Apologies, I'm getting confused with the new tool. The answer would be that they used patient-reported outcomes, but whether that indicates a high risk of bias is a judgement call, and I'd say no in this case due to the things I've mentioned
-
-
Replying to @GidMK @Students4BE
"Is it likely that assessment of the outcome was influenced by knowledge of intervention received?" with patient-reported symptoms in trials of homeopathy given as example of 'yes', & you'd answer 'no' for this CBT trial relying on subjective self-report? I don't understand why.
2 replies 0 retweets 0 likes -
Replying to @MEMilitant1 @Students4BE
No it is certainly likely that the assessment was influenced. Whether this biases the study, however, is another question, and that's the point. They implemented a number of controls to reduce this bias, so I'd argue that it's not a high risk
1 reply 0 retweets 0 likes -
Replying to @GidMK @Students4BE
But if you agree assessment was likely influenced then following RoB2 you necessarily get a high risk of bias in that domain. I'm not so impressed by their other controls to reduce bias, but subjective judgements on that are outside of this assessment tool anyway.pic.twitter.com/N71k4eBTat
1 reply 0 retweets 0 likes -
Replying to @MEMilitant1 @Students4BE
Mmm that's the default, but as I said RoB assessment isn't a hard rule. Look at the sentence underneath the title of that graphicpic.twitter.com/ZhtiXjh7bq
1 reply 0 retweets 1 like -
Replying to @GidMK @Students4BE
That line wasn't in the copy I was looking at: https://drive.google.com/file/d/1jjOMDIZMtDcXS9QFD-n8DIRlp06CRLEc/view … My 'necessarily' may have been too strong, but I don't see any clear reason for over-riding the default for this trial. It fulfils that RoB2 criteria for 'high risk', and with good reason imo.
1 reply 0 retweets 0 likes -
Replying to @MEMilitant1 @Students4BE
Oh odd. I just went to the Cochrane site and navigated to the RoB tool. I've said why I don't think it's high risk so I guess we'll just have to agree to disagree
1 reply 0 retweets 0 likes -
That being said, we agree that it's likely not possible to run a trial at less risk of bias on this topic so why does it matter? This is probably the best you can get, doesn't mean CBT is 100% effective but it does provide some fairly strong evidence
1 reply 0 retweets 0 likes -
Replying to @GidMK @Students4BE
I don't agree! eg I linked to this old RCT of CBT for CFS that included a placebo & drug arm, helping to account for problems with bias: https://www.amjmed.com/article/0002-9343(93)90183-P/abstract … Also, if A vs A+B trials & subjective self-report outcomes provide fairly strong evidence for CBT, they do for CAM too.pic.twitter.com/DHXuYOlk4q
1 reply 0 retweets 0 likes -
Rgardless of the merits of this trial design, researchers should make clear to readers, patients & journalists that reported improvements may merely reflect problems with bias. We can't be certain how much this is a problem but people should be informed of risk.
1 reply 0 retweets 0 likes
Well I can't disagree with that. But frankly, I don't think communicating these complex issues is going to happen any time soon in most media
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.