Creating API Calls in Large Clinical Datasets for Surgery Outcomes Research.

Area of Science:

• Health Informatics
• Clinical Research Informatics
• Data Science in Healthcare

Background:

• Application Programming Interfaces (APIs) facilitate software system interoperability and data exchange.
• APIs offer efficient access to large datasets for clinical research, supporting reproducible workflows.
• Automated data extraction and analysis are crucial for modern health research.

Purpose of the Study:

• To demonstrate the utility of APIs for accessing and analyzing large-scale clinical and research datasets.
• To showcase reproducible data extraction methods using Python and publicly available health data.
• To highlight the potential of API-driven workflows to improve research efficiency.

Main Methods:

• Constructed Python scripts to query the NIH RePORTER and CMS Medicare datasets via APIs.
• Utilized JSON payloads for NIH API calls to retrieve project funding data.
• Employed URL parameters for CMS API calls to filter Medicare enrollment data by state and year.
• Automated data collection across multiple years and saved outputs in CSV format.

Main Results:

• Successfully extracted project-level funding data from NIH RePORTER.
• Retrieved filtered Medicare enrollment data from CMS datasets.
• Demonstrated scalability and automation of data collection using API calls.
• Generated data in CSV format suitable for further analysis.

Conclusions:

• API-based data extraction provides a scalable and reproducible approach for clinical research.
• Leveraging APIs from sources like NIH and CMS enables automated data retrieval and customized filtering.
• Widespread adoption of API workflows can significantly enhance data accessibility and research efficiency in healthcare.