OHDSI2025 Tutorials

OHDSI 2025 Tutorials

Morning Session (8 am - 12 pm ET)

An Introduction to the Journey from Data to Evidence Using OHDSI

The journey from data to evidence can be challenging alone but is greatly enabled through community collaboration. In this half-day tutorial, we will introduce newcomers to OHDSI. Specifically, about the tools, practices, and open-science approach to evidence generation that the OHDSI community has developed and evolved over the past decade.

Lead: Erica Voss

Afternoon Session (1 pm - 5 pm ET)

Developing and Evaluating Your Extract, Transform, Load (ETL) Process to the OMOP Common Data Model

In this tutorial, students will learn about the tools and practices developed by the OHDSI community to support the journey to establish and maintain an ETL to standardize your data to OMOP CDM and enable standardized evidence generation across a data network.

Lead: Clair Blacketer

Using the OHDSI Standardized Vocabularies for Research

In this tutorial, students will learn how to take advantage of the OHDSI standardized vocabularies as an analytic tool to support your research, including searching for relevant clinical concepts, navigating concept relationships, creating conceptsets and understanding source codes that map within these expressions. Students will also learn where the OHDSI standardized vocabularies are used throughout OHDSI’s standardized analytic tools.

Lead: Anna Ostropolets

Clinical Characterization Applications to Generate Reliable Real-World Evidence

Clinical characterization—descriptive statistics to summarize disease natural history, treatment utilization, and outcome incidence—are the at heart of many real-world data applications, including study feasibility and quality improvement. In this tutorial, students will learn how to design and implement observational network studies for characterization, and how to apply tools and practices developed by the OHDSI community to ensure the evidence generated is reliable.

Lead: Patrick Ryan

Population-Level Effect Estimation Applications to Generate Reliable Real-World Evidence

Population-level effect estimation–causal inference methods for comparative effectiveness and safety surveillance–enables researchers to understand how exposure to medical interventions are expected to impact health outcomes. In this tutorial, students will learn how to design causal inference studies and how to apply tools (such as CohortMethod) and practices (such as objective diagnostics) developed by the OHDSI community to ensure the evidence generated is reliable.

Lead: George Hripcsak

Patient-Level Prediction Applications to Generate Reliable Real-World Evidence

Patient-level prediction—the use of machine learning to train, test, and apply predictive models for disease interception and precision medicine—offers the potential to personalize healthcare by enabling individualized risk prediction based on personal health history. In this tutorial, students will learn how apply tools and practices developed by the OHDSI community, including the PatientLevelPrediction HADES R package, to design and implement network studies capable of learning and externally validating prediction models, and how to apply these models to your population.