Did he predict the fish’s death or cause it? Was it his saying the words, seeing the vision, or having the thought? Despite his remarkable predictive success, these causal pathways alter the implications for appropriate responsive actions. Draw these DAG. Bruno’s life depends on it.

I did a prediction model vs causation talk with an extended metaphor using Bruno from Encanto but by year 2 like a third of the twenty-something’s in the room had seen the movie.

Weren't the dreadful helmets just the Maryland flag? A flag which, along with being hideous, is designed to celebrate that Maryland was 50/50 secessionist in the Civil War.

So submit relevant papers to the JGIM special issue on Learning Health Systems in General Medicine! Abstracts Due: September 30, 2024 Final Manuscripts Due: December 16, 2024 www.sgim.org/news/call-fo... Thx for providing the lead-in. I appreciate you.

1. Rapid time-frame. 2. Tend to be well-cited, in part because of being near related papers. 3. Perception of more likely acceptance.

If not for Napoleon, we might be tracking 10-day follow-up.

Pretty sure it’s the Babylonians.

From a Bayesian perspective my prior is all ICU trials are negative.

Think about the funnel plot. If it shows that small papers matter, then they conclude they’re a problem and exclude them. So why did they bother?

As an aside, even without fraud meta-analysis involves spending huge amounts of time searching for lowest-impact and on average lowest-quality papers that barely alter the precision but might worsen bias. The whole endeavor might be better served with rules that exclude these and more.