David Laursen

Thomas Wakley, the original editor of @thelancet.com was indeed no friend of Dr John Snow.
In happier news, the journal published a 'corrected' obituary in 2013:
www.thelancet.com/journals/lan...

We're glad you had a refreshing visit to the John Snow pub @pwgtennant.bsky.social‬ 😊

a cat wearing sunglasses and a chain around its neck is sitting on a blue couch . Alt: a cat wearing sunglasses and a shiny chain around its neck is sitting on a blue couch looking fabulous.

Typo in student paper: Single-bling randomized control trial

@spacecowboybooks.bsky.social Thanks for a very useful feed! Only minor quibble: The feed seems to "refresh" every time I go away from it (to another feed) and back - losing my current "reading progress". Other similar feeds do not appear to refresh as often. Possible to reduce "refresh rate"?

Thanks, I'll take a closer look!

Casual causal?

Causal inference coming to an unblinded trial near you!

Hoping to read it more carefully soon and understand how it relates to previous efforts in measuring and handling unblinding (e.g. blinding indices) and to choice of control (e.g. active placebo)

#Blinding #Masking #Unblinding #Unmasking

... (in an estimation setting, this would probably be estimating differences in beliefs between arms, either with conventional 2x2 measures, or blinding indices). This appears to be a central discussion in many fields, so would appreciate your reflections.

Thanks in advance!

3/3

I noticed your warning against conditioning on post-treatment belief (since it is a collider). Just wondering, does this reservation extend more generally to cautioning against testing for success of blinding at all, regardless of doing a stratified analysis of treatment effects by belief? ...

2/3

(cross-posting from Disqus at MedRxiv now that I found this post: hope that's okay!)

Thanks for an interesting preprint, which I hope to read more carefully soon. I am not particularly well versed within causal inference so apologies if the question is unclear.

...

1/3

Previous post about this study:
bsky.app/profile/davi...

3 If feasible and meaningful, have another person check procedures that are not done in duplicate (e.g. trial overlap removal).

5/5

Lessons learned:
1 Use automatic data checks (e.g., that all trials are included in analysis, and no more than once), and ensure that they are working as intended.
2 Add useful extra info the appendix for easy validation (e.g. forest plots for all main analyses with outcomes listed).

4/5

Correcting these mistakes leads to small changes in estimates but does not change our overall conclusions (NB. two analyses of heterogeneity become statistically significant, but we interpret these with caution).

3/5

Some details: One mistake related to incorrect exclusion of a few trials during removal of overlapping trials across meta-analyses). The other mistake was misclassifying a meta-analysis (about attrition but classified as co-intervention) which then entered the wrong analysis.

2/5

Errare humanum est ...
(To err is human ...)

While preparing my PhD thesis, I discovered some (minor) errors in which trials and meta-analyses were included for analysis in our meta-epidemiological study.

1/5

Sounds great! Would be interesting to compare # original posts across platforms

Thanks for the status! Do you have separate numbers for original posts vs reposts and how these compare between Bluesky and X?

Webinar - read more and sign up: www.cochrane.org/events/how-t...

On 17 February 2026, Andreas Lundh and Asbjørn Hrobjartsson will host a Cochrane Learning Live webinar about the TACIT tool.

... IN MICE 🐭

jamesheathers.medium.com/in-mice-expl...

Vigtigt med retvisende overskrifter og passende forbehold for ekstrapolering til mennesker.

#JustSaysInMice #InMice #IMus

Many seemingly overlapping terms: MCID, MID, minimally relevant different (MiReDif), smallest effect size of interest (SESOI), smallest worthwhile difference (SWD) ...

Different traditions? Would be nice with a good overview.

Thanks, I'll check it out!

Any specific critical literacy resources from philosophy you would recommend?

Happening TODAY at 3pm GMT 👇

Introduction to INSPECT-SR Training Workshop December An introductory 2-hour online workshop will introduce participants to the INSPECT-SR tool for assessing trustworthiness of randomised controlled...

Have increased capacity for this December INSPECT-SR online training workshop following a successful 1st event today. Book here: www.trybooking.com/uk/FKHV

Recent trial with potential data problems @bmj.com. Great with IPD available for checking.

Helpful/feasible with early standard checks in submission? E.g. onlinelibrary.wiley.com/doi/full/10....) or INSPECT-IPD (in development, www.sciencedirect.com/science/arti..., @jdwilko.bsky.social)

In Danish, it is similar with the word for male cousin (perhaps more in humorous context). E.g., "I saw this weird cousin in the train today ..."

@tweis.bsky.social

Just to clarify our definition: active placebo = similarly looking control + imitating the perceptible side effects of the exp drug + not therapeutic. Agree that standard of care needs to be considered for both/control arm(s), e.g. therapeutically active control.

Example of a trial sequential analysis plot. From: https://ctu.dk/tsa/

What: Webinar on trial sequential analysis (TSA) in meta-analyses
When: 28 Oct 2025 13:00 to 13:45 (CEST)
Where: syddanskuni.zoom.us/j/61110310530

Christian Gunge Riberholt, associate professor at the University of Copenhagen, will talk about TSA and the potential errors of using it.

#MetaAnalysis

Ah, OK! Stopping rules are not my field so forgive me if mistaken. I think active placebos would likely not mask true disease progression itself, but potentially affect "subjective" assessments, e.g. chronic pain, depression, PTSD. (But ethics important re. imitated side effects' acceptability)