Table 1: A summary of various aspects of the Sentinel, CNODES, and PCORnet distributed networks.
Network Sentinel CNODES PCORnet
Primary goal Act as a resource for the US government and other interested parties to evaluate safety and expand the role of real-world data in regulatory decisions Use collaborative, population-based approaches to provide rapid answers to questions about drug safety and effectiveness Improve US capacity to conduct health research, particularly comparative effectiveness research (CER) by creating a large, highly representative network for conducting clinical outcomes research
Main users of the network The US Food and Drug Administration The Canadian Drug Safety and Effectiveness Network PCORnet is a national resource allowing a wide array of investigators and research funders to request queries
Total individuals covered and data partners More than 228 million individuals at 16 data partners including fee-for-service claims databases, hospital networks, and large health plans in the US More than 100 million individuals from 10 data partners: 8 Canadian provinces, 1 US fee-for-service database, and the UK CPRD system Integrated record or claims data for more than 70 million individuals across 8 large Clinical Research Networks and 2 Health Plan Research Networks in the US
Data format Data is maintained in the Sentinel Common Data Model in SAS Data format varies, with many partners adopting a common data model All PCORnet Network Partners operate under a shared PCORnet Common Data Model
Analysis SAS code is written at the Sentinel Operations Center and then distributed as a “package” to each data partner. After the code is run, data sets with outcome and treatment are returned to SOC for final analysis After finalizing queries, CNODES creates a central project team. This team creates a scientific protocol that provides detailed analytic instructions and receives ethics board approval. This scientific protocol is then used to write site-specific analytic protocols and code Study teams and analysis plans are assembled to meet the needs of the individual study. Analysis of the distributed query data may be performed centrally, by participating network partners, or by individual participating sites
Consolidation and reporting Results are presented as network-wide estimates, with the option to present partner-specific estimates; analytic code and results are posted online Partner-specific estimates are returned to the central team. These results can be presented separately or combined with meta-analytic techniques Results of the distributed queries are returned to the central team for a given study. Results may be reported separately or aggregated together centrally, depending on the protocol