Differences between healthcare datasets in structure, content, and coding systems are widely recognized as significant barriers to generating robust evidence for regulatory and medical decision making. As a result, there is a growing interest in using common data models embedded within large data networks. By standardizing the structure, contents, and semantics of disparate healthcare databases, common data models like the Observational and Medical Outcomes Partnerships common data model (OMOP-CDM) enable multidatabase studies to be undertaken at speed and in a transparent way. To date, little attention has been given to their potential role in health technology assessment (HTA).

We identify the uses of observational data in generating evidence in HTA, some common analytical challenges faced in their estimation, and the infrastructural, technical, and data reusability constraints that limit its wider use. We discuss where and how the OMOP-CDM could overcome these barriers in relation to different types of evidence requirements.

The OMOP-CDM increases the interoperability of otherwise disparate datasets, allowing reliable evidence to be generated from multidatabase studies at speed and transparently. The current analytical tools are best suited for clinical characterization and population-level effect estimation. Further developments to these tools are required to support analyses common in HTA like parametric survival modeling. Differences in costing methods as well as the structure of healthcare delivery between countries may limit the feasibility and value of standardization.

The OMOP-CDM has the potential to support reliable and timely evidence generation in HTA. The analytical tools should be further developed to support common HTA use cases.