Jessica Van Den Broeck

	It is already generally acknowledged that recruitment is one the most important parts of a clinical trial and one of the major reasons for this is that without the appropriate number of patients there is no possible scientifically sound research. However, when literature is reviewed it seems that not reaching initial recruitment targets within initial timelines is more the rule than the exception. This indicates that current recruitment strategies are shortcoming and need to be reviewed. Furthermore, improving recruitment will have scientific, ethical and economic benefits on a global scale.

 defines social media as “the internet and mobile technology based channels of communication in which people share content with each other.” The best-known examples of such social media are social networking sites like Facebook and Twitter. The definition also notes that social media blur the line between producers and consumers-because of the ability to share content between users-and it is in this characteristic that social media is distinct from traditional media channels.

 The evolution of the internet (from Web 1.0 to Web 2.0, allowing websites to run the write and read process at the same time) and the ever increasing internet availability are key drivers in the rise of social media and we are now at a point where social media are a new prime source of information. Furthermore, social media has also created a new way of searching information: instead of actively searching for information, we now use social media to ask for information or we wait for information to reach us via social media. 

	Recruitment exists in essence only of spreading study specific information to potential participants so that they can make an informed decision on whether or not they want to participate. In a way, then it seems logical to investigate whether it is possible to tailor this activity to the new way of searching information. This led to our developing an online survey of social media and the recruitment of study subjects for clinical trials focused on the Belgian clinical trial landscape. For Belgium, improving recruitment will be even more necessary as the country will lose its major advantage of very short ethics committee and Regulatory Authorities approval timelines because of the new European Clinical Trial Regulation. Logically, the country’s strong points--level of expertise, quality of research and number of trials per million habitants--need to be maintained while improving the current weak points.

	The online, cross-sectional survey was conducted to systematically assess the opinions of clinical research professionals from different stakeholders about using social media as a targeting recruitment tool. As the survey was only conducted in Belgium, all questions were specifically about the Belgian clinical trial landscape. The survey consisted of 31 questions, in eight categories, and presented in a fixed way so that every participant completed the same survey. The survey was live between February 3, 2014 and April 20, 2014.

	Invitations to participate were sent to professionals from Clinical Trial Units (CTU), Contract Research Organizations (CRO), sponsors, Ethics Committees (EC) and the authorities (FAMHP). Furthermore, two inclusion criteria were used for inviting participants: they had to be a member of one of the stakeholders mentioned above and they had to be professionally located in Belgium. These inclusion criteria ensured that the invited participants had sufficient knowledge about clinical research and that they were familiar with the Belgian clinical research landscape. Participants from sponsors and CROs also had to be part of a clinical operations/research management team or had to be part of a recruitment department, as this ensured that they had knowledge about trends in the industry and/or trends in recruitment. Potential participants were contacted via e-mail and a standard text was used to reduce bias in the willingness to participate.

	The descriptive nature of the study forced the authors not to define a required sample size up front, as people who participated were deemed to be interested in the topic and/or in developing clinical research and this makes the total number of participants a descriptive statistic on itself. The goal of reaching a number of 120-150 completed surveys was set and completed surveys had to come from members of all stakeholders. The number of completed surveys per stakeholder was however not expected to be equal. Another consequence of the descriptive nature of the study is that statistical analysis is limited and results are presented as mean percentages or as an average numeral score of these mean percentages.

	Of 405 invited professionals invited, 154 participated in the survey, which is a response rate of 38%, and ultimately 135 completed surveys were included in the analysis. This means that 19 surveys were excluded from the analysis, mainly due to incompliance with the survey rules, equalling a drop-out rate of 12%.

	All stakeholders were represented across the 135 analyzed surveys and most of surveys were completed professionals from CTUs (52%). The majority (62%) of these CTU participants were principal- and/or sub-investigators. Almost half of the respondents were professionally located in or around Brussels, which accurately reflects the composition of Belgium. Respondents were almost equally divided in gender and 42% were between 40 and 50 years old. Within this cohort of survey respondents, LinkedIn is the most popular social media platform.

	The majority-81%-of survey participants considers recruitment to be important for the successful conduct of a clinical trial and this reflects the general idea about recruitment. More surprisingly is that 99% of the survey respondents considers recruitment to be a difficulty. The largest part (78%) thinks that it is a difficulty that can be overcome, but a similar portion of the respondents (79%) is still convinced that improving recruitment in Belgium is necessary. Also remarkable is that 82% think that the best way to improve recruitment involves an option that includes developing new recruitment strategies.

	These ciphers indicate that clinical research professionals in Belgium are well aware of the necessity to improve recruitment and they have an open mind toward developing new recruitment strategies. There is, however, a clear distinction between hospital-related research professionals (respondents from CTUs and ECs) and non hospital-related respondents from CROs, sponsors and the FAMHP in the extent on how important recruitment and improving recruitment are to them, with hospital-related thinking of both as less important. Furthermore, the majority of EC respondents believe that improving recruitment can best be done by further developing current strategies, which is opposite of the majority sentiment. From all this information, it can be concluded that the first step toward improving recruitment will be to get all clinical research professionals from all stakeholders looking at improving recruitment in the same way.

	In general, there is a positive attitude toward using social media for recruitment purposes among Belgian clinical research professionals. This can be concluded from the fact that there is always a majority who agrees with one of the following three statements: ‘social media will help in reaching initial recruitment goals within initial timelines,’ ‘potential subjects will be enthusiastic to receive recruitment messages via social media,’ and ‘social media are the best new recruitment strategy of all possible new recruitment strategies.’ The division between hospital-related and non hospital-related respondents is again present. Graph 1 shows that the hospital-related are less convinced that social media will help in reaching initial recruitment goals within initial timelines and the same is true for the two other statements.

	Response per stakeholder + illustration of the relative positive attitude per stakeholder. The relative positive attitude number per stakeholder is the sum of the products of the percentage per answer and the appreciation factor (applied appreciation factors: I strongly disagree = 1, I disagree = 2, I Agree = 3 and I strongly agree = 4). Applying this formula, the minimum relative positive attitude number has a value of 100 while the maximum has a value of 400. The FAMHP is not included in this kind of analysis because of the too small sample size. The average relative positive attitude number for this graph is 263.

	The format of the survey did not allow to assess the absolute use of social media for recruitment purposes and, therefore, we attempted to get an idea of to what extent social media are already used for this cause. The result of this assessment can be viewed in graph 2. The key conclusions from this graph are that social media are already used in Belgium for recruitment purposes and the most stated reason for rejecting social media recruitment strategies is ethical/privacy concerns.

	The percentage of the parameter ‘respondents’ is given in function of the total number of individual participants included in the analysis (n = 135). The percentages for the parameters ‘yes’ and no’ are given in function of the number of individual responses per question (n = 106, n = 101 and n = 102 respectively).

	Even more important than the result regarding the actual use of social media for recruitment purposes, are the results regarding the ideas about the future use of social media for recruitment purposes. Forty percent of the participants think that social media are already used for recruitment purposes but 87% think that they will be used to recruit study subjects within the next 10 years. This percentage even augments to 90% when participants were asked if social media would ever be used for recruitment purposes. This indicates that Belgian professionals think it is inevitable that social media will be used as recruitment tool in the near future, which corresponds to a generally positive attitude towards social media recruitment and developing new recruitment strategies . It also points out that efforts should be done to make clinical research ready for the introduction of social media.

	A general trend among people involved in clinical research is that they expect problems with implementing or using social media for recruitment purposes, reflected by the 67% who think that CTUs do not have the knowledge to perform a social media recruitment campaign. This belief is justified, as 73% of CTU respondents stated that they do not have the knowledge to do so. Furthermore, 78% is also convinced that the current regulations and guidelines regarding recruitment aren’t sufficient to regulate the use of social media for recruitment purposes. Next to these structural problems, there also seems to be some ethical concern as there is a 21% difference (83% versus 62% respectively) between respondents who consider it acceptable to recruit healthy volunteers with social media and respondents who consider it acceptable to recruit patients with social media. Also remarkable is that hospital-related CRPs have greater concerns about ethical and privacy issues compared to not hospital-related CRPs. However, from comparing graphs 3 and 4 with graphs 5 and 6, it can be concluded there is a greater concern about the structural problems. It is also notable that this is the case for each stakeholder group.

	Comparison between the relative positive attitude numbers per stakeholder for the statements written with each graph. The mean relative positive attitude numbers are 233, 212, 307 and 271 respectively. These indicate that there is a less positive attitude towards the structural problems (graphs 3 and 4) and that there is a more positive attitude towards recruiting healthy volunteers with social media compared to recruiting patients (graphs 5 and 6).

	Also important to point out is that almost half of the respondents (47%) believe that ECs will be the greatest barrier to the implementation of social media into the recruitment process. Respondents from EC seem to understand that they could be a barrier, as a quarter pointed themselves out as the greatest barrier. A majority of the EC respondents, however, point toward the organizations (Belgian Council of Medical Doctors (30%) and the FAMHP (25%)) that create the guidelines and regulations they have to take into account when forming an opinion as the greatest barrier. This indicates again that the current regulations and guidelines are a major obstacle to implementing social media recruitment and that a thorough review is necessary before advanced recruitment tools, like social media, can be used in a safe and ethical way.

	The paragraphs above give an overview of the ideas of Belgian clinical research professionals about social media recruitment but these ideas and doubts aren’t necessarily correct or justified. Sixteen percent of the survey respondents indicated that they already did research on social media recruitment and half of these respondents belong to a CTU. In order to check whether these respondents actually did research on social media recruitment some general control questions about social media were included in the survey. The result of this is illustrated in graph 7 and confirms that the ‘have done research’ group has a better social media knowledge than the ‘haven’t done research’ group.

	Comparison between the ‘have done research subgroup’ and the ‘haven’t done research subgroup’ for the control questions on social media knowledge. These control questions were based on general statements found in social media literature. The ‘have done research’ subgroup gives answers that are in better agreement with what is found in literature.

	Over the whole result of the survey, the ‘have done research’ group is more positive about social media recruitment and also gives-according to what was found in literature-more correct answers about it. This is confirmed by the relative positive attitude analysis, visualized in graph 8. It can thus be concluded that any reluctance or misjudgment towards social media recruitment is mainly the product of a lack of knowledge. The ‘have done research’ group also has less ethical or privacy concerns and shows for example only a marginal difference between their opinions on the acceptability of recruiting patients or healthy volunteers via social media. This indicates that privacy and/or ethical concerns are really not a major problem of social media recruitment and this collides with the fact that the most stated reason for rejecting social media recruitment tactics are privacy/ethical concerns. The opinion of this group about the structural problems does not differ a lot from the ‘haven’t done research’ group, which confirms that these problems are indeed the greatest barriers to the implementation of social media into the recruitment process.

	Comparison between the ‘have done research subgroup’ and the ‘haven’t done research subgroup’ for the survey statements on the problems with social media recruitment.

	From this research, it can be concluded that patient recruitment should be improved and that this certainly applies to the Belgian clinical research landscape. Via a literature study and this survey, it was assessed that social media is a promising tool to enhance recruitment. It can be used as a targeting recruitment tool but also as a medium to create more knowledge and awareness about clinical research. The latter is a new major responsibility of the pharmaceutical/clinical research industry that can best be done by providing correct, understandable and patient-centric information via the patients’ new information sources, like social media. Interaction between the industry and patients via these media channels is also something to be considered.

	The survey pointed out that Belgian clinical research professionals from different stakeholders believe that social media can help in improving recruitment and that they expect that social media will be used in the near future to recruit patients. The major obstacles that are withholding us from recruiting patients via social media are a lack of knowledge on the matter and an insufficiency of the current regulations and guidelines about the use of social media in clinical trials. Ethical and privacy issues seem not to be of great concern.

Vincent Schroyens, Quintiles Belgium – Department Site Management; Jessica Van Den Broeck, Quintiles Belgium – Department Site Management; Yves Geysels, Quintiles Belgium – Department  Site Management, University of Namur – Faculty of Medicine: department Biomedical Sciences

	1. Gearhart, C. (2012). IRB review of the use of social media in research. The Monitor vol. 26 issue 7, 39-42.

	2. Coman, D., and Spaeder, J.A. (2013). Harnessing the power of the digital patient. (

http://www.quintiles.com/library/white-papers/harnessing-the-power-of-the-digital-patient-white-paper.pdf?aid=31899

	3. Taylor, P. (2014). The rules of digital patient engagement – more needs to be done to recruit patients to clinical trials.

	4. Burrell,E.R., Pines,H.A., Robbie,E., Coleman,L., Murphy,R.D., Hess,K.L., Anton,P., and Gorbach,P.M. (2012). Use of the location-based social networking application GRINDR as a recruitment tool in rectal microbicide development research. AIDS Behav. 

	5. Fenner,Y., Garland,S.M., Moore,E.E., Jayasinghe,Y., Fletcher,A., Tabrizi,S.N., Gunasekaran,B., and Wark,J.D. (2012). Web-based recruiting for health research using a social networking site: an exploratory study. J. Med. Internet. Res. 

	6. Watson,J.M. and Torgerson,D.J. (2006). Increasing recruitment to randomised trials: a review of randomised controlled trials. BMC. Med. Res. Methodol. 

	7. Mackenzie,I.S., Wei,L., Rutherford,D., Findlay,E.A., Saywood,W., Campbell,M.K., and Macdonald,T.M. (2010). Promoting public awareness of randomised clinical trials using the media: the 'Get Randomised' campaign. Br. J. Clin. Pharmacol. 

	8. O'Connor,A., Jackson,L., Goldsmith,L., and Skirton,H. (2013). Can I get a retweet please? Health research recruitment and the Twittersphere. J. Adv. Nurs.

	9. Wood, L. (2013). Presentation: Can patients drive better recruitment? – A case study in patients recruitment from the bottom up. From ACRP 16th European Conference – Late breaking Clinical trial News 2013.

	10. Gehring,M., Taylor,R.S., Mellody,M., Casteels,B., Piazzi,A., Gensini,G., and Ambrosio,G. (2013). Factors influencing clinical trial site selection in Europe: the Survey of Attitudes towards Trial sites in Europe (the SAT-EU Study). BMJ Open. 

	11. The Initiative. (2012). PwC study: Clinical research footprint and strategic plan to promote clinical trials in Belgium. (

http://www.pharma.be/assets/files/3229/3229_129857736290100673.pdf

	12. Financial Times Lexicon, definition of social media (

http://lexicon.ft.com/Term?term=social-media

	13. Van de Belt,T.H., Engelen,L.J., Berben,S.A., and Schoonhoven,L. (2010). Definition of Health 2.0 and Medicine 2.0: a systematic review. J. Med. Internet. Res. 

	14. Emmanuel, S., and Day, K. (2011). Using social media to facilitate patient-provider interaction. HCIRO, 23-30

	15. CISCRP. (2013). The 2013 CISCRP Perceptions & Insights Study – Report on Clinical Trail Information Seekers. (

https://www.ciscrp.org/wp-content/uploads/2014/01/2013-CISCRP-Study-Clinical-Trial-Information-Seekers.pdf

	16. Fisher,J. and Clayton,M. (2012). Who gives a tweet: assessing patients' interest in the use of social media for health care. Worldviews. Evid. Based. Nurs. 

Articles

A survey of Belgian clinical research professionals offers insights into the future of social media

Jessica Van Den Broeck

Articles

Is There a Link Between Recruitment and Social Media?