Cancer Epidemiology Unit CeRMS and CPO-Piemonte University of Turin Turin, Italy firstname.lastname@example.org (Richiardi, Pivetta, Merletti)
The NINFEA study was supported by Compagnia di San Paolo/FIRMS, the Piedmont Region, and the Italian Ministry for University and Research. The authors reported no other financial interests related to this research.
To the Editor:
Although the scientific community is beginning to recognize the potential of Internet-based recruitment for epidemiologic studies (especially for cohort and randomized intervention trials), methodological research on how to validly and efficiently recruit populations through the Internet is still in its infancy. The NINFEA study is an ongoing Internet-based mother-child cohort in Italy, which to date has recruited 4100 pregnant women.1,2 We assessed the efficiency of an advertising campaign on the social network Facebook as a tool for study recruitment.
The NINFEA cohort is made up of children born to Internet users who become aware of the study and, at any time during their pregnancy, register to participate in the study through the NINFEA project Web site and complete the first questionnaire. We advertise the study in several ways, including active collaboration with health personnel involved in pregnancy and childbirth who distribute leaflets or introduce the study to pregnant women, links to the project Web site posted on other Web sites dedicated to pregnant women, and participation in online pregnancy-related forums.
In December 2009, we created a Facebook page, which is updated weekly. The page is both a means to communicate with participants and a tool to enhance study recruitment. On 6 May 2011, we began piloting an advertising campaign on Facebook (www.facebook.com/advertising). The invitation to participate in the NINFEA study was displayed on the personal Facebook page of women aged 18 to 45 years with an IP address in Italy whom we selected by applying 2 Italian keywords (in English: “pregnancy” and “delivery”) to “likes” and “interests” listed in Facebook profiles. We used a cost-per-click approach (ie, we paid a fee each time someone clicked on the advertisement). A total budget of 200 euros was allocated to the campaign, which ended on 20 June 2011 after the advertisement had been displayed 1,829,115 times and was clicked on 624 times (corresponding to a cost of 0.32 euros per click).
At the beginning of the advertising campaign, the NINFEA Facebook page had 283 fans (acquired between December 2009 and May 2011). This number increased to 450 fans by 20 June 2011. According to the Facebook report, 150 of the 167 new fans were obtained owing to the advertising campaign. Google analytics indicated that during the 5-week advertising campaign, 8 participants came to the study through Facebook, compared with 15 women in the 17 months before the campaign. Eight women reported to us that they learned of the NINFEA study through Facebook, thus validating the figures obtained from Google analytics.
Assuming that all 8 women were recruited because of the advertising campaign, the cost was 25 euros per participating woman. For the sake of comparison, the NINFEA study recruited about 1200 women in 1 year between 20 June 2010 and 19 June 2011. We roughly estimate that we used one-half person-year of work for recruitment. Considering a cost of 30,000 euros per person-year, this translates into approximately 12.5 euros per participating woman. Although our per-person recruitment cost through the pilot advertising campaign on Facebook was not cheap, it also increased interest in the NINFEA study Facebook page, which, in turn, may have translated into further registrations in the study.
Cancer Epidemiology Unit
CeRMS and CPO-Piemonte
University of Turin
1. Richiardi L, Baussano I, Vizzini L, Douwes J, Pearce N, Merletti F. Feasibility of recruiting a birth cohort through the Internet: the experience of the NINFEA cohort. Eur J Epidemiol. 2007;22:831–837.