JW

Pharmacoepidemiology and miscellany. @OHDSI collaborator.

[JW @_jamie_weaver_]

https://t.co/IGeKpH8Qzl

[JW @_jamie_weaver_]

[7/7] Phenotype algorithm performance, and by extension study validity, is what we care about. This is data model agnostic.

[JW @_jamie_weaver_]

[1/7] I recently discussed current approaches to federated analysis with Judith Maro, Michael Paterson @ICESOntario, and @robertwplatt @CanadianNODES. It was hosted by @hdrn_rrds during their learning series, Federated Analysis: State of the Science. https://t.co/1aMrQtKu2B

[JW @_jamie_weaver_]

[6/7] Rather, meaning comes from standardized derived elements that use codes + disease model assumptions to approximate a well-defined clinical description of a phenotype. It's from carefully developed and evaluated phenotype algorithms that we derive meaning.

[OHDSI @OHDSI]

The necessity of validity diagnostics when drawing causal inferences from observational data #OHDSISocialShowcase

Lead: James Weaver
Team: Erica A Voss, Guy Cafri, Kathleen Beyrau, Michelle Nashleanas, Robert Suruki

Learn more: https://t.co/zXKXXUGwV4

#JoinTheJourney

[HDRN Canada - RRDS Canada @hdrn_rrds]

🧐 Want to learn more about current approaches for distributed analysis? Join our next #FederatedAnalysis webinar for a panel discussion on distributed analysis w. practical use cases from @CanadianNODES, @OHDSI & Sentinel. Register ➡️ https://t.co/J8PdHZa5MO

[JW @_jamie_weaver_]

👀

https://t.co/O7mQV9wGYz

[JW @_jamie_weaver_]

🙌

[Jeff Slater 🍊 @jslaterTO]

Congrats to @clarehacksel on being nominated in Toronto–Danforth! Looking forward to you, along with other Toronto @ndp candidates being elected in the next election.

[JW @_jamie_weaver_]

This approach leads to a large amount of information being available to review, but surely reducing transparency to decrease reviewer burden is not a tenable solution to ensuring scientific integrity.

[JW @_jamie_weaver_]

https://t.co/IGeKpH8Qzl

[JW @_jamie_weaver_]

All studies should share a protocol containing the complete prespecification of the analyses to be done, as well as the associated source code that implements the protocol.

[Steven Ge @StevenXGe]

Happy holidays! Introducing https://t.co/Ub3wgs1KVz, an AI-powered app that lets you chat with your data in English! RTutor uses Davinci (#ChatGPT’s sibling) to turn requests into R code, which is executed & results are shown instantly, available as a HTML report in seconds. 1/8

[JW @_jamie_weaver_]

Nothing quite like a full focus, dedicated research sprint!

[EHDEN @IMI_EHDEN]

✨📖Recommended read: EHDEN/@OHDSI just had a paper published in #JAMIAOpen that evaluates the novel approach of study-a-thons to stimulate open science collaborations that are significantly speeding up the research process. Full paper here 👇 https://t.co/a8dhG9ncnC

[Daniel Morales @DanMoralesEpi]

Great to see @EMA_News announcing the first data partners onboarded to the DARWIN EU network for #RWE studies in its first year.
@ErasmusMC @IMI_EHDEN

https://t.co/ZyA3tjKimq