SEARCH

SEARCH BY CITATION

Keywords:

  • social network;
  • medical genetics;
  • human subjects

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

Social networking sites (SNS) have potential value in the field of medical genetics as a means of research subject recruitment and source of data. This article examines the current role of SNS in medical genetics research and potential applications for these sites in future studies. Facebook is the primary SNS considered, given the prevalence of its use in the United States and role in a small but growing number of studies. To date, utilization of SNS in medical genetics research has been primarily limited to three studies that recruited subjects from populations of Facebook users [McGuire et al. (2009); Am J Bioeth 9: 3–10; Janvier et al. (2012); Pediatrics 130: 293–298; Leighton et al. (2012); Public Health Genomics 15: 11–21]. These studies and a number of other medical and public health studies that have used Facebook as a context for recruiting research subjects are discussed. Approaches for Facebook-based subject recruitment are identified, including paid Facebook advertising, snowball sampling, targeted searching and posting. The use of these methods in medical genetics research has the potential to facilitate cost-effective research on both large, heterogeneous populations and small, hard-to-access sub-populations. © 2013 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

More than 80% of adults in the United States engage in one or more online activities [Pew Research Center, 2012c]. Two-thirds of internet users participate in online social networking, using websites such as Facebook, MySpace, or Twitter [Pew Research Center, 2012b]. Facebook is the most popular social networking site (SNS) in the United States with more than 150 million users [Hui, 2011; Cosenza, 2012].

SNS are online services that allow members to create a profile, list connections with other members and explore their own connections and those of others [Boyd and Ellison, 2007]. They are one of many “web 2.0” internet-based applications that allow users to create and interact with online content [Pew Research Center, 2012a]. Members' ability to view each other's networks is a defining characteristic of SNS [Boyd and Ellison, 2007]. This not only facilitates connections between individuals with common interests or experiences, but also provides opportunities for individuals to search for and connect with members of their existing social networks [Boyd and Ellison, 2007; Lee and Crawley, 2009].

These sites often contain publicly available data in the form of user posts. In addition, unlike in-person social interactions, they allow individuals to connect with one another regardless of geographic proximity. Thus, they present opportunities for individuals living in different locations to form disease-specific networks and communities. For researchers conducting studies on rare diseases, SNS may provide access to larger networks of patients and family members than geographically-based networks.

These qualities and the prevalence of SNS usage have led medical and public health researchers to explore the role of SNS in research. At a 2011 National Institute of Child Health and Human Development Scientific Vision Workshop on Pregnancy and Pregnancy Outcomes, the need for strategies to effectively utilize social media networks, both as potential sources of research data and for subject recruitment, was identified as a component of broader research themes [Catalano et al., 2011].

In recent years, medical genetics has developed an increasing internet presence. Thus, SNS-based research may hold particular value for research in this field. A wide range of conditions, including many genetic diseases, is represented on Facebook. Facebook Pages and Groups provide opportunities for patients, families, and advocates to interact, and in some cases they are used to communicate information about research. In addition, web 2.0 applications such as forums and blogs are often settings for online conversations regarding advances in medical genetics. For example, a search on several pregnancy-oriented websites (Babble.com, BabyandBump.com, BabyCenter.com, TheBump.com, and WhattoExpect.com) for public posts regarding MaterniT21, a recently released, minimally invasive prenatal genetic test (PGT) for trisomy 21, 18, and 13, reveals that it has been a topic of discussion in these online communities [Sequenom Center for Molecular Medicine, 2012].

Direct-to-consumer (DTC) genetic testing is an area of medical genetics that is largely internet-based. DTC tests are primarily marketed online and some DTC companies, such as 23andme, provide opportunities for their consumers to participate in research [Hogarth et al., 2008; 23andMe, 2012]. These characteristics suggest that internet-based recruitment may be particularly effective for research related to DTC testing. Moreover, they may help explain the early adoption of SNS-based sampling in studies of consumer understanding of DTC products.

To date, published research utilizing SNS-based recruitment methods is limited. This article examines two studies of DTC genetic testing, both of which surveyed Facebook users [McGuire et al., 2009; Leighton et al., 2012]. In addition, it reviews studies from other research fields that used Facebook as a context for recruiting research subjects. It identifies multiple Facebook-based recruitment methodologies and explores their applicability to medical genetics. Facebook is the SNS discussed given the prevalence of its use and role in a small but growing number of studies.

The studies reviewed were identified through searching PubMed.gov through August 2012 for articles containing terms such as “Facebook,” “social networking,” “social network,” “social media,” or “subject recruitment.” Search results were then evaluated to determine their relevancy to this analysis.

BACKGROUND

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

Characteristics of SNS Users

A key consideration in exploring the potential utilization of SNS in medical genetics research lies in understanding characteristics of SNS users. Factors that affect SNS usage include sex and age. While rates of internet usage do not differ between men and women, female internet users are more likely to use SNS [Madden and Zickuhr, 2011; Pew Research Center, 2012c]. According to 2011 survey data from the Pew Internet and American Life Project (PIALP), 69% of women use SNS, compared to 60% of men [Madden and Zickuhr, 2011]. SNS utilization is highest among those under the age of 30. Eighty-three percent of adults in this age range use SNS compared to 70% in the 30–49 age range. Factors that do not significantly affect SNS participation include household income level, educational level, race/ethnicity, and geographic location [Madden and Zickuhr, 2011].

In addition to social networking, obtaining health information online is a common practice among internet users. According to 2010 PIALP estimates, 80% of adults who use the internet go online to gather health information [Fox, 2011]. Rates of health information searching among adults increase with educational level and household income, decrease with age and are higher among non-Hispanic whites than among other race/ethnicity groups [Fox, 2011]. In addition, females are more likely than males to use the internet to search for health information; 65% engage in online health information seeking. Among female internet users, 24% specifically seek information regarding pregnancy and childbearing [Fox, 2011]. As of January 2012, there were over 5,000 Facebook Pages related to pregnancy alone.

Samples drawn from SNS may be likely to over-represent females and young adults given the high rate of SNS participation among adults under age 30. In addition, given that rates of online health information searches are lower for lower-income individuals, those with less education, and those who identify themselves as Hispanic or non-Hispanic black, internet-based recruitment of subjects for medical research studies may result in underrepresentation of these groups.

Characteristics of Facebook

While a variety of SNS exist, many have similar components, including the capability to customize profiles with photos and applications as well as features that allow users to post publicly on one another's profiles and to message one another privately [Boyd and Ellison, 2007]. Boyd and Ellison [2007] state that SNS primarily facilitate connections between members of pre-existing social networks. Lee and Crawley [2009] note, however, that SNS such as Facebook enable the formation of new linkages.

On Facebook, each user's profile contains a count and list of connections known as “friends.” Inviting another user to be a friend is commonly referred to as “friending.” Each profile has a section known as a “wall” where visible interactions between friends occur.

Facebook allows the construction of “Pages” and “Groups.” Facebook Pages differ from individual users' profiles in that they are designed for “businesses, organizations and brands to share their stories and connect with people” [Facebook Help Center, 2012c]. They may be visible to both Facebook users and non-users. Fans of Pages can “like” them and each Page has a count of the number of “likes” it has received. According to Facebook, “liking” a Page is a form of networking:

When you click Like on a Facebook Page…you are making a connection. A story about your like will appear on your Wall (timeline) and may also appear in News Feed. You may be displayed on the Page you connected to, in advertisements about that Page, or in social plugins next to the content you like [Facebook Help Center, 2012b].

Some Pages contain walls where individuals can post public comments and messages. Facebook Groups are designed for small groups to communicate about common interests; unlike Pages, many are not for public sharing of information [Facebook Help Center, 2012a].

A wide range of health conditions is represented among Facebook Pages and Groups including many genetic syndromes. Organizations dedicated to supporting research and advocacy for patients with a particular condition often have Pages to supplement their websites. As of April 2012, more than 400 Pages were dedicated to trisomy 21, of which 20 had at least 1,500 likes. The National Down Syndrome Society Page had the largest fan base with nearly 33,000 likes [National Down Syndrome Society, 2012]. Twenty-five pages were dedicated to trisomy 18, the most popular of which was the Trisomy 18 Foundation Page with close to 33,000 likes [Trisomy 18 Foundation, 2012]. Less common genetic disorders are also represented. For example, Cri-du-chat syndrome, is represented on multiple Facebook Pages. The largest of these is affiliated with the organization “Friends of Cri du Chat” and had nearly 500 likes as of April 2012 [Cri du chat/5p-syndrome, 2012].

In addition to raising awareness of conditions, promoting fundraising for research and helping to build a community of affected individuals and their family members, Facebook Pages provide a forum for sharing information regarding advances in research. In some instances, Pages may contain posts advertising research for which participants are needed. For example, a search of content on the Facebook Page for the the Myotonic Dystrophy Foundation showed a research recruitment posting in June 2012 for individuals residing in Sacramento, CA [Myotonic Dystrophy Foundation, 2012]. Several comments accompanying the post addressed a desire for additional studies in other locations within and outside the US and provided insight into the geographic heterogeneity of site users. Thus, Facebook Pages may provide a context for medical genetics researchers to recruit participants from a wide range of locations for studies on conditions of varying prevalence.

EXISTING SNS-BASED LITERATURE

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

Recruitment of Research Subjects

A small but growing number of medical and public health research studies have utilized Facebook in recruiting research subjects. Multiple recruitment methods have been used, the most common of which is paid Facebook advertising (Table 1).

Table 1. Studies Using SNS-Based Subject Recruitment
Recruitment methodStudy descriptionStudy typeSample recruited via FacebookDuration of recruitmentNumber recruited via Facebook-based methodsa
  1. a

    Depending on how recruitment data were reported, includes either the total number of individuals who consented to participate or the number who completed the study (the latter is shown in parentheses).

  2. b

    Total sample recruited using both methods.

  3. c

    The number of participants recruited from Facebook is not known as recruitment was conducted on multiple websites and studies did not report responses by site.

Advertising
Fenner et al. [2012]Online health-related survey of young adult females completed at a study site or remotelyCross-sectionalFemale Facebook users, ages 16–25, from Victoria, Australia4.5 months426
Richiardi et al. [2012]Internet-based study of a mother-child cohortLongitudinalPregnant Facebook users, ages 16–456 weeks8
Ramo and Prochaska [2012]Online survey regarding use of tobacco and other substances among young adultsCross-sectionalEnglish-speaking Facebook users, ages 18–25, with recent reported cigarette use and residing in the U.S.13 months5,237
Vrangalova and Savin-Williams [2012]Online survey regarding sexual orientation among adultsCross-sectionalAdults, ages 18 or older, residing in the U.S.5 days(1,784)
Rogers et al. [2009]Online survey regarding emotional self-disclosure and preference for online vs. face-to-face therapy among young adultsCross-sectionalEnglish-speaking, U.S.-registered Facebook users, ages 21–30Not provided413
Targeted Facebook search
Mychasiuk and Benzies [2012]Follow-up in a study of a preschool intervention programLongitudinalSubjects enrolled in the programNot provided(19)
Jones et al. [2012]Follow-up study on interventions to prevent reductions in physical activity among adolescent femalesFollow-up study11th grade females previously enrolled in the Trial of Activity for Adolescent GirlsNot provided43
Moreno et al. [2012]Evaluation of undergraduate students' Facebook profile alcohol references and online survey regarding problem drinkingCross-sectionalFacebook users, ages 18–20, at two universitiesNot provided(224)
Snowball sampling
Leighton et al. [2012]Online survey of Facebook users and genetic counselors regarding DTC genetic testingCross-sectionalFacebook users, age 18 and older>1 month(145b)
Strasser et al. [2012]Online survey of intimate partner violence among gay men in Atlanta, GeorgiaCross-sectionalGay, male Facebook users living in Atlanta, Georgia16 days(100b)
Facebook posting
Janiec et al. [2012]Online survey of European football fans to determine knowledge of health-related travel advice provided by public health organizations for the 2012 European football tournamentCross-sectionalEURO 2012 fans2 weeksUnknownc
Leighton et al. [2012]Online survey of Facebook users and genetic counselors regarding DTC genetic testingCross-sectionalFacebook users, age 18 and older>1 month(145b)
Strasser et al. [2012]Online survey of intimate partner violence among gay men in Atlanta, GeorgiaCross-sectionalGay, male Facebook users living in Atlanta, Georgia16 days(100b)

Fenner et al. [2012] used paid Facebook advertising to recruit young women living in Victoria, Australia to participate in a survey regarding health issues and their willingness to participate in a larger study. Costs were determined by the number of clicks the advertisements received. Researchers chose a price per click (bid) based on information provided by Facebook on the range of bids for advertisements of similar content targeting the population of interest. The bid, relative to those of competing advertisements, helped determine the probability of an advertisement's display. Because bid ranges varied for demographic groups within the target population, the researchers conducted separate advertising campaigns for age and geography-based sub-populations. Participants were given the opportunity to complete the survey in-person at a designated site or online. Seventy percent of those who enrolled agreed to travel to the study site though more than half were subsequently lost to follow-up or opted to take the survey remotely. Fenner et al. [2012] found the geographic distribution of the final sample to be consistent with the general population. They concluded that targeted Facebook advertising is not only cost effective but may be particularly useful for recruiting study participants often underrepresented in research, such as young women or individuals living in rural areas.

Three other cross-sectional studies have used paid Facebook advertising to recruit participants for online surveys [Rogers et al., 2009; Ramo and Prochaska, 2012; Vrangalova and Savin-Williams, 2012] (Table 1). In Ramo and Prochaska [2012], the authors found that Facebook recruiting costs were substantially lower than other methods used in previous research [Ramo et al., 2010]. Vrangalova and Savin-Williams [2012] reported that 12% of survey participants were recruited from sources outside of Facebook, suggesting that information regarding the study had been shared by Facebook users.

One cross-sectional study used targeted Facebook searches, rather than Facebook advertising, to recruit survey participants. Moreno et al. [2012] conducted a random search on Facebook for undergraduate students at two universities. Identified, eligible students were recruited to take part in an online survey and their profiles were evaluated for alcohol references.

Facebook posting is an additional method by which researchers directly post on a Facebook Page or Group. Unlike paid Facebook advertising, posting is free. Strasser et al. [2012] used this method to recruit a defined number of participants for a survey of intimate partner violence among gay males living in Atlanta, GA. In addition to posting on a Facebook Group, the investigators employed snowball sampling by emailing members of the Group to encourage them to explore the survey link and to share it with others. Janiec et al. [2012] assessed dissemination of public health travel information targeted at European football fans who attended the EURO 2012 football tournament by posting survey links on multiple Facebook Pages and other sites. They noted that use of Facebook facilitated low-cost, rapid recruitment.

Richiardi et al. [2012] used Facebook advertising to recruit participants for a longitudinal internet-based study. The authors found Facebook advertising to be more expensive and to yield fewer participants than other methods. They note, however, that advertising increased fans of the study's Facebook Page and thus may have facilitated later recruitment that was not captured.

Follow-Up of Research Subjects

Facebook has also been used to locate participants lost to follow-up in longitudinal research. Through Facebook, Mychasiuk and Benzies [2012] contacted caregivers who could not be contacted via phone, post, or email to set up a home visit following completion of a pre-school intervention program. Using Facebook to contact missing individuals increased participant retention at follow-up from 45.8% to 61.7%. The authors noted that the children of families contacted via Facebook had higher receptive language scores than children in other families. Thus, their exclusion would have resulted in a lower overall score. Jones et al. [2012] used Facebook to find and recruit adolescent females whom they were not able to contact in-person or by mail for a follow-up study. Costs associated with this study were limited to the investigators' time, as the use of Facebook was free.

These studies provide evidence that Facebook can be used in recruiting and following up with subjects in cross-sectional and longitudinal studies that span a range of research areas. In five of ten studies, subjects were young adults or adolescents [Rogers et al., 2009; Fenner et al., 2012; Jones et al., 2012; Moreno et al., 2012; Ramo and Prochaska, 2012]. Jones et al. [2012] suggest that because of the pervasive use of Facebook among adolescents, effective recruitment from this population may require incorporation of SNS-based methods.

Cost-Effectiveness

Evidence regarding the cost effectiveness of recruitment conducted via paid Facebook advertising is somewhat mixed. While two studies found Facebook advertising to be low cost, Richiardi et al. [2012] found this method to be expensive and less successful than other recruitment approaches [Fenner et al., 2012; Ramo and Prochaska, 2012]. These differences may reflect the variation in length and number of advertising campaigns in these studies. While Richiardi et al. [2012] carried out a campaign for 6 weeks aimed at a broad population, Fenner et al. [2012] conducted more targeted advertising with multiple campaigns for more than 4 months. Targeted searches and posts using Facebook are highly cost-effective.

THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

The use of SNS in medical genetics research has thus far been limited. A small number of published studies have used SNS-based recruitment methodologies and some organizations have begun to use Facebook as a forum for discussion of research. Leighton et al. [2012] and McGuire et al. [2009] surveyed Facebook users to help elucidate consumer understanding and views of DTC testing. A third study used Facebook in combination with other websites to recruit participants for an online survey of parents with children with Trisomy 13 or Trisomy 18 [Janvier et al., 2012]. Because details of the SNS-based recruitment methodology were not provided in this latter study, it is not discussed here in detail.

Many DTC genetic tests screen single-nucleotide polymorphisms for their associations with a number of health conditions and traits [Hogarth et al. 2008; Dvoskin and Kaufman, 2011]. Fetal gender tests using cell-free fetal DNA extracted from a maternal blood sample have also been developed [Bianchi, 2006]. These tests are primarily marketed and sold online and the majority can be obtained without the involvement of a healthcare provider [Hogarth et al., 2008]. McGuire et al. [2009] hypothesize that social networkers may be early consumers of DTC genetic tests because of the internet presence of these products and because some DTC companies have a social networking component.

Leighton et al. [2012] assessed understanding of sample DTC test results among the general public using an online survey made available to Facebook users. As a comparison group, they also surveyed genetic counselors. Survey recruitment was carried out in part by a snowball sampling method, whereby a recruitment message with a survey link was sent to the principal investigator's Facebook friends, and each friend was asked to forward the message to five individuals. Messages were also posted on the Pages of some large Facebook groups (e.g., “Yankee Fans”).

The majority of non-counselor respondents came from the principal investigator's extended Facebook network. Among those who provided demographic information (n = 105), three-quarters or more were under the age of 30, female, white, non-Hispanic or had an educational level of bachelor's or graduate degree or higher. Fifty-six percent reported employment or education in healthcare or biological sciences. Although Facebook users served as a proxy for the general public in this study, the generalizability of the results to the larger population may be limited due to the recruiting methods employed.

The survey contained four scenarios related to test results [Leighton et al., 2012]. Respondents were asked to interpret their risk for developing certain medical conditions. In addition, they rated their concern regarding each set of results and the results' importance to their future medical care. Nearly half of non-counselor respondents indicated they had never heard of DTC testing and only 10% had “a great deal” of familiarity with it. For three out of four scenarios, non-counselors' interpretation of risk differed significantly from that of counselors'. Moreover, compared to counselor respondents, non-counselors reported that results were more helpful.

McGuire et al. [2009] sampled Facebook users to assess social networkers' knowledge and perceptions of DTC genetic testing companies as well as their understanding of testing results. Participants were not recruited from Facebook, but from among a panel of survey respondents recruited and validated by MarketTools, a company that provides online research services [MarketTools, 2012]. The use of such services has been shown in other research to be more expensive than methods that directly use Facebook [Ramo et al., 2010; Ramo and Prochaska, 2012].

Within 36 hr of the study's initiation, a total of 1,087 surveys were completed. The majority of respondents were white or female and 60% had an educational level of bachelor's degree or higher. Slightly more than half were under the age of 35. Only 7% were healthcare professionals. The online survey contained questions regarding subjects' “knowledge and awareness of personal genome testing companies” and “opinions and attitudes” towards these companies and DTC genetic test results.

Although just over half of respondents reported having no knowledge of DTC testing companies, nearly two-thirds reported that they might consider a future purchase of DTC products [McGuire et al., 2009]. Six percent reported having purchased DTC testing services. Approximately two-thirds of participants who had or would consider using DTC testing services felt test results would impact their future decisions regarding medical care.

Leighton et al. [2012] and McGuire et al. [2009] utilized different methods for recruiting research subjects from populations of Facebook users. Compared with survey respondents in McGuire et al., respondents in Leighton et al. were more likely to be under the age of 30 and to have a bachelor's or graduate degree. A large percentage had exposure to healthcare or biological sciences through employment or education. The methods used in Leighton et al. also yielded a much smaller sample size (n = 145) than that obtained in McGuire et al. (n = 1,087).

Despite these differences, both studies suggest that a substantial share of social networkers have some familiarity with DTC genetic testing, while only a small percentage have used or have a great deal of familiarity with DTC products. In addition, both studies indicate that individuals may interpret results to have significant implications regarding disease risk and future medical care.

APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

While the use of SNS as a recruitment tool in the medical genetics literature has thus far been limited to a small number of studies, these sites have the potential to facilitate research on a range of topics if used effectively. Importantly, they are already being used by some organizations to make families and patients aware of research opportunities, for example through Facebook Page wall postings. The studies discussed above suggest possibilities for Facebook to be used as a setting for recruiting both large, population-based samples and smaller samples of hard-to-reach groups. The establishment of SNS as a recruitment tool will depend on assessment of these methodologies in study populations relevant to medical genetics.

Facebook-based subject recruitment may have particular value for research on PGT given the prevalence of SNS use among both women and adults of childbearing age. A perusal of publicly available forums on pregnancy websites such as BabyGaga.com, Babble.com, and TheBump.com, all of which have active Facebook Pages, reveals that PGT is a frequent topic of discussion and debate.

An online survey of pregnant Facebook users could be used to assess patient understanding and perceptions of PGT. The results could help inform physicians' and genetic counselors' knowledge of patient concerns regarding testing. Such research may become increasingly important as advances in PGT provide more testing options and the complexity of decision-making regarding testing increases for patients and their families. To obtain a robust sample size, reduce sample bias and maximize cost-effectiveness, targeted, paid Facebook advertising, as discussed previously, may be the most effective approach for recruiting participants.

Facebook may also enhance future research on DTC genetic testing. No study to date has examined consumer use and perceptions of DTC tests for fetal gender. As with research on PGT, Facebook may be a rich context for recruitment of study participants on this topic.

SNS also hold potential for research on rare genetic diseases. Snowball sampling may be a particularly effective approach for research in this area as it has been used successfully in recruiting hard-to-reach sub-populations [Sadler et al., 2010]. Study recruitment messages could be posted on the Facebook Pages of patient and family support and advocacy organizations with a request for interested individuals to share study information with other groups. While some organizations are already facilitating communication regarding research via SNS, the opportunity exists for more groups to use sites such as Facebook in this way.

In most studies that have utilized Facebook-based recruiting, study participation has been limited to completion of remote, online surveys. Only one study has attempted to recruit participants for research involving travel to a study site and results were mixed: a majority of enrollees indicated they were interested in traveling to the site but many were subsequently lost to follow-up [Fenner et al., 2012]. More research is needed to determine the effectiveness of Facebook recruiting for studies involving travel to a study site. In addition, research is needed to assess whether Facebook can be used in recruitment for clinical research. Given the geographic range of Facebook users, the site has potential to be useful in recruiting subjects for multi-site studies at geographically diverse clinical centers.

In considering the possible uses of SNS in medical genetics research, the limitations of these sites must be taken into account. The inability to randomly sample SNS users is a weakness of SNS-based recruitment and may create sample bias. Snowball sampling is inherently non-random [Sadler et al., 2010]. Facebook advertising may result in some individuals seeing recruitment advertisements more than others. Posting of study information on popular Facebook Pages, the fans of which may compose a large and diverse sample of SNS users, warrants examination as a potentially lower-bias recruitment method.

In addition, because there is variation in SNS use by age and gender, studies using SNS as a context for recruitment may be at risk for underrepresentation of men and elderly individuals. To the extent that individuals who are more likely search for health information online are also more likely to participate in medical research, SNS-based samples may also be likely to overrepresent white, non-Hispanic individuals as well as those with higher income or education status. The use of multiple recruitment methods may reduce the potential for sample bias in studies utilizing SNS.

Difficulties identifying duplicate or false subjects in studies using online surveys may also be an issue in the recruitment of SNS-based samples. For studies in which sample validity is a particular concern, using an online research service may be preferable to recruiting directly from a SNS. Finally, because recruiting research subjects from SNS is a relatively new phenomenon, some SNS users may be reticent to participate due to concerns regarding the veracity of recruitment messages and the security of personal information that is shared. A Facebook Page, with information describing the research and its affiliated institutions, may add to a study's credibility.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

SNS provide researchers with a global portal through which they can recruit research participants. Effectively leveraging the strengths of these sites has the potential to facilitate high-quality, lower-cost research on both large populations and small, difficult-to-access sub-populations. The field of medical genetics may lend itself to SNS-based research given the internet presence of DTC genetic testing services and the role of web 2.0 applications in accelerating the flow of information between patients, family members, healthcare providers and the general public regarding genetic syndromes and their diagnosis and treatment. A number of studies illustrate methodologies for recruiting SNS-based samples of research subjects, and additional research is needed to test them in the context of medical genetics.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES

The authors would like to thank Laurie Demmer and Avery Reaves for their critical reading of this manuscript and helpful comments and suggestions.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. BACKGROUND
  5. EXISTING SNS-BASED LITERATURE
  6. THE ROLE OF SNS IN CURRENT MEDICAL GENETICS RESEARCH
  7. APPROACHES FOR USE OF SNS IN FUTURE MEDICAL GENETICS RESEARCH
  8. CONCLUSIONS
  9. ACKNOWLEDGMENTS
  10. REFERENCES