Kids are going back to school this month, and that return of scholarly concerns raises a not-always-pleasant topic: grades. Students of all levels are tromping off to class, bracing themselves for homework, exercises, and the frequent question, how am I doing? This atmosphere, combined with our recent attention on the informed consent form process, has me wondering: how are we doing, as researchers and ethics committees, on our writing assignments? The FDA has sent out a draft of its expectations for informed consent, and it has provided a not-too-successful example of its own consent form language. We know we need to create consent forms that explain experiments clearly to the people who will be involved in them. But if that were a school writing assignment, how well would the research community score?
A recent survey by CISCRP (The Center for Information & Study on Clinical Participation) suggests that the answer is, “Not very well.” According to the results of that global survey the research community might be lucky to get a low B in Understandable Consent Forms.
CISCRP is a nonprofit organization dedicated to understanding and improving the clinical trial process. In particular, they concentrate on the participants’ experience (and run the laudable Medical Heroes program). CISCRP’s survey of 5,700 sought to learn more about the experiences around and attitudes towards clinical research. One topic of the survey was the perceived difficulty of an informed consent form. In answer to “How difficult was the consent form?” 81% of the people in the survey said they were “Not At All Difficult” or “Not Very Difficult.” That left 19% of the people describing this crucial-to-understand document as “Somewhat Difficult” or “Very Difficult.”
And what if that score were a letter grade? According to a handout that my son received on his first day of high school, that 81% score rates a B-.
In a similar survey in 2005, CISCRP reported that 87% of the survey takers understood their consent forms, so performance apparently has dropped. In addition, these results were self-reported: that is, the participants themselves were assessing how well they understood what they read, and other research suggests that patients over-rate their comprehension of medical explanations. So possibly these survey responders did not get as much from the informed consent document as they thought.
And that B- came from those who made it all the way through a clinical study. Those who did not complete the study had an even harder time with the consents. Only 65% of them thought the consent form was either “Not At All Difficult” or “Not Very Difficult.” My son’s high school would call that a D.
Another observer, ethicist Nicholas Steneck, also hands out some tough grades on consent forms. Dr. Steneck is the Director of the Research Ethics and Integrity Program at the University of Michigan, and in a presentation at this year’s DIA conference he shared his assessment of an informed consent project. His office studied how well volunteers understood the issues in a bio-repository program. That program went beyond the typical medium of a printed consent form to explain things; it used a pamphlet layout, with graphics and a carefully-written question and answer format. And with the consent-form writers doing all of that extra-credit work, 89% of the volunteers understood the materials.
While the high school would grade that as a B+, for Dr. Steneck it was not good enough. In the presentation, he emphasized that this accomplishment still meant that 11% of the people in the program did not understand what they were agreeing to.
While these scores might seem discouraging, we should not consider them a final grade. Informed consent involves more than a written document, and it is important to note that in another section of that CISCRP study, most survey responders said they would sign up for another study. We’ve seen how hard it is to reach target reading levels, and we know the challenges of explaining complex issues. With those challenges in mind, what can researchers do to improve these scores? A future blog post will look closer at some possibilities.