JMIR Publications

ABSTRACT

Background:

Recruitment and retention of research participants is challenging. Social media, particularly Facebook, has emerged as a tool for connecting with participants due to its high uptake in the community. The Environmental Determinants of Islet Autoimmunity (ENDIA) study is an Australia-wide prospective pregnancy-birth cohort following children who have a first-degree relative with type 1 diabetes (ACTRN1261300794707). A dedicated Facebook page was established for the ENDIA study in 2013 with the aim to enhance recruitment and support participant retention.

Objective:

The purpose of this investigation was to evaluate the long-term impact of Facebook as a recruitment and retention tool. The hypotheses were that (1) Facebook was an important source of referral to the ENDIA study, (2) the sociodemographic characteristics of participants recruited by Facebook would be different from those of participants recruited by other means (i.e., ‘conventional recruits’), and (3) recruitment by Facebook would be associated with long-term retention. We also evaluated the most effective types of Facebook content based on post engagement.

Methods:

Recruitment of 1511 ENDIA participants was completed in December 2019. Characteristics of participants recruited through Facebook were compared to conventional recruits using linear, logistic, and multinomial logistic regression models. A logistic regression model was used to determine the risk of study withdrawal. Data pertaining to 794 Facebook posts over 7.5 years from June 2013 until December 2020 were extracted using the Facebook ‘Insights’ function for thematic analysis.

Results:

Facebook was the third largest source of referral to the ENDIA study (300/1511; 19.9%) behind in-person clinics (500/1511, 33.1%) and healthcare professional referrals (347/1511, 23.0%). The ENDIA Facebook page had 2337 followers at the close of recruitment. Approximately 20% of these could be identified as participating parents. Facebook recruits were more frequently Australian-born (P<.001) enrolling postnatally (P=.01) and withdrew from the study at a significantly lower rate compared to conventional recruits (4.7% vs 12.3%; P<.001) after a median of follow-up of 3.3 years.

Conclusions:

Facebook was a valuable recruitment tool for the ENDIA study and participants recruited through Facebook were three times less likely to withdraw during long-term follow-up. The sociodemographic characteristics of Facebook recruits were different to conventional recruits, but perhaps in unintended ways. Facebook content featuring stories and images of participants received the highest engagement despite the fact that most Facebook followers were not enrolled in the study. These findings should inform social media strategies for future cohort studies involving pregnant women and young families, and for type 1 diabetes risk studies. Clinical Trial: Australia New Zealand Clinical Trials Registry: ACTRN1261300794707