The OHDSI Oncology Working Group aims to provide a foundation for representing cancer data within the OMOP CDM at the levels of granularity and abstraction required to support observational cancer research.
Oncology support in OMOP is a work in progress. We welcome your participation!
TODO: update the following table with the new 2024 meeting links. Until then, and even after then, it is highly recommended that you add the meetings series to your own calendar, which are found within the oncology teams calendar, instead of reyling on this table.
subgroup | schedule | meeting details | Team Lead |
---|---|---|---|
CDM/Vocabulary & Development | First Thu 1PM EST Third Thu 9AM EST |
link | Thomas Falconer, Michael Gurley, & Robert Miller |
Outreach/Research | Second Tue 3PM EST Fourth Wed 9AM EST |
link | Asieh Golozar & Christian Reich |
Genomic | Second and Fourth Tue 9AM EST | link | Asieh Golozar |
In a typical observational study, the definition of the study population (cohort), exposures and outcomes are usually based on diagnostic codes in addition to drug exposures, procedure occurrences or lab measurements. For cancer studies, this information is typically not sufficient, as more details are required for the proper identification of the study population, treatment and subsequent outcomes.
Appropriate characterization of cancer requires details such as anatomical site, morphology, local penetration, affected lymph nodes, metastatic spread, biomarkers, and disease staging and grading. In typical observational data sources, this necessary level of detail is not regularly present. Patient results from diagnostic procedures are collected but may not be available within the given data source or what is collected cannot appropriately serve as a surrogate for the above attributes. Correct identification of cancer treatment regimens also tends to be more complex compared to other disease modalities within observational data. Most cancer treatments are administered in chemotherapy regimens with complex dosing and scheduling in multiple cycles and are often combined with targeted therapies, immunotherapies, surgery or radiotherapy. None of these attributes follow standard definition to be applied to observational data, as most regimens are personalized to the individual patient need, making a priori standardized definitions more complex. Additionally, clinically relevant information on disease, treatment and outcomes that appropriately reflects a patient’s journey including information on the time of diagnosis, response to treatments, time to treatment failure, disease progression, recurrence and (disease-free and overall) survival requires data abstraction and is rarely available in the source data and has not been traditionally supported in OMOP CDM.
The Oncology CDM Extension of the OMOP CDM aims to provide a foundation for representing cancer data at the levels of granularity and abstraction required to support observational cancer research.
The extension has been tested in EHR and Cancer Registry data against a number of typical use cases.
This site contains the following sections:
Problem Space | High level summary of working group mission |
Publications/Presentations | Links to some relevant publications & presentations |
Development Overview
Purpose Goals Notable Challenges Context Scope What we need Project Management |
Overview of current development effort |
Strategy Delta Vocabulary Validation Framework |
Details regarding key components of the development strategy |
Progress Map Miro Map Roadmap |
Miro Map of completed and outstanding work within scope (with links) |
Github Project
Orientation Architecture Strategy Project Navigation Example Walkthrough |
Documentation about navigating and understanding the Github Project and approach |
Getting Involved
Join Collaboration Channels Submit Collaborator Form Review Project Documentation Contributing |
Suggestions and links for getting started in the effort! |
Conventions Diagnostic Treatment |
Overview of current development effort |
Model | |
Genomics |