Identifying Compound Effect of Drugs on Rheumatoid Arthritis Treatment Based on the Association Rule and a Random Walking-Based Model

Abstract

Rheumatoid arthritis (RA) is a chronic autoimmune disorder that is diagnosed mainly on the basis of patient signs, symptoms, and laboratory indices. However, the exact causes of RA are unclear. Moreover, there is a lack of any method of dynamically evaluating the efficacy of the medication administered to treat RA. Here, we applied a random walk model to reveal the compatibility among the various constituents of traditional Chinese medicine and evaluate their therapeutic efficacy against RA. Drugs commonly used to treat RA were investigated using cluster analysis. The association rule analysis was applied to identify compatibilities among the constituents. A random walk model was developed to evaluate drug efficacy based on an in-house database comprising the clinical records of 9,408 RA patients. Frequently administered medicines were combined into three correlated sets. The evaluation based on the random walk method showed that the drug combination improved ESR, CRP, C3, C4, and IgA more effectively than any single drug. The present study demonstrated that the TCM constituents complement each other and various combinations of them produce different therapeutic effects on RA treatment.