Valentijn de Jong

Assistant professor, @UMCUtrecht. SNE @EMA_News. Statistics & Epidemiology. Opinions are my own.

[Valentijn de Jong @VMTdeJong]

@BasedScience @jd_wilko The IPD reviewers/authors typically can't share the data that was shared with them either

[Valentijn de Jong @VMTdeJong]

@nwerre @sharectd Please do!

[Valentijn de Jong @VMTdeJong]

We have 11 positions for PhD programs @sharectd on sharing and analysis data. Interested and have a MSc in epi, stats, data science, or related? Apply this month. https://t.co/3bdp0O5SnZ

[Valentijn de Jong @VMTdeJong]

@nwerre @sharectd Yes we accept internationals. In fact, the eu mobility rule requires that applicants move to the country. The masters needs to be finished by September, but preferably earlier

[Valentijn de Jong @VMTdeJong]

This includes one PhD project on Methodology for cross-design synthesis with me & @CarlMoons, at @UMCUtrecht and collab with @TPA_Debray @Stanford & https://t.co/jd7fdcWRki

[Bada Yang @yang_bada]

🚨Course announcement🚨

Cochrane Netherlands is once again organizing an online course on how to perform Diagnostic Test Accuracy (DTA) reviews on 29 Nov to 1 Dec 2023.

Learn from experts such as @leeflang_m and @hans_reitsma!

Info and registration at: https://t.co/gg46TIsEAX

[Bada Yang @yang_bada]

🔥Course announcement🔥

Cochrane Netherlands is organizing an online course on how to perform Diagnostic Test Accuracy (DTA) reviews on 7, 8 and 9th Dec 2022.

Learn from experts such as @leeflang_m and @hans_reitsma!

More info and registration at:
https://t.co/AN0s5uZBd5

[Laure Wynants @laurewynants.bsky.social @laure_wynants]

⚠️Just published:
UPDATE to the living review of prognostic models for covid-19

*over 600 model appraised
*low risk of bias models for several applications
*calibration heterogenous across settings, even for low rob models
*common pitfalls discussed

https://t.co/XME4UHIWdu

[Maarten van Smeden @MaartenvSmeden]

NEW PAPER
Large external validation of covid-19 prediction models for mortality @bmj_latest
TL;DR: performance highly heterogenous -- often pretty bad --, but some positive exceptions too that require some recalibration
https://t.co/iYThZtXpFj

[Valentijn de Jong @VMTdeJong]

@Richard_D_Riley I recently asked the deadline for my peer review to be extended by a week. Response: "within a month would be great"

[Valentijn de Jong @VMTdeJong]

@DekbedV @wajongersactie Als er frauduleus een nepvaccin is gezet is een ander verhaal. Maar dat is algemeen bekend

[Valentijn de Jong @VMTdeJong]

@DekbedV @wajongersactie Maar die worden niet als covidvaccinatie geregistreerd bij het RIVM. Ze houden bij welk vaccin en welke batch er is gebruikt.

[Valentijn de Jong @VMTdeJong]

@DekbedV @wajongersactie Nee, natuurlijk kan je met een placebo prik in een experimenteel onderzoek geen QR code krijgen. Hoe kom je daar bij?

[Thomas Debray @TPA_Debray]

@VMTdeJong proposes new methods for developing generalizable prediction models from pooled studies and large clustered data sets . Corresponding software available from CRAN - https://t.co/rXTMPq7qhz https://t.co/NxeRVPGstN

[Laure Wynants @laurewynants.bsky.social @laure_wynants]

UPDATE of our review of covid 19 models for diagnosis and prognosis published in @bmj_latest https://t.co/4XQOvEuZwY
https://t.co/XME4UHIWdu

A few additional models (Qcovid) are not in this version but on our website https://t.co/Dxa4yj2BEq

[Valentijn de Jong @VMTdeJong]

@denisevanhout Haha yes 3 groups, except 155 + 81 patients (1/3 of the treatment patients) received a different treatment than they were assigned. There's nothing prospective about that

[Valentijn de Jong @VMTdeJong]

What do #epitwitter #epistats think of this article?

https://t.co/QKARaFnBMu

1/6

[Valentijn de Jong @VMTdeJong]

Finally, they conclude that the positive effect was only observed for HCQ and not CQ, so that ‘these drugs cannot be regarded as interchangeable’ without ever testing for a difference between the outcomes in HCQ vs CQ

6/6

[Valentijn de Jong @VMTdeJong]

So in both groups HALF of the patients received treatment, and could then still be removed from the analysis if they died or were moved to the ICU that first day, before the study inclusion started!

5/6