JMIR Publications

ABSTRACT

Background:

With the increased popularity of mobile menstrual tracking applications and boosted Facebook posts, there is a unique opportunity to recruit research study participants from across the globe via these modalities to evaluate women’s health. However, no studies to date have assessed the feasibility of using these recruitment sources for epidemiological research on ovulation and menstruation.

Objective:

The objective of this study was to assess the feasibility of recruiting a diverse sample of women into an epidemiological study of ovulation and menstruation health (OM Global Health Study) using digital recruitment sources. The feasibility and diversity were assessed via click and participation rates, geographic location, body mass index (BMI), smoking status, and other demographic information.

Methods:

Participants were actively recruited via in-app messages using the menstrual tracking app Clue and a boosted Facebook post by DivaCup. ‘Other’ passive recruitment activities also took place throughout the recruitment period (e.g., email communications, blog, other social media). Click and participation rates were used to quantify success of recruiting participants to the study website and study survey completion, respectively. Participation rate was defined as the number of completed surveys per number of completed consents and eligibility screeners. Survey completion was defined as finishing the pregnancy and birth history section of the OM Study questionnaire.

Results:

The recruitment period occurred from February 27, 2018 through January 24, 2020. In-app messages and the boosted Facebook post were seen by 104,000 and 21,400 people, respectively. Overall, 215 participants started the OM Global Health Study survey, of which 140 (65%), 39 (18%), and 36 (17%) participants were recruited via the app, the boosted Facebook post, and ‘other’ passive recruitment activities, respectively. The click rate via the app was 19% (19,700 clicks/104,000 ad views) and 1.6% via the boosted Facebook post (340 clicks/ 21,400 ad views.) The overall participation rate was 45% (198/444) , and the average participant age was 22 years. In terms of geographic and racial/ethnic diversity, 44% of participants resided outside of the US and 29% identified as non-Hispanic White. In-app recruitment produced the most geographically diverse stream with 32.8% in Europe, 57.5% in North America, and 9.8% in other parts of the world. Both human error and non-human procedural breakdowns occurred during the recruitment process, including a computer programming error related to age eligibility and a hacking attempt by an internet bot.

Conclusions:

In-app messages using the menstrual tracking app Clue was the most successful method for recruiting participants from many geographic regions and producing the greatest numbers of started and completed surveys. This study demonstrates the utility of digital recruitment to enroll participants from diverse geographic locations and provides some lessons to avoid technical recruitment errors in future digital recruitment strategies for epidemiological research. Clinical Trial: N/A