Gun violence studies can be among the most sensitive for a participant’s emotional risk. Researchers and IRBs should take the following steps to ensure participants’ protection:
• Find experts. “An IRB always does well and wise to hope there is a mental health expert or consultant on the research team, even if the focus of the study is not mental health,” says David C. Clark, PhD, director of the human research protections program at the Medical College of Wisconsin in Milwaukee.
“It’s always wise to have at least one mental health expert on the IRB review team because there always are these psychological aspects to the study,” he adds.
Investigators should have mental health professionals on their team to review surveys and identify items that might be upsetting, Clark says.
“In this area of gun-related violence or death, a huge number of people are going to have trouble thinking about and talking about these events. The investigator needs a sophisticated screening process, and will be willing to say, ‘I don’t think you are ready for an interview. Let’s call it a day and make sure you’re getting the healthcare you need,’” Clark adds.
• Be open with participants. “Make sure that informed consent is complete and thorough,” says Patrick M. Carter, MD, assistant professor of emergency medicine at the University of Michigan Medical School. “Address potential issues that might come up in the course of the study,” he adds.
For example, if research involves gun violence victims who are involved in gangs or criminal activity, researchers need to disclose that they must notify authorities if participants reveal plans to harm themselves or other people after recovering from injuries, Carter says.
“We worry about retaliation, and so we assess that and address it with appropriate protocols,” he adds.
When gun violence research involves surveys, it is important to let people know what they are going to be reading and answering questions about, says Kimberly J. Mitchell, PhD, research associate professor of psychology and senior research scientist at the Crimes Against Children Research Center at the University of New Hampshire.
“We let them know that especially if they have any personal experiences, it could be upsetting,” Mitchell says.
Researchers should give participants an opportunity to skip questions they find upsetting, she adds.
“Most of our questions are ‘yes,’ ‘no,’ ‘not sure,’ or ‘refuse to answer,’” Mitchell says. “And they just click on that, and go on.”
Investigators then look to see how many times participants chose to not answer a question.
“If it’s a large amount that is missing, that’s interesting in itself,” she adds.
• Research using de-identified data can be low risk. When researchers studied neighborhood gun violence and birth outcomes in Chicago, they collected de-identified data from birth certificates, census information, and police reports. The study found that race and ethnicity were associated with birth outcomes, regardless of exposure to gun violence.2
“None of the institutional review boards had any concerns about the study or raised ethical issues, probably because it’s all de-identified data,” says Nana Matoba, MD, MPH, assistant professor of pediatrics in the division of neonatology, department of pediatrics at the Ann & Robert H. Lurie Children’s Hospital of Chicago.
The study did not highlight specific communities, so there were no concerns about stigma or the research shedding an unfavorable light on particular groups or neighborhoods, she adds.
• Hold community meetings. Matoba and co-investigators did not hold community meetings before performing their study about gun violence and birth outcomes, but it could have been helpful, she says.
“The violent neighborhoods are really all over the city, interspersed in neighborhoods,” Matoba notes.
Since there was not one compact community that researchers could target for focus groups or meetings, and since the risk of stigma was low as the neighborhoods were not identified, they did not believe it was necessary to hold community meetings.
But even in this case, some sort of community meeting could have been positive, Matoba says.
For instance, after the study closed and researchers concluded that gun violence was not a significant factor alone in poor birth outcomes, Matoba spoke with community members and learned that they were affected by many more stressors that could harm birth outcomes. These included poverty, unemployment, lack of education, and daily life in Chicago.
“A woman would say, ‘There are gunshots, but that won’t impact my pregnancy by itself,’” Matoba says.
“Maybe if I had worked with a community from the outset, it might have been clear that it’s not just gun violence,” she adds. “This was a humbling experience, and I’d probably do it differently next time. I’d try to find out what are the other factors within the community.”
• Consider nuances in the effects of gun violence. Many gun deaths are the result of suicide. Participants who have lost loved ones to gun homicides or suicides experience a type of trauma that researchers and IRBs should consider when approaching study participants and writing survey questions, Clark says.
“It’s a very troubling sort of grief,” he explains. “An investigator thinks, ‘I’ll contact the next of kin a month or two after the coroner calls it a death by firearms, whether it’s suicide or homicide.’”
Often, family members are so traumatically affected that the interviews can be counterproductive, harmful, and exceedingly troublesome for them, Clark says.
“Sometimes, they’re altruistic and want to help, but they’re not ready and they don’t know how to gauge their readiness to answer questions,” he says. “They’ve never been investigated before, and the investigator wants more details. These are details that can be very upsetting.”
Even when people are willing to speak with investigators, they might find it difficult to recount details about the trauma and their feelings about what happened, he notes.
“Sometimes, people are willing to speak with investigators because they assume the investigator is a mental health professional, and they see this as some free time with a counselor or mental health professional who will answer their questions or help them with their mental health burden,” Clark says. “But rarely are research teams able to go into the field, do those interviews, and be prepared to make on-the-spot counseling or do referrals.”
When researchers investigate the effect of gun violence on shooting survivors, the same emotional issues can occur. Also, people who attempted a gun suicide often are not reliable sources of information about what they were thinking or feeling before they made the suicide attempt, Clark says.
“They often can’t remember what happened weeks or months later,” he says. “They can only guess what was in their minds. They are the same person, but they were in an altered state, so the investigator is pressing them for details. The person can’t recreate their mental state, and they can get pretty upset.”
• Employ well-trained staff to conduct interviews. The people hired to interview gun violence victims should have some experience with mental health and in asking sensitive screening questions, Clark says.
“Even if they are not talking to people within days, months, or weeks after the traumatic event, they are often dealing with people who have complicated grief,” he explains.
“It’s one thing if someone dies tragically of a natural cause — the grief can be full and especially difficult if the person was young,” he adds. “But when it’s a suicide or homicide, the violence adds a layer of being stuck with wondering about the violence done to the body and what it felt like.”
The research staff screening these participants must be sensitive to the fine line that separates people who are not emotionally ready to spend a lot of time going through the details and those who are, Clark says.
What IRBs and researchers do not want is for their interviewers to be “mops,” which are people who leave a mess for someone else to clean up when they leave, Clark suggests.
For instance, when research teams interview students who have been affected by gun violence, they might find the youths to be as expected: They are experiencing some grief, but are steady and capable of handling the questions. Then, a few days later, researchers learn that the surveys started something that resulted in the students becoming inflamed, very upset, and excited. For some days they are emotionally distraught, making things worse in the school, Clark explains.
Researchers who work with youth directly affected by gun violence are very sensitive to the emotional impact of the research.
“Our interventions are done with an approach that emphasizes empathy for what the patient/participant has been through,” Carter says. “Our therapists are trained to assess whether patients are experiencing any distress, and address that by referring them to available resources.”