A small but growing number of researchers are turning to a controversial approach to recruit hard-to-reach populations. They are contacting social media influencers with thousands of followers to help spread the word about clinical trials. “These are people who have a standing in the social media space who can help you connect with the population you are trying to study,” says Katherine Wentzell, PhD, PNP, pediatric nurse practitioner at Joslin Diabetes Center in Boston.

Wentzell and colleagues outlined steps for researchers to follow, including IRB approval and engaging with influencers.1 “All of the authors have been successful [when] recruiting for our studies using this method. It is new and innovative, so there are not a lot of details out there,” Wentzell reports.

When recruiting for a diabetes study, Wentzell contacted about 20 influencers. Eight of them agreed to partner with her to recruit. “Though my study was an anonymous survey, I did ask how participants found out about my survey,” Wentzell explains.

About 150 participants (more than half the sample) reported learning about the survey through a post by a person or hashtag they followed on Instagram. After successfully using social media influencers to recruit, the authors realized many other investigators probably were interested in this approach. However, there was no clear guidance on ethical considerations and IRB issues. “We felt like we were reinventing the wheel each time. After doing that work, we felt that we should share what we learned,” Wentzell says.

Many researchers struggle to recruit hard-to-reach populations. During the COVID-19 pandemic, it became even more difficult since researchers could not recruit in clinics. “We decided to write a ‘how to’ paper on how to interact in the social media world as a researcher, and to reach people who you would not be able to reach otherwise,” Wentzell says.

A good example is when researchers recruit young adults with type 1 diabetes. “We often lose that population to follow-up during the transition from pediatric care to adult care. It is so hard to find ways to recruit them in clinical spaces,” Wentzell explains.

Other researchers struggled to recruit racially diverse, older adults with type 2 diabetes. Those patients tend to seek care in many different places; some go to primary care physicians, and others see endocrinologists. “There are all these people out there, but they don’t necessarily have one clinical home where we can find them,” Wentzell observes.

Other populations are even harder for researchers to reach. When investigators partner with a social influencer to recruit their followers, it effectively adds a third party to the recruitment process. “You are recruiting twice,” Wentzell says.

First, researchers have to convince the social media influencer to partner with them. Next, investigators have to follow up with potential participants who found out about the study through the influencer. “The power of influencer culture is that followers look to the influencer for advance or word-of-mouth information, as we’ve seen in marketing,” Wentzell notes.

In the same way, when the influencer posts about a research study, it confers legitimacy to the people who follow that influencer. “It’s a different way to connect with your study population,” Wentzell says.

To do this effectively, researchers must become familiar with the social media environment relevant to their study population, find the right influencers, and contact those people. Wentzell started with hashtags that represented the study population. Next, she saw which people were using those hashtags, learned who had accumulated many followers, and who posted content that connected with the study population of interest.

Wentzell only sought out influencers who disclosed they were living with diabetes and posted about their condition. “This is one of the limitations of this recruitment approach, because we know that not all young people disclose their diagnosis on social media,” Wentzell says.

For researchers who lack social media know-how, the first step is creating accounts themselves. “It might be necessary to create an Instagram account when they normally wouldn’t have one,” Wentzell offers.

It is necessary to learn how to connect with their study population. Concurrently, researchers have to keep the IRB’s perspective in mind every step of the way.

“One of the big things we recommend is building a rapport with the IRB. We have found it does take some time for IRBs to get comfortable with this innovative approach,” Wentzell says.

Probably the biggest hurdle is lack of control over what people are going to post about the study. Usually, IRBs want to approve the precise language and images used. When researchers are approaching someone else (i.e., social media influencers) to recruit, it is another story. The language is going to be more conversational, and communication happens informally through private messages or public comments. “You can’t just send them a form letter,” Wentzell says.

Once the influencer agrees to post something about the study, researchers can offer suggested wording, but it will not be followed word-for-word. For example, an influencer might share stories of feeling overwhelmed living with diabetes. Then, the influencer would post, “I am partnering with a researcher [and include @username for transparency] to recruit for a study. Please see the approved language in the comments.”

At that point, Wentzell would post the IRB-approved language: “Help us understand what it feels like to live with T1D as a young adult! And get a chance to win 1 of 5 $100 Amazon gift cards. All young adults (ages 18-30) with T1D are eligible. Tag your friends! Click the link in bio.”

What makes the influencer appealing to the followers is his or her unique style. “The power of influencers is in their brand, so the images and text should be the influencer’s own creative content,” Wentzell says. “The IRB can get a little fussy about that.”

IRBs will ask, “What is the influencer going to say about the study?” Researchers cannot answer that. “As researchers, our role is to be cautious about editing or revising what the influencer is saying,” Wentzell cautions.

IRBs are accustomed to approving the exact language used to promote the study. “Though some IRBs may ultimately require this, it is important to note that this may make the influencer partnership less effective,” Wentzell stresses.

IRBs often want to hear the researcher will write the influencer’s post in its entirety. Researchers cannot give those assurances because the influencer’s authentic voice is how he or she connects with followers. “The IRBs are not used to hearing that someone who is not research-trained is promoting the study,” Wentzell observes.

IRBs also will want to know how the research team is going to monitor comments made by other people about the original post. “It’s important to let the IRB know you will be following the post, looking at comments, and keeping a log documenting all the communications,” Wentzell says.

Researchers can assure IRBs they have a plan to track comments. If any comments are incorrect or unsafe, researchers can ask the post be taken down or comments turned off, and the IRB will be notified. “But it is out there in the world. You do have less control over that,” Wentzell says.

For investigators, “the important thing is to emphasize that the influencer’s creative content is simply the flashing lights, and is just to get people interested. It is not the recruitment language, which is separate, and is IRB-approved,” Wentzell says.

It is not unlike placing recruitment posters on brightly colored paper in an attention-getting spot. After reading an influencer’s post, anyone interested in participating still has to click on a link, call a number, or receive email correspondence. At that point, the researcher will use only IRB-approved language and IRB-approved consent processes.

Another ethical concern is researchers must be transparent about who they are and their intentions when engaging with people on social media. The same always was true for in-person interactions. “But the risk with social media is that there is something in between you,” Wentzell says.

To mitigate this, researchers use clear language on social media profiles used for their research, such as “I am a scientist interested in diabetes research.”

Above all, researchers interested in social media recruitment via influencers must be proactive in building a rapport with the IRB even before submitting protocols. “Start the conversation,” Wentzell offers. “Find out what the IRB is most worried about.”

REFERENCE

  1. Wentzell K, Walker HR, Hughes AS, Vessey JA. Engaging social media influencers to recruit hard-to-reach populations. Nurs Res 2021; Jul 30. [Online ahead of print].