The challenge
UVA Health, a world-class medical center and health system based in Virginia, United States, have just started their journey with OMOP/OHDSI. The analytics group envisions a centralized database of available health data for internal research purposes and a way to easily collaborate in research studies with external partners. The OMOP Common Data Model (CDM) has been tested on a few occasions but the overall path to adopt the OHDSI suite in the organization tool stack is yet to be defined.
How we solved it
The Hyve has been working with University of Virginia Health (UVAH) to streamline and facilitate the adoption process of the various open-source components of the OHDSI stack by providing hands-on training and tailored education.
The Hyve provided four online sessions to the technical staff at UVAH. The workshop was given by three of our OHDSI experts, Maxim Moinat, Sofia Bazakou and Anne van Winzum. Prior to the workshop, our experts sat down with the key stakeholders at UVAH to learn about the ongoing efforts within OHDSI, internal and external data workflows and current data architecture. The workshop provided the participants with quick access to resources and foundational knowledge on the following topics:
The OHDSI community resources, including the forum, working groups, EHDEN Academy and real-time collaborative research such as the Covid-19 studyathon.
Current OMOP Common Data Model (OMOP CDM V5.3) including a deep dive into the OMOP standardized vocabularies, structure and conventions, as well as new and ongoing CDM developments including new OMOP CDM versions (5.4 and 6.0) and the oncology extension.
A comparison of the OMOP CDM to other healthcare standards such as FHIR, CDISC, Sentinel and i2b2.
An introduction to the OHDSI ETL and data quality tooling: White Rabbit, Rabbit in a Hat, Usagi, Data Quality Dashboard, CDM Inspection Report and Achilles. Including live demo and exercises for participants.
ETL building, best practices and Delphyne.
The OHDSI analytic toolset, focussing on the main features of Atlas and how to use these to build your own observational health study.
During the last session, our experts and the UVAH team interactively built an OHDSI adoption roadmap for UVAH. The main question was: What is the next step for UVAH in their OHDSI journey? To address this question, The Hyve asked the views and opinions of all stakeholders via multiple interactive polls. This provided a better understanding of the direction that the UVAH technical staff wanted to go and ultimately pinpointed areas where additional effort was needed first. After the workshop, the participants were equipped with all the materials mentioned as well as the session recordings.
The outcome
The workshop enabled UVAH to reach a consensus on a clear and concrete plan on how to move forward with their OHDSI journey.
Having gained a deeper understanding of the steps that need to be taken in the data conversion process to align them with the corresponding components of the OHDSI stack, the team identified and prioritized tasks to be taken up. Leading factors for the prioritization were (1) the highest impact on data quality and (2) an improvement on collaboration processes within the team.
The workshop was received with great enthusiasm and engagement from the UVAH technical staff and we are excited to see that workshop participants have become regular attendees at OHDSI meetings and actively contribute to the working groups.