Xuefeng Shan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xuefeng Shan has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Xin Wang
Department of Epidemiology and Biostatistics, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
2 shared publications Yang Yang
Health Management Center, The First Affiliated Hospital of Chongqing Medical University, Chongqing, 400016, China
2 shared publications- Ben Zhang
Department of Epidemiology and Health Statistics, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
1 shared publication - Chao Xia
Chongqing Public Health Medical Center, Chongqing, China
1 shared publication Chi Zhang
Department of Prevention, Tianjin Medical University Cancer Institute and Hospital, Tianjin, China
1 shared publication- Dongqing Gu
Department of Infectious Diseases, First Affiliated Hospital, Army Medical University, Chongqing, China
1 shared publication
Latest Publications
- Open Access
- 3 June 2026
- Published by Springer Nature in International Journal of Clinical Pharmacy
- Vol. 48 (4) , 1660-1671
Development and validation of a machine learning-based clinical decision support tool for stratifying intravenous medication risk in hospitalized patients with heart failure
Hospitalized patients with heart failure (HF) frequently receive multiple high-risk intravenous (IV) medications, placing them at a substantial risk of clinically significant drug-related problems (DRPs). Timely severity-based stratification of IV medication-related risks remains challenging, particularly in the context of complex regimens and organ dysfunction. To develop and externally validate a clinically applicable machine learning-based stratification tool for classifying IV medication-related risk severity in hospitalized patients with HF and to support pharmacist-led medication safety management through a web-based clinical decision support tool. This multicenter retrospective study included 1,884 adult patients hospitalized with HF from seven tertiary hospitals, with an independent external validation cohort of 100 patients. IV medication-related DRPs were identified and classified by senior clinical pharmacists using the Pharmaceutical Care Network Europe (PCNE) DRP classification (version 9.1) and stratified by severity using the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) index. Medication risk severity (levels C-F) was used as a multiclass stratification outcome. Candidate clinical and medication-related variables reflecting patient characteristics, organ function, and medication burden were selected using least absolute shrinkage and selection operator (LASSO) regression, and six machine learning algorithms for multiclass risk stratification were developed and compared. Overall, 1405 patients (74.6%) experienced at least one IV medication-related DRP, with treatment safety problems predominating. Eleven clinically interpretable predictors, including neutrophil percentage, fibrinogen, serum albumin, and creatinine clearance were retained in the final models. In internal validation, the random forest (RF) model achieved the highest discriminative performance (AUC = 0.934), whereas in external validation, the artificial neural network (ANN) demonstrated the best performance (AUC = 0.921). Considering its consistent performance across datasets, ANN was selected as the final model, achieving AUC of 0.889 and 0.921 in the internal and external validation, respectively. Based on this model, a web-based clinical decision support tool was developed to provide individualized IV medication risk stratification at the point-of-care. A machine learning-based clinical decision support tool that incorporates routinely available clinical and medication-related variables can accurately stratify IV medication-related risk severity in hospitalized patients with HF.
- 3 June 2026
- Published by Springer Nature in International Journal of Clinical Pharmacy
- Vol. 48 (4) , 1660-1671
Open AccessDevelopment and validation of a machine learning-based clinical decision support tool for stratifying intravenous medication risk in hospitalized patients with heart failure
Hospitalized patients with heart failure (HF) frequently receive multiple high-risk intravenous (IV) medications, placing them at a substantial risk of clinically significant drug-related problems (DRPs). Timely severity-based stratification of IV medication-related risks remains challenging, particularly in the context of complex regimens and organ dysfunction. To develop and externally validate a clinically applicable machine learning-based stratification tool for classifying IV medication-related risk severity in hospitalized patients with HF and to support pharmacist-led medication safety management through a web-based clinical decision support tool. This multicenter retrospective study included 1,884 adult patients hospitalized with HF from seven tertiary hospitals, with an independent external validation cohort of 100 patients. IV medication-related DRPs were identified and classified by senior clinical pharmacists using the Pharmaceutical Care Network Europe (PCNE) DRP classification (version 9.1) and stratified by severity using the National Coordinating Council for Medication Error Reporting and Prevention (NCC MERP) index. Medication risk severity (levels C-F) was used as a multiclass stratification outcome. Candidate clinical and medication-related variables reflecting patient characteristics, organ function, and medication burden were selected using least absolute shrinkage and selection operator (LASSO) regression, and six machine learning algorithms for multiclass risk stratification were developed and compared. Overall, 1405 patients (74.6%) experienced at least one IV medication-related DRP, with treatment safety problems predominating. Eleven clinically interpretable predictors, including neutrophil percentage, fibrinogen, serum albumin, and creatinine clearance were retained in the final models. In internal validation, the random forest (RF) model achieved the highest discriminative performance (AUC = 0.934), whereas in external validation, the artificial neural network (ANN) demonstrated the best performance (AUC = 0.921). Considering its consistent performance across datasets, ANN was selected as the final model, achieving AUC of 0.889 and 0.921 in the internal and external validation, respectively. Based on this model, a web-based clinical decision support tool was developed to provide individualized IV medication risk stratification at the point-of-care. A machine learning-based clinical decision support tool that incorporates routinely available clinical and medication-related variables can accurately stratify IV medication-related risk severity in hospitalized patients with HF.
- Open Access
- 30 March 2023
- Published by Frontiers Media SA in Frontiers in Public Health
- Vol. 11
Prevalence of nonalcoholic fatty liver disease and the related risk factors among healthy adults: A cross-sectional study in Chongqing, China
Background: Epidemiological characteristics of nonalcoholic fatty liver disease (NAFLD) in Chongqing, a west-central city of China, remain unclear. The objective of this study was to investigate the prevalence of NAFLD and the related risk factors among healthy adults for physical examination in Chongqing.Methods: A total of 110,626 subjects were enrolled in the present study. Each of the participants underwent physical examination, laboratory measurements, and abdominal ultrasonography. The chi-square test was employed to compare differences in the NAFLD prevalence, and logistic regression analysis was used to estimate the odds ratio for risk factors of NAFLD.Results: The prevalence of NAFLD in individuals in the population of Chongqing was 28.5%, and the prevalence in men (38.1%) was significantly higher than that in women (13.6%) (OR = 2.44; 95% CI: 2.31–2.58). NAFLD was more common in men aged 51–60 years and women over 60 years. Approximately 79.1% of the people with obesity and 52.1% of the people with central obesity had NAFLD. The prevalence of NAFLD in people with hypertension and cholelithiasis was 48.9 and 38.4%, respectively. Logistic regression showed that gender, age, body max index (BMI), central obesity, hypertension, impaired fasting glucose/diabetes mellitus (DM), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), hyperuricemia (HUA), alanine transaminase (ALT), and cholelithiasis were independently associated with the presence of NAFLD.Conclusion: The prevalence of NAFLD among healthy adults in Chongqing was high. To improve the prevention and management of NAFLD, special attention should be paid to the factors associated with the presence of NAFLD, including higher BMI, higher waist circumference, higher blood glucose, hypertension, hypertriglyceridemia, hyperuricemia, cholelithiasis, and elevated ALT.
- 30 March 2023
- Published by Frontiers Media SA in Frontiers in Public Health
- Vol. 11
Open AccessPrevalence of nonalcoholic fatty liver disease and the related risk factors among healthy adults: A cross-sectional study in Chongqing, China
Background: Epidemiological characteristics of nonalcoholic fatty liver disease (NAFLD) in Chongqing, a west-central city of China, remain unclear. The objective of this study was to investigate the prevalence of NAFLD and the related risk factors among healthy adults for physical examination in Chongqing.Methods: A total of 110,626 subjects were enrolled in the present study. Each of the participants underwent physical examination, laboratory measurements, and abdominal ultrasonography. The chi-square test was employed to compare differences in the NAFLD prevalence, and logistic regression analysis was used to estimate the odds ratio for risk factors of NAFLD.Results: The prevalence of NAFLD in individuals in the population of Chongqing was 28.5%, and the prevalence in men (38.1%) was significantly higher than that in women (13.6%) (OR = 2.44; 95% CI: 2.31–2.58). NAFLD was more common in men aged 51–60 years and women over 60 years. Approximately 79.1% of the people with obesity and 52.1% of the people with central obesity had NAFLD. The prevalence of NAFLD in people with hypertension and cholelithiasis was 48.9 and 38.4%, respectively. Logistic regression showed that gender, age, body max index (BMI), central obesity, hypertension, impaired fasting glucose/diabetes mellitus (DM), triglyceride (TG), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), hyperuricemia (HUA), alanine transaminase (ALT), and cholelithiasis were independently associated with the presence of NAFLD.Conclusion: The prevalence of NAFLD among healthy adults in Chongqing was high. To improve the prevention and management of NAFLD, special attention should be paid to the factors associated with the presence of NAFLD, including higher BMI, higher waist circumference, higher blood glucose, hypertension, hypertriglyceridemia, hyperuricemia, cholelithiasis, and elevated ALT.
- Open Access
- 2 March 2022
- Published by Springer Nature in Lipids in Health and Disease
- Vol. 21 (1) , 1-12
Blood lipid metabolism and the risk of gallstone disease: a multi-center study and meta-analysis
Background: Gallstone disease (GSD) is a common and costly biliary disorder. Multiple studies have investigated the associations between blood lipid metabolism and GSD risk; however, the results are inconsistent. This research aimed to comprehensively evaluate the relationships among serum total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides and GSD risk. Methods: Firstly, a multi-center cross-sectional study was carried out. Subjects who participated in the health examination in three hospitals between January 2015 and May 2020 were recruited. Multivariable logistic regression was used to investigate blood lipid metabolism associated with GSD risk. Then, a meta-analysis was performed to verify the associations further. Medline and Embase databases were systematically searched before June 10, 2021. The DerSimonian and Laird random-effect model was utilized when the heterogeneity was high; otherwise, fixed-effect model was adopted. Results: There were 548,934 eligible participants included in the multi-center study, and 45,392 of them were diagnosed with GSD. The results demonstrated that total cholesterol and HDL cholesterol were negatively associated with GSD risk in both high vs. low model and per mmol/L increase model, while triglyceride was positively associated with GSD risk in the per unit increase model. In the meta-analysis, 104 studies with approximately 3 million participants were finally included. The results verified that HDL cholesterol [odds ratio (OR) = 0.636, P = 5.97 × 10− 16 in high vs low model; OR = 0.974, P = 6.07 × 10− 05 in per unit model] and triglyceride (OR = 1.192, P = 3.47 × 10− 05 in high vs. low model; OR = 1.011, P = 5.12 × 10− 05 in per unit model) were related to GSD risk in the two models. Conclusions: The findings indicated that low HDL cholesterol levels and high triglyceride levels were risk factors for GSD. This study provides a basis for identifying the population at high risk for GSD and implementing tertiary prevention strategies for GSD, thus contributing to GSD prevention as well as disease burden relief.
- 2 March 2022
- Published by Springer Nature in Lipids in Health and Disease
- Vol. 21 (1) , 1-12
Open AccessBlood lipid metabolism and the risk of gallstone disease: a multi-center study and meta-analysis
Background: Gallstone disease (GSD) is a common and costly biliary disorder. Multiple studies have investigated the associations between blood lipid metabolism and GSD risk; however, the results are inconsistent. This research aimed to comprehensively evaluate the relationships among serum total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides and GSD risk. Methods: Firstly, a multi-center cross-sectional study was carried out. Subjects who participated in the health examination in three hospitals between January 2015 and May 2020 were recruited. Multivariable logistic regression was used to investigate blood lipid metabolism associated with GSD risk. Then, a meta-analysis was performed to verify the associations further. Medline and Embase databases were systematically searched before June 10, 2021. The DerSimonian and Laird random-effect model was utilized when the heterogeneity was high; otherwise, fixed-effect model was adopted. Results: There were 548,934 eligible participants included in the multi-center study, and 45,392 of them were diagnosed with GSD. The results demonstrated that total cholesterol and HDL cholesterol were negatively associated with GSD risk in both high vs. low model and per mmol/L increase model, while triglyceride was positively associated with GSD risk in the per unit increase model. In the meta-analysis, 104 studies with approximately 3 million participants were finally included. The results verified that HDL cholesterol [odds ratio (OR) = 0.636, P = 5.97 × 10− 16 in high vs low model; OR = 0.974, P = 6.07 × 10− 05 in per unit model] and triglyceride (OR = 1.192, P = 3.47 × 10− 05 in high vs. low model; OR = 1.011, P = 5.12 × 10− 05 in per unit model) were related to GSD risk in the two models. Conclusions: The findings indicated that low HDL cholesterol levels and high triglyceride levels were risk factors for GSD. This study provides a basis for identifying the population at high risk for GSD and implementing tertiary prevention strategies for GSD, thus contributing to GSD prevention as well as disease burden relief.
- Open Access
- 28 August 2021
- Published by Springer Nature in BMC Endocrine Disorders
- Vol. 21 (1) , 1-11
Prevalence and associated metabolic factors for thyroid nodules: a cross-sectional study in Southwest of China with more than 120 thousand populations
Objective: To explore the prevalence and its associated metabolic factors of thyroid nodules (TNs) among subjects who participated in the physical examinations in Chongqing, China. Methods: The participants from the Health Management Center of JinShan Hospital of Chongqing Medical University, between September 2015 and May 2020, were included in this study. All of the participants underwent thyroid ultrasonography, anthropometric measurements, and serum examinations. Differences in the TNs prevalence were compared with the chi-square test or Wilcoxon rang-sum test. Multivariable logistic regression analyses were used to estimate the metabolic factors associated with TNs and multiple thyroid nodules (MTNs). Results: Of the included 121,702 participants, 41,547 had TNs, and 20,899 had MTNs, with the prevalence of 34.1 and 17.0 %, respectively. Women had a significantly higher prevalence of TNs than men (40.6 % vs. 29.8 %; χ2 = 1517.33, P < 0.001), and TNs prevalence was gradually increased with age (P for trend < 0.001). Female gender, advanced age, and metabolic factors including central obesity, hypertension, diabetes and fatty liver were positively associated with TNs; BMI, hyperlipoidemia and hyperuricemia were not independent risk factors of TNs. While female gender, advanced age, central obesity, hypertension and diabetes were independent risk factors of MTNs. Conclusions: The prevalence of thyroid nodules was relatively high. The associated factors identified in this study could help the clinicians to detect the high-risk patients and make targeted screening strategies for the preventing of the occurrence of TNs.
- 28 August 2021
- Published by Springer Nature in BMC Endocrine Disorders
- Vol. 21 (1) , 1-11
Open AccessPrevalence and associated metabolic factors for thyroid nodules: a cross-sectional study in Southwest of China with more than 120 thousand populations
Objective: To explore the prevalence and its associated metabolic factors of thyroid nodules (TNs) among subjects who participated in the physical examinations in Chongqing, China. Methods: The participants from the Health Management Center of JinShan Hospital of Chongqing Medical University, between September 2015 and May 2020, were included in this study. All of the participants underwent thyroid ultrasonography, anthropometric measurements, and serum examinations. Differences in the TNs prevalence were compared with the chi-square test or Wilcoxon rang-sum test. Multivariable logistic regression analyses were used to estimate the metabolic factors associated with TNs and multiple thyroid nodules (MTNs). Results: Of the included 121,702 participants, 41,547 had TNs, and 20,899 had MTNs, with the prevalence of 34.1 and 17.0 %, respectively. Women had a significantly higher prevalence of TNs than men (40.6 % vs. 29.8 %; χ2 = 1517.33, P < 0.001), and TNs prevalence was gradually increased with age (P for trend < 0.001). Female gender, advanced age, and metabolic factors including central obesity, hypertension, diabetes and fatty liver were positively associated with TNs; BMI, hyperlipoidemia and hyperuricemia were not independent risk factors of TNs. While female gender, advanced age, central obesity, hypertension and diabetes were independent risk factors of MTNs. Conclusions: The prevalence of thyroid nodules was relatively high. The associated factors identified in this study could help the clinicians to detect the high-risk patients and make targeted screening strategies for the preventing of the occurrence of TNs.
- Open Access
- 14 August 2020
- Published by Cold Spring Harbor Laboratory in medRxiv
Clinical characteristics of COVID-19 and the model for predicting the occurrence of critically ill patients: a retrospective cohort study
Background: The present study aim to comprehensively report the epidemiological and clinical characteristics of the COVID-19 patients and to develop a multi-feature fusion model for predicting the critical ill probability.Methods: It was a retrospective cohort study that incorporating the laboratory-confirmed COVID-19 patients in the Chongqing Public Health Medical Center. The prediction model was constructed with least absolute shrinkage and selection operator (LASSO) logistic regression method and the model was further tested in the validation cohort. The performance was evaluated by the receiver operating curve (ROC), calibration curve and decision curve analysis (DCA).Results: A total of 217 patients were included in the study. During the treatment, 34 patients were admitted to intensive care unit (ICU) and no developed death. A model incorporating the demographic and clinical characteristics, imaging features and laboratory findings were constructed to predict the critical ill probability and it was proved to have good calibration, discrimination ability and clinic use.Conclusions: The prevalence of critical ill was relatively high and the model may help the clinicians to identify the patients with high risk for developing the critical ill, thus to conduct timely and targeted treatment to reduce the mortality rate.
- 14 August 2020
- Published by Cold Spring Harbor Laboratory in medRxiv
Open AccessClinical characteristics of COVID-19 and the model for predicting the occurrence of critically ill patients: a retrospective cohort study
Background: The present study aim to comprehensively report the epidemiological and clinical characteristics of the COVID-19 patients and to develop a multi-feature fusion model for predicting the critical ill probability.Methods: It was a retrospective cohort study that incorporating the laboratory-confirmed COVID-19 patients in the Chongqing Public Health Medical Center. The prediction model was constructed with least absolute shrinkage and selection operator (LASSO) logistic regression method and the model was further tested in the validation cohort. The performance was evaluated by the receiver operating curve (ROC), calibration curve and decision curve analysis (DCA).Results: A total of 217 patients were included in the study. During the treatment, 34 patients were admitted to intensive care unit (ICU) and no developed death. A model incorporating the demographic and clinical characteristics, imaging features and laboratory findings were constructed to predict the critical ill probability and it was proved to have good calibration, discrimination ability and clinic use.Conclusions: The prevalence of critical ill was relatively high and the model may help the clinicians to identify the patients with high risk for developing the critical ill, thus to conduct timely and targeted treatment to reduce the mortality rate.
- Open Access
- 11 January 2019
- Published by Springer Nature in Cancer Cell International
- Vol. 19 (1) , 13
Rosiglitazone metformin adduct inhibits hepatocellular carcinoma proliferation via activation of AMPK/p21 pathway
Rosiglitazone metformin adduct (RZM) is a novel compound, synthesized from rosiglitazone (Ros) and metformin (Met) combined at a molar mass ratio of 1:1. Met and Ros are widely used together for treating type 2 diabetes to improve drug effectiveness and reduce adverse drug reactions. Recent studies reported that both Met and Ros may possess antineoplastic properties in several cancers, including hepatocellular carcinoma (HCC). However, the effects of RZM in HCC and its underlying mechanisms remain unknown. RZM was synthesized from Ros and Met at an equal molar ratio and identified by infrared spectroscopy. MTS and colony formation assays were performed to detect proliferative repression of RZM, the mixture, Met and Ros, respectively. Tumorigenesis assay in vivo was used to confirm the anti-tumorigenesis potential of RZM and Met. Moreover, cellular apoptosis caused by RZM was analyzed by hoechst staining assay and flow cytometry. RT-qPCR and western blotting were performed to reveal mechanisms for the function of RZM. Both in vitro and in vivo data showed that low doses of RZM enhanced inhibitory effect on HCC cells growth compared with Met. Flow cytometry analysis confirmed that treatment with RZM at 1 mM for 48 h triggered HCC cells apoptosis. RT-qPCR and western blotting analyses showed that p21 was upregulated in response to 1 mM RZM treatment. Furthermore, RZM could increase AMPK activation compared with Met. The increased p21 expression induced by RZM treatment was attenuated by an AMPK inhibitor compound C. All these observations demonstrate that RZM increases the antiproliferative effect of Met in HCC via upregulating p21 expression in an AMPK-dependent manner. Our results suggest that RZM has the potential to be an adjuvant for HCC therapy.
- 11 January 2019
- Published by Springer Nature in Cancer Cell International
- Vol. 19 (1) , 13
Open AccessRosiglitazone metformin adduct inhibits hepatocellular carcinoma proliferation via activation of AMPK/p21 pathway
Rosiglitazone metformin adduct (RZM) is a novel compound, synthesized from rosiglitazone (Ros) and metformin (Met) combined at a molar mass ratio of 1:1. Met and Ros are widely used together for treating type 2 diabetes to improve drug effectiveness and reduce adverse drug reactions. Recent studies reported that both Met and Ros may possess antineoplastic properties in several cancers, including hepatocellular carcinoma (HCC). However, the effects of RZM in HCC and its underlying mechanisms remain unknown. RZM was synthesized from Ros and Met at an equal molar ratio and identified by infrared spectroscopy. MTS and colony formation assays were performed to detect proliferative repression of RZM, the mixture, Met and Ros, respectively. Tumorigenesis assay in vivo was used to confirm the anti-tumorigenesis potential of RZM and Met. Moreover, cellular apoptosis caused by RZM was analyzed by hoechst staining assay and flow cytometry. RT-qPCR and western blotting were performed to reveal mechanisms for the function of RZM. Both in vitro and in vivo data showed that low doses of RZM enhanced inhibitory effect on HCC cells growth compared with Met. Flow cytometry analysis confirmed that treatment with RZM at 1 mM for 48 h triggered HCC cells apoptosis. RT-qPCR and western blotting analyses showed that p21 was upregulated in response to 1 mM RZM treatment. Furthermore, RZM could increase AMPK activation compared with Met. The increased p21 expression induced by RZM treatment was attenuated by an AMPK inhibitor compound C. All these observations demonstrate that RZM increases the antiproliferative effect of Met in HCC via upregulating p21 expression in an AMPK-dependent manner. Our results suggest that RZM has the potential to be an adjuvant for HCC therapy.