Facebook Recruitment of Young Adult Smokers for a Cessation Trial: Methods, Metrics, and Lessons Learned

Abstract

Further understanding is needed of the functionalities and efficiency of social media for health intervention research recruitment. Facebook was examined as a mechanism to recruit young adults for a smoking cessation intervention. An ad campaign targeting young adult smokers tested specific messaging based on market theory and successful strategies used to recruit smokers in previous clinical trials (i.e. informative, call to action, scarcity, social norms), previously successful ads, and general messaging. Images were selected to target smokers (e.g., lit cigarette), appeal to the target age, vary demographically, and vary graphically (cartoon, photo, logo). Facebook's Ads Manager was used over 7 weeks (6/10/13 - 7/29/13), targeted by age (18-25), location (U.S.), and language (English), and employed multiple ad types (newsfeed, standard, promoted posts, sponsored stories) and keywords. Ads linked to the online screening survey or study Facebook page. The 36 different ads generated 3,198,373 impressions, 5,895 unique clicks, at an overall cost of $2,024 ($0.34/click). Images of smoking and newsfeed ads had the greatest reach and clicks at the lowest cost. Of 5,895 unique clicks, 586 (10%) were study eligible and 230 (39%) consented. Advertising costs averaged $8.80 per eligible, consented participant. The final study sample (n=79) was largely Caucasian (77%) and male (69%), averaging 11 cigarettes/day (SD=8.3) and 2.7 years smoking (SD=0.7). Facebook is a useful, cost-effective recruitment source for young adult smokers. Ads posted via newsfeed posts were particularly successful, likely because they were viewable via mobile phone. Efforts to engage more ethnic minorities, young women, and smokers motivated to quit are needed.

Keywords: Facebook; participant recruitment; smoking cessation; social media; tobacco; young adult.

Conflict of interest statement

Conflict of Interest statement: All authors declare no completing interests associated with this study.

Figures

Figure 1

Examples of two successful ads (1a), moderately successful ads (1b), and unsuccessful (1c) ads from the Facebook campaign based on the cost per unique click metric. Ad type included Standard, Newsfeed, Sponsored stories, and Promoted posts; Total Reach is the number of unique users who saw an ad; Total Unique Clicks is the number of unique clicks an ad received during the time it was turned on; Unique Click-Through Rate is the number of unique clicks divided by the number of unique users who saw an ad.

Figure 2

Facebook ad campaign reach and recruitment process. Unless otherwise indicated, percentages in each box are the number reported in that box out of the number reported in the box above. Broad targeting included “interests” specified in a Facebook profile including cigarette, tobacco, and smoking. Specific targeting included broad target keywords and additional smoking-related keywords (e.g., nicotine, health effects of tobacco, electronic-cigarette). Excluded participants were either found to be ineligible for reasons listed in the figure or left the online survey before reaching the consent page.