This Week In OHDSI

Feb. 20, 2024

Each community call during “Phenotype Phebruary” features a set of Workgroup 2024 Objectives & Key Result (OKR) announcements, as well an update from that week’s Phenotype Phebruary activities and findings. Workgroups that presented during this call were Themis, Healthcare Systems, Generative AI and Foundational Models, Oncology, Vaccine Vocabulary, Patient-Level Prediction (PLP), ATLAS, Open-Source Community, Psychiatry, and Natural Language Processing (NLP). The Phenotype Phebruary update, led by Anna Ostropolets and Jamie Weaver, reflected on findings from the first two weeks, as well as early research and critical questions around the Week 3 phenotype, major depressive disorder.

Community Updates

• Congratulations to the team of Martin Boeker, Daniela Zöller, Romina Blasini, Philipp Macho, Sven Helfer, Max Behrens, Hans-Ulrich Prokosch and Christian Gulden on the publication of Effectiveness of IT-supported patient recruitment: study protocol for an interrupted time series study at ten German university hospitals in Trials.

• Congratulations to the team of Moshe Zisser and Dvir Aran on the publication of Transformer-based time-to-event prediction for chronic kidney disease deterioration in JAMIA.

• The 2024 OHDSI Global Symposium will be held Oct. 22-24 at the Hyatt Regency Hotel in New Brunswick, NJ. The tentative symposium format will feature tutorials and workshops on Oct. 22, the main conference on Oct. 23, and workgroup activities on Oct. 24. Registration has not opened yet.

• Registration is now OPEN for the 2024 OHDSI Europe Symposium, which will be held June 1-3 in Rotterdam, Netherlands. There will be tutorials and workshops June 1-2 at the Erasmus University Medical Center, and the main conference will be held Monday, June 3, on the Steam Ship Rotterdam. Please visit the event homepage for more information and registration details.

• Henrik John recently announced a new network study that he is leading with Chungsoo Kim, Jenna Reps and Egill Fridgeirsson on “Deep Learning Comparison.” The aim is to assess the value of deep learning methods over conventional methods for the development of clinical prediction models. The specific diseases under consideration are dementia in individuals over 55, lung cancer in those over 45, and bipolar disorder in patients misdiagnosed with major depressive disorder. If you would like to join this effort, please read this forum post for more information and reach out to the study leads by March 1.

• James Weaver, an Associate Director of Observational Health Data Analytics at Janssen Research and Development, will speak during a panel session on Current Approaches for Distributed Analysis on Thursday, March 14 (1 pm ET) during a Health Data Research Network Canada event. This will be a virtual conversation; more information and a registration link are available here.

• Under the leadership of Azza Shoaibi, Anna Ostropolets, Gowtham Rao and James Weaver, Phenotype Phebruary 2024 focuses on assessing consistency in phenotype definition components, phenotype representation structure, and phenotype validation methods. The month-long activity empowers OHDSI collaborators to engage with each other while advancing the science of phenotyping and gaining education and training around phenotype development and evaluation. You can check out the event homepage here.

• Research from the 2023 Global Symposium Collaborator Showcase can be viewed on the Global Symposium Showcase page. Research is also being shared daily on OHDSI’s LinkedIn, Twitter/X and Instagram feeds as part of the #OHDSISocialShowcase. Below are posters (with study leads) that are featured this week:

Monday — Challenges and opportunities in adopting OMOP-CDM in Brazilian healthcare: a report from Hospital Israelita Albert Einstein (Maria Abrahao)
Tuesday — Developing a pregnancy algorithm in ATLAS: Applying start date offset (Rupa Makadia)
Wednesday — Creating parsimonious patient-level prediction models using feature selection (Aniek Markus)
Thursday — From Complexity to Clarity: Reproducible and Scalable Phenotype Development and application of LLM in a support role (Asieh Golozar)
Friday — Establishing and Operating the OHDSI Dentistry Workgroup: A Model for Other Disciplines (Danielle Boyce)

Job Openings

• Alex Asiimwe shared three openings at Gilead that could be of interest to OHDSI collaborators. There are openings for a Director, RWE – Data Science, a Director, Data Acquisition, and a Senior Director, Head of Data Office. You can learn more about each position and apply using the links above.

• Linying Zhang shared openings for both a Postdoc and a Senior Data Analyst at the Washington University School of Medicine in St. Louis. Successful candidates will work on causal machine learning and responsible AI for reliable real-world evidence generation. Interested applicants should send a CV and cover letter to linyingz@wustl.edu.

• Nathan Hall introduced a summer internship at Johnson & Johnson for an Epidemiology UX/Web Design Intern. This internship provides a unique opportunity to merge design principles with epidemiological research, contributing to the advancement of real-world evidence applications. In this role, you will have the opportunity to blend your passion for user experience (UX) and web design with the field of epidemiology, contributing to impactful projects that enhance our ability to derive insights from health data. More information and an application link are available here.

• There is an opening for an Epidemiology Graduate Intern at Johnson & Johnson. Among the responsibilities for this remote position are assisting in managing epidemiologic studies and literature reviews to characterize incidence, prevalence, risk factors, and associated comorbidities and treatment patterns for lung, head, and neck cancers across interventional oncology, contributing to the development of protocols for analyzing real-world data cohorts, such as SEER Medicare, and conducting quantitative analyses using both public and private data sources. More information and an application link are available here.

The University of North Carolina Chapel Hill CTSA is hiring a Research Informatics Specialist. This is a remote position (US only, with mostly East Coast hours). This person will join a skilled and highly collaborative team of data analysts, software developers, and data scientists within our CTSA. The core purpose of this position is to support the All of Us Center for Linkage and Acquisition of Data project. We are looking for folks with SQL and health care and/or claims data experience, especially OMOP.

There is an opening for a Data Steward position at the EBMT. Among the responsibilities is the design, implementation and testing of new data collection processes including data collection forms (DCFs) development, as well as the mapping of new items from DCFs to the OMOP CDM.

Slides

Workgroup OKRs + Community Updates

Videos

Workgroup Updates (Themis, Healthcare Systems, Generative AI and Analytics in Healthcare (GAIA), Oncology, Vaccine Vocabulary, Patient-Level Prediction, ATLAS, Open Source Community, Psychiatry, and NLP)

Phenotype Phebruary Update #3