Observational Health Data Sciences and Informatics

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== OHDSI Collaborative Studies FAQ ======


**What is an OHDSI collaborative research study?**

A study in which collaborators with data in CDM format run common analytic code to generate results that are then shared with the OHDSI community.

**Will any patient-level data shared?**

No.  OHDSI studies produce only population level aggregate summary results.

**Are IRB protocols provided?**

Individual study proposals may include an IRB.  However, several collaborators have had success with a CDM-based research [[https://dl.dropboxusercontent.com/u/29827409/IRB%20OHDSI%20generic.docx|Generic IRB Protocol]].  For individual studies that require custom IRBs, this may serve as a starter template as well.

**How are collaborative research studies proposed?**

Proposals may be submitted by any named collaborator in the OHDSI consortium.  The submission process involves posting a preliminary overview and protocol to the [[research:protocols_in_development|Protocols in Development]] page.   A notification of this posting should also be made on the [[http://forums.ohdsi.org/c/researchers|OHDSI Researchers Forum]].  The study proposal is then reviewed by the community.  This review period, typically 2-3 weeks, will determine level of interest as well as generate suggestions for modification to the study design.

**What are the required elements?**

The required elements for a Proposal can be viewed on the [[research:project_proposal_template_2|Project Proposal Sample Template]].  Some elements may of course change over the review process.

**Is there a project approval process?**

For a study to move from //Proposed// to //Active// status, it must have:

  * At least 2 sites willing to participate
  * Working code that has been tested at at least 2 sites


**How are projects executed at local sites?**

Projects are executed via SQL and/or R code.  In order to make projects runnable at multiple sites running different database dialects (e.g., SQL Server, Oracle, Postgres), we employ a tool called SqlRender.  A great way to learn about how these studies are written is to review an [[https://github.com/OHDSI/StudyProtocols/tree/master/Study%201%20-%20Treatment%20Pathways|example study]] on GitHub as well as the SqlRender documentation.  Also feel free to contact us at the user forums.

**How are results returned to the OHDSI community?**

Each study needs a study coordinator who will serve as the contact point.  Results-- which should be both de-identified and aggregate-- should be returned to this individual.  They are then responsible for organizing and analyzing data with the study leadership team for publication and sharing back to the OHDSI community.
Observational Health Data Sciences and Informatics

User Tools

Site Tools

Page Tools
