Metaresearch on patient-reported outcomes in trial protocols and results publications suggested large outcome reporting bias, 2025, Heravi, Busse et

[Sean]

But it does make one significant source of bias less likely to be an issue.

Or at least easier to point out.

Getting it taken into account and acted on is another matter entirely. As we have learned the hard way.

Things like the publishing of protocols

And data.

 

[Creekside]

There are things that could be done to improve PRO inclusions, such as reporting what % of participants fully filled out the questionnaire. A low number might indicate selection bias (or fatalities, or "too sick to fill out a form"). Whichever improvements are in the guidelines, the problem lies in motivating people to follow them properly, and in dealing with the people who find ways to abuse them. Guidelines may work for ethical people, but they're not the problem the guidelines are meant to solve.

 

[bicentennial]

... Whichever improvements are in the guidelines,

the problem lies in motivating people to follow them properly,

and in dealing with the people who find ways to abuse them.

Guidelines may work for ethical people, but they're not the problem the guidelines are meant to solve.

Academic institutions teach and train the teacher-trainers who got their students into such bad habits, and perpetuated it, by forming the concepts and practices of the next generation.

Do their employers know what they did? Are their employers versed in the scientific requirements of standardised research, or just relying on the marketed approval by publishing media? Is that what it takes for a thesis to pass an exam? Who set such low standards?

I don't know how the qualification of professionals is supposed to work, but it looks like ungoverned treacher-trainers allowing their students to play with loose canon.