Medicare Payments Analysis Through an Adaptive Neural Fuzzy Inference System

Abstract

There has been a great disparity in payments between different hospitals over the same diagnosis. This paper aims to identify cost change patterns for patients who are covered by Medicare and to reveal the hidden structures about costs for the same diagnosis and treatments from different healthcare providers. It deals with the study of an Adaptive Neural Fuzzy Inference System for Medicare payment data in order to understand these variations in hospital payments. Clustering algorithms have been utilized in order to identify the payment differences and reveal the hidden structures that make the amounts vary. Experiment results show that cost change patterns were clearly understood using hierarchical clustering algorithms.