Patterns and Related Influencing Factors of Multiple Chronic Diseases Among the Elderly Based on Data Mining in Shanxi, China

Background: The presence of multiple chronic diseases has become a public health issue worldwide. The study aimed to explore the patterns and related influencing factors of multiple chronic diseases among the elderly so as to provide a scientific reference for the health management strategy, which is dedicated to the prevention and intervention for multiple chronic diseases. Methods: A cross-sectional study, which used a multi-stage random sampling way, was conducted among 3266 elderly aged 60 years and above in Shanxi. A chi square test was used to compare the sociodemographic factors. The data mining methods, including association rule and decision tree, were performed to explore the patterns and related influencing factors of multiple chronic diseases in the elderly.Results: The prevalence of multiple chronic diseases was 30.31%. According to the results of the association rules, there were 10 association patterns screened out. The first five multiple chronic disease patterns were atherosclerosis and hypertension, coronary heart disease and hypertension, diabetes and hypertension, stroke and hypertension, eye disease and hypertension. The decision tree model selected 5 related influencing factors of multiple chronic diseases, including age, smoking, family genetic history, fresh fruit intake frequency and education level, simultaneously generating 9 influence rules.Conclusions: Multiple chronic diseases were prevalent among the elderly, which patterns mainly consisted of two or three chronic diseases. Regarding the unmodifiable related influencing factors, including an older age, a family genetic history, and low education level, relevant departments should help the elderly to establish prevention knowledge and raise awareness of strongly related chronic diseases, so as to reduce the incidence and adverse consequences.