“Data-driven approaches have the potential to improve efficiency in the identification of the most important drivers of [heterogeneity of treatment effect] from RCTs.” Full article by Rishi J Desai et al.: eviden.cc/4cupGhU Muthu Vaduganathan

** New paper from the Sentinel Innovation Center ** Wyss et al. show that targeted learning can effectively tackle covariate non-overlap when fitting large scale PS models in sparse high-dimensional settings academic.oup.com/aje/advance-ar… Shirley Wang Darren Toh Mark van der Laan

** NEW PAPER from Sentinel Innovation Center ** A simulation study establishing validity of the diagnostics proposed in the smdi package for identifying likely missingness mechanisms is now published in Clinical Epidemiology and available open access dovepress.com/getfile.php?fi…

Very glad to have this paper out in JACC Journals showing suboptimal local use patterns of triple GDMT for HFrEF correlating with indicators of socioeconomic disadvantage.. jacc.org/doi/10.1016/j.… Shout out to a great team Ankeet S. Bhatt, MD, MBA Muthu Vaduganathan

LLMs are undoubtedly going to be used increasingly in extracting and synthesizing information from free-text EHRs and other sources of data. Very glad share some considerations published in JAMA Network Open led by Michael Matheny.. jamanetwork.com/journals/jaman…

It has been tremendously fun and rewarding to be a part of the FDA's Sentinel Innovation Center, and to work with Rishi J Desai and a great coauthor group on this thought piece. Colin Walsh Mohammed Al-Garadi Sharon E Davis Daniel Fabbri Kevin B. Johnson, MD, MS

Researchers suggest FDA could use artificial intelligence to enhance Sentinel, the agency's safety surveillance system for drugs and devices. trib.al/qbgthis

Thrilled to share that the vision we described 4 years ago is now realized with unveiling of the RWE DE- a fully functional system of 25+ mil individuals with linked EHR and claims data ready to serve the FDA for post marketing surveillance of drugs! onlinelibrary.wiley.com/doi/full/10.10…

In 2019, the U.S. FDA requested a team of investigators to create a new data infrastructure to improve drug safety surveillance research studies. 5 years later, Rishi J Desai and the team introduce the Real-World Evidence Data Enterprise. Read more:buff.ly/3C9qT0g

In a study npj Digital Medicine led by Bowen Gu and Jie Yang, PhD, FAMIA, we find that LLMs struggle to convey uncertainty and can be overly confident in their answer even when it is wrong. Imp area for research and improvement for detection of hallucinations.. nature.com/articles/s4174…

Excited to share our latest paper in Am J Epidemiology led by Janick Weberpals We introduce high-dimensional multiple imputations (hdMI) to handle MNAR through data-driven identification of auxiliary covariates and show that this approach outperforms standard MI academic.oup.com/aje/advance-ar…

Vericiguat Adherence for Heart Failure With Reduced Ejection Fraction in Routine Clinical Care in the U.S New RL JACC Journals Heart Failure, led by star med student Naira Ikram Muthu Vaduganathan Ankeet S. Bhatt, MD, MBA jacc.org/doi/abs/10.101…

#VICTORIA demonstrated benefits of sGC stimulator vericiguat in recent worsening HF & led to its #FDA approval in '21 New JACC Journals nationwide analysis finds use thus far has been limited jacc.org/doi/abs/10.101… Will #VICTOR (no recent worsening HF) support its broader use?

In a world filled with madness and uncertainty, I’m glad we have the purity and serenity of test cricket to hold on to! Skill, determination, belief on full display in London. A cracking opening session on day 4 on the cards tomorrow..

So to summarize, 5 tests played over 25 days, run fests, tense chases, fifers, handshakes (or lack thereof), sendoffs, blokes showing up on one working foot or with one working arm, and yeah, no winner, still nobody complaining! A series for the ages #ENGvIND

Very pleased to see this out in JMIR Publications - can we extract timely mortality information from public sources to bridge the gap in capture of death in EHRs for research applications? Promising findings... jmir.org/2025/1/e71113