Tom Chalmers in 1977: 'Randomize the first patient!'. He called for assessment of innovative treatments using strong designs early in the process. We believe RISAT makes that more possible. RISAT: Random Invitation Single-Arm Trial @AnnalsofIM https://t.co/K8lnp34t00

#JAMA editor Drummond Rennie who died recently acted as the champion of the JAMA Users’ Guides to the Medical Literature. Without his advocacy the series would never have happened and the #EBM movement might never have got off the ground. JAMA published this tribute to Drummond.

What does a "weak recommendation" actually mean? Evidence-based Medicine 101 by @GuyattGH Wonderful!

Anyone interested in determining the effect of a drug in a single patient by conducting a #RandomizedTrial in a that patient – an N of 1 trial – will find this step-by-step guide enormously useful. https://t.co/4lpKzoM98J

If you are using #GRADE to rate #CertaintyofEvidence for diagnostic tests this article that deals with study design, #RiskofBias, and #indirectness provides crucial guidance. https://t.co/3Vc9UR75Aw

Clinicians can enhance patient understanding by using numerical data instead of verbal probabilities, consistent denominators, absolute risk comparisons, and clear context for unfamiliar data types. https://t.co/1cAtzXCbSn

https://t.co/UZAydQwobf Nice read: „…. advanced applications should only be used in combination with sufficient background knowledge. Over time, consistently verifying LLM outputs may lead to an appropriately calibrated trust in these tools among users.“

Please apply

Join a wonderful team transforming clinical research. We are embedding trials into care, creating real-world evidence, and innovating at the intersection of data science and patient impact. 100% postdoc @ Department Clinical Research @ U Bern Apply:

#GRADE’s primary focus in #SystematicReviews has been on rating certainty in evidence on interventions. But in this article GRADE also provides guidance for #CertaintyofEvidence in #prognosis in which #ObservationalStudies start as high certainty evidence https://t.co/wXH6r6eOVO

Truth starts with accurate words. 'Emulate' means to copy. Target trial emulation does not copy RCT design other than defining the inception cohort. It does not copy randomization, blinding, prospective data collection, specifying adjustment variables before collecting data ..

https://t.co/Rtab6DKJbn Huge Impact of immortal Time Bias: Across 25 topics (182 studies), 44% studies were affected by ITB "In 23.8% (5/21), the overall summary results changed from stat. significant to non-stat. significant or vice versa after excluding studies with ITB"

Just published critique of #GRADE notes the system’s limitations: excessive complexity and need for more practical guidance and examples. Seven Core GRADE papers published in #BMJ this spring go a long way toward addressing these issues. https://t.co/qcdQuyhHTO

Google’s research on a Personal Health Large Language Model (PH-LLM) is now published in @NatureMedicine Magazine!

This is wonderful!

!

💡New free course! We’ve developed a self-paced course to help the public better understand randomised trials – how they work, why they matter, and their role in advancing healthcare. Start here 👉 https://t.co/8qSlaHkZIE @hrbtmrn @hrbireland

Developed by an international team of clinicians, researchers, and journal editors, the Chatbot Assessment Reporting Tool (CHART) is a guideline that includes 12 items and 39 subitems developed to provide reporting recommendations for studies evaluating the performance of

#RealWorldData advocates facing challenges addressing sources of bias suggest causal modelling as a solution. Doesn’t work. Comparison of 19 modelling studies with #RCTs showed 42% differed in direction and 47% of confidence intervals didn’t include RCT estimate.

“This trial is registered with PROSPERO (CRD42024529706).” ???

Impressive. Moderate certainty evidence despite “The findings..should be interpreted in light of limitations, such as the small number of studies available for most outcomes, a lack of age-specific analysis and biases at the individual study level, including residual confounding”