Using kernel density function to evaluate the health risk from multi-diseases

Abstract

Many previous studies have explored the risk factors and the prediction model for a specific disease. However, the focus of the preventive care is not only treating a disease, but also how to maintain health. This study developed a health risk index for the prevention of multi-diseases of the healthy people. The kernel density technique was proposed to estimate the distribution of the common risk factors of multi-diseases and to define the health risk index. A data set of the hypertension, hyperlipidemia and hyperglycemia from National Health Insurance Research Database in Taiwan was used to explain the proposed analysis process. The five common risk factors include Fasting plasma glucose (FPG), T-CHO, Triglyceride (TG), SBP, DBP were used to calculate the health risk index. When FPG, T-CHO, TG, SBP and DBP average reduced 0.2 mg/dl, 5.2 mg/dl, 13.4 mg/dl, 5.3 mmHg and 3.7 mmHg, separately. The analysis showed that the subjects could reduce 7.29% of health risk to suffer from the three diseases.