Research On The Spatial Distribution Characteristics Of Arthritis In Middle-aged And Old People In China And The Risk Assessment Model Of Arthritis Based On Influencing Factors

Arthritis is a public health problem of global concern.As the most common chronic disease,arthritis disease has a high incidence in China and even the world.At the same time,arthritis also shows the characteristics of increasing incidence with increasing age.In recent years,due to the gradual increase in the number of middle-aged and older adults in my country,the incidence of arthritis in the middle-aged and elderly is also rising.Arthritis poses a significant threat to the health,safety,and stability of middle-aged and older adults.Therefore,the current status,risk,and related risk factors of arthritis in middle-aged and older adults deserve special attention.Due to the severe incidence of arthritis in middle-aged and older adults in my country,many significant problems are to be solved.Traditional statistical research methods are no longer able to meet the research needs.Therefore,this paper uses2011,2013,and 2015 China Health and Elderly Follow-up Surveys The middle-aged and older adults in the data(China Health and Retirement Longitudinal Study,CHARLS)are research samples,combined with spatial epidemiology,a variety of spatial information technologies and machine learning algorithms,to analyze the evolution of arthritis among middle-aged spatially and older adults in my country.In-depth study of its spatial distribution characteristics,spatial autocorrelation,development trends,and spatial aggregation patterns to understand and explore the spatial evolution and epidemiological aspects of arthritis.Then,use correlation analysis to explore the risk factors related to the onset of arthritis.Then,multivariate logistic regression analysis was used to determine the risk factors related to arthritis onset.Finally,statistically significant risk factors are incorporated into machine learning algorithms such as the Logistic regression model and random forest model to construct a risk assessment model that considers not only the impact of risk factors but also reflects the spatial characteristics of arthritis and selects the best assessment results from them the model for risk assessment and verification of the onset of arthritis.The main conclusions of the study are as follows:(1)In 2011 and 2013,the overall incidence of arthritis reached the highest value in the 65-69 age group.Women are also most likely to develop the disease in the65-69 age group,while men are most vulnerable in the 70-74 age group.In 2015,the overall incidence of arthritis,the incidence of arthritis in women,and arthritis incidence in men reached the highest values in the 70-74 age group.Regardless of the age group,the incidence of arthritis in middle-aged and older adults is lower than the incidence of arthritis in middle-aged and older women.(2)The high-risk areas of arthritis in middle-aged and older adults are mainly concentrated in the southwest.Moreover,there is apparent spatial autocorrelation in the spatial distribution of arthritis cases in 2011,2013,and 2015.Besides,in the local spatial autocorrelation,the regions with HH clustering characteristics are mainly concentrated in the northwest,southwest,south,and central areas of China.On the contrary,the regions with LL clustering characteristics are mainly concentrated in North China,East China,Southeast China,and Northeast China.Trend surface analysis also shows that arthritis in middle-aged and older adults in the southwest is more serious.(3)The spatial scan identified three spatial clusters of arthritis(P<0.01).The first type of agglomeration area includes 22 locations with a radius of 749.69 km,mainly distributed in China’s Sichuan Province(10 locations),Yunnan Province(6 locations),Guizhou Province(2 locations),and Chongqing City(1 location),Guangxi Province(1 location),Hubei Province(1 location),and Shaanxi Province(1 location).The second cluster contains 7 locations,mainly in China’s Jiangxi Province(1 location)and Hunan Province(6 locations).The third cluster contains 3 locations,all concentrated in Anhui Province.(4)Multivariate logistic regression analysis results show that lung disease,heart disease,stroke,kidney disease,liver disease,digestive system disease,gender,education level,BMI,CES-D,age,altitude,geographic distribution,and average temperature are all related to middle-aged and elderly people.The onset of human arthritis is closely related.(5)Multivariate Logistic regression model and random forest model were used to model and evaluate arthritis incidence in the elderly in China.Through comparison,it is found that the sensitivity,evaluation accuracy,AUC,and Youden index of the random forest risk assessment model are much higher than the multi-factor Logistic regression risk assessment model.It shows that the random forest risk assessment model’s assessment accuracy is higher than that of the multi-factor Logistic regression risk assessment model,and the overall classification effect is also better than that of the Logistic regression risk assessment model.Therefore,the random forest risk assessment model has achieved better modeling and assessment results.(6)At present,the epidemic situation of arthritis disease in my country is relatively tricky.This study uses various research methods to conduct an in-depth exploration of arthritis disease’s spatial epidemiological characteristics and the risk of arthritis in my country’s middle-aged and older adults from multiple perspectives.It provides a scientific basis and information-based decision-making tools to help public health departments manage arthritis.