Blinding Checks for labels in a clinical study

[psyched1]

I have been asked to create a blinding check for labels in a clinical study. Does anyone have an idea for how this would work?

[Scott Catron]

An explanation of what a blinding check is may be in order - a Google search didn't turn up much.

[psyched1]

Yes I googled first and got a lot on blinding studies but nothing on ensuring that materials do not unblind the study. We are using both human and visual systems to inspect while I can show consistancy in th manufacturing process I still need to account for estitic features to ensure that the packaging does not unblind the study.

[Jim Wynne]

Originally Posted by psyched1

Yes I googled first and got a lot on blinding studies but nothing on ensuring that materials do not unblind the study. We are using both human and visual systems to inspect while I can show consistancy in th manufacturing process

Can you explain a bit more about what you're trying to do? It sounds like you're doing some sort of blind study on some sort of labels in a clinical setting, but that's just a guess.

Originally Posted by psyched1

I still need to account for estitic features to ensure that the packaging does not unblind the study.

Do you mean aesthetic features?

[Tim Folkerts]

Here is my guess (but it is only a guess).

The difference between the true item and the placebo must to clear and obvious to someone who is designing the study, but not clear an obvious to the doctors and patients.

For example, you wouldn't want to put the true items and the placebos in different sized containers, or use red labels for the true items and blue labels for the placebos. Even if the labels were quite similar, if they came from two different printers it might be possible to detect subtle differences and hence guess which one was the placebo.


If this is the case, the first step in a "blinding check" could start with a check list of features to watch for in the pills themselves or the packaging they come in. I'm not an expert in that field, but size & shape & colors of the pills, the packages, and the labels come to mind.

After checking all the obvious things yourself, perhaps you could gather a small group of people like the doctors who will be involved (but not the ones who would be participating in the actual study!) and ask them if they can detect and differences.


Tim F

[psyched1]

I am only concerned with the labeling aspects of the study and ensuring we do not unblind the study. I have been asked to design a blinding/unblinding test of the labels that will be used. Being relatively new to pharmaceutical and clinical manufacturing I am looking for a scientific approach to determine if the labels with unblind the study.

Size and shape and colors (Black) are consistent. In printing labels we always have some minor defects for paper flaws or dust in the printing process so I guess the question is do I attempt to demonstrate how random these are as to not cause a concern for unblinding the study while still trying to consistently produce identical labels or do I just display how consistent the labels are?

Keep in mind we do employ vision systems(OCR/bar code scanning and a robotic proof reader) through out the processes.