george davey smith

1/n A webinar from @CDC_Genomics on Mendelian randomization (MR) highlights some under-appreciated aspects of MR; this thread outlines these youtube.com/watch?v=pc3uQz…

Investigating potential causal effects of maternal smoking on offspring handedness using a proxy gene-by-environment (GxE) Mendelian randomization design out today: link.springer.com/article/10.100… with Daisy Crick @mendel_random @medland_sarah

2/n according to @yudapearl application of the front door criteria will become a "serious competitor to RCTs". To make such claims he must, surely, have some evidence to support it. Can he please provide one (1 single) example of how this competitor has performed in this regard?

the very antithesis of this notion of combining evidence from different sources is the below, from #thebookofwhy which sees the front door criteria as a "serious competitor" to RCTs - this notion of competition should have disappeared with the ark, but sadly, here it is

Really important reply on recent claims being made in the Covid inquiry by some unelected/ unaccountable scientists.. important that science advice is not dressed up as political ideology

Absolutely NO causal relationships can be drawn from this paper due to the high risk of selection bias in the design. The 'control group' consists of only 2,509 children compared to the overall cohort of 73,254 children, and only 5 were diagnosed with autism.

Triangulation of evidence in medical research explained, in work led by @SirenaGtz and @MariaGlymour link.springer.com/article/10.100… Will do a thread when I get the time

My favourite visual representation of Mendelism is below, but as always the story is more complex than it seems ieureka.blogs.bristol.ac.uk/2022/07/15/the…

Reposted with suggested listening for playing when reading the editorial youtube.com/watch?v=GGeOo2…

The advent of Mendelian randomization followed the myriad cases of observational studies suggesting vitamins would prevent many diseases, with RCTs quashing these hopes. Sadly fallible MR methods simply recapitulate the spurious epidemiological findings academic.oup.com/ije/article/54…

What’s the point of the Labour party? theguardian.com/commentisfree/…

It’s 5 years from intensification of covid policy in the UK, and the mantra of “lives vs economy” becoming common place. Here a long thread from soon after, reflecting on the fact that there was so much more at stake that “lives vs economy”. It was “lives vs lives” that mattered

Thought for the day, from Conrad Waddington in 1969 “Life might be defined as the art of getting away with it; and Theoretical Biology as the attempt painstakingly to explicate just how it is done”

the "butter bad dark chocolate bad" authors have been producing nonsense for 40 years .. see ieureka.blogs.bristol.ac.uk/tag/george-dav… and youtube.com/watch?v=8IgpTT…

Se lanza la 3ra Edición del libro que formalizó un enfoque de curso de vida para la epidemiología contemporánea 👇🏽

Lifecourse epidemiology debuted in 1997 with “A Lifecourse Approach to Chronic Disease Epidemiology” edited by Di Kuh & Yoav Ben-Shlomo. The 3rd edition of the ground breaking book is being launched 12 March 15.30 UK time bristol.ac.uk/integrative-ep… & foreword bristol.ac.uk/media-library/…

Is there a future for Mendelian randomization? 10th March 11.00 Canadian time, online and in person @mcgillu, I'll discuss this and why, if I had been better read, I would have used the term "meiotic randomization" rather than "Mendelian randomization" in the initial papers

Contralateral cancers are an enigma: the unaffected organ is perfectly matched on exposures across life. Yet at older ages, risk of cancer in the unaffected organ is not much higher than in a randomly selected organ from the same population. How & why? youtu.be/7rbsGpz9-XE

Liveposting on Bluesky: @mendel_random (GS): Mendelian randomization: what it was, what it is, and what it should become bsky.app/profile/michae…