Hydroxychloroquine update

Following up on our earlier post, James “not the cancer cure guy” Watson writes:

I [Waston] wanted to relay a few extra bits of information that have come to light over the weekend.

The study only has 4 authors which is weird for a global study in 96,000 patients (and no acknowledgements at the end of the paper). Studies like this in medicine usually would have 50-100 authors (often in some kind of collaborative group). The data come from the “Surgical Outcomes Collaborative”, which is in fact a company. The CEO (Sapan Desai) is the second author. One of the comments on the blog post is “I was surprised to see that the data have not been analyzed using a hierarchical model”. But not only do they not use hierarchical modelling and they do not appear to be adjusting by hospital/country, they also give almost no information about the different hospitals: which countries (just continent level), how the treated vs not treated are distributed across hospitals etc. A previous paper by the same group in NEJM says that they use data from UK hospitals (no private hospitals are treating COVID so must be from the NHS). Who is allowing some random company to use NHS data and publish with no acknowledgments. Another interesting sentence is about patient consent and ethical approval:

The data collection and analyses are deemed exempt from ethics review.

We emailed them to ask for the data, in particular to look at the dose effect which I think is key in understanding the results. They got back to us very quickly and said

Thanks for your email inquiry. Our data sharing agreements with the various governments, countries and hospitals do not allow us to share data unfortunately. I do wish you all the very best as you continue to perform trials since that is the stance we advocate. All we have said is to cease and desist the off label and unmonitored and uncontrolled use of such therapy in hospitalized patients.“

So unavailable data from unknown origins . . .

Another rather remarkable aspect is how beautifully uniform the aggregated data are across continents:

For example, smoking is almost between 9.4-10% in 6 continents. As they don’t tell us which countries are involved, hard to see how this matches known smoking prevalences. Antiviral use is 40.5, 40.4, 40.7, 40.2, 40.8, 38.4%. Remarkable! I didn’t realise that treatment was so well coordinated across the world. Diabetes and other co-morbidities don’t vary much either.

I [Watson] am not accusing the authors/data company of anything dodgy, but as they give almost no details about the study and “cannot share the data”, one has to look at things from a skeptical perspective.

Again, I have not looked into this at all. I’m sharing this because open data is a big deal. Right now, hydroxychloroquine is a big deal too. And we know from experience that Lancet can make mistakes. Peer review is nothing at all compared to open review.

The authors of the paper in question, or anyone else who knows more, should feel free to share information in the comments.