Statin-related side effects: the recent Lancet publication is biased toward false-negatives

Dear Erin,

First of all, I want to make it clear that I, too, am convinced of your genuine intentions in writing your comment.

That said, I disagree with almost everything you wrote. To begin with, if the prior assumption is that plausible AEs are virtually absent based on the existing literature, then there is no need to resort to a series of statistical rituals grounded in nullism, dichotomization, reification, and the significance fallacy to claim as much. If there is an initiative aimed at evaluating adverse effects, then this should be conducted with the highest level of methodological rigor. I find it incorrect - and dangerous - to criticize papers that point out inconsistencies based on the very strong assumption “any method is acceptable since the results will ultimately be interpreted in light of the belief that AEs are rare.”

Second, I cannot endorse your general narrative for three main reasons: i) the accounts of clinicians and methodologists I know - some of whom have worked directly for regulatory agencies; ii) what has already occurred in the history of medicine (including striking examples, like the Vioxx case); and iii) the evidence that has emerged in recent years regarding phenomena such as regulatory capture and the large influence of large pharmaceutical companies on decision-making bodies (e.g., 1], 2], 3], [[4]). The current system of evidence generation is not neutral and tends to overstate the strength of the evidence produced according to very clear incentives - ranging from [securing funding to the need to take [political positions.

To give just one good example among many, consider the case of the WHO, which at the beginning of the pandemic stated, “it is very clear right now that we have no sustained human-to-human transmission” with reference to SARS-CoV-2, and published outreach posts containing statements such as “FACT: #COVID19 is NOT airborne.”

There would be further aspects to address (e.g., the establishment of cost-benefit assessments based not on surrogate endpoints but on patient-centered and multivariate endpoints). Nonetheless, I think what I have outlined above is more than sufficient to argue that, for the good faith I believe guides your comment to be fully justified methodologically, a far greater effort is required than simply adhering to priors that are convenient for the most powerful and robust stakeholders.

Openly calling out what one considers to be methodological flaws in research - regardless of the topic under examination - is, and must remain, one of the indispensable ethical and epistemic principles of scientific conduct, in order to safeguard both public health and the production of knowledge.