The effectiveness of social media (Facebook) compared with more traditional advertising methods for recruiting eligible participants to health research studies: a randomized, controlled clinical trial

Background: Recruiting participants for research studies can be difficult and costly. The popularity of social media platforms (eg, Facebook) has seen corresponding growth in the number of researchers turning to social networking sites and their embedded advertising frameworks to locate eligible participants for studies. Compared with traditional recruitment strategies such as print media, social media advertising has been shown to be favorable in terms of its reach (especially with hard-to-reach populations), cost effectiveness, and usability. However, to date, no studies have examined how participants recruited via social media progress through a study compared with those recruited using more traditional recruitment strategies.

Objectives: (1) Examine whether visiting the study website prior to being contacted by researchers creates self-screened participants who are more likely to progress through all study phases (eligible, enrolled, completed); (2) compare conversion percentages and cost effectiveness of each recruitment method at each study phase; and, (3) compare demographic and smoking characteristics of participants recruited through each strategy to determine if they attract similar samples.

Methods: Participants recruited to a smoking cessation clinical trial were grouped by how they had become aware of the study: via social media (Facebook) or traditional media (eg, newspaper, flyers, radio, word of mouth). Groups were compared based on throughput data (conversion percentages and cost) as well as demographic and smoking characteristics.

Results: Visiting the study website did not result in individuals who were more likely to be eligible for (P = 0.24), enroll in (P = 0.20), or complete (P = 0.25) the study. While using social media was more cost effective than traditional methods when we examined earlier endpoints of the recruitment process (cost to obtain a screened respondent: AUD $22.73 vs $29.35; cost to obtain an eligible respondent: $37.56 vs $44.77), it was less cost effective in later endpoints (cost per enrolled participant: $56.34 vs $52.33; cost per completed participant: $103.66 vs $80.43). Participants recruited via social media were more likely to be younger (P = 0.001) and less confident in their quit attempts (P = 0.004) compared to those recruited via traditional methods.

Conclusions: Our study suggests that while social media advertising may be effective in generating interest from potential participants, this strategy's ability to attract conscientious recruits is more questionable. Researchers considering using online resources (eg, social media advertising, matrix codes) should consider including prescreening questions to promote conversion percentages. Ultimately, researchers seeking to maximize their recruitment budget should consider using a combination of advertising strategies.