rcprd: An R package to simplify the extraction and processing of Clinical Practice Research Datalink (CPRD) data, and...

Area of Science:

• Health Informatics
• Biostatistics
• Data Science

Background:

• The Clinical Practice Research Datalink (CPRD) is a vital UK electronic health record resource.
• Extracting and managing CPRD data is computationally intensive, especially using R.
• Raw CPRD data in .txt files presents querying inefficiencies.

Purpose of the Study:

• To introduce the rcprd R package for simplifying CPRD data extraction and processing.
• To enable researchers to efficiently build datasets for statistical analysis from CPRD data.
• To reduce duplicated effort in CPRD data management for research.

Main Methods:

• The rcprd package converts raw CPRD .txt files into an efficient SQLite database.
• It employs a four-stage process: cohort definition, data import, database querying, and dataset creation.
• The package allows extraction of common variable types and custom queries without requiring SQL knowledge.

Main Results:

• rcprd significantly simplifies the extraction and processing of CPRD data.
• The package creates queryable SQLite databases for efficient data retrieval.
• It generates datasets ready for statistical analysis, usable entirely within R.

Conclusions:

• The rcprd package offers a user-friendly and efficient solution for researchers using CPRD data.
• It reduces the computational burden and time required for data preparation.
• rcprd facilitates more focused research by streamlining data management tasks.