,
5 tweets,
2 min read
Read on Twitter
1/ Okay, I'm going to revisit this as a thread. I like mortality. Alive or dead. Very easy to diagnose, very easy to measure.
Why don't I like non-fatal "events"? Two very big reasons...
Why don't I like non-fatal "events"? Two very big reasons...
2/ First reason – it's more arbitrary. While most things like MIs and strokes are easily agreed on, there are some in the margins where it is less certain. I've had cardios flat out tell me that cholesterol levels of the pt can be a contributing factor for their decision making.
3/ Which means the only real way to do a truly double-blinded experiment would be to remove lipid level awareness of the patients from themselves, their medical team, and the trialists until completion of a study (quite doubtful this would ever happen)...
4/ But the second, and arguably bigger reason -- is that trials don't account for all *non*-CVD, non-fatal events. Is having lower LDL increasing your chances for, say, infection or bacterial disease susceptibility? Before answering, you might want to read cholesterolcode.com/lipoprotein-po…
5/ They'll never likely be a study that is powered enough to track everything bad that happens to the participants, CVD or non-CVD. So we only see one side of the ledger. And that's why it just isn't as interesting to me. I like hard endpoints over soft ones for these reasons.
