Machine learning algorithms are comparable to conventional regression models in predicting distant metastasis of follicular thyroid carcinoma.

OBJECTIVE: Distant metastasis often indicates a poor prognosis, so early screening and diagnosis play a significant role. Our study aims to construct and verify a predictive model based on machine learning (ML) algorithms that can estimate the risk of distant metastasis of newly diagnosed follicular thyroid carcinoma (FTC). DESIGN: This was a retrospective study based on the Surveillance, Epidemiology, and End Results (SEER) database from 2004 to 2015. PATIENTS: A total of 5809 FTC patients were included in the data analysis. Among them, there were 214 (3.68%) cases with distant metastasis. METHOD: Univariate and multivariate logistic regression (LR) analyses were used to determine independent risk factors. Seven commonly used ML algorithms were applied for predictive model construction. We used the area under the receiver-operating characteristic (AUROC) curve to select the best ML algorithm. The optimal model was trained through 10-fold cross-validation and visualized by SHapley Additive exPlanations (SHAP). Finally, we compared it with the traditional LR method. RESULTS: In terms of predicting distant metastasis, the AUROCs of the seven ML algorithms were 0.746-0.836 in the test set. Among them, the Extreme Gradient Boosting (XGBoost) had the best prediction performance, with an AUROC of 0.836 (95% confidence interval [CI]: 0.775-0.897). After 10-fold cross-validation, its predictive power could reach the best [AUROC: 0.855 (95% CI: 0.803-0.906)], which was slightly higher than the classic binary LR model [AUROC: 0.845 (95% CI: 0.818-0.873)]. CONCLUSIONS: The XGBoost approach was comparable to the conventional LR method for predicting the risk of distant metastasis for FTC.

Klotho reduces the risk of osteoporosis in postmenopausal women: a cross-sectional study of the National Health and Nutrition Examination Survey (NHANES).

BACKGROUND: Osteoporosis (OP) is one of the diseases that endanger the health of the elderly population. Klotho protein is a hormone with anti-aging effects. A few studies have discussed the relationship between Klotho and OP. However, there is still a lack of research on larger populations. This study aims to evaluate the association between OP and Klotho in American postmenopausal women. METHODS: This is a retrospective study. We searched the National Health and Nutrition Examination Survey (NHANES) database and collected data of 3 survey cycles, finally involving 871 postmenopausal women over 50 years old in the present study. All participants took dual-energy X-ray absorptiometry examination and serum Klotho testing at the time of investigation. After adjusting the possible confounding variables, a multivariate regression model was employed to estimate the relationship between OP and Klotho proteins. Besides, the P for trend and restricted cubic spline (RCS) were applied to examine the threshold effect and calculate the inflection point. RESULTS: Factors influencing the occurrence of OP included age, ethnicity, body mass index and Klotho levels. Multivariate regression analysis indicated that the serum Klotho concentration was lower in OP patients than that in participants without OP (OR[log2Klotho] = 0.568, P = 0.027). The C-index of the prediction model built was 0.765, indicating good prediction performance. After adjusting the above-mentioned four variables, P values for trend showed significant differences between groups. RCSs revealed that when the Klotho concentration reached 824.09 pg/ml, the risk of OP decreased drastically. CONCLUSION: Based on the analysis of the data collected from the NHANES database, we propose a correlation between Klotho and postmenopausal OP. A higher serum Klotho level is related to a lower incidence of OP. The findings of the present study can provide guidance for research on diagnosis and risk assessment of OP.

Combined T4 + T3 therapy versus T4 monotherapy effect on psychological health in hypothyroidism: A systematic review and meta-analysis.

PURPOSE: To evaluate whether T4 + T3 combination therapy had advantages in improving psychological health compared with T4 monotherapy. METHODS: We searched PubMed, Embase, Cochrane Library, and Web of Science from January 2000 to March 2021, and updated in September 2021. The inclusion criteria (prospective study, published in English, had a T4 + T3 combination therapy test group and a T4 monotherapy control group, patients aged ≥18 years and with overt primary hypothyroidism, and published after January 2000) were applied by two reviewers; any disagreement was resolved by a third reviewer. The two reviewers independently extracted data using a standard data form and assessed the risk of bias using the Cochrane risk of bias tool. Coprimary outcomes included the psychological health measures of depression, fatigue, pain, anxiety and anger, measured using validated and reliable instruments. RESULTS: Eighteen of 2029 studies (883 patients) were included in the meta-analysis. No significant difference was found between T4 + T3 combination therapy and T4 monotherapy with regard to depression (standardized mean difference [SMD]: -0.06, 95% confidence interval [CI]: -0.18; 0.07), fatigue (SMD: 0.06, 95% CI: -0.13; 0.26), pain (SMD: -0.01, 95% CI: -0.24; 0.22), anxiety (SMD: 0.01, 95% CI: -0.15; 0.17) and anger (SMD: 0.05, 95% CI: -0.15; 0.24). Methodological heterogeneity had no influence on the results. The patients preferred combination therapy significantly. CONCLUSIONS: Compared with T4 monotherapy, T4 + T3 combination therapy had no significant advantage in improving psychological health. For patients who are unsatisfied with LT4 monotherapy, the patient and the physician should make a joint decision concerning therapy.

Exploring risk factors for cervical lymph node metastasis in papillary thyroid microcarcinoma: construction of a novel population-based predictive model.

BACKGROUND: Machine learning was a highly effective tool in model construction. We aim to establish a machine learning-based predictive model for predicting the cervical lymph node metastasis (LNM) in papillary thyroid microcarcinoma (PTMC). METHODS: We obtained data on PTMC from the SEER database, including 10 demographic and clinicopathological characteristics. Univariate and multivariate logistic regression (LR) analyses were applied to screen the risk factors for cervical LNM in PTMC. Risk factors with P < 0.05 in multivariate LR analysis were used as modeling variables. Five different machine learning (ML) algorithms including extreme gradient boosting (XGBoost), random forest (RF), adaptive boosting (AdaBoost), gaussian naive bayes (GNB) and multi-layer perceptron (MLP) and traditional regression analysis were used to construct the prediction model. Finally, the area under the receiver operating characteristic (AUROC) curve was used to compare the model performance. RESULTS: Through univariate and multivariate LR analysis, we screened out 9 independent risk factors most closely associated with cervical LNM in PTMC, including age, sex, race, marital status, region, histology, tumor size, and extrathyroidal extension (ETE) and multifocality. We used these risk factors to build an ML prediction model, in which the AUROC value of the XGBoost algorithm was higher than the other 4 ML algorithms and was the best ML model. We optimized the XGBoost algorithm through 10-fold cross-validation, and its best performance on the training set (AUROC: 0.809, 95%CI 0.800-0.818) was better than traditional LR analysis (AUROC: 0.780, 95%CI 0.772-0.787). CONCLUSIONS: ML algorithms have good predictive performance, especially the XGBoost algorithm. With the continuous development of artificial intelligence, ML algorithms have broad prospects in clinical prognosis prediction.

Survival benefit of postoperative radioiodine therapy among patients with intermediate-risk differentiated thyroid carcinoma.

BACKGROUND: The 2015 American Thyroid Association (ATA) guidelines proposed the use of the ATA Risk Stratification System and American Joint Committee on Cancer Tumor-Node-Metastasis (AJCC/TNM) Staging System for postoperative radioiodine decision-making. However, the management of patients with intermediate-risk differentiated thyroid carcinoma (DTC) is not well defined. In this study, we aimed to evaluate the therapeutic efficacy of radioactive iodine therapy (RAIT) among various subgroups of patients with intermediate-risk DTC after surgery. METHODS: This was a retrospective study based on the Surveillance, Epidemiology, and End Results (SEER) database (2010-2015). The DTC patients with intermediate risk of recurrence were divided into two groups (treated or not treated with radioactive iodine (RAI)). As the treatment was not randomly assigned, stabilized inverse probability treatment weighting (sIPTW) was used to reduce selection bias. We used the Kaplan-Meier method and log-rank test to analyze overall survival (OS) and cancer-specific survival (CSS). RESULTS: Kaplan-Meier analysis after sIPTW found a significant difference in OS and CSS between no RAIT and RAIT (log-rank test, P < 0.0001; P = 0.0019, respectively). The Kaplan-Meier curves of CSS in age cutoff of 55 years showed a significant association between no RAIT and RAIT (log-rank test, P = 0.0045). Univariate and multivariate Cox regression showed RAIT was associated with a reduced risk of mortality compared with no RAIT (hazard ratio [HR] 0.59, 95% confidence interval [95% CI 0.44-0.80]). Age (≥ 55) years showed a worse CSS regardless of whether or not a patient was treated or not treated with RAI ([HR] 8.91, 95% confidence interval [95% CI 6.19-12.84]). CONCLUSIONS: RAIT improves OS and CSS in patients with intermediate-risk DTC after surgery. 55 years is a more appropriate prognostic age cutoff for the relevant classification systems and is a crucial consideration in RAI decision-making. Therefore, we need individualized treatment plans.

Socioeconomic disparities and regional environment are associated with cervical lymph node metastases in children and adolescents with differentiated thyroid cancer: developing a web-based predictive model.

Purpose: To establish an online predictive model for the prediction of cervical lymph node metastasis (CLNM) in children and adolescents with differentiated thyroid cancer (caDTC). And analyze the impact between socioeconomic disparities, regional environment and CLNM. Methods: We retrospectively analyzed clinicopathological and sociodemographic data of caDTC from the Surveillance, Epidemiology, and End Results (SEER) database from 2000 to 2019. Risk factors for CLNM in caDTC were analyzed using univariate and multivariate logistic regression (LR). And use the extreme gradient boosting (XGBoost) algorithm and other commonly used ML algorithms to build CLNM prediction models. Model performance assessment and visualization were performed using the area under the receiver operating characteristic (AUROC) curve and SHapley Additive exPlanations (SHAP). Results: In addition to common risk factors, our study found that median household income and living regional were strongly associated with CLNM. Whether in the training set or the validation set, among the ML models constructed based on these variables, the XGBoost model has the best predictive performance. After 10-fold cross-validation, the prediction performance of the model can reach the best, and its best AUROC value is 0.766 (95%CI: 0.745-0.786) in the training set, 0.736 (95%CI: 0.670-0.802) in the validation set, and 0.733 (95%CI: 0.683-0.783) in the test set. Based on this XGBoost model combined with SHAP method, we constructed a web-base predictive system. Conclusion: The online prediction model based on the XGBoost algorithm can dynamically estimate the risk probability of CLNM in caDTC, so as to provide patients with personalized treatment advice.

Everyday Activities of Older Adults in Tianjin, China: Coupling Constraints, Gendered Mobilities and Social Context.

Many Western studies have indicated that older women are generally more vulnerable in terms of mobility compared to older men, particularly regarding driving. However, the situation may differ in the context of China. This study, based on activity diaries and semi-structured interviews, focuses on the spatiotemporal behavior of older adults in Tianjin and explores how the constraints posed by activity companions (in terms of type, size, and composition) shape the mobilities of older men and women, including activity locations, travel distances, and transportation modes. The key findings are as follows: First, older women are more engaged with their families due to a higher percentage and longer duration of activities spent with family members. Second, older men tend to have more concentrated travel distances near their homes compared to older women. Third, older women exhibit a broader range of activities in different locations and engage in longer-distance leisure travel with family members when compared to older men. In the context of Western literature, this study discusses older women's enhanced social interactions, their earlier retirement in China, and the impact of COVID-19 as factors that help explain these findings. This study contributes to a deeper understanding of accompanied mobilities among Chinese older adults using geographical theory and methods, emphasizing the importance of flexible work schedules for the workforce and the organization of community-based activities to promote the social interactions and mobilities of older adults.

Establishment and validation of a nomogram model for predicting distant metastasis in medullary thyroid carcinoma: An analysis of the SEER database based on the AJCC 8th TNM staging system.

Objective: Medullary thyroid carcinoma (MTC) patients with distant metastases frequently present a relatively poor survival prognosis. Our main purpose was developing a nomogram model to predict distant metastases in MTC patients. Methods: This was a retrospective study based on the Surveillance, Epidemiology, and End Results (SEER) database. Data of 807 MTC patients diagnosed from 2004 to 2015 who undergone total thyroidectomy and neck lymph nodes dissection was included in our study. Independent risk factors were screened by univariate and multivariate logistic regression analysis successively, which were used to develop a nomogram model predicting for distant metastasis risk. Further, the log-rank test was used to compare the differences of Kaplan-Meier curves of cancer-specific survival (CSS) in different M stage and each independent risk factor groups. Results: Four clinical parameters including age > 55 years, higher T stage (T3/T4), higher N stage (N1b) and lymph node ratio (LNR) > 0.4 were significant for distant metastases at the time of diagnosis in MTC patients, and were selected to develop a nomogram model. This model had satisfied discrimination with the AUC and C-index of 0.894, and C-index was confirmed to be 0.878 through bootstrapping validation. A decision curve analysis (DCA) was subsequently made to evaluate the feasibility of this nomogram for predicting distant metastasis. In addition, CSS differed by different M stage, T stage, N stage, age and LNR groups. Conclusions: Age, T stage, N stage and LNR were extracted to develop a nomogram model for predicting the risk of distant metastases in MTC patients. The model is of great significance for clinicians to timely identify patients with high risk of distant metastases and make further clinical decisions.

Value of machine learning algorithms for predicting diabetes risk: A subset analysis from a real-world retrospective cohort study.

AIMS/INTRODUCTION: To compare the application value of different machine learning (ML) algorithms for diabetes risk prediction. MATERIALS AND METHODS: This is a 3-year retrospective cohort study with a total of 3,687 participants being included in the data analysis. Modeling variable screening and predictive model building were carried out using logistic regression (LR) analysis and 10-fold cross-validation, respectively. In total, six different ML algorithms, including random forests, light gradient boosting machine, extreme gradient boosting, adaptive boosting (AdaBoost), multi-layer perceptrons and gaussian naive bayes were used for model construction. Model performance was mainly evaluated by the area under the receiver operating characteristic curve. The best performing ML model was selected for comparison with the traditional LR model and visualized using Shapley additive explanations. RESULTS: A total of eight risk factors most associated with the development of diabetes were identified by univariate and multivariate LR analysis, and they were visualized in the form of a nomogram. Among the six different ML models, the random forests model had the best predictive performance. After 10-fold cross-validation, its optimal model has an area under the receiver operating characteristic value of 0.855 (95% confidence interval [CI] 0.823-0.886) in the training set and 0.835 (95% CI 0.779-0.892) in the test set. In the traditional LR model, its area under the receiver operating characteristic value is 0.840 (95% CI 0.814-0.866) in the training set and 0.834 (95% CI 0.785-0.884) in the test set. CONCLUSIONS: In the real-world epidemiological research, the combination of traditional variable screening and ML algorithm to construct a diabetes risk prediction model has satisfactory clinical application value.

Economic effect between surgery and thermal ablation for patients with papillary thyroid microcarcinoma: a systemic review and meta-analysis.

BACKGROUND: At present, there are still many controversies regarding the treatment of papillary thyroid microcarcinoma (PTMC). It is worth noting that thermal ablation has become a viable alternative for patients at high risk of surgery or those who wish to receive more active treatment than active surveillance. OBJECTIVE: To investigate the economic benefits, efficacy, and safety of thermal ablation for patients with PTMC. METHODS: We searched PubMed, Cochrane Library, Web of Science, and Embase for articles published from the database establishment to August 2021. All of the studies included were retrospective cohort analyses. In addition, meta-analysis was performed using the Stata15.1 versions and Review Manager5.3. Data were extracted according to the PRISMA guidelines by two professional doctors. RESULTS: A total of 7 articles (1582 patients) were included in our systematic review and meta-analysis. We discovered that thermal ablation could shorten the hospitalization time (SMD, -3.82[95%CI, -5.25 to -2.40]), operation time (SMD, -5.67[95%CI, -6.65 to -4.68]), cost (SMD, -3.40 [95%CI, -5.06 to -1.75]) and reduce postoperative complications (OR, 0.07 [95%CI, 0.01 to 0.44]) than surgical treatment. In addition, our study also found that compared with surgery, there was no significant difference in the recurrence rate (OR, 1.53 [95% CI, 0.63 to 3.73]) and metastasis rate (OR, 0.47 [95% CI, 0.12 to 1.86]). CONCLUSION: On the premise of being safe and effective, thermal ablation has better economic benefits than surgical treatment for patients with PTMC.

Surgical Methods and Social Factors Are Associated With Long-Term Survival in Follicular Thyroid Carcinoma: Construction and Validation of a Prognostic Model Based on Machine Learning Algorithms.

Background: This study aimed to establish and verify an effective machine learning (ML) model to predict the prognosis of follicular thyroid cancer (FTC), and compare it with the eighth edition of the American Joint Committee on Cancer (AJCC) model. Methods: Kaplan-Meier method and Cox regression model were used to analyze the risk factors of cancer-specific survival (CSS). Propensity-score matching (PSM) was used to adjust the confounding factors of different surgeries. Nine different ML algorithms,including eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), Random Forests (RF), Logistic Regression (LR), Adaptive Boosting (AdaBoost), Gaussian Naive Bayes (GaussianNB), K-Nearest Neighbor (KNN), Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP),were used to build prognostic models of FTC.10-fold cross-validation and SHapley Additive exPlanations were used to train and visualize the optimal ML model.The AJCC model was built by multivariate Cox regression and visualized through nomogram. The performance of the XGBoost model and AJCC model was mainly assessed using the area under the receiver operating characteristic (AUROC). Results: Multivariate Cox regression showed that age, surgical methods, marital status, T classification, N classification and M classification were independent risk factors of CSS. Among different surgeries, the prognosis of one-sided thyroid lobectomy plus isthmectomy (LO plus IO) was the best, followed by total thyroidectomy (hazard ratios: One-sided thyroid LO plus IO, 0.086[95% confidence interval (CI),0.025-0.290], P<0.001; total thyroidectomy (TT), 0.490[95%CI,0.295-0.814], P=0.006). PSM analysis proved that one-sided thyroid LO plus IO, TT, and partial thyroidectomy had no significant differences in long-term prognosis. Our study also revealed that married patients had better prognosis than single, widowed and separated patients (hazard ratios: single, 1.686[95%CI,1.146-2.479], P=0.008; widowed, 1.671[95%CI,1.163-2.402], P=0.006; separated, 4.306[95%CI,2.039-9.093], P<0.001). Among different ML algorithms, the XGBoost model had the best performance, followed by Gaussian NB, RF, LR, MLP, LightGBM, AdaBoost, KNN and SVM. In predicting FTC prognosis, the predictive performance of the XGBoost model was relatively better than the AJCC model (AUROC: 0.886 vs. 0.814). Conclusion: For high-risk groups, effective surgical methods and well marital status can improve the prognosis of FTC. Compared with the traditional AJCC model, the XGBoost model has relatively better prediction accuracy and clinical usage.

Predictive value of HbA1c for in-hospital adverse prognosis in COVID-19: A systematic review and meta-analysis.

BACKGROUND AND AIMS: Clinical and laboratory predictors of adverse clinical course and death in COVID-19 patients urgently need to be identified. So far, the association between HbA1c and in-hospital mortality of COVID-19 remains a controversial issue. The aim of this study is to analyze predictive value of HbA1c for adverse prognosis in COVID-19. METHODS: Both Chinese and English databases were systematically searched using specific keywords associated with the aims until November 21th, 2020. The Newcastle-Ottawa Scale (NOS) was used for quality assessment. A Statistical analysis was carried out using Review Manager 5.3 and STATA 15.1. RESULTS: Nine clinical trials were included in this study involving 2577 subjects. The results indicate that the association between elevated HbA1c referred as a continuous variable and adverse prognosis of COVID-19 was not significant (OR, 1.02; 95%CI, 0.95-1.09). However, higher HbA1c levels regarded as a dichotomous variable contributed to an increase mortality of COVID-19 (OR, 2.300; 95%CI, 1.679-3.150). Results were stable in a sensitivity analysis. More studies are needed to demonstrate the effect of HbA1c on hospital mortality. CONCLUSION: Prolonged uncontrolled hyperglycemia increases the risk of adverse prognosis in COVID-19. Patients with higher HbA1c should be monitored strictly to minimize the risk of adverse prognosis in COVID-19.

Novel nomogram for predicting the 3-year incidence risk of osteoporosis in a Chinese male population.

OBJECTIVE: To establish a rapid, cost-effective, accurate, and acceptable osteoporosis (OP) screening model for the Chinese male population (age ≥ 40 years) based on data mining technology. MATERIALS AND METHODS: This was a 3-year retrospective cohort study, which belonged to the sub-cohort of the Chinese Reaction Study. The research period was from March 2011 to December 2014. A total of 1834 subjects who did not have OP at the baseline and completed a 3-year follow-up were included in this study. All subjects underwent quantitative ultrasound examinations for calcaneus at the baseline and follow-ups that lasted for 3 years. We utilized the least absolute shrinkage and selection operator (LASSO) regression model to select feature variables. The characteristic variables selected in the LASSO regression were analyzed by multivariable logistic regression (MLR) to construct the predictive model. This predictive model was displayed through a nomogram. We used the receiver operating characteristic (ROC) curve, C-index, calibration curve, and clinical decision curve analysis (DCA) to evaluate model performance and the bootstrapping validation to internally validate the model. RESULTS: The predictive factors included in the prediction model were age, neck circumference, waist-to-height ratio, BMI, triglyceride, impaired fasting glucose, dyslipidemia, osteopenia, smoking history, and strenuous exercise. The area under the ROC (AUC) curve of the risk nomogram was 0.882 (95% CI, 0.858-0.907), exhibiting good predictive ability and performance. The C-index for the risk nomogram was 0.882 in the prediction model, which presented good refinement. In addition, the nomogram calibration curve indicated that the prediction model was consistent. The DCA showed that when the threshold probability was between 1 and 100%, the nomogram had a good clinical application value. More importantly, the internally verified C-index of the nomogram was still very high, at 0.870. CONCLUSIONS: This novel nomogram can effectively predict the 3-year incidence risk of OP in the male population. It also helps clinicians to identify groups at high risk of OP early and formulate personalized intervention measures.

Impact and efficacy of mobile health intervention in the management of diabetes and hypertension: a systematic review and meta-analysis.

With the continuous development of science and technology, mobile health (mHealth) intervention has been proposed as a treatment strategy for managing chronic diseases. In some developed countries, mHealth intervention has been proven to remarkably improve both the quality of care for patients with chronic illnesses and the clinical outcomes of these patients. However, the effectiveness of mHealth in developing countries remains unclear. Based on this fact, we conducted this systematic review and meta-analysis to evaluate the impact of mHealth on countries with different levels of economic development. To this end, we searched Pubmed, ResearchGate, Embase and Cochrane databases for articles published from January 2008 to June 2019. All of the studies included were randomized controlled trials. A meta-analysis was performed using the Stata software. A total of 51 articles (including 13 054 participants) were eligible for our systematic review and meta-analysis. We discovered that mHealth intervention did not only play a major role in improving clinical outcomes compared with conventional care, but also had a positive impact on countries with different levels of economic development. More importantly, our study also found that clinical outcomes could be ameliorated even further by combining mHealth with human intelligence rather than using mHealth intervention exclusively. According to our analytical results, mHealth intervention could be used as a treatment strategy to optimize the management of diabetes and hypertension in countries with different levels of economic development.

Risk factors for 2019 novel coronavirus disease (COVID-19) patients progressing to critical illness: a systematic review and meta-analysis.

IMPORTANCE: With the rising number of COVID-19 cases, global health resources are strained by the pandemic. No proven effective therapies or vaccines for this virus are currently available. In order to maximize the use of limited medical resources, distinguishing between mild and severe patients as early as possible has become pivotal. OBJECTIVE: To systematically review evidence for the risk factors of COVID-19 patients progressing to critical illness. EVIDENCE REVIEW: We conducted a comprehensive search for primary literature in both Chinese and English electronic bibliographic databases. The American agency for health research and quality tool was used for quality assessment. A meta-analysis was undertaken using STATA version 15.0. RESULTS: Twenty articles (4062 patients) were eligible for this systematic review and meta-analysis. First and foremost, we observed that elderly male patients with a high body mass index, high breathing rate and a combination of underlying diseases (such as hypertension, diabetes, cardiovascular disease, and chronic obstructive pulmonary disease) were more likely to develop severe COVID-19 infections. Second, compared with non-severe patients, severe patients had more serious symptoms such as fever and dyspnea. Besides, abnormal laboratory tests were more prevalent in severe patients than in mild cases, such as elevated levels of white blood cell counts, liver enzymes, lactate dehydrogenase, creatine kinase, C-reactive protein and procalcitonin, as well as decreased levels of lymphocytes and albumin. INTERPRETATION: This is the first systematic review exploring the risk factors for severe illness in COVID-19 patients. Our study may be helpful for clinical decision-making and optimizing resource allocation.