Lars G. Hemkens

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 acpjournals.org/doi/10.7326/AN…

#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.…

>100000 single-arm trials exist. Most use convenience samples & give weak non-generalizable evidence. We can do better. We propose Random Invitation Single-Arm Trials (RISAT) - a simple way to get randomized real-world evidence from SATs. @AnnalsofIM acpjournals.org/doi/10.7326/AN…

Can administrative real-world data substitute for RCTs and offer propensity matching to reduce bias? Results often fail to correspond & @LGHemkens et al study showed direction of effect differed in 31% & CI failed to include RCT estimate in 56% pubmed.ncbi.nlm.nih.gov/26858277/

Win Ratio: A Seductive But Potentially Misleading Method for Evaluating Evidence from Clinical Trials | Circulation ahajournals.org/doi/10.1161/CI… This thought-provoking paper critiques the win ratio in a convincingly articulated manner. I'm curious whether the Bayesian approach with…

Wonderful quote: “…one of the problems …is that we may think that our customers are grant review panels or maybe…editors,” …“Those may be our short-term customers, but …not … ultimate customers. Our ultimate customers are people who have to make decisions about healthcare”

Such research has extraordinary return of investment. Funders need to know, consider and implement.

A single-page digital recruitment letter increased clinical trial enrollment compared to longer letter formats, with no impact from color schemes. ja.ma/40ga7Wo

Thanks @LGHemkens for the positive feedback on our recently published ‘RCT within a large RCT’. We need to randomize to understand how we can improve processes in our society. The link to our ‘RCT within a RCT’ can be found 👇 jamanetwork.com/journals/jama/… @CTCPR_ @niklasdj1

Excellent meta-research: A trial within a large trial shows that consent form length impacts recruitment. Please consider too much text can exclude vulnerable groups / We need to rethink consent forms, too much information may harm. jamanetwork.com/journals/jama/…

Personalized treatment decision algorithms for the clinical application of serum neurofilament light chain in #MultipleSclerosis. A modified Delphi Study from @LGHemkens @JensKuhle @UniBasel_en @RC2NB ▶️ journals.sagepub.com/doi/full/10.11…

Today in #BMJ, 6th in 7 part series on the essentials of #GRADE – Core GRADE: creating excellent Summary of Findings tables allowing clinicians and guideline developers to easily and comprehensively grasp evidence that will assist patients to make optimal decisions.…

I’m proud to welcome Dr. Vinay Prasad, MD, MPH, as the new head of FDA’s Center for Biologics Evaluation & Research. With 500+ peer-reviewed publications and two books, Dr. Prasad brings the kind of scientific rigor, independence, and transparency we need at CBER—a significant…

This article is the first in a series describing the fundamental points that authors of systematic reviews, health technology assessment reports, and clinical practice guidelines need to use the Core GRADE approach bmj.com/content/389/bm…

A major update to the CONSORT reporting guidance for clinical trials was published last week. 📌 CONSORT 2025 replaces all previous versions and should be used from now on. So what’s new and what’s different? 1/7 #MethodologyMonday #116 (COI - I am a co-author)

#GRADE, indeed #EBM, will never be the same. GRADE papers have become too complex and difficult to navigate. The remedy: a series of 7 papers in the #BMJ laying out the essentials of GRADE: Core GRADE. This week, an overview of Core GRADE. bmj.com/content/389/bm…

The GRADE approach is the dominant system of rating certainty of evidence and grading healthcare recommendations worldwide. This article introduces a series of papers on new guidance for the GRADE approach bmj.com/content/389/bm…

You can write clinical trial protocols meeting all formal requirements for regulators and ethic committees that produce zero scientific value. Thats a problem. Clinical study protocols are not primarily regulatory documents.