Special Report: Improving Informed Consent

Cognitive interviewing becoming part of process

Researcher provides foundation for questions

Sometimes the biggest problem after a clinical trials professional sits down with a participant to provide informed consent is that the participant doesn't know what he doesn't understand.

According to a new study, one solution might be cognitive interviewing in which clinical trial professionals or investigators probe samples of volunteer subjects to discover thought processes that might otherwise be hidden for the goal of improving the science of informed consent, to pretest materials for a study, or to assess participant's thought processes in the course of obtaining informed consent.1

"There is a silent misunderstanding where people think everything is fine," says Gordon Willis, PhD, a cognitive psychologist with the National Cancer Institute in Rockville, MD, and the author of a study on the topic, which was published in the premier issue of the Journal of Empirical Research on Human Research Ethics.

"The participant might believe that everyone involved in a study is going to be receiving medication that is likely to have a therapeutic benefit to them, but that's not what the consent form says," Willis says. "It may say that half will be assigned to a placebo, and that's not likely to have any therapeutic benefit."

Then there are the cases were a participant will receive a lot of information about a study, and they'll read everything and know that they're confused, but they fall victim to the tyranny of self-presentation, Willis says.

"It's the notion that we're all worried about how we come across to others, so who really wants to admit they don't understand what's going on," he says. "So this poor person who doesn't understand the materials will typically be asked only one question by investigators: Do you have any questions?"

That's a hard question for someone to answer if he or she is confused, Willis says.

"There are a number of things we can do to try to get under the surface of either eradicating silent misunderstandings or plain old confusion," Willis says. "One is to ask more questions explicitly of potential study subjects to see what they really understand."

Willis proposes doing this by borrowing the techniques used in studying survey questions.

"I've spent the last 30 years using cognitive interviewing methods to understand how people who receive survey questions understand intent, design, and why they give the answers they give back to us," Willis says. "One theme that comes out is that standardized written materials, survey questions, and instructions are very frequently misunderstood in subtle ways."

For example, when the average citizen is asked whether anyone in the family has had a dental sealant within the last 12 months, they'll quickly say, "Sure," Willis says. "Then you probe them and ask them, 'What is a dental sealant?' and they'll say it's getting a cavity filled, but that's not what the survey question means," Willis says. "Dental sealant means a plastic covering that goes on the teeth, but people hear the term and think it means something else."

So the investigator and the participant think they're connecting, but they're not because of a silent misunderstanding, he notes.

The idea is to ask a question that can be more clearly understood.

"I've spent a long, long time trying to eradicate these little devils from survey questions," Willis says. "Then I was fortunate enough to be offered the opportunity to be an IRB chair for my own institution, and at that point it became very clear that there was a connection between standardized survey questions and standardized consent materials that we used to communicate with study subjects, especially in biomedical research."

Here are the questions investigators could ask participants as part of cognitive interviewing during the informed consent process, Willis suggests:

  • What does the term 'placebo' means to you?
  • What does participation in this research study involve?
  • Can you tell me what it is you are being asked to do?
  • What are we going to do to you through the course of the study?
  • What kind of payment or remuneration are you going to receive?
  • What are the potential benefits of participation to your or to someone else?

"The idea is to see whether people can answer those questions in a way that illustrates they have reasonable comprehension of informed consent," Willis says.

When clinical trials professionals are not satisfied with participants' answers then there are different ways in which cognitive interviewing can be applied:

1. Conduct an academic study.

"The first way is to study the basic science of informed consent where we want to know in general what kinds of formulations and wording are best understood by study subjects," he says. "This would be an academic study that is not connected to a particular protocol."

A researcher would conduct a study that scripts different versions that convey a concept and then try out the versions on different individuals, using one-on-one cognitive interviews, Willis explains.

The investigator would review the material, use follow-up probing questions, and make judgments about what type of wording is most effective in conveying information effectively, he adds. Over time, researchers would develop a core data base of study results that would inform future investigators on how to convey basic information meaningfully, Willis notes.

"We'd have research literature to rely on when we develop consent materials, rather than just knowing what we did last year or making it up as we go along," Willis says.

2. Pre-test materials.

"The second thing we can do is akin to what we do in testing survey questions, and that's pre-test materials," Willis says. "We are judging the conduct of a study, and we want to see how our own materials will function for this study."

Investigators would test a consent form on individuals who are representative of those in the main study, he says.

For instance, a researcher could take a consent brochure and give it to a small set of individuals, who would participate in intensive, one-on-one cognitive interviews, Willis explains.

After the material is presented, investigators would conduct intensive probing to understand the participants' level of decision making. They might ask participants, "Given what you just read, does this sound like the type of study that is useful, and would you be interested in participating in it?" Willis says.

Researchers would learn what participants' misconceptions were and whether there are problems with the consent forms and materials, such as poor organization of ideas, too many technical terms, etc. "If we have some indication that something is not working because it's not understood, then optimally what we'll do is revise the materials," Willis says.

While this may seem burdensome to many in the research community, it's been proven effective in other contexts, such as in the field of survey research. "We routinely pretest questionnaires through cognitive interviewing, and it's well-accepted as a standard practice," Willis says. "It's almost expected that one has to do some kind of cognitive pre-testing of a survey questionnaire before going out in the field and collecting your data."

3. Embellish the ongoing study.

"The third way one can use cognitive interviews is not as basic science or pre-testing before the study, but actually as a way to embellish the study as it's ongoing," Willis says. "We can use cognitive probing within the consent process itself to find out what it is they understand."

An investigator presents the consent form, asks the participant to read it, and then asks questions.

One probing question would be, "Tell us in your own words what it is we're asking you to do as part of participation in this study," Willis suggests. If the participant cannot answer adequately, then this provides an opportunity for corrective action, he says.

"This is our way of establishing that attempts at communication really are working, and if they are not, then we need to go back and remediate the situation," Willis adds. "Cognitive interviewing can make the whole process more flexible and have a built-in qualitative check."

Sure, it will take more time, but that's a fair trade-off for having effective informed consent, Willis says.


  1. Willis G. Cognitive interviewing as a tool for improving the informed consent process. JERHRE. 2006:9-24.