The public and researchers are often reminded of the 17-year time lag between the translation of research findings into practice (Balas & Boren, 2000; Grant, Green, & Mason, 2003; Wratschko, 2009). Although several models for implementing evidence-based research into practice have emerged (e.g., Kanter, Schottinger, & Whittaker, 2017), a variety of barriers remain, such as a total reduction in public research dollars, competition with big pharma trials, and a shortage of research volunteers. Moreover, current estimates indicate that nearly 80% of clinical trials (Carlisle, Kimmelman, Ramsay, & MacKinnon, 2015) and between 10% and 60% (Kitterman, Cheng, Dilts, & Orwoll, 2011) of other research projects fail to meet their enrollment goal and timeline.
Recruitment and enrollment present as an even larger challenge when vulnerable populations—like pregnant women—are involved. Nonetheless, recruitment can be a burdenless undertaking if innovative recruitment strategies are employed (Frew et al., 2014; Goff et al., 2016). For example, Shere, Zhao, and Koren (2014) found the use of social media to increase recruitment efforts of periconceptual women to increase 12-fold. Similarly, Admon et al. (2016) reported that compared to traditional methods yielding 219 participants, social media recruitment yielded 1,178 participants. Although these are just a few of the available studies in the literature that used social media as a recruitment modality, it remains clear that social media may be a beneficial resource for the recruitment of pregnant women into research (Adam, Manca, & Bell, 2016; Kira, Glover, Walker, & Bauld, 2016; Richardson et al., 2016). Furthermore, it may be beneficial for researchers to view a detailed Facebook recruitment plan that was used to successfully recruit pregnant women into a research study. Therefore, the purpose of this brief is to share an experience of diversifying recruitment strategies with the use of Facebook to recruit pregnant women into research.
On the basis of a power analysis, a total of 82 participants were needed in order to complete the parent study. Like many studies, the principal investigator (PI) experienced several challenges in recruiting pregnant women (Adam et al., 2016; Chansamouth et al., 2017; Frew et al., 2014; Goff et al., 2016). A series of traditional recruitment strategies were attempted before the full potential of Facebook recruitment was understood and implemented. Traditional recruitment strategies utilized were based on methods reported in the literature for the recruitment of pregnant women (e.g., Manca et al., 2013; Muggli, Curd, Nagle, Forster, & Halliday, 2018). Traditional recruitment methods in the parent study included posting flyers in community locations, such as grocery stores, recreation centers, libraries, and other places pregnant women may frequent. Additionally, the PI attended prenatal care courses around the surrounding area of the university to speak to course participants about the study. Prior to speaking to any potential participants or conducting any of the research protocol, university institutional review board (IRB) approval was obtained.
To be eligible, pregnant women had to be in the second (13–27 weeks) or third (28–40 weeks) trimester, at least 18 years old, able to speak and understand English, and live in the Northeast Ohio area to participate in face-to-face data collection. Individuals were excluded if they resided out of the geographic area and could therefore not participate in face-to-face data collection; they self-identified as a substance user (cigarettes, alcohol, or illegal or recreational drugs); or if they disclosed any of the following: previous or current treatment or diagnosis of depression, history of or current gestational diabetes, and current treatment for a respiratory condition (i.e., bronchodilator). Additionally, potential participants were screened for hypertension and would be excluded from the study if their systolic blood pressure was >140 mm Hg. The parent study from which this brief was derived aimed to describe relationships among social determinants of health, psychophysiological stress, and mental health in pregnant women. Therefore, the eligibility criteria above were chosen in order to control for potential confounding variables in the parent study.
Before Facebook recruitment began, an electronic version of the recruitment flyer was created. University IRB approved the flyer. The original flyer was in bright colors and had only necessary details about the study, including the study purpose, procedures, time expected, and contact information. The contact information on the flyer was the PI's university e-mail address and a phone number. The PI also created a Facebook page that displayed the study logo, information about the PI, the study flyer, and contact information. The PI created a separate Facebook page for the study so interested participants could refer to the page for more information or to invite their friends to “like” the page to generate interest in the study.
Next, the PI “liked” and “followed” several Facebook groups in which mothers and expectant mothers within the geographic area of interest may be members. Facebook groups were found by searching a list of key terms within the search bar imbedded in Facebook. Example search terms included mom, mother, mommy, mama, pregnant, expecting, pregnancy, perinatal, and parent. A list of search terms were compiled based on a SmartText search in Academic Search Complete, where similar terms to mother were generated. Additionally, search terms specific to the geographic area were also searched within Facebook (e.g., Cleveland and Northeast Ohio). Once the PI was given access to the Facebook groups, the PI introduced herself to the group moderator via private message and explained the purpose of the study and asked permission to post the study flyer in the Facebook group.
Detailed records of all recruitment efforts (traditional and Facebook) were kept in an encrypted Excel file. The spreadsheet included the amount of time spent recruiting, frequency and location of flyer postings, names of the Facebook groups, and category of the Facebook groups. All data reported and analyzed in this brief are frequencies of Facebook activity. Data points are the following: likes/shares, comments, referrals, and participants enrolled. “Likes” are intentional reactions by Facebook users to indicate they approve of a post. “Shares” refer to when a Facebook user uses the image you post and posts it to their own profile. “Comments” are only comments on the study recruitment flyer. Referrals are when a Facebook user “tags” another user to see the study recruitment flyer. Participants enrolled are the frequency of enrollees who indicated they were recruited into the study because of the study recruitment flyer on Facebook.
Target enrollment for the parent study was met with a total enrollment of 82 participants. Traditional recruitment methods including posting a flyer in the community and the PI speaking to antenatal classes yielded 20 participants in 75 days. Facebook recruitment yielded the remaining 62 participants in 48 days. Overall, the majority of the targeted study sample (N = 82) was recruited from Facebook, with 75% of the sample indicating they enrolled via Facebook. Flyers were posted in private Facebook groups at a frequency of one time for the length of the study. One hundred percent of the private groups consented to the PI posting information about the research study.
The study recruitment flyer was posted in a total of 61 private groups on Facebook. Facebook groups were categorized based on the content of the messages and posts within the private group. The majority of recruitment flyer postings were in private groups focused on discussion related to parenting topics (n = 20) (e.g., advice for moms from moms), followed by parent special interest (n = 17) (e.g., natural parenting and organic), sell/trade/jobs boards (n = 10), general pregnancy (n = 7) (e.g., breastfeeding support), miscellaneous (n = 3), events (n = 2), and nonparent discussion (n = 2).
The greatest number of likes, shares, comments, and referrals by the recruitment flyer were generated from the sell/trade/jobs page and the greatest number of participants enrolled saw the flyer in Facebook groups focused on parent discussion. The nonparent discussion groups yielded the fewest likes and shares; however, the events groups yielded the fewest comments and referrals. The miscellaneous, events, and nonparent discussion groups yielded no participants enrolled. Table 1 displays a breakdown of activity generated by the study flyer in the private Facebook groups.
The target enrollment goal (n = 82 participants) for the parent study was met in approximately 4 months. An overwhelming majority of the sample, or 75%, indicated that they enrolled in the study because of the flyer they viewed on Facebook. Although traditional recruitment methods were utilized at the beginning of the study, Facebook recruitment yielded a larger sample in a shorter amount of time. A total of 61 private Facebook groups were contacted and 100% of private groups consented to the PI posting in the group. This high rate of consent by the Facebook groups could be because the PI sought permission before posting and was transparent about the research. Furthermore, after enrollment for the study had ceased, the PI continued to receive calls and e-mails from approximately 10 interested participants over a 2-month time period.
Surprisingly, the greatest amount of Facebook activity was from private Facebook pages that were aimed at buying, selling, and trading items. Some of these Facebook pages were also job boards in which users freely posted about volunteer and for-pay opportunities. The sell/trade/jobs Facebook pages have far more users and followers than any of the other group categories (e.g., parent discussion and general pregnancy). However, although the sell/trade/jobs pages attracted the most amount of Facebook activity, the greatest participants enrolled were derived from the parent discussion group, followed by parent special interest.
No matter the recruitment modality utilized, it is important to recognize the inherent selection bias. Although social media historically has attracted a female, young, wealthy, Caucasian sample (Alessi & Martin, 2010; Gu, Skierkowski, Florin, Friend, & Ye, 2016; Ince, Cuijpers, van't Hof, & Riper, 2014), recent data suggest that participants who are recruited via traditional methods (e.g., flyers) are similar to those recruited with social media (International Telecommunication Union, 2017; Pew Research Center, 2018; Smith & Anderson, 2018). This similarity is due in part because as many as 68% of adults in the United States use Facebook regularly (Smith & Anderson, 2018), and in 2018, as much as 89% of the population has access to the Internet (Pew Research Center, 2018).
Although a large proportion of the population has access to the Internet and social media, it is important to illustrate that not all study populations are reachable by social media, nor is all research recruitment feasible using social media. For example, social media has been an ineffective strategy for the recruitment of parents (Dworkin, Hessel, Gliske, & Rudi, 2016), and social media recruitment can be expensive depending on whether researchers elect to purchase the paid advertising option (Gu et al., 2016; Moreno et al., 2017). Furthermore, it should be noted that the target sample of pregnant women in the current parent study were relatively healthy with no comorbidities. Therefore, recruiting pregnant women with certain health conditions may not be feasible. For example, Williamson, O'Connor, Chamberlain, and Halpin (2018) were unsuccessful in recruiting pregnant asthmatic women with social media.
Findings from this brief illustrate the importance of diversifying traditional recruitment strategies by including innovative tactics like the use of Facebook. Researchers in the future who are interested in implementing Facebook as a recruitment strategy should post their study recruitment flyer in a variety of Facebook groups that are both specific and nonspecific to their population to generate a wide array of interest. This brief is evidence that Facebook recruitment is a contemporary strategy that may aid in meeting recruitment goals and thereby decreasing the time lapse between research findings and translation to practice.
Adam L. M., Manca D. P., & Bell R. C. (2016). Can Facebook
be used for research? Experiences using Facebook
to recruit pregnant women
for a randomized controlled trial. Journal of Medical Internet Research
, 18, e250. doi:10.2196/jmir.6404
Admon L., Haefner J. K., Kolenic G. E., Chang T., Davis M. M., & Moniz M. H. (2016). Recruiting pregnant patients for survey research: A head to head comparison of social media
-based versus clinic-based approaches. Journal of Medical Internet Research
, 18, e326. doi:10.2196/jmir.6593
Alessi E. J., & Martin J. I. (2010). Conducting an internet-based survey: Benefits, pitfalls, and lessons learned. Social Work Research
, 34, 122–128.
Balas E. A., & Boren S. A. (2000). Managing clinical knowledge for health care improvement. Yearbook of Medical Informatics 2000: Patient-Centered Systems
, 1, 65–70.
Carlisle B., Kimmelman J., Ramsay T., & MacKinnon N. (2015). Unsuccessful trial accrual and human subjects protections: An empirical analysis of recently closed trials. Clinical Trials
, 12, 77–83. doi:10.1177/1740774514558307
Chansamouth V., McGready R., Chommanam D., Homsombath S., Mayxay M., & Newton P. N. (2017). Enrolling pregnant women
in research: Ethical challenges encountered in Lao PDR (Laos). Reproductive Health
, 14, 167. doi:10.1186/s12978-017-0428-9
Dworkin J., Hessel H., Gliske K., & Rudi J. H. (2016). A comparison of three online recruitment strategies for engaging parents. Family Relations
, 65, 550–561. doi:10.1111/fare.12206
Frew P. M., Saint-Victor D. S., Isaacs M. B., Kim S., Swamy G. K., Sheffield J. S., … Ault K. (2014). Recruitment and retention of pregnant women
into clinical research trials: An overview of challenges, facilitators, and best practices. Clinical Infectious Diseases
, 59, S400–S407. doi:10.1093/cid/ciu726
Goff S. L., Youssef Y., Pekow P. S., White K. O., Guhn-Knight H., Lagu T., … Lindenauer P. K. (2016). Successful strategies for practice-based recruitment of racial and ethnic minority pregnant women
in a randomized controlled trial: The IDEAS for a Healthy Baby Study. Journal of Racial and Ethnic Health Disparities
, 3, 731–737. doi:10.1007/s40615-015-0192-x
Grant J., Green L., & Mason B. (2003). Basic research and health: A reassessment of the scientific basis for the support of biomedical science. Research Evaluation
, 12, 217–224. doi:10.3152/147154403781776618
Gu L. L., Skierkowski D., Florin P., Friend K., & Ye Y. (2016). Facebook
, Twitter, & Qr codes: An exploratory trial examining the feasibility of social media
mechanisms for sample recruitment. Computers in Human Behavior
, 60, 86–96. doi:10.1016/j.chb.2016.02.006
Ince B. U., Cuijpers P., van't Hof E., & Riper H. (2014). Reaching and recruiting Turkish migrants for a clinical trial through Facebook
: A process evaluation. Internet Interventions
, 1, 74–83. doi:10.1016/j.invent.2014.05.003
Kira A., Glover M., Walker N., & Bauld L. (2016). Recruiting pregnant indigenous women who smoke into a high contact incentivized cessation trial: A feasibility study. Nicotine & Tobacco Research
, 18, 2036–2040. doi:10.1093/ntr/ntw106
Kitterman D. R., Cheng S. K., Dilts D. M., & Orwoll E. S. (2011). The prevalence and economic impact of low-enrolling clinical studies at an academic medical center. Academic Medicine: Journal of the Association of American Medical Colleges
, 86, 1360–1366. doi:10.1097/ACM.0b013e3182306440
Manca D. P., O'Beirne M., Lightbody T., Johnston D. W., Dymianiw D. L., Nastalska K.… APrON study team (2013). The most effective strategy for recruiting a pregnancy cohort: A tale of two cities. BMC Pregnancy and Childbirth
, 13, 75. doi:10.1186/1471-2393-13-75
Moreno M. A., Waite A., Pumper M., Colburn T., Holm M., & Mendoza J. (2017). Recruiting adolescent research participants: In-person compared to social media
approaches. Cyberpsychology, Behavior and Social Networking
, 20, 64–67. doi:10.1089/cyber.2016.0319
Muggli E., Curd H., Nagle C., Forster D., & Halliday J. (2018). Engaging pregnant women
in observational research: A qualitative exploratory study. BMC Pregnancy and Childbirth
, 18, 334. doi:10.1186/s12884-018-1966-z
Richardson J. L., Stephens S., Thomas S. H. L., Jamry-Dziurla A., de Jong-van den Berg L., Zetstra-van der Woude P., … Blackburn S. C. (2016). An international study of the ability and cost-effectiveness of advertising methods to facilitate study participant self-enrolment into a pilot pharmacovigilance study during early pregnancy. JMIR Public Health and Surveillance
, 2, e13. doi:10.2196/publichealth.5366
Shere M., Zhao X. Y., & Koren G. (2014). The role of social media
in recruiting for clinical trials in pregnancy. PLOS ONE
, 9, e92744. doi:10.1371/journal.pone.0092744
Williamson G. R., O'Connor A., Chamberlain C., & Halpin D. (2018). mHealth resources for asthma and pregnancy care: Methodological issues and social media
recruitment. A discussion paper. Journal of Advanced Nursing
, 74, 2442–2449. doi:10.1111/jan.13773
Wratschko K. (2009). Empirical setting: The pharmaceutical industry. In: Strategic orientation and alliance portfolio configuration
(pp. 87–96). Wiesbaden: Gabler. doi:10.1007/978-3-8349-9459-2_5