Abstract
Motivation
As prescription drug prices have drastically risen over the past decade, so has the need for real-time drug tracking resources. In spite of increased public availability to raw data sources, individual drug metrics remain concealed behind intricate nomenclature and complex data models. Some web applications, such as GoodRX, provide insight into real-time drug prices but offer limited interoperability. To overcome both obstacles we pursued the direct programmatic operation of the stateless Application Programming interfaces (HTTP REST APIs) maintained by the Food and Drug Administration (FDA), Medicaid, and National Library of Medicine. These data-intensive resources represent an opportunity to develop Software Development Kits (SDK) to streamline drug metrics without downloads or installations, in a manner that addresses the FAIR principles for stewardship in scientific data—Findability, Accessibility, Interoperability, and Reusability. These principles provide a guideline for continual stewardship of scientific data.
Results
MedicaidJS SDK was developed to orchestrate API calls to three complementary data resources: Medicaid (data.medicaid.gov), Food and Drug Administration (open.fda.gov), and the National Library of Medicine RxNorm (lhncbc.nlm.nih.gov/RxNav). MedicaidJS synthesizes response data from each platform into a zero-footprint JavaScript modular library that provides data wrangling, analysis, and generation of embeddable interactive visualizations. The SDK is served on github with live examples on observableHQ notebooks. It is freely available and can be embedded into web applications as modules returning structured JSON data with standardized identifiers.
Availability and implementation
Open source code publicly available at https://github.com/episphere/medicaid, live at episphere.github.io/medicaid, supplementary interactive Observable Notebooks at observablehq.com/@medicaidsdk/medicaidsdk....

MedicaidJS: a FAIR approach to real-time drug analytics

10.1093/bioadv/vbad170