Randomised controlled trials evaluating artificial intelligence in clinical practice: a scoping review - The Lancet Digital Health thelancet.com/journals/landi…

Interesting study.
Agree to be very cautious, but conclusion 'The findings of sRCTs should be considered hypothesis-generating and should not contribute to guidelines.' seems to go to far.
This is are very special sample, not only characterized by being single centric.

Our new flagship project: #MultiSCRIPT

We made randomization routine care.

We aim to randomize all persons with MS in our routine care in a series of nationwide pragmatic trials based on 'real world data' - and almost all we invite participate!

Thank you!!

Study: Many surrogate markers used for pivotal trials of products to treat non-oncologic chronic diseases lack meta-analyses linking them to outcomes, raising uncertainty of their validity Joshua D. Wallach Yale_CRRIT raps.org/news-and-artic…

We tested how we can best collaborate with AI to do systematic reviews, meta-research or asses study designs - fantastic team and teamwork, thank you Tim Woelfle et al!!

Overly complex methods may impair pragmatic use of core evidence-based medicine principles

#Editorial by Rebecca Kuehn, Ying Wang, Gordon H. Guyatt

Highlight of the EBHC Conference 2023

Link: bit.ly/4aNWGzR

Two articles highlight big problems when platform trials include non-contemporaneous controls. In addressing A vs no A, analysis should only use randomizations that included possibility of A. Should never use controls of B vs non-B.
nejm.org/doi/full/10.10…
sciencedirect.com/science/articl…

Long anticipated, it's here now

🌐 Exploring the frontier of #DigitalBiomarkers in #healthcare ! Learn how devices like smartphones and wearables are revolutionizing #healthdata collection.

Link to Article: bit.ly/4cWkJhM

Tim Woelfle Lars G. Hemkens #healthtech #smartdevices #wearabledevices

Ioannidis JPA.
Leading researchers in the leadership of leading research universities: meta-research analysis.
bioRxiv 2024.04.02.587840; doi: doi.org/10.1101/2024.0…. biorxiv.org/content/10.110…

Do elements with rounded shapes enhance click-through rates?
Maybe; the two controlled experiments that supposedly justify this claim are not trustworthy. A fun exercise is to see if you can identify why the two A/B tests in the paper are flawed, just from the relevant paragraphs…

Data Checks Before Registering Study Protocols for Health Care Database Analyses. A very interesting JAMA viewpoint:

jamanetwork.com/journals/jama/…

A new #RC2NB study in BMJ Health Care & Informatics explores digital biomarkers. There is no consensus in the field, with 127 out of 128 articles providing different definitions. The study aims to guide the development of a more widely accepted definition. informatics.bmj.com/content/31/1/e…

We found in 128 studies 127 definitions of 'digital biomarker'.

Consensus needed to facilitate use of this promising technology !!

RC2NB BMJ Health & Care Informatics

informatics.bmj.com/content/31/1/e…

Voting for death?
The Swiss were the only people in the world voting on public health measures to control COVID-19.
We estimated excess mortality at the municipal level for the year 2020 in Switzerland.
tinyurl.com/mwstuj58
1/x

Great example of importance to consider socio-economic factors for real-world evidence studies!

Just imagine interpretation without having that data.

This is fun

Isn’t that exactly what makes 'usual care' usual? Unfortunately, in real world, care is rarely optimal, so it would be totally artificial if care would be optimal or standardized everywhere, no?