thanks, I was not aware of this recent controversy at Cochrane. I think this supports the idea that Cochrane was considered the authoritative meta-review group. Did this change now? I looked up – there is nothing substantial coming later after this 2018 controversy, hat is partially personal, partially scientific. The only other later articles are by the person who lost in 2018 and continues writing about it on sites with such intriguing names as “mad in america”…
On substance, the discussion is
1) are RCTs biased by design as they are often funded by industry
2) do RCTs reflect clinical experience, as they often provide a simplistic set up and weed out complex patients, preferring patients with only one diagnosis, and ignore clinician input.
I think both points may be valid, but to what degree? RCT meta-analysis still seems to be the best way to get noise out of the data. Hopefully, analysis of real data will find way into meta-analysis. This seem to be happening now with COVID. Some of the real data analysis, including from Israel, uses post-processing that emulates RCTs – by pairing patients from 2 groups based on similar demographic attributes.