Most research informed consent forms are written at a high school reading level when many people who are being recruited for studies might need language nearer to a fifth-grade level, according to a new study.
Researchers reviewed more than 200 informed consent forms, finding that the mean readability was at the 10th-grade level for the IRB-approved forms.1
For the study, researchers developed a plain language informed consent template that was at the fifth-grade level.1
“We aim for a fourth- to eighth-grade reading level for informed consent forms,” says Edith Paal, MSJourn, MPH, IRB program manager at the University of Arkansas for Medical Sciences (UAMS) in Little Rock.
“That level is generally appropriate for our population,” Paal says. “We need to make the information very clear to people in the consent process.”
The need is there: “We were frequently getting requests to help simplify informed consent,” says Kristie Hadden, PhD, associate professor in the college of medicine and executive director of the UAMS Center for Health Literacy.
The Center for Health Literacy worked with the UAMS IRB and the clinical translational research center to improve the institution’s research informed consent.
“I researched best practices, problems, and did research with patients with low health literacy,” Hadden says. “There is quite a lot of literature out there that demonstrates that informed consent for research is difficult for most people to understand.”
While the average reading level in the United States is ninth grade, multiple studies have demonstrated that informed consent forms are written at a higher reading level, she adds.
“If you want people to understand the study, and if you want to provide informed consent from an ethical standpoint, then they need to be written at a sixth- to eighth-grade level,” Hadden says. “When we established a baseline, looking back at informed consent from 2013 to 2015, we found the mean readability was 10th grade, so we knew we had room to improve.”
People who have low health literacy and numeracy, meaning they struggle understanding numbers and quantitative information, have difficulty understanding a study’s risks and benefits, she says.
“Risks are often expressed in percentages and proportions, and this can be difficult for a lot of people to understand,” Hadden explains. “They might have a completely different view of the study than what the investigator intended.”
The UAMS IRB works with investigators who initiate studies on improving their informed consent language.
“A lot of times, especially with investigator-initiated studies, we see the principal investigator, who is writing the consent form, is very well-versed in his or her subject, but is not cognizant that the consent’s audience is not as familiar with the language,” Paal says. “It’s language that investigators use every day when talking with research staff and clinicians, but their words are not how their subjects would describe it.”
For example, in the section on risks, the investigator might use the word “pruritus” to mean “itching,” when “itching” would work just as well, Paal says.
Another common word informed consents use is “hypertension.” This is a common word to investigators and healthcare professionals, but the average person might not know what it means, Hadden says.
“You can put ‘hypertension’ in a title and explain what it is at the beginning, but from then on use the words ‘high blood pressure’ because that’s the language patients will use,” Hadden suggests. “Write informed consent from the perspective of the participant.”
Sponsored research also has informed consent problems, but it’s easier for an IRB to encourage readability changes when it involves an investigator-initiated study.
“Industry-sponsored informed consents are the most challenging ones to read, but we don’t have a lot of flexibility in what we can tell them to change,” Paal says.
Or, even if an IRB has flexibility, there’s the logistics problem, says Jennifer Holland, JD, IRB director at UAMS IRB.
“It would require a complete rewrite of the consent form,” Holland says. “We can ask sponsors for certain terms to be put into lay language, but we just don’t have the manpower to rewrite it, and they’re usually not willing to make an extensive change.”
Giving feedback to sponsors on their informed consent readability can lead to some positive changes. Some sponsors have said they would work on improving the language and did make changes that improved upon the original version, Holland notes.
One possibility, which the IRB has not yet implemented, is to supplement the industry sponsor form with a brief summary with highlights tailored to the local audience. It would be something easy to read quickly and it would specifically refer to the main consent form. Or, it could be a brochure that uses pictures and simple language, Paal says.
The informed consent study found that about half of investigators voluntarily used the the plain language template, Hadden says.
“More importantly, there was a 658% increase in written informed consents that were written at eighth-grade level or below,” she says. “If they didn’t use the template to improve the reading level, it was too high — a 10th-grade level.”
The next step in improving informed consent readability is to study the results of improvements.
UAMS has not yet tested the simplified template in a randomized trial, but when the institution conducts this study, the results could help the IRB decide on next steps.
“One of the things I found in designing our next step for the research agenda is that a lot of studies have looked at a simplified informed consent vs. a standard informed consent, and not many demonstrated a difference in comprehension,” Hadden says.
“Participants may prefer the simplified version, but they rarely found meaningful differences in comprehension,” she adds.
Hadden theorizes this is because participants were asked questions that were difficult to comprehend — even at a high reading level.
“When we do a study, we’ll use the simplified template and include the teach-back method of confirming understanding,” Hadden says. “We’ll ask people to explain the study in their own words to the research coordinator.”
This should help the study discern comprehension differences between the plain language informed consent and the standard consent. “We’ll find out how the template works for participants of low health literacy,” she adds.
1. Hadden K, Moore T, James L, et al. Understandable to the subject: Plain language IRB informed consents for research. Poster presented at the PRIM&R Advancing Ethical Research Conference, Nov. 5-8, 2017, in San Antonio. Abstract: 26.