Check Your Protocol Charts Posted on December 3, 2007 Gary S. Sachs (bio) explains the importance of standardizing protocol "cheat sheets" across sites in a study.

The idea that things would be complicated and there were charts drawn up, that was something that at the end of the study, I had a different view of than while we were doing it. The idea of having little laminated charts drawn up, cheat sheets, all the different mnemonics that people made, aids for your memory. Those things got out of control.

And it turns out that as we were analyzing data and trying to figure things out, that sometimes the cheat sheets that would be made up — some of them were local, some of them distributed from our national coordination center — they weren’t always in agreement with each other.

And what was done as a shortcut in one instance became confused with what the protocol really said. And the differences would be tiny. They wouldn’t always be caught. But we saw several of them at the end of the study, and that actually did cause us a little bit of problem, need for data cleaning, sometimes need to set aside some of the data.

At the end of the study and we’re writing up the papers, a clinician who was actually involved in authoring one of these papers, as we’re writing our methods section, they would say, “Well, actually, no. I randomized somebody who didn’t meet that inclusion criteria.” “Really? Why?”

Well, because the list of the medications that were specifically allowed would be part of the cheat sheet somewhere, and somewhere else it might just say, “Mood stabilizers.” So in that person’s mind what was a mood stabilizer, and therefore allowable, was different than what the actual list of them was, and they didn’t see them side-by-side.

So little things like that can make a difference. And if you’re doing a very small study, having a few people who are disallowed could be catastrophic. In a big study, it turns out only to be a nuisance.

« Back to Article