Kineret: Israel's Largest Hospital Network Transformed into the OMOP common data model for collaborative research

Abstract

Background In 2021, the Directorate of Government Medical Centers at the Israeli Ministry of Health launched the Kineret initiative to standardize clinical data across its network of public medical centers and facilitate its secondary use for research and innovation. The primary goals were to streamline data extraction, cleaning, and sharing processes, thereby enabling efficient reuse of clinical data for translational and collaborative research. The Directorate oversees a national network of 25 government healthcare institutions, including 11 general medical centers, 9 mental health centers, and 5 geriatric care facilities. Methods Following an evaluation of existing data models, the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM) was selected as the standard framework for semantic harmonization across institutions. A dedicated instance of ATLAS, the OHDSI open-source platform for observational research, was deployed on a secure cloud environment accessible to authorized researchers within the network. This infrastructure enables efficient feasibility assessment and exploratory data analysis buy the end users. Approved research projects are conducted within a secure, cloud-based virtual environment that supports diverse computational needs. Results As of 2025, six medical centers have been successfully integrated into the Kineret data infrastructure, with full harmonization of their clinical data into the OMOP CDM. A seventh center is currently in the final stages of integration and is expected to join the network by the end of the year. Conclusion Kineret initiative provides a scalable, secure, and standardized data infrastructure that supports both intra-national multi-center research and international collaborative studies. By enabling streamlined access to high-quality, harmonized clinical data, Kineret holds significant potential to advance both local and global healthcare research. A detailed description of the data and platform is available at https://kineret.health.gov.il/.

Fig 1. Screenshot of the centralized ATLAS instance of Kineret.

A screenshot from the Cohort Definition tab showing the different available hospitals.

Fig 2. A diagram describing Kineret system.

The process include multipel steps including on-premise (in medical centers) computing, on a centralized cloud service, and private virtual research room managed by Kineret.

Fig 3. Number of persons by year of birth.

Histogram of number of persons in Kineret born in every year from 1900 to 2025.

Fig 4. Illustration of process flow in Kineret.

Internal users can access a secured environment with an ATLAS instance to generate cohorts from all sources at once or independently. To extract a cohort’s data and upload it to the secured study room, IRB approval is required. Kineret stuff supports every step if required by the user.