OHDSI symposium 2021: Impact of the COVID-19 Pandemic and the Journey to reliable evidence

The OHDSI Symposium is an annual event where members of the OHDSI community present their ongoing research, share the lessons learned and discuss the science that is evolving throughout the community. It is a great opportunity to understand where the OHDSI community stands, where they have been and where they are going in the next year. In addition, members of the OHDSI community from different institutions and universities all over the world get the chance to connect and interact with each other.

The Hyve has been attending the OHDSI symposiums since 2016. Every year, our colleagues look forward to presenting their work on OHDSI to the community and engaging in meaningful discussions that help us implement The Hyve's mission and vision.

This year the Collaborator Showcase received a record number of poster submissions. The OHDSI community members also submitted software demos, and lightning talks that comprised the 2021 Collaborator Showcase. The presentations featured OHDSI advancements in observational data standards and management, methodological research, open-source analytics development, and clinical applications.

The first day of the 2021 OHDSI Symposium started with the State of the Community presentation, updates from the global OHDSI network, and then a series of talks about OHDSI’s impact on the COVID-19 pandemic. The second day of the 2021 OHDSI Symposium focused on the journey to reliable evidence.

The Hyve’s contribution: Talks and posters

This year Kees van Bochove (Founder, The Hyve) gave a talk on the results of the PIONEER study-a-thon and we presented two posters, about the UK biobank mapping and our work on the Data Quality Dashboard. Both posters were very well received during the interactive poster session held in Gather Town (web-conferencing software). Even though it is hard to compare in-person conferences − like the OHDSI symposium in Washington that Kees and Maxim attended in 2016 − to the ones that are held online, the attendees of the 2021 OHDSI Symposium felt that the event was very well organized. Members of our OHDSI team who hadn’t attended an in-person conference before felt that Gather Town was a great choice for presenting the posters, having discussions and for meeting other members of the community. They enjoyed Gather Town’s video-game environment that allowed them to walk around with their avatar and interact with videos, posters and other attendees.

Reproducibility challenge

For our colleagues, the first OHDSI Reproducibility challenge workshop was the highlight of the conference as it was eye-opening for everyone. Nine teams were tasked with reproducing Yasser Albogami et. al.’s study on Glucagon-Like Peptide 1 Receptor Agonists and Chronic Lower Respiratory Disease Exacerbations Among Patients With Type 2 Diabetes. None of the teams was successful. This hands-on challenge proved that, despite having defined standards within the community, reproducing the same study cohorts and results, as papers can be an uphill task. It underscores that the narrative descriptions in scientific papers are not precise enough to communicate the details of the analysis, and that medical papers need to be accompanied with code. The plenary presentation held on the second day of the symposium took a deep dive into the subject of reproducibility and addressed the issues that had arisen during the challenge the previous day. At the end of the day, the challenge helped reinforce our goal of generating reliable real-world evidence data.

Maxim’s award

During the symposium our data engineer, Maxim Moinat was awarded the OHDSI Titan Award in the category Data Standards. We are thrilled about this distinction. The award recognizes his extraordinary contributions towards developing and maintaining the OHDSI ETL tools. Read more about Maxim’s award here. Even though Maxim believes there is still much more to accomplish, this award serves as validation and appreciation for his work within the OHDSI community. This award is equally important for The Hyve's OHDSI team as it confirms that they are on the right track and it motivates our team members to continue improving tools that are used within the OHDSI community.

Overall experience

All in all, the OHDSI Symposium was a good and inspiring experience for our colleagues. We are happy and proud to be acknowledged for the very hard work that Maxim Moinat, and the entire OHDSI team, put into the community. The Reproducibility challenge surprised us; even though the study cohorts and results by Yasser Albogami were very well documented, none of the groups that participated in the challenge was able to reproduce the study results. This gives us perspective and motivates us to keep learning, improving our work and be an active part of the community and working groups. The closing ceremony and the presentation by Jamie Weaver (epidemiologist at Janssen R&D and OHDSI member) captivated us; the personal story from Jamie Weaver about his journey with cancer and how he used OHDSI tooling to decide on his treatment inspired us and reminded us that we are together in this journey; we felt the shared sense of community.

What we do

At The Hyve we specialise in transforming data contained in different healthcare databases into a harmonised format (Common Data Model), as well as using a common representation (terminologies, vocabularies, coding schemes). OHDSI provides a systematic process and tool suite to evaluate and disseminate large-scale patient-level analysis and prediction models using observational data in a network. The open-source suite is backed and promoted worldwide by a community of patient organisations, clinicians, data scientists and epidemiologists, and The Hyve actively participates in the OHDSI community.

Let's start collaborating

We offer a range of services: consultancy, OMOP conversion, deployment, development, training and workshops. For inquiries about our OMOP/OHDSI services, feel free to contact us via this form.

Fill in the form and we will get in touch

Choose a subject