A recent preprint paints a grim picture of the effect of role of obesity in COVID-19 infection. A high role of C-reactive protein (CRP) is specifically indicated as a risk for critical illness.

It is possible to reduce the level of CRP by statin administration, but statin administration raises the level of ACE2. It is not known whether that increases or decreases the risk from the virus. The use of statins for coronavirus patients has been suggested - so has the possibility of the discontinuation of statin use by infected patients.

Obviously I would hope to see a study of statin use in COVID-infected patients, and one does begin recruiting this month. The entry, however, says that it will be completed in August 2021. My thought is that surely somebody could run around the New York COVID wards with a clipboard, though I'd hope for a wiser way to do this, asking patients whether they use statins versus whether they have ever turned down the suggestion of taking statins when their doctor mooted the issue, and seeing how many of which group ended up in the ICU. What are the factors preventing data like this, which could have such influence on the way people respond to the outbreak, from being collected and published within a couple of weeks?

  • 2
    $\begingroup$ Because this is not the way a clinical trial is done. If you are not blinding the patients/doctors you at least want to randomize the patients into treatment groups, otherwise you will possible severely skew the data you get out. Or select already much sicker patients for your "study" in one of the arms which will either affect the outcome or mask small effects of a drug. $\endgroup$
    – Chris
    Apr 14, 2020 at 6:16
  • 3
    $\begingroup$ I’m voting to close this question because this is a suggestion for a clinical trial of a drug, not a question about biology in the terms of SE Biology. $\endgroup$
    – David
    Apr 14, 2020 at 8:22

1 Answer 1


The problem with your setup is that by targeting an ICU ward you are already narrowing your study to those who developed severe COVID-19. A more robust design would be a retrospective cohort study, where the exposure of interest has already been recorded.

E.g. the UK Biobank (~500,000 participants) has recently incorporated SARS-CoV-2 test results into its dataset, so you could use this to:

  1. create a case/control split based on exposure (i.e. patients' historic use of statins)*
  2. study the effect this has on outcome (i.e. development of severe COVID-19)

In this study, we implemented rapid dynamic linkage, which allows us to provide a regular feed of new COVID-19 (SARS-CoV-2) test results to UK Biobank to facilitate rapid and urgent research into the epidemiological and human genetic risk factors for severe infection in the cohort.

* Or any of the other thousands of variables recorded in UKB.
** Disclaimer: I have worked on this project.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .