1
|
Zhang Y, Zhang L, Lv H, Zhang G. Ensemble machine learning prediction of hyperuricemia based on a prospective health checkup population. Front Physiol 2024; 15:1357404. [PMID: 38665596 PMCID: PMC11043598 DOI: 10.3389/fphys.2024.1357404] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2023] [Accepted: 03/11/2024] [Indexed: 04/28/2024] Open
Abstract
Objectives: An accurate prediction model for hyperuricemia (HUA) in adults remain unavailable. This study aimed to develop a stacking ensemble prediction model for HUA to identify high-risk groups and explore risk factors. Methods: A prospective health checkup cohort of 40899 subjects was examined and randomly divided into the training and validation sets with the ratio of 7:3. LASSO regression was employed to screen out important features and then the ROSE sampling was used to handle the imbalanced classes. An ensemble model using stacking strategy was constructed based on three individual models, including support vector machine, decision tree C5.0, and eXtreme gradient boosting. Model validations were conducted using the area under the receiver operating characteristic curve (AUC) and the calibration curve, as well as metrics including accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. A model agnostic instance level variable attributions technique (iBreakdown) was used to illustrate the black-box nature of our ensemble model, and to identify contributing risk factors. Results: Fifteen important features were screened out of 23 clinical variables. Our stacking ensemble model with an AUC of 0.854, outperformed the other three models, support vector machine, decision tree C5.0, and eXtreme gradient boosting with AUCs of 0.848, 0.851 and 0.849 respectively. Calibration accuracy as well as other metrics including accuracy, specificity, negative predictive value, and F1 score were also proved our ensemble model's superiority. The contributing risk factors were estimated using six randomly selected subjects, which showed that being female and relatively younger, together with having higher baseline uric acid, body mass index, γ-glutamyl transpeptidase, total protein, triglycerides, creatinine, and fasting blood glucose can increase the risk of HUA. To further validate our model's applicability in the health checkup population, we used another cohort of 8559 subjects that also showed our ensemble prediction model had favorable performances with an AUC of 0.846. Conclusion: In this study, the stacking ensemble prediction model for HUA was developed, and it outperformed three individual models that compose it (support vector machine, decision tree C5.0, and eXtreme gradient boosting). The contributing risk factors were identified with insightful ideas.
Collapse
Affiliation(s)
- Yongsheng Zhang
- Health Management Center, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Institute of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Shandong Engineering Laboratory of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Li Zhang
- Department of Pharmacology, Jinan Central Hospital Affiliated to Shandong First Medical University, Jinan, China
| | - Haoyue Lv
- Health Management Center, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Institute of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Shandong Engineering Laboratory of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| | - Guang Zhang
- Health Management Center, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Institute of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
- Shandong Engineering Laboratory of Health Management, The First Affiliated Hospital of Shandong First Medical University and Shandong Provincial Qianfoshan Hospital, Jinan, China
| |
Collapse
|
2
|
Kurajoh M, Akari S, Nakamura T, Ihara Y, Imai T, Morioka T, Emoto M. Seasonal variations for newly prescribed urate-lowering drugs for asymptomatic hyperuricemia and gout in Japan. Front Pharmacol 2024; 15:1230562. [PMID: 38292940 PMCID: PMC10825023 DOI: 10.3389/fphar.2024.1230562] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/29/2023] [Accepted: 01/08/2024] [Indexed: 02/01/2024] Open
Abstract
Background: Urate-lowering drugs (ULDs) have been approved for treatment of asymptomatic hyperuricemia and gout in Japan. Although serum urate levels and rates of gout onset are known to have seasonal variations, no survey results regarding the seasonality of ULD prescriptions for asymptomatic hyperuricemia and gout have been reported. Methods: A large-scale database of medical claims in Japan filed between January 2019 and December 2022 was accessed. In addition to total size of the recorded population for each month examined, the numbers of patients every month with newly prescribed ULDs for asymptomatic hyperuricemia and gout were noted, based on the International Classification of Diseases, 10th Revision, codes E79.0 and M10. Results: The results identified 201,008 patients with newly prescribed ULDs (median age 49.0 years, male 95.6%). Of those, 64.0% were prescribed ULDs for asymptomatic hyperuricemia and 36.0% for gout. The proportion of new ULD prescriptions was seasonal, with that significantly (p < 0.001) higher in summer (June-August) [risk ratio (RR) 1.322, 95% CI 1.218 to 1.436] and autumn (September-November) (RR 1.227, 95% CI 1.129-1.335) than in winter (December-February), whereas the proportion in spring (March-May) was not significantly different from winter. There was no significant difference after stratification by drug type (uric acid production inhibitor/uricosuric agent) or size of the medical institution, nor subgrouping by age or sex (p for interaction = 0.739, 0.727, 0.886, and 0.978, respectively). On the other hand, the proportions of new ULD prescriptions for asymptomatic hyperuricemia were significantly lower and for gout significantly higher in spring than winter, while those were similar in summer and autumn for both groups (p for interaction<0.001). Conclusion: The present findings indicate that new prescriptions for ULDs to treat asymptomatic hyperuricemia or gout in Japan show seasonal differences, with higher rates noted in summer and autumn as compared to winter.
Collapse
Affiliation(s)
- Masafumi Kurajoh
- Department of Metabolism, Endocrinology and Molecular Medicine, Osaka Metropolitan University Graduate School of Medicine, Osaka, Japan
| | - Seigo Akari
- Medical Affairs Department, Sanwa Kagaku Kenkyusho Co., Ltd., Nagoya, Aichi, Japan
| | - Takashi Nakamura
- Medical Affairs Department, Sanwa Kagaku Kenkyusho Co., Ltd., Nagoya, Aichi, Japan
| | - Yasutaka Ihara
- Department of Medical Statistics, Osaka Metropolitan University Graduate School of Medicine, Osaka, Japan
| | - Takumi Imai
- Department of Medical Statistics, Osaka Metropolitan University Graduate School of Medicine, Osaka, Japan
| | - Tomoaki Morioka
- Department of Metabolism, Endocrinology and Molecular Medicine, Osaka Metropolitan University Graduate School of Medicine, Osaka, Japan
| | - Masanori Emoto
- Department of Metabolism, Endocrinology and Molecular Medicine, Osaka Metropolitan University Graduate School of Medicine, Osaka, Japan
| |
Collapse
|
3
|
Huang X, Chen X, Liu Q, Zhang Z, Miao J, Lai Y, Wu J. The relationship between education attainment and gout, and the mediating role of modifiable risk factors: a Mendelian randomization study. Front Public Health 2024; 11:1269426. [PMID: 38259784 PMCID: PMC10800502 DOI: 10.3389/fpubh.2023.1269426] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/03/2023] [Accepted: 12/14/2023] [Indexed: 01/24/2024] Open
Abstract
Objective To investigate the causal relationship between educational attainment (EA) and gout, as well as the potential mediating effects of individual physical status (IPS) such as body mass index (BMI) and systolic blood pressure (SBP) and lifestyle habits (LH) including alcohol intake frequency (drinking), current tobacco smoking (smoking), and time spent watching television (TV). Methods Utilizing two-sample Mendelian randomization (MR), we analyzed the causal effects of EA on gout risk, and of IPS (BMI and SBP) and LH (smoking, drinking, and TV time) on gout risk. Multivariable MR (MVMR) was employed to explore and quantify the mediating effects of IPS and LH on the causal relationship between EA and gout risk. Results An elevation of educational attainment by one standard deviation (4.2 years) exhibited a protective effect against gout (odds ratio 0.724, 95% confidence interval 0.552-0.950; p = 0.020). We did not observe a causal relationship between smoking and gout, but BMI, SBP, drinking, and TV time were found to be causal risk factors for gout. Moreover, BMI, SBP, drinking, and TV time acted as mediating factors in the causal relationship between EA and gout risk, explaining 27.17, 14.83, 51.33, and 1.10% of the causal effects, respectively. Conclusion Our study indicates that having a genetically predicted higher level of EA may provide protection against gout. We found that this relationship is influenced by IPS factors such as BMI and SBP, as well as LH including drinking and TV time.
Collapse
Affiliation(s)
- Xin Huang
- Department of Orthopedics, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Xin Chen
- Department of Urology, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Qixi Liu
- Department of Nursing, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Zhiwei Zhang
- Department of Orthopedics, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Juan Miao
- Department of Nursing, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Yuchan Lai
- Department of Nursing, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| | - Jinqing Wu
- Department of Orthopedics, Mindong Hospital Affiliated to Fujian Medical University, Fuan, Fujian Province, China
| |
Collapse
|
4
|
Čypienė A, Gimžauskaitė S, Rinkūnienė E, Jasiūnas E, Laucevičius A, Ryliškytė L, Badarienė J. Effect of Alcohol Consumption Habits on Early Arterial Aging in Subjects with Metabolic Syndrome and Elevated Serum Uric Acid. Nutrients 2023; 15:3346. [PMID: 37571284 PMCID: PMC10421141 DOI: 10.3390/nu15153346] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/04/2023] [Revised: 07/25/2023] [Accepted: 07/26/2023] [Indexed: 08/13/2023] Open
Abstract
BACKGROUND Hyperuricemia is perceived as one of the risk factors for developing and progressing cardiovascular disease and metabolic syndrome through various pathological mechanisms. Endogenous synthesis and exogenous factors such as diet and beverages consumed play a major role in determining serum uric acid (sUA) levels. The aim of this study was to evaluate the effect of alcohol consumption on early arterial aging in middle-aged patients with metabolic syndrome (MetS) and hyperuricemia. MATERIALS AND METHODS This study included 661 middle-aged subjects (241 men and 420 women) from the Lithuanian High Cardiovascular Risk (LitHiR) primary prevention program. Characteristics of subjects such as blood pressure, laboratory testing, and the specialized nutrition profile questionnaire were evaluated. As an early marker of arterial stiffness, carotid-femoral pulse wave velocity (cfPWV) was assessed using a non-invasive applanation tonometry technique. RESULTS Hyperuricemia was present in 29% of men and 34% of women. Hyperuricemic men reported 1.6 times higher rates of alcohol drinking compared to men with normal sUA levels. After analyzing the correlation between alcohol consumption and cfPWV, no statistically significant relationships were found at a significance level of α = 0.05 but lowering the significance level to 0.06 revealed significant associations in men with normal sUA (ε2ordinal = 0.05, p = 0.06) and in women with increased sUA levels (ε2ordinal = 0.05, p = 0.08). Regression analysis showed that hyperuricemic men, consuming more than one unit of alcohol per week, had a significant impact on increasing cfPWV, while men with normal sUA levels, abstaining from alcohol entirely, resulted in a statistically significant decrease in cfPWV. Our results showed statistically significant relationships only among a group of men, although the women in the hyperuricemic group had a statistically higher cfPWV than women with normal sUA levels. CONCLUSIONS Drinking alcohol is associated with increased arterial stiffness among hyperuricemic middle-aged men with MetS.
Collapse
Affiliation(s)
- Alma Čypienė
- State Research Institute Centre for Innovative Medicine, 08406 Vilnius, Lithuania; (A.Č.); (A.L.)
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| | - Silvija Gimžauskaitė
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| | - Egidija Rinkūnienė
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| | - Eugenijus Jasiūnas
- Center of Informatics and Development, Vilnius University Hospital Santaros Klinikos, 08661 Vilnius, Lithuania;
| | - Aleksandras Laucevičius
- State Research Institute Centre for Innovative Medicine, 08406 Vilnius, Lithuania; (A.Č.); (A.L.)
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| | - Ligita Ryliškytė
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| | - Jolita Badarienė
- Faculty of Medicine, Vilnius University, 03101 Vilnius, Lithuania; (E.R.); (L.R.); (J.B.)
| |
Collapse
|