Diabetes Control in China

Abstract

In 2016, Healthy China 2030 has been announced and with it its five specific goals: controlling major risk factors, increasing the capacity of the health service, enlarging the scale of the health industry, perfecting the health service system and improving the health nationwide. Nevertheless, Diabetes continues to be a leading public health challenge in China. In this thesis, the author explores reasons for the rapid diabetes prevalence, as well as how a simple supervised machine learning model build in Python, based on the China Health and Retirement Longitudinal Study (CHARLS), can predict the risk of a Chinese citizen aged 45 and above having diabetes. Three different algorithms, Random Forest, Support Vector Machine and Logistic Regression are compared. The model is built with the help of SciKit-learn and imbalanced-learn. Findings obtained are correlations between diabetes and dyslipidemia, as well as correlations among diabetes and education level among other things. The most suited algorithm for prediction is Support Vector Machine, after introducing over-sampling. Generally, the findings demonstrate that the blueprint Healthy China 2030 is a thought-out and needed strategy change.