Common Data Models (CDMs) to Enhance International Big Data Analytics: A Diabetes Use Case to Compare Three CDMs

Area of Science:

• Health Informatics
• Pharmacovigilance
• Real-World Evidence Research

Background:

• Common Data Models (CDMs) are crucial for analyzing diverse, large-scale health datasets.
• Three prominent CDMs, OMOP, Sentinel, and PCORI, are widely used in observational research.
• Evaluating CDM suitability is essential for specific research use cases, such as cohort studies.

Purpose of the Study:

• To assess the performance of three leading CDMs (OMOP, Sentinel, PCORI) against predefined criteria for a diabetes cohort study.
• To evaluate the capability of these CDMs in analyzing medication benefits, risks, and cost-effectiveness.
• To identify gaps in CDM functionality concerning privacy, clinical outcomes, and cost data.

Main Methods:

• A literature review identified OMOP, Sentinel, and PCORI as relevant CDMs.
• Fifteen predefined criteria were established for a diabetes cohort study use case.
• The three CDMs were tested against these criteria to assess their performance in capturing benefit (diabetes control), risk (hypoglycaemia), and cost-effectiveness data for new medications.

Main Results:

• All three evaluated CDMs (OMOP, Sentinel, PCORI) demonstrated utility in planning collaborative, cross-jurisdictional research.
• OMOP met 14 out of 15 criteria, Sentinel met 13 out of 15, and PCORI met 10 out of 15.
• None of the CDMs met the specified privacy level, and significant gaps were identified in clinical and cost outcome data capture.

Conclusions:

• CDMs like OMOP, Sentinel, and PCORI are valuable for real-world evidence research and collaborative studies.
• OMOP demonstrated the highest performance against the tested criteria for the diabetes cohort use case.
• Further development is needed to address privacy requirements and enhance the capture of clinical and economic outcome data within CDMs.