The trusted source for
healthcare information and
Behavioral research risks may not be life or death, but subjects could suffer
Think outside of biomedical box when assessing risks
IRB members who are accustomed to dealing with the serious potential physical risks associated with many biomedical research projects may give little thought to the risks inherent in social-behavioral research.
But this attitude is a mistake, according to two social-behavioral research experts, who are scheduled to speak at the Contemporary Issues in Human Research Protections conference on Nov. 17 at the Iowa Methodist Medical Center in Des Moines.
IRBs that deal primarily with medical research sometimes lack an appreciation for the risk possibilities in social-behavioral research projects, notes Brenda Ruotolo, CIP, team manager for one of the institutional review boards at the Columbia University Health Sciences in New York City.
"It’s obvious when you’re administering an investigational drug that you are doing research," Ruotolo says. "You have a new drug and are testing it for safety and efficacy, often collecting quantifiable data."
In social-behavioral research, the methodology and data collection may be very different, she adds. "However, subjects deserve the same protections, and IRBs need to know how to apply the same review criteria."
Also, biomedical IRBs may focus on physical risks and overlook the range of risks that includes social, economical, financial, and psychological, Ruotolo reports.
Some researchers and IRB members erroneously believe that the federal regulations for protection of human subjects do not apply to social-behavioral research, says George Pospisil, MA, CIP, public health analyst for the division of education and development at the Office for Human Research Protections in Rockville, MD, which is cosponsoring the meeting.
However, it’s a myth to think the federal regulations did not include social-behavioral research because the documents that preceded the regulations definitely had behavioral and social research in mind, Pospisil says.
Historical examples of research abuses include social-behavioral studies: for example, one study about stuttering, he notes.
"There was a study about stuttering involving young people, some who stuttered and some who did not stutter but were induced to stutter," Pospisil says. "Some of those people who did not stutter before the study continued to stutter the rest of their lives and suffered emotional and personal consequences of that."
Decades later, some of the families of those research subjects are bringing litigation against the research institution, he adds.
Other examples of social-behavioral research that had greater than minimal risk included a study in which research subjects were asked to play the role of guards or prisoners in a mock prison.
"The study was forced to be terminated in the middle of the study because of the emotional damage done to some subjects," Pospisil recalls.
Another classic example is a study in which subjects were placed at a console and told they had control over giving electrical stimulation to another person, he says.
"They were led into a situation in which they thought they were delivering lethal levels of electricity to another person to kill them; and because of the guilty feelings and embarrassment they felt, some of those people have needed psychological help," Pospisil says.
"Those are just some examples that illustrate that social-behavioral research risks can be more than minimal, and that psychological risks are real risks," he adds.
In more recent times, the risk of disclosure of private health or personal information can create a significant risk to some research subjects.
"If you’re doing a survey over the telephone, depending on the nature of the questions and focus of the study, there can be some information that’s collected that could put the respondent at risk if that information was disclosed," Ruotolo says. "And yet it can be routine in a protocol that involves a telephone interview to request a waiver of consent."
Privacy increasingly is an important risk consideration as researchers make greater use of the Internet for surveying subjects and visiting chat rooms, Pospisil says.
Concerns about privacy and other risks pertinent to social-behavioral research have led to the creation of the Social Behavioral Research Working Group of the National Human Research Protection Advisory Committee, Pospisil says.
The working group has been developing a set of principles that would apply more specifically to behavioral research, he adds.
"Quite frankly, in cases of behavioral research most of the activities involve minimal risk, and there is flexibility built into the regulations," Pospisil says. "So that even if an IRB is reviewing the research, the flexibility allows the work to be reviewed a little faster and with fewer restrictions being applied."
IRBs that deal with social-behavioral research can make good use of exemptions, waivers, and expedited review procedures, Pospisil advises.
However, IRBs should keep in mind that sometimes when investigators submit a proposal for exempt research they may not include pertinent information that would help the IRB determine whether the proposal truly does pose minimal risk, Ruotolo says.
"They may not make it clear whether they are collecting identifiers," she adds.
A solution would be for IRBs to require investigators to list all of the necessary information and to make it clear that if investigators leave out pertinent information about the study then it will take the IRB longer to review the proposal, Ruotolo states.
"The focus comes down to the IRB to have clear policies, and these need to be communicated to investigators and staff," she says.
"If an institution has a social-behavioral research board, then that’s great. But if one IRB is reviewing both biomedical and social-behavioral research, then the board needs to be well educated and have members with appropriate expertise," Ruotolo advises.
If it’s really research
IRBs and institutions need to avoid both treating social-behavioral research the same way they would treat biomedical research, and they need to avoid the other extreme of ignoring it altogether, Ruotolo and Pospisil say.
"I see two practices out there in the field that are the extremes away from the actual definition [of human subjects research] itself," Pospisil says. "Sometimes IRBs are very zealous in application of the regulations and want to extend the jurisdiction of regulations to activities that may not be research."
In these cases, OHRP officials will try to let IRBs know that they don’t need to call every classroom exercise, for example, research unless it is conducted in the outside world or the institution’s own policies require them to review such protocols.
"The IRB has to know the regulations, and the institution has to have firm policies on how they are applying the regulations," Ruotolo says.
Likewise, IRBs and institutions need to make certain researchers and faculty are aware of the difference between research that needs to be reviewed by the IRB and classroom exercises that do not.
"On the other side of the pendulum, there are some principal investigators who are professors who assign research to students that becomes real research and is designed to contribute to generalizable knowledge throughout the world and is designed to be published," Pospisil says. "And sometimes those people don’t feel the IRB has jurisdiction over their work since they’re making it part of their curriculum in their classroom."
Deception and debriefing
Another risk area to consider when reviewing social-behavioral research involves the use of human subjects in a study that employs deception, he says.
"Deception is a thoroughly legitimate area of study, and it involves using human subjects who either, a) may not know that they are subjects or, b) know that they are subjects, but are misled about the purpose of their participation," Pospisil explains.
This type of research needs very careful review, and precautions need to be built into the study so that if people begin to experience emotional discomfort or severe reactions, there are interventions planned to assist them, he says.
Also, IRBs should consider whether and how investigators would debrief the subjects. In some cases an IRB may decide that it would be better to not let subjects know the true nature of the research in which they were involved, while in other cases subjects would be told the truth in a very careful and sensitive manner, Pospisil says.
"A scientific study might create situations in which people would do things they wouldn’t normally do or approve of doing," he adds.
To minimize risk in social-behavioral studies, Pospisil makes these recommendations: