CONSORT 2025 RCT standards document

data analysis reporting
,
1

"Readers should not have to infer what was probably done; they should be told explicitly.” Douglas G Altman

These are the opening words for the newly promulgated standard re: RCT but searching I could not find a checklist component for the acceptable characteristics of measurement. Given that CONSORT 2025 is a standard for reporting a mathematical function, this is an important oversight.

During my initial tenure on the ASTM (FDA standards) committee, there was no standard for detection of the dynamic patterns of SPO2. This was corrected. A standard which is missing a critical component risks rendering ineffective products (here pathological RCT such as RCT mimics). The fact that a defective RCT meets the standard provides a false sense of security, false trust and false EBM. In particular I note that “RCT mimics” (such as the PettyBone RCT) (which can render false “high grade” EBM) will pass the standard.

I suggest that CONSORT 2025 be promptly amended to correct this deficiency in the checklist.

2 Likes
2

Thanks for sharing, Ilynn. I am interested to know if readers on this platform have any comments/thoughts on the CONSORT update. I am surprised estimands did not reach consensus, and I suspect the same would be for the SPIRIT update.

CONSORT-DEFINE item 12d, though, (for early phase dose finding trials) asks strategies for handling intercurrent events be defined.

1 Like
3

Many confusing things are being done about estimands (intercurrent events being one of them), so I’m glad the update didn’t go too much into them. It would have been better had the statement included something like “It is important to have very specific estimands identified and for the estimands to identify with observables that have an intent-to-treat interpretation. Beware of special treatments of intercurrent events that attempt to not count those events as bad outcomes.”

3 Likes