1/x Mini-Tweetorial on External Validity vs. Competing Risk.
Based on an ongoing discussion with @drjohnm , @ATSandhu , @gcfmd, @mikejohansenmd and others, I have put together this mini-tweetorial on how to think about external validity of trials vs. competing risks.
2/x A lot of these concepts are ones that I have learned over the years from disease-simulation modeling, but haven’t necessarily seen put together in a case example. I hope folks find this helpful.
3/x I think the easiest way to think about this issue of external validity vs. competing risk is with an example like an ICD trial. Historically, the primary prevention ICD trials have shown benefit with respect to a primary endpoint of all-cause mortality.
My first ever tweetorial! In the last 48 hours, there has been a lot of buzz in the lay press (WSJ, CNN) about a new observational study of hydroxychloroquine (HCQ) in COVID-19 (ijidonline.com/article/S1201-…).
In case you haven't seen it, the single-center study demonstrates a very strong association between early treatment with HCQ for hospitalized patients and a substantial (50% or greater) reduction in the risk of 28-day mortality.
I don’t think I can match @ProfDFrancis in the use of clever GIFs and polls. But my main goal here is to educate people about all the types of issues that come up in observational comparative effectiveness studies.
Although Italy had its worst day to date (6557 new cases, 793 deaths), in the "silver lining" department, it’s clear that COVID-19 cases and deaths are not growing exponentially in Italy anymore.
I made a simple spreadsheet using data from worldometers.info/coronavirus/co… to track the total number of cumulative and incident cases and deaths over the last 2 weeks (since 3/8).
If you look at the last column, which shows the “5-day moving average” you can see clearly that the relative daily increase is continuing to decrease. (3/n)