Big Data Strategies for Enhancing Transparency in U.S. Healthcare Pricing

Abstract

The high price opacity of the United States healthcare persistently weighs down on consumers, as evidenced by studies indicating that more than 65% of patients are surprised by the cost of the treatment after getting care. Finally, it provides a solution to enhanced ‘transparency’ and cost predictability leveraging big data technologies. The analysis of a dataset comprising of more than 5 million hospital billing records from more than 1,200 US hospitals, obtained from the Centers for Medicare & Medicaid Services (CMS) data, in this research helps analyze how big data strategies can improve price transparency. The study employs a data mining technique, regression analysis and visualization tools to quantify price discrepancies exceeding 300% for identical procedures between different regions. Both predictive analytics models were useful in consumer facing pricing tools and were able to predict average procedure costs with an average 92% accuracy. The results verify that big data can revolutionize the process of identifying pricing anomalies and informing regulatory strategies. However, one of the challenges present today is the lack of standardization of data and interoperability. This dissertation presents a scalable framework for bringing big data into national healthcare pricing systems that better advance transparency, equity, and consumer trust.