Insights
Epidemiology, causal inference, machine learning and the distance between a benchmark and a ward. Written in the language the idea arrived in — some pieces are in Spanish, some in English.
Hay un error de ajuste peor que bloquear un efecto: crear uno que no existía. Y los datos no te van a avisar.
Four things that decide the deal, none of which are the model — written from the side of the table that has to sign and answer for it.
The connection drops. The patient is still in front of you. That sentence was the whole specification.
Ajustar por un mediador «por si acaso» no es prudencia: es sabotear el efecto que querías medir.
Why diagnostic AI keeps winning benchmarks and losing wards — and the four places it dies in between.
Ten years of watching health organizations sit on mountains of data and decide with a fraction of it.