A Cloud-Native Approach to Hospital Price Transparency: Leveraging AWS for Scalable Data Processing and Presentation
The healthcare business is undergoing a paradigm shift with ongoing hospital data-sharing initiatives across the country. The Federal Health Care PRICE Transparency Act which took effect in January 2021 requires hospitals and payers to publicly disclose and periodically update standard services rates online, in order to promote informed healthcare decisions and guard against overcharging and fraudulent medical billing. This paper addresses this urgent need by presenting a novel unified data platform to enable the dissemination of vast amounts of hospital pricing and related data, which are currently being formulated from over 6000 hospitals across the United States. The platform’s key components are: 1) A hybrid architecture that combines on-premises and Amazon Web Services (AWS) cloud environments to manage erratic amounts of continuous data streams and perform web-based information searches, 2) An Airflow-based data orchestration and scheduling framework to gather pricing data in machine-readable formats concurrently and crosswalk diagnoses to medical procedures, 3) A Natural Language Processing (NLP)-based diagnostic query interface that improves patient interaction with pertinent medical procedures and pricing recommendations. Results show that the platform was successfully deployed, currently supporting data from hospitals in all 50 states. The system processes over 8,078 pricing files from 6,553 hospitals, covering a significant portion of all registered hospitals in the United States. This work serves as a solid foundation for the future addition of new data elements and formats, and significantly advances the ongoing pursuit of transparent pricing, with the potential to influence healthcare practices and policy decisions.
Authors
- Ali Ahmed (ORCID: https://orcid.org/0000-0003-4476-2384)
- Catherine Jayapandian
Institutions
- Case Western Reserve University (US)
Publication Details
- Published
- 2024-12-15
- DOI
- https://doi.org/10.1109/bigdata62323.2024.10825994
- Citations
- 1
- Primary Topic
- Big Data and Business Intelligence
- Type
- article
- Field-Weighted Citation Impact
- 0.53