Predictive Analytics: A Case Study in Machine-Learning and Claims Databases

Date: 07/16/2019

Journal: The American Journal of Pharmacy Benefits

Date published: November/December 2016

Authors: David A. Kvancz, MS, RPh, FASHP; Marcus N. Sredzinski, PharmD; and Celynda G. Tadlock, PharmD, MBA


This study focused the power of modern analytics on hereditary angioedema (HAE), a single rare disease, because it exhibits features of diseases associated with high costs: rare, hard to diagnose, progressive, and takes a long time from diagnosis to appropriate treatment. Despite the availability of effective therapies, misdiagnoses and underdiagnosis of HAE result in significant burden to the healthcare system.


This study successfully demonstrated the ability of this state-of-the-art predictive analysis to find rare-disease patients in a large and complex database. This information could be valuable to claims managers and employers who may realize savings by helping physicians bring these patients to appropriate treatment sooner.