Using Facebook Advertising as a Recruitment Tool for an mHealth Alcohol Intervention Clinical Trial -Abstract Only

Abstract

Smartphone-delivered interventions improve access to tailored, timely and private support for adults seeking to manage their alcohol use. Researchers attempting to identify and recruit eligible participants to test alcohol interventions encounter barriers reaching this hidden population, making achievement of sample size a costly and lengthy endeavor. We aimed to recruit 150 adults, drinking above recommended limits, to a 3-month mHealth study, comparing two versions of an evidence-based alcohol app with a control condition. Facebook ads were run every two-weeks until recruitment goals were met. Interested individuals clicked on the ad and were linked to the study website, where they were screened for eligibility and consented. Eligibility criteria: Alcohol Use Disorders Identification Test (AUDIT) score between 6-20 for females and 8-20 for males; not enrolled in, or mandated to, alcohol treatment; U.S. citizen; English literacy; and smartphone possession. Facebook analytics were reviewed weekly, and ad content and frequency adjusted to improve reach. Demographic characteristics of eligible and ineligible individuals were reviewed for representativeness and to assure criteria were not systematically excluding otherwise eligible participants. Within 3-months 1,417 participants were prescreened, with 415 eligible. Of those eligible, 259 completed consent and 191 completed baseline surveys. Total advertising cost was $2,184.48 or $11.44 per enrolled participant. Participant characteristics were 54% female; mean age = 40.3 (SD = 12.6), possibly indicating preferential use of Facebook by women and older aged adults. Iterative review of demographics found that our conservative AUDIT score cut-off (i.e., 20) eliminated otherwise eligible adults who could benefit. Raising the cut-off slightly (i.e., 24) greatly accelerated achieving enrollment goals. In addition, iterative review of enrollment data identified numerous consents with the same IP address, an indicator of phishing. Once identified, we added a manual IP review for completed consents and emailed eligible participants their baseline surveys, which replaced automatic routing from consent to survey. Facebook provided an efficient, low-cost method to reach a hidden population. Frequent analytic review to maximize ad appeal, and processes to authenticate participants are recommended. Recruitment using other social media platforms may improve representation of younger adults and males.