and IL-8 release by human airway smooth muscle cells. Am J Physiol Lung Cell Mol Physiol, 2012. 303 (3): p. L272-8. Heijink, I.H., et al., Cigarette smoke impairs airway epithelial barrier function and cell-cell contact recovery. Eur Respir J, 2012. 39 (2): p. 419-28. Godtfredsen, N.S., et al., Risk of hospital admission for COPD following smoking cessation and reduction: a Danish population study. Thorax, 2002. 57 (11): p. 967-72. Wu, J. and D.D. Sin, Improved patient outcome with smoking cessation: when is it too late? Int J Chron Obstruct Pulmon Dis, 2011. 6 : p. 259-67. Bai, J.W., et al., Smoking cessation affects the natural history of COPD. Int J Chron Obstruct Pulmon Dis, 2017. 12 : p. 3323-3328. Tønnesen, P., Smoking cessation and COPD. Eur Respir Rev, 2013. 22 (127): p. 37-43. Toelle, B.G., et al., Respiratory symptoms and illness in older Australians: the Burden of Obstructive Lung Disease (BOLD) study. Med J Aust, 2013. 198 (3): p. 144-8. Wang, Y., et al., Chronic obstructive pulmonary disease across three decades: trends, inequalities, and projections from the Global Burden of Disease Study 2021. Front Med (Lausanne), 2025. 12 : p. 1564878. Yang, H.Y., et al., Increased Risk of Chronic Obstructive Pulmonary Disease in Patients with Hyperlipidemia: A Nationwide Population-Based Cohort Study. Int J Environ Res Public Health, 2022. 19 (19). Chen, W., et al., Risk of cardiovascular comorbidity in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis. Lancet Respir Med, 2015. 3 (8): p. 631-9. Marcuccio, G., et al., Endothelial dysfunction in chronic obstructive pulmonary disease: an update on mechanisms, assessment tools and treatment strategies. Front Med (Lausanne), 2025. 12 : p. 1550716. Bolton, C.E., et al., Insulin resistance and inflammation - A further systemic complication of COPD. Copd, 2007. 4 (2): p. 121-6. Anghel, L., et al., Chronic Obstructive Pulmonary Disease and Type 2 Diabetes Mellitus: Complex Interactions and Clinical Implications. J Clin Med, 2025. 14 (6). Agusti, A. and J.B. Soriano, COPD as a systemic disease. Copd, 2008. 5 (2): p. 133-8. Moermans, C., et al., Local and systemic cellular inflammation and cytokine release in chronic obstructive pulmonary disease. Cytokine, 2011. 56 (2): p. 298-304. O'Conor, R., et al., Effects of health literacy and cognitive abilities on COPD self-management behaviors: A prospective cohort study. Respir Med, 2019. 160 : p. 105630. Eisner, M.D., et al., Socioeconomic status, race and COPD health outcomes. J Epidemiol Community Health, 2011. 65 (1): p. 26-34. Lee, Y.S., et al., The association between living below the relative poverty line and the prevalence of chronic obstructive pulmonary disease. J Thorac Dis, 2019. 11 (2): p. 427-437. Atassi, M., et al., Association between chronic airflow obstruction and socio-economic position in Morocco: BOLD results. Int J Tuberc Lung Dis, 2020. 24 (2): p. 202-206. Giannico, O.V., et al., Educational level, marital status and sex as social gender discharge determinants in chronic obstructive pulmonary disease exacerbations: a time-to-event analysis. Monaldi Arch Chest Dis, 2019. 89 (2). Trivedi, R.B., et al., The influence of informal caregivers on adherence in COPD patients. Ann Behav Med, 2012. 44 (1): p. 66-72. Liu, Y.R., et al., Path analysis of illness perception, medication beliefs, family support on inhaler adherence in elderly COPD patients: Based on triadic reciprocal determinism. Patient Educ Couns, 2025. 130 : p. 108465. Sun, Y., et al., BMI is associated with FEV(1) decline in chronic obstructive pulmonary disease: a meta-analysis of clinical trials. Respir Res, 2019. 20 (1): p. 236. Zhou, Y., et al., The association between BMI and COPD: the results of two population-based studies in Guangzhou, China. Copd, 2013. 10 (5): p. 567-72. Wang, M., X. Ni, and F. Yu, Impact of Body Mass Index on Risk of Exacerbation in Patients With COPD: A Systematic Review and Meta-Analysis. Chronic Obstr Pulm Dis, 2024. 11 (5): p. 524-533. Çolak, Y., et al., Sex differences in COPD in relation to smoking exposure: a population-based cohort study. Thorax, 2025. 80 (8): p. 512-519. Raghavan, D., A. Varkey, and T. Bartter, Chronic obstructive pulmonary disease: the impact of gender. Curr Opin Pulm Med, 2017. 23 (2): p. 117-123. Forsslund, H., et al., Gender differences in the T-cell profiles of the airways in COPD patients associated with clinical phenotypes. Int J Chron Obstruct Pulmon Dis, 2017. 12 : p. 35-48. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 08 Apr, 2026 Editor invited by journal 15 Mar, 2026 Editor assigned by journal 16 Jan, 2026 Submission checks completed at journal 16 Jan, 2026 First submitted to journal 11 Jan, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8574459","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":622038391,"identity":"9d376a71-3789-46f7-815e-5cf3c87570df","order_by":0,"name":"Tianyi Yang","email":"","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Tianyi","middleName":"","lastName":"Yang","suffix":""},{"id":622038392,"identity":"e41fa9bb-10d3-4b81-8dc6-585557ffbd06","order_by":1,"name":"Qiang Li","email":"","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Qiang","middleName":"","lastName":"Li","suffix":""},{"id":622038393,"identity":"88cfe941-8402-463b-bbaf-4a3a74bc718a","order_by":2,"name":"Yuemu Li","email":"","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yuemu","middleName":"","lastName":"Li","suffix":""},{"id":622038394,"identity":"e7e6b135-0ec5-40d6-aeab-f6d5e72cb9c2","order_by":3,"name":"Songhao Du","email":"","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Songhao","middleName":"","lastName":"Du","suffix":""},{"id":622038395,"identity":"8c5b220c-717d-44dc-9d56-e15488060cee","order_by":4,"name":"Baihua Jiang","email":"","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Baihua","middleName":"","lastName":"Jiang","suffix":""},{"id":622038396,"identity":"f98af07e-b679-4820-bec3-553ad6e44fbe","order_by":5,"name":"Hanwei Zhao","email":"","orcid":"","institution":"The First Affiliated Hospital of Heilongjiang University of Chinese Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hanwei","middleName":"","lastName":"Zhao","suffix":""},{"id":622038397,"identity":"44481c85-b026-43e9-b2d1-d20edb22d193","order_by":6,"name":"Guannan Jin","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAz0lEQVRIiWNgGAWjYBACxmYEs/FBQkUNSVqYmw0enDlGkoXsbZIPW5gJq2Nu5z38mqfmTh7/7Ma2isQGNgb+9u4EAg7jS7PmOfasWOLOwbYbiTtkGCTOnN1AQAuPmTEP2+HEhhuJQC1n2BgMJHKJ0fLvcOJ8oJaCxDZmorQYP+ZtO5y4AaiFgVgtZoxz+w4nbryR2CyRcOYYD0G/GPafMf7w5tvhxHk30h9+/FFRI8ff3ktASwMDmwSyAA9e5SAgD4yaDwRVjYJRMApGwcgGAKI1TQl7pH39AAAAAElFTkSuQmCC","orcid":"","institution":"Heilongjiang Academy Of Sciences Of Traditional Chinese Medicine","correspondingAuthor":true,"prefix":"","firstName":"Guannan","middleName":"","lastName":"Jin","suffix":""}],"badges":[],"createdAt":"2026-01-11 15:23:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8574459/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8574459/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107242054,"identity":"2087f1f9-31e2-458f-8078-5272f35d55c0","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":74227,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe flow chart of this stud\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/ce31ddacc1fac267051f1e02.png"},{"id":107242055,"identity":"6f8da11c-536e-4094-8c8b-fd11932a64cf","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":144120,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePlots of the LASSO Algorithm.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/8335b81de623f5ec3e38be19.png"},{"id":107242059,"identity":"00f069a9-53c1-4e13-a48f-9be0050b4c47","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":90561,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest Plot of Univariate Logistic Regression.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/37fea7b530a2377964452994.png"},{"id":107242057,"identity":"2d0aa0e0-b3f4-4283-938e-11cf23f58f05","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":99080,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eForest Plot of Multivariate Logistic Regression.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/aae4e2fb44c36dfb7248e1b3.png"},{"id":107705287,"identity":"27f67884-86ec-4294-927c-e9ccdfee4af6","added_by":"auto","created_at":"2026-04-24 09:10:46","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":131181,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eNomogram of the Training Set.\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/7ced704b4e63028b7a9a55ac.png"},{"id":107704931,"identity":"18af9773-a020-4686-8489-043a54deed64","added_by":"auto","created_at":"2026-04-24 09:04:17","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":93022,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCalibration Curves of the Nomogram for the Training Set (Left) and Validation Set (Right).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/5e965543f47494d17a9a3f33.png"},{"id":107242061,"identity":"20f50915-9080-4bfe-96cd-87eab24e43dd","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":48671,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDecision Curves for the Training Set (Left) and Validation Set (Right).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/74c8c12b72e37d3c0a25ae54.png"},{"id":107242058,"identity":"3ca39ec7-1a95-484b-a3e8-c3c683f44408","added_by":"auto","created_at":"2026-04-19 07:19:27","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":74062,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eROC Curves of the Nomogram for the Training Set (Left) and Validation Set (Right).\u003c/strong\u003e\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/bb277aac4a7d2ccf53a1b80e.png"},{"id":107712655,"identity":"184daeb6-09c9-4219-9551-c71767e5f4d0","added_by":"auto","created_at":"2026-04-24 09:49:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1034081,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8574459/v1/4ad8ffbc-7c20-4c9f-8183-04f73752cc7e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine learning driven-tracing the risk predictors for Chronic Obstructive Pulmonary Disease patients from a retrospective study","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eChronic Obstructive Pulmonary Disease (COPD) is a progressive and irreversible respiratory disease that ranks among the leading causes of morbidity and mortality worldwide, posing substantial challenges to public health systems and individual health [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The Global Burden of Disease Study 2019 estimated that in 2019, there were over 251\u0026nbsp;million COPD patients globally, with the disease causing approximately 3.23\u0026nbsp;million deaths annually\u0026mdash;accounting for 5.6% of total global deaths[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The World Health Organization (WHO) further emphasizes that the public health threat of COPD is constantly escalating, noting that driven by population aging and continued tobacco use, COPD is projected to become one of the leading causes of death by 2030[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. In the United States, COPD was the sixth leading cause of death in 2021, with approximately 14.2\u0026nbsp;million people living with the disease. Notably, the prevalence of COPD is significantly higher in rural areas and among low-income populations than in urban areas and among high-income groups, revealing significant health disparities[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eCOPD is characterized by persistent airflow limitation, chronic symptoms (e.g., dyspnea, sputum production), and impaired quality of life. It is also frequently accompanied by comorbidities, which exacerbate disease progression and increase the demand for healthcare services [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Studies have shown that over 97% of COPD patients have at least one comorbidity, and approximately 53% are diagnosed with four or more comorbidities [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Common comorbidities include ischemic heart disease, heart failure, osteoporosis, anemia, lung cancer, and depression. These comorbidities not only complicate COPD management and increase healthcare costs but also elevate patient mortality[\u003cspan additionalcitationids=\"CR8 CR9\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]。\u003c/p\u003e \u003cp\u003eEarly identification of high-risk populations is crucial for mitigating the impact of COPD, as timely interventions\u0026mdash;including smoking cessation, pulmonary rehabilitation, and targeted pharmacotherapy in accordance with the Global Initiative for Chronic Obstructive Lung Disease guidelines\u0026mdash;have been proven to reduce COPD-related mortality[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Specifically, smoking cessation can slow the annual decline in lung function among COPD patients from 42\u0026ndash;82 mL/year to 0\u0026ndash;49 mL/year[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. For patients with recent acute exacerbations, pulmonary rehabilitation (including aerobic exercise and breathing training) can reduce readmission rates and mortality while significantly improving quality of life[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, the implementation of these interventions requires a prior diagnosis of COPD. Unfortunately, existing diagnostic tests have limited sensitivity for early-stage patients with no obvious symptoms; many individuals are only diagnosed with COPD after disease progression, resulting in missed opportunities for early intervention[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Therefore, the development of new diagnostic methods and early risk assessment tools is essential for improving patient prognosis and alleviating the overall disease burden.\u003c/p\u003e \u003cp\u003eBased on 2023 National Health Interview Survey (NHIS) data, this study screened predictors using The Least Absolute Shrinkage and Selection Operator (LASSO) regression and univariate/multivariate logistic regression, and constructed and validated a nomogram. The aim of this study is to provide a reliable tool for the clinical early identification of high-risk populations for COPD.\u003c/p\u003e \u003cp\u003eThe flow chart of this study is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e"},{"header":"2. Material and methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Research Participants and Research Design\u003c/h2\u003e \u003cp\u003eAll data and materials are publicly available from the Centers for Disease Control and Prevention (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.cdc.gov/nchs/nhis/documentation/index.html\u003c/span\u003e\u003cspan address=\"https://www.cdc.gov/nchs/nhis/documentation/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and can be accessed online. The NHIS study was approved by the Disclosure Review Board of the National Center for Health Statistics, and all participants provided written informed consent. This study analyzed all subjects in 2023. The data exclusion criteria were as follows: 1. Subjects with missing covariate information were excluded; 2. Subjects with unclear COPD diagnosis were excluded. A total of 27,388 subjects were finally recruited.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Variable Definition\u003c/h2\u003e \u003cp\u003eThe definition of COPD was determined based on participants' responses to the variable corresponding to Question ID (CON.0010.00.1): \"Have you EVER been told by a doctor or other health professional that you had Chronic Obstructive Pulmonary Disease, C.O.P.D.?\" Participants were categorized into two groups: the COPD group (n\u0026thinsp;=\u0026thinsp;1,499) and the non-COPD group (n\u0026thinsp;=\u0026thinsp;25,889).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Covariates\u003c/h2\u003e \u003cp\u003eCovariates of COPD were categorized into four major groups: sociodemographic factors, health status, comorbid conditions, and healthcare data, as shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eTable of Variables and Their Definitions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariables\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eVariable Definitions\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge (AGE_P)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e18\u0026ndash;30, 31\u0026ndash;45, 46\u0026ndash;65, \u0026gt;\u0026thinsp;65\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex (SEX)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMale, Female\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRace (HISDETP_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHispanic (Mexican/Mexican American), Hispanic (all other groups), Not Hispanic\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEducational Attainment (EDUCP_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eBelow high school education, High School Graduate, Above high school education\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarital Status (R_MARITL)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMarried, Unmarried Cohabitation, Neither of the above\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEconomic Status (RATCAT_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003epoor, Relatively poor, Not poor\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSmoking Status (SMKCIGST_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSmokes daily, Occasionally smokes, Previously smoked, Never smoked\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRegion (REGION)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNortheast, Midwest, South, West\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBody Mass Index (BMICAT_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnderweight, Healthy weight, Overweight, Obesity\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHigh Cholesterol Status (CHLEV_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes, No\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDiabetes Mellitus (PREDIB_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes, No\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypertension Status (HYPEV_A)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eYes, No\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Statistical Analysis\u003c/h2\u003e \u003cp\u003eA total of 27,388 eligible participants were included in this study based on predefined inclusion and exclusion criteria. The data were randomly split into a training set and a validation set in a 7:3 ratio. All statistical analyses were performed using R software.\u003c/p\u003e \u003cp\u003eNormally distributed data are presented as mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (mean\u0026thinsp;\u0026plusmn;\u0026thinsp;SD), non-normally distributed data as median (interquartile range, IQR), and categorical variables as counts (percentages). For statistical comparisons, the t-test was used for normally distributed continuous variables, the Wilcoxon rank-sum test for non-normally distributed continuous variables, and the chi-square test for categorical variables.\u003c/p\u003e \u003cp\u003eLASSO regression model was applied to the training set data. A 10-fold cross-validation was performed to determine the optimal lambda (λ) value for selecting risk factors for COPD. The selected variables were then incorporated into univariate and multivariate logistic regression analyses for further variable refinement. Risk factors with a P-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were included in the nomogram model. After constructing the nomogram, the Receiver Operating Characteristic (ROC) curve and calibration curve were used to evaluate its predictive performance, and Decision Curve Analysis (DCA) was employed to assess its clinical utility.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Result","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1 Baseline Characteristics of Study Participants\u003c/h2\u003e\n \u003cp\u003eA total of 27,388 eligible participants were included in this study based on predefined inclusion and exclusion criteria. To assess whether there were significant differences in 12 baseline characteristics (covariates, see Table \u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) between participants with and without COPD, the included participants were categorized into two groups: the COPD group and the non-COPD group. The baseline table was generated using the tableone R package [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Differences in various baseline characteristics between the two groups were analyzed using the t-test or chi-square test, with a significance level set at \u003cem\u003eP\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003c/em\u003e (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The results indicated that all 12 characteristics were significantly associated with the grouping (COPD vs. non-COPD).\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eAnalysis of Participants\u0026apos; Baseline Characteristics\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"5\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eLevel\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003eNO COPD\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003eCOPD\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colname=\"c5\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003en\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e25889\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e1499\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eAGEP_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e18\u0026ndash;30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e1536 (5.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e6 (0.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e31\u0026ndash;45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8071 (31.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e98 (6.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e46\u0026ndash;65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e7529 (29.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e419 (28.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003e\u0026gt;\u0026thinsp;65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8753 (33.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e976 (65.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eSEX_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eMale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e12033 (46.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e620 (41.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eFemale\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e13856 (53.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e879 (58.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eHISDETP_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eHispanic (Mexican/Mexican American)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e2187 (8.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e38 (2.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eHispanic (all other groups)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e1737 (6.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e45 (3.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNot Hispanic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e21965 (84.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e1416 (94.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eEDUCP_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eBelow high school education\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e2053 (7.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e239 (15.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eHigh School Graduate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e6421 (24.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e504 (33.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eAbove high school education\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e17415 (67.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e756 (50.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eMARITAL_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eMarried\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e12080 (46.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e499 (33.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eUnmarried Cohabitation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e1760 (6.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e70 (4.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNeither of the above\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e12049 (46.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e930 (62.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eRATCAT_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003epoor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e2471 (9.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e298 (19.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eRelatively poor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e4526 (17.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e451 (30.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNot poor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e18892 (73.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e750 (50.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eSMKCIGST_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eSmokes daily\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e1859 (7.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e387 (25.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eOccasionally smokes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e686 (2.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e65 (4.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003ePreviously smoked\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e6304 (24.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e696 (46.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNever smoked\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e17040 (65.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e351 (23.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eREGION (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNortheast\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e3983 (15.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e221 (14.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eMidwest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e5685 (22.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e366 (24.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eSouth\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e9570 (37.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e605 (40.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eWest\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e6651 (25.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e307 (20.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eCHLEV_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8151 (31.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e855 (57.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e17738 (68.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e644 (43.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eBMICAT_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eUnderweight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e370 (1.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e55 (3.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eHealthy weight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8001 (30.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e369 (24.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eOverweight\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8977 (34.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e453 (30.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eObesity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e8541 (33.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e622 (41.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003ePREDIB_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e4387 (16.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e488 (32.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e21502 (83.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e1011 (67.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\n \u003cp\u003eHYPEV_A (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e9373 (36.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e976 (65.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c2\"\u003e\n \u003cp\u003eNo\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c3\"\u003e\n \u003cp\u003e16516 (63.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c4\"\u003e\n \u003cp\u003e523 (34.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2 Screening of Risk Factors for COPD Symptoms\u003c/h2\u003e\n \u003cp\u003eA total of 27,388 participants were randomly divided into a training set (n\u0026thinsp;=\u0026thinsp;19,171) and a validation set (n\u0026thinsp;=\u0026thinsp;8,217) at a ratio of 7:3 for subsequent analyses.\u003c/p\u003e\n \u003cp\u003eTo screen for risk factors associated with COPD, the LASSO regression model was applied to all samples in the training set. Based on the aforementioned covariates, the LASSO regression model was constructed using the glmnet R package[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e], with 10-fold cross-validation performed. Using the optimal lambda value of 0.00007, 24 dummy variables (representing 12 candidate risk factors) were identified (Fig. \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Results of the LASSO regression indicated that all 12 candidate risk factors were important features, which were thus designated as candidate features.\u003c/p\u003e\n \u003cp\u003eIn Fig. \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, LASSO coefficient path plot (top): This plot illustrates the changes in feature coefficients under different values of the regularization parameter (\u0026lambda;). The x-axis represents the values of the regularization parameter, while the y-axis denotes the absolute values or scaled values of the feature coefficients. LASSO regularization path plot (bottom): This plot demonstrates the model fitting performance corresponding to different values of the regularization parameter (\u0026lambda;) in the LASSO algorithm. The x-axis indicates the values of the regularization parameter, and the y-axis represents the model performance metrics.\u003c/p\u003e\n \u003cp\u003eTo further identify COPD-related risk factors, univariate logistic regression analysis was applied to all samples in the training set to screen the candidate features derived from the LASSO regression. Odds ratios (OR) and 95% confidence intervals (95% CI) were calculated, and a forest plot was generated to display the 12 candidate features (Fig. \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, P\u0026thinsp;\u003cem\u003e\u0026lt;\u0026thinsp;0.05\u003c/em\u003e). Results of the univariate logistic regression showed that never smoking (SMKCIGST_A4) significantly reduced the likelihood of COPD, whereas increasing age significantly increased the likelihood of COPD.\u003c/p\u003e\n \u003cp\u003eIn Fig. \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, The x-axis represents the value of OR (Odds Ratio). The interpretation of OR values is as follows: 1) When OR\u0026thinsp;=\u0026thinsp;1: The predictor variable has no effect on the probability of event occurrence (neutral effect); 2) When OR\u0026thinsp;\u0026gt;\u0026thinsp;1: The predictor variable is positively associated with the event occurrence; specifically, an increase in or the presence of the variable increases the likelihood of the event happening; 3) When OR\u0026thinsp;\u0026lt;\u0026thinsp;1: The predictor variable is negatively associated with the event occurrence; specifically, an increase in or the presence of the variable decreases the likelihood of the event happening.\u003c/p\u003e\n \u003cp\u003eMultivariate logistic regression analysis was then used to screen the 12 candidate features derived from the univariate logistic regression analysis, with OR and 95% CI calculated. This yielded 11 predictors associated with COPD occurrence (excluding REGION [region]): AGE_P (age), SEX (gender), race (HISDETP_A), education level (EDUCP_A), HYPEV_A (presence of hypertension), marital status (R_MARITL), BMI (BMICAT_A), SMKSTAT2 (smoking status), presence of high cholesterol (CHLEV_A), presence of diabetes (PREDIB_A), and RATCAT (economic status) (\u003cstrong\u003eFigure 4\u003c/strong\u003e, \u003cem\u003eP\u0026lt;0.05\u003c/em\u003e).\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3.3 Nomogram Construction\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eTo further evaluate the overall predictive ability of the combined predictors, a nomogram was constructed using the rms R package based on the predictors identified from the LASSO regression model and univariate/multivariate logistic regression analyses, with all samples in the training set included (\u003cstrong\u003eFigure 5\u003c/strong\u003e). Each predictor was assigned a corresponding score (Point) via the nomogram. The sum of scores for all risk factors yields the Total Point, which enables the estimation of COPD incidence. A higher total score indicates a greater likelihood of developing COPD.\u003c/p\u003e\n \u003cp\u003eIn \u003cstrong\u003eFigure 5\u003c/strong\u003e, predictors are displayed on the left side. On the right side, scales are marked on the line corresponding to each predictor, representing the range of possible values for that predictor. The length of the line segment reflects the magnitude of the predictor\u0026apos;s contribution to the disease. The \u0026beta;(X\u0026minus;m) terms in the plot denote the individual scores corresponding to each predictor at different values. \u0026quot;Total score\u0026quot; represents the cumulative score obtained by summing the individual scores corresponding to the values of all predictors. \u0026quot;Pr\u0026quot; denotes the risk of developing the disease.\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e3.4 Evaluation of the Nomogram\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eTo assess the predictive performance of the nomogram, its validity and discriminative ability were evaluated using various metrics in both the training and validation sets.\u003c/p\u003e\n \u003cp\u003eCalibration curves for the training and validation sets were generated using the calibrate R package[17] to evaluate the predictive accuracy of the nomogram (\u003cstrong\u003eFigure 6\u003c/strong\u003e). A calibration curve closer to the diagonal line indicates better predictive accuracy of the model. The high degree of overlap between the predicted probabilities of the nomogram and the reference line demonstrated that the nomogram had high predictive accuracy.\u003c/p\u003e\n \u003cp\u003eIn \u003cstrong\u003eFigure 6\u003c/strong\u003e, the x-axis represents the predicted event rate, and the y-axis represents the observed actual event rate, both ranging from 0 to 1 (which can be interpreted as event rates, expressed as percentages). The dashed diagonal line serves as the reference line, representing the scenario where predicted values are equal to actual values. The \u0026quot;Ideal\u0026quot; line denotes the scenario where the model\u0026rsquo;s predictions are perfectly consistent with actual outcomes under ideal conditions. The \u0026quot;Bias-corrected\u0026quot; line represents the performance of the model trained after repeated bootstrap sampling of the samples. The \u0026quot;Apparent\u0026quot; line represents the performance of the dataset.\u003c/p\u003e\n \u003cp\u003eDecision curves for the training and validation sets, generated using the rmda R package, showed that the net benefit of the nomogram was higher than both the \u0026quot;all\u0026quot; line and \u0026quot;none\u0026quot; line, indicating the model has practical value (\u003cstrong\u003eFigure 7\u003c/strong\u003e). Additionally, ROC curves were plotted using the pROC R package[18]. The results revealed that the nomogram model exhibited excellent predictive performance in both the training set (the area under the curve, AUC = 0.84) and the validation set (AUC = 0.84) (\u003cstrong\u003eFigure 8\u003c/strong\u003e).\u003c/p\u003e\n \u003cp\u003eIn \u003cstrong\u003eFigure 7\u003c/strong\u003e, the x-axis represents the threshold probability (Pt). In a risk assessment tool, the probability that Patient i is diagnosed with the disease is denoted as Pi. When Pi reaches a certain threshold (denoted as Pt), the patient is defined as positive, and treatment is administered. At this point, there are benefits of treating patients (considered as \u0026quot;benefits\u0026quot;), as well as harms of treating non-patients and losses resulting from failing to treat patients (considered as \u0026quot;harms\u0026quot;). The y-axis represents the net benefit (NB), calculated as benefits minus harms.\u0026nbsp;Different curved lines in the plot represent different clinical diagnostic models (see the legend for identification). Among these lines, two represent extreme scenarios: 1) The \u0026quot;None\u0026quot; curve indicates that all samples are classified as negative (Pi \u0026lt; Pt), meaning no one receives treatment, and the net benefit is 0; 2)The \u0026quot;All\u0026quot; curve indicates that all samples are classified as positive, meaning everyone receives treatment, and the net benefit is represented by a line with a negative slope.\u003c/p\u003e\n \u003cp\u003eIn\u003cstrong\u003e\u0026nbsp;Figure 8\u003c/strong\u003e, the x-axis represents the false positive rate (1 - specificity), and the y-axis represents the true positive rate (sensitivity). Each point on the curve corresponds to the sensitivity and specificity of a screening test under a specific positive cutoff value.The AUC reflects the value of the diagnostic test: the larger the AUC (i.e., the closer it is to 1.0), the higher the diagnostic accuracy; conversely, the closer the AUC is to 0.5, the lower the diagnostic accuracy.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eSmoking status was the dominant predictive factor in this study, which aligns with evidence from decades of epidemiological and mechanistic research. Such studies have confirmed that tobacco smoke is the single most significant modifiable cause of COPD[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan additionalcitationids=\"CR20\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Mechanistically, cigarette smoke induces irreversible damage to airway epithelial cells, activates oxidative stress pathways, and triggers chronic inflammation via pro-inflammatory cytokines\u0026mdash;all of which are core processes in the pathogenesis of COPD[\u003cspan additionalcitationids=\"CR23 CR24\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Our findings further underscore the critical role of smoking cessation interventions, which remain one of the core components of the 2023 Global Initiative for Chronic Obstructive Lung Disease guidelines [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Notably, research indicates that even smoking cessation in the advanced stages of COPD can reduce the rate of acute disease exacerbations[\u003cspan additionalcitationids=\"CR27 CR28\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], a conclusion that further validates the practical value of this nomogram in identifying smoking populations for targeted interventions.\u003c/p\u003e \u003cp\u003eAge (AGE_P) emerged as a strong independent predictive factor for COPD, reflecting the cumulative impact of aging on lung structure and function. Studies have shown that the prevalence of COPD is common among individuals aged 40 years and older and increases with advancing age[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. This is consistent with our analytical results, highlighting the value of this nomogram in prioritizing screening for middle-aged and elderly populations. Implementing early detection in this group can slow the rapid progression of the disease.\u003c/p\u003e \u003cp\u003eIn this study, hypertension (HYPEV_A), diabetes mellitus (PREDIB_A), and hypercholesterolemia (CHLEV_A) were identified as independent predictive factors, supporting the notion that comorbidity status in COPD is a key driver of adverse outcomes[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Furthermore, chronic airway inflammation in COPD can spread to the systemic circulation, leading to endothelial dysfunction, insulin resistance, and dyslipidemia. These abnormalities not only exacerbate COPD but also worsen comorbid conditions [\u003cspan additionalcitationids=\"CR35 CR36 CR37\" citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e], further underscoring the clinical relevance of this nomogram.\u003c/p\u003e \u003cp\u003eThe study also confirms that socioeconomic factors, including educational level (EDUCP_A), economic status (RATCAT), and marital status (R_MARITL), are key predictive factors for COPD. Previous research has demonstrated that lower educational levels are associated with insufficient health literacy, which contributes to delayed diagnosis of COPD and poor adherence to inhaled therapy [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Additionally, individuals with lower incomes face a higher risk of developing COPD[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan additionalcitationids=\"CR41\" citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e], a finding corroborated by our analysis. Marital status has received relatively little attention as a predictive factor in COPD research; however, this study found an association between marital status and COPD risk, which may be linked to social support. Spousal support, for instance, can improve medication adherence[\u003cspan additionalcitationids=\"CR44\" citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Conversely, single or divorced participants experience higher levels of psychological stress\u0026mdash;a known trigger for acute COPD exacerbations. This emphasizes the need to address social isolation in COPD risk assessment.\u003c/p\u003e \u003cp\u003eBody Mass Index (BMI, BMICAT_A) was significantly associated with COPD, consistent with previous studies [\u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. Moreover, the observed gender (SEX)-related differences in COPD risk align with a growing body of evidence regarding gender-specific susceptibility to lung injury[\u003cspan additionalcitationids=\"CR50\" citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Additionally, race (HISDETP_A) was included as a predictive factor in our analysis.\u003c/p\u003e \u003cp\u003eThis study offers several methodological and clinical strengths: 1) It utilized data from the 2023 NHIS, which features a large, representative sample\u0026mdash;minimizing selection bias to the greatest extent possible; 2) The statistical methods employed were rigorous; 3) Comprehensive validation was conducted; 4) The predictive factors are easily accessible. However, the study also has limitations: 1) The NHIS is a cross-sectional survey, precluding causal inference; 2) COPD diagnosis was based on self-report, which may lead to underdiagnosis of early asymptomatic COPD or overdiagnosis due to recall bias; 3) Confounding factors such as air pollution, occupational exposure, and genetic factors were not measured; 4) Participants with missing covariate data were excluded, which may have introduced selection bias.\u003c/p\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThe aim of this study was to identify predictive factors for the development of COPD. Through baseline analysis, data were divided into training and validation sets, followed by the identification of predictive factors for COPD development using LASSO regression, univariate, and multivariate logistic regression. A nomogram model was constructed to predict the risk of disease occurrence, and its predictive value was validated in both the training and validation sets using calibration curves, DCA curves, and ROC curves.\u003c/p\u003e \u003cp\u003eIn summary, multiple statistical methods were used to analyze the impact of various covariates on the disease, ultimately confirming that AGE_P (age), SEX (gender), race (HISDETP_A), education level (EDUCP_A), HYPEV_A (presence of hypertension), marital status (R_MARITL), BMI (BMICAT_A), SMKSTAT2 (smoking status), presence of high cholesterol (CHLEV_A), presence of diabetes (PREDIB_A), and RATCAT (economic status) can serve as predictors of COPD development. A nomogram was constructed based on these 11 predictors to predict the risk of disease occurrence. The predictive model we developed holds promise as a potential tool to guide clinicians in predicting the probability of COPD development, which may facilitate the implementation of early intervention measures and improve clinical outcomes. However, biological systems are highly dynamic and complex, and this model may not fully capture all relevant biological processes and interactions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003e6. Ethics approval and consent to participate\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study analyzed publicly available data from Centers for Disease Control and Prevention (https://www.cdc.gov/nchs/nhis/documentation/index.html). As this study involved the secondary analysis of de-identified public data, further ethical approval and consent to participate were not required.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e7.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e8.\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets generated and/or analyzed during the current study are available in the Centers for Disease Control and Prevention (https://www.cdc.gov/nchs/nhis/documentation/index.html).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e9.Competing interests\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e10.Funding\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that no funding was received for this study.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eWang, H., et al., \u003cem\u003eGlobal, regional, and national burden of chronic obstructive pulmonary disease from 1990 to 2019.\u003c/em\u003e Front Physiol, 2022. \u003cstrong\u003e13\u003c/strong\u003e: p. 925132.\u003c/li\u003e\n \u003cli\u003eBoers, E., et al., \u003cem\u003eGlobal Burden of Chronic Obstructive Pulmonary Disease Through 2050.\u003c/em\u003e JAMA Netw Open, 2023. \u003cstrong\u003e6\u003c/strong\u003e(12): p. e2346598.\u003c/li\u003e\n \u003cli\u003eLi, H.Y., et al., \u003cem\u003eGlobal, regional and national burden of chronic obstructive pulmonary disease over a 30-year period: Estimates from the 1990 to 2019 Global Burden of Disease Study.\u003c/em\u003e Respirology, 2023. \u003cstrong\u003e28\u003c/strong\u003e(1): p. 29-36.\u003c/li\u003e\n \u003cli\u003eLiu, Y., et al., \u003cem\u003eTrends in the Prevalence of Chronic Obstructive Pulmonary Disease Among Adults Aged \u0026ge;18 Years - United States, 2011-2021.\u003c/em\u003e MMWR Morb Mortal Wkly Rep, 2023. \u003cstrong\u003e72\u003c/strong\u003e(46): p. 1250-1256.\u003c/li\u003e\n \u003cli\u003eRaju, S., et al., \u003cem\u003eRural Residence and Poverty Are Independent Risk Factors for Chronic Obstructive Pulmonary Disease in the United States.\u003c/em\u003e Am J Respir Crit Care Med, 2019. \u003cstrong\u003e199\u003c/strong\u003e(8): p. 961-969.\u003c/li\u003e\n \u003cli\u003eNegewo, N.A., P.G. Gibson, and V.M. McDonald, \u003cem\u003eCOPD and its comorbidities: Impact, measurement and mechanisms.\u003c/em\u003e Respirology, 2015. \u003cstrong\u003e20\u003c/strong\u003e(8): p. 1160-71.\u003c/li\u003e\n \u003cli\u003eSmith, M.C. and J.P. Wrobel, \u003cem\u003eEpidemiology and clinical impact of major comorbidities in patients with COPD.\u003c/em\u003e Int J Chron Obstruct Pulmon Dis, 2014. \u003cstrong\u003e9\u003c/strong\u003e: p. 871-88.\u003c/li\u003e\n \u003cli\u003eDivo, M. and B.R. Celli, \u003cem\u003eMultimorbidity in Patients with Chronic Obstructive Pulmonary Disease.\u003c/em\u003e Clin Chest Med, 2020. \u003cstrong\u003e41\u003c/strong\u003e(3): p. 405-419.\u003c/li\u003e\n \u003cli\u003eBazell, C., et al., \u003cem\u003eVariation in Prevalence and Burden of Chronic Obstructive Pulmonary Disease by State and Insurance Type in the United States.\u003c/em\u003e Chronic Obstr Pulm Dis, 2025. \u003cstrong\u003e12\u003c/strong\u003e(2): p. 158-174.\u003c/li\u003e\n \u003cli\u003eAgust\u0026iacute;, A., et al., \u003cem\u003eGlobal Initiative for Chronic Obstructive Lung Disease 2023 Report: GOLD Executive Summary.\u003c/em\u003e Eur Respir J, 2023. \u003cstrong\u003e61\u003c/strong\u003e(4).\u003c/li\u003e\n \u003cli\u003evan Eerd, E.A., et al., \u003cem\u003eSmoking cessation for people with chronic obstructive pulmonary disease.\u003c/em\u003e Cochrane Database Syst Rev, 2016. \u003cstrong\u003e2016\u003c/strong\u003e(8): p. Cd010744.\u003c/li\u003e\n \u003cli\u003eGreening, N.J., et al., \u003cem\u003eAn early rehabilitation intervention to enhance recovery during hospital admission for an exacerbation of chronic respiratory disease: randomised controlled trial.\u003c/em\u003e Bmj, 2014. \u003cstrong\u003e349\u003c/strong\u003e: p. g4315.\u003c/li\u003e\n \u003cli\u003eStolz, D., et al., \u003cem\u003eTowards the elimination of chronic obstructive pulmonary disease: a Lancet Commission.\u003c/em\u003e Lancet, 2022. \u003cstrong\u003e400\u003c/strong\u003e(10356): p. 921-972.\u003c/li\u003e\n \u003cli\u003eWang, L., et al., \u003cem\u003eConstruction and validation of a risk prediction model for chronic obstructive pulmonary disease (COPD): a cross-sectional study based on the NHANES database from 2009 to 2018.\u003c/em\u003e BMC Pulm Med, 2025. \u003cstrong\u003e25\u003c/strong\u003e(1): p. 317.\u003c/li\u003e\n \u003cli\u003ePanos, A. and D. Mavridis, \u003cem\u003eTableOne: an online web application and R package for summarising and visualising data.\u003c/em\u003e Evid Based Ment Health, 2020. \u003cstrong\u003e23\u003c/strong\u003e(3): p. 127-130.\u003c/li\u003e\n \u003cli\u003eFriedman, J., T. Hastie, and R. Tibshirani, \u003cem\u003eRegularization Paths for Generalized Linear Models via Coordinate Descent.\u003c/em\u003e J Stat Softw, 2010. \u003cstrong\u003e33\u003c/strong\u003e(1): p. 1-22.\u003c/li\u003e\n \u003cli\u003eGraffelman, J. and F. van Eeuwijk, \u003cem\u003eCalibration of multivariate scatter plots for exploratory analysis of relations within and between sets of variables in genomic research.\u003c/em\u003e Biom J, 2005. \u003cstrong\u003e47\u003c/strong\u003e(6): p. 863-79.\u003c/li\u003e\n \u003cli\u003eRobin, X., et al., \u003cem\u003epROC: an open-source package for R and S+ to analyze and compare ROC curves.\u003c/em\u003e BMC Bioinformatics, 2011. \u003cstrong\u003e12\u003c/strong\u003e: p. 77.\u003c/li\u003e\n \u003cli\u003eGan, H., et al., \u003cem\u003eSmoking: a leading factor for the death of chronic respiratory diseases derived from Global Burden of Disease Study 2019.\u003c/em\u003e BMC Pulm Med, 2022. \u003cstrong\u003e22\u003c/strong\u003e(1): p. 149.\u003c/li\u003e\n \u003cli\u003eRegan, E.A., et al., \u003cem\u003eGenetic epidemiology of COPD (COPDGene) study design.\u003c/em\u003e Copd, 2010. \u003cstrong\u003e7\u003c/strong\u003e(1): p. 32-43.\u003c/li\u003e\n \u003cli\u003eMuro, S., \u003cem\u003e[Cigarette smoking is the most important causal factor for developing chronic obstructive pulmonary disease (COPD)].\u003c/em\u003e Nihon Rinsho, 2011. \u003cstrong\u003e69\u003c/strong\u003e(10): p. 1735-40.\u003c/li\u003e\n \u003cli\u003eSchumacher, F.R., et al., \u003cem\u003eRNAi Screen for NRF2 Inducers Identifies Targets That Rescue Primary Lung Epithelial Cells from Cigarette Smoke Induced Radical Stress.\u003c/em\u003e PLoS One, 2016. \u003cstrong\u003e11\u003c/strong\u003e(11): p. e0166352.\u003c/li\u003e\n \u003cli\u003eZhang, Y., et al., \u003cem\u003eNrf2/HO-1 signaling activation alleviates cigarette smoke-induced inflammation in chronic obstructive pulmonary disease by suppressing NLRP3-mediated pyroptosis.\u003c/em\u003e J Cardiothorac Surg, 2024. \u003cstrong\u003e19\u003c/strong\u003e(1): p. 58.\u003c/li\u003e\n \u003cli\u003ePera, T., et al., \u003cem\u003eRole for TAK1 in cigarette smoke-induced proinflammatory signaling and IL-8 release by human airway smooth muscle cells.\u003c/em\u003e Am J Physiol Lung Cell Mol Physiol, 2012. \u003cstrong\u003e303\u003c/strong\u003e(3): p. L272-8.\u003c/li\u003e\n \u003cli\u003eHeijink, I.H., et al., \u003cem\u003eCigarette smoke impairs airway epithelial barrier function and cell-cell contact recovery.\u003c/em\u003e Eur Respir J, 2012. \u003cstrong\u003e39\u003c/strong\u003e(2): p. 419-28.\u003c/li\u003e\n \u003cli\u003eGodtfredsen, N.S., et al., \u003cem\u003eRisk of hospital admission for COPD following smoking cessation and reduction: a Danish population study.\u003c/em\u003e Thorax, 2002. \u003cstrong\u003e57\u003c/strong\u003e(11): p. 967-72.\u003c/li\u003e\n \u003cli\u003eWu, J. and D.D. Sin, \u003cem\u003eImproved patient outcome with smoking cessation: when is it too late?\u003c/em\u003e Int J Chron Obstruct Pulmon Dis, 2011. \u003cstrong\u003e6\u003c/strong\u003e: p. 259-67.\u003c/li\u003e\n \u003cli\u003eBai, J.W., et al., \u003cem\u003eSmoking cessation affects the natural history of COPD.\u003c/em\u003e Int J Chron Obstruct Pulmon Dis, 2017. \u003cstrong\u003e12\u003c/strong\u003e: p. 3323-3328.\u003c/li\u003e\n \u003cli\u003eT\u0026oslash;nnesen, P., \u003cem\u003eSmoking cessation and COPD.\u003c/em\u003e Eur Respir Rev, 2013. \u003cstrong\u003e22\u003c/strong\u003e(127): p. 37-43.\u003c/li\u003e\n \u003cli\u003eToelle, B.G., et al., \u003cem\u003eRespiratory symptoms and illness in older Australians: the Burden of Obstructive Lung Disease (BOLD) study.\u003c/em\u003e Med J Aust, 2013. \u003cstrong\u003e198\u003c/strong\u003e(3): p. 144-8.\u003c/li\u003e\n \u003cli\u003eWang, Y., et al., \u003cem\u003eChronic obstructive pulmonary disease across three decades: trends, inequalities, and projections from the Global Burden of Disease Study 2021.\u003c/em\u003e Front Med (Lausanne), 2025. \u003cstrong\u003e12\u003c/strong\u003e: p. 1564878.\u003c/li\u003e\n \u003cli\u003eYang, H.Y., et al., \u003cem\u003eIncreased Risk of Chronic Obstructive Pulmonary Disease in Patients with Hyperlipidemia: A Nationwide Population-Based Cohort Study.\u003c/em\u003e Int J Environ Res Public Health, 2022. \u003cstrong\u003e19\u003c/strong\u003e(19).\u003c/li\u003e\n \u003cli\u003eChen, W., et al., \u003cem\u003eRisk of cardiovascular comorbidity in patients with chronic obstructive pulmonary disease: a systematic review and meta-analysis.\u003c/em\u003e Lancet Respir Med, 2015. \u003cstrong\u003e3\u003c/strong\u003e(8): p. 631-9.\u003c/li\u003e\n \u003cli\u003eMarcuccio, G., et al., \u003cem\u003eEndothelial dysfunction in chronic obstructive pulmonary disease: an update on mechanisms, assessment tools and treatment strategies.\u003c/em\u003e Front Med (Lausanne), 2025. \u003cstrong\u003e12\u003c/strong\u003e: p. 1550716.\u003c/li\u003e\n \u003cli\u003eBolton, C.E., et al., \u003cem\u003eInsulin resistance and inflammation - A further systemic complication of COPD.\u003c/em\u003e Copd, 2007. \u003cstrong\u003e4\u003c/strong\u003e(2): p. 121-6.\u003c/li\u003e\n \u003cli\u003eAnghel, L., et al., \u003cem\u003eChronic Obstructive Pulmonary Disease and Type 2 Diabetes Mellitus: Complex Interactions and Clinical Implications.\u003c/em\u003e J Clin Med, 2025. \u003cstrong\u003e14\u003c/strong\u003e(6).\u003c/li\u003e\n \u003cli\u003eAgusti, A. and J.B. Soriano, \u003cem\u003eCOPD as a systemic disease.\u003c/em\u003e Copd, 2008. \u003cstrong\u003e5\u003c/strong\u003e(2): p. 133-8.\u003c/li\u003e\n \u003cli\u003eMoermans, C., et al., \u003cem\u003eLocal and systemic cellular inflammation and cytokine release in chronic obstructive pulmonary disease.\u003c/em\u003e Cytokine, 2011. \u003cstrong\u003e56\u003c/strong\u003e(2): p. 298-304.\u003c/li\u003e\n \u003cli\u003eO\u0026apos;Conor, R., et al., \u003cem\u003eEffects of health literacy and cognitive abilities on COPD self-management behaviors: A prospective cohort study.\u003c/em\u003e Respir Med, 2019. \u003cstrong\u003e160\u003c/strong\u003e: p. 105630.\u003c/li\u003e\n \u003cli\u003eEisner, M.D., et al., \u003cem\u003eSocioeconomic status, race and COPD health outcomes.\u003c/em\u003e J Epidemiol Community Health, 2011. \u003cstrong\u003e65\u003c/strong\u003e(1): p. 26-34.\u003c/li\u003e\n \u003cli\u003eLee, Y.S., et al., \u003cem\u003eThe association between living below the relative poverty line and the prevalence of chronic obstructive pulmonary disease.\u003c/em\u003e J Thorac Dis, 2019. \u003cstrong\u003e11\u003c/strong\u003e(2): p. 427-437.\u003c/li\u003e\n \u003cli\u003eAtassi, M., et al., \u003cem\u003eAssociation between chronic airflow obstruction and socio-economic position in Morocco: BOLD results.\u003c/em\u003e Int J Tuberc Lung Dis, 2020. \u003cstrong\u003e24\u003c/strong\u003e(2): p. 202-206.\u003c/li\u003e\n \u003cli\u003eGiannico, O.V., et al., \u003cem\u003eEducational level, marital status and sex as social gender discharge determinants in chronic obstructive pulmonary disease exacerbations: a time-to-event analysis.\u003c/em\u003e Monaldi Arch Chest Dis, 2019. \u003cstrong\u003e89\u003c/strong\u003e(2).\u003c/li\u003e\n \u003cli\u003eTrivedi, R.B., et al., \u003cem\u003eThe influence of informal caregivers on adherence in COPD patients.\u003c/em\u003e Ann Behav Med, 2012. \u003cstrong\u003e44\u003c/strong\u003e(1): p. 66-72.\u003c/li\u003e\n \u003cli\u003eLiu, Y.R., et al., \u003cem\u003ePath analysis of illness perception, medication beliefs, family support on inhaler adherence in elderly COPD patients: Based on triadic reciprocal determinism.\u003c/em\u003e Patient Educ Couns, 2025. \u003cstrong\u003e130\u003c/strong\u003e: p. 108465.\u003c/li\u003e\n \u003cli\u003eSun, Y., et al., \u003cem\u003eBMI is associated with FEV(1) decline in chronic obstructive pulmonary disease: a meta-analysis of clinical trials.\u003c/em\u003e Respir Res, 2019. \u003cstrong\u003e20\u003c/strong\u003e(1): p. 236.\u003c/li\u003e\n \u003cli\u003eZhou, Y., et al., \u003cem\u003eThe association between BMI and COPD: the results of two population-based studies in Guangzhou, China.\u003c/em\u003e Copd, 2013. \u003cstrong\u003e10\u003c/strong\u003e(5): p. 567-72.\u003c/li\u003e\n \u003cli\u003eWang, M., X. Ni, and F. Yu, \u003cem\u003eImpact of Body Mass Index on Risk of Exacerbation in Patients With COPD: A Systematic Review and Meta-Analysis.\u003c/em\u003e Chronic Obstr Pulm Dis, 2024. \u003cstrong\u003e11\u003c/strong\u003e(5): p. 524-533.\u003c/li\u003e\n \u003cli\u003e\u0026Ccedil;olak, Y., et al., \u003cem\u003eSex differences in COPD in relation to smoking exposure: a population-based cohort study.\u003c/em\u003e Thorax, 2025. \u003cstrong\u003e80\u003c/strong\u003e(8): p. 512-519.\u003c/li\u003e\n \u003cli\u003eRaghavan, D., A. Varkey, and T. Bartter, \u003cem\u003eChronic obstructive pulmonary disease: the impact of gender.\u003c/em\u003e Curr Opin Pulm Med, 2017. \u003cstrong\u003e23\u003c/strong\u003e(2): p. 117-123.\u003c/li\u003e\n \u003cli\u003eForsslund, H., et al., \u003cem\u003eGender differences in the T-cell profiles of the airways in COPD patients associated with clinical phenotypes.\u003c/em\u003e Int J Chron Obstruct Pulmon Dis, 2017. \u003cstrong\u003e12\u003c/strong\u003e: p. 35-48.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-pulmonary-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pulm","sideBox":"Learn more about [BMC Pulmonary Medicine](http://bmcpulmmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pulm/default.aspx","title":"BMC Pulmonary Medicine","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Chronic Obstructive Pulmonary Disease, a retrospective study, Machine learning, Risk predictors, Risk populations","lastPublishedDoi":"10.21203/rs.3.rs-8574459/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8574459/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Chronic Obstructive Pulmonary Disease (COPD) is a major respiratory disorder, and identifying its predictive factors is essential for early intervention to improve clinical outcomes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e This study analyzed 2023 data from the National Health Interview Survey (NHIS), including 27,388 eligible participants (1,499 with COPD, 25,889 without). Data were split into a 70% training set and a 30% validation set. LASSO regression, univariate, and multivariate logistic regression were applied to screen predictors. A nomogram was constructed, and its performance was assessed using ROC curves, calibration curves, and Decision Curve Analysis (DCA).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Eleven independent predictors of COPD were identified in this study. Based on these predictors, a nomogram was constructed to assess the risk of COPD. Results showed good agreement between the predicted probabilities of COPD and the actual incidence rates, and DCA confirmed the clinical utility of this nomogram.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion: \u003c/strong\u003eThe nomogram based on these 11 predictors can assist clinicians in estimating COPD risk for early intervention. However, it may not fully capture all complex biological processes and interactions related to COPD.\u003c/p\u003e","manuscriptTitle":"Machine learning driven-tracing the risk predictors for Chronic Obstructive Pulmonary Disease patients from a retrospective study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-19 07:19:22","doi":"10.21203/rs.3.rs-8574459/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-04-08T04:51:49+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-15T09:34:16+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-16T11:38:17+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-16T11:32:37+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Pulmonary Medicine","date":"2026-01-11T15:12:05+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-pulmonary-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"pulm","sideBox":"Learn more about [BMC Pulmonary Medicine](http://bmcpulmmed.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/pulm/default.aspx","title":"BMC Pulmonary Medicine","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"10947ef7-73fd-4433-b868-52ec1ca4b034","owner":[],"postedDate":"April 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-19T07:19:23+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-19 07:19:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8574459","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8574459","identity":"rs-8574459","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}