A Rationale For Pilot Testing

AER15 Blog Squad: Michael Kraten

Many thanks to my Blog Squad colleague Courtney Zweig for exploring the proposed changes to the Common Rule involving single IRB review processes. That particular proposal is Issue #6 of 19 of the Office of Human Research Protections’ (OHRP) summary of the proposed regulatory changes in ANPRM.

If you’re reviewing the other proposed changes as well, don’t miss Issue #5 of 19 involving informed consent. According to the OHRP, modifications may be necessary because consent documents are currently “too long and hard to understand.” Under the proposed regulations, these documents would be “shorter, more readily understood [and] less confusing, [and would] contain all of the key information … to help someone make a good decision about whether to participate in a study.”

It’s difficult to find fault with such a proposal, eh? The OHRP also complains that the current regulations “provide only basic information about the elements of informed consent and how consent documents should be written.”

But as I discussed in my two previous posts regarding the recent Social, Behavioral, and Educational Research (SBER) Conference session about the Facebook emotional contagion study and the AER Conference session about big data studies, when projects utilize massive data sets, researchers often find it impracticable to conduct any type of consent activity at all.

Many traditional studies require waivers of consent activities on grounds of impracticality as well. Nevertheless, as I suggested in my previous posts, the next best alternative to a consent activity may be a debriefing activity.

What type of debriefing activity? Anything that conveys information using language that is brief, understandable, and comprehensive, naturally.

Sounds familiar? Of course it does. The OHRP’s proposed revision for consent language also serves as appropriate guidance for debriefing language. But how might that insight help someone like Professor Jeff Hancock, the researcher on the controversial emotional contagion study that utilized Facebook’s massive data set?

Well, even when a consent activity is not practicable, a researcher might find it helpful to draft brief, understandable, and comprehensive language about the study any way. And if feasible, that language could be utilized for a debriefing activity.

And if a traditional debriefing activity is unrealistic, a researcher might opt to utilize the language in a consent or debriefing document during a limited pilot test. After all, any reasonable consent or debriefing activity should be practicable during a limited pilot. And who knows? Such an activity might provide insights about consent methods that are surprisingly replicable on a much larger scale.

If not, the researcher may still obtain useful feedback about the concerns of future participants during the pilot test. Such feedback may have helped Professor Hancock prepare himself for the public debate regarding his study by helping him develop the explanations that he eventually furnished to the media.

As the IRB chair at Providence College, I had not thoroughly considered this issue prior to attending this year’s SBER and AER Conferences. Fortunately, I’m quite certain that my attendance has prepared me to more effectively assess consent waiver requests in the future.

Michael Kraten, chair of the IRB at Providence College, is a member of the PRIM&R Blog Squad for the 2015 AER Conference. The PRIM&R Blog Squad is composed of PRIM&R members who blogged here, on Ampersand, to give our readers an inside peek of what happened at the conference in Boston, MA.