The role of social media in clinical research
In November I was delighted to be invited to speak in the Great Debate at the Clinfield Annual Conference about the use of social media in clinical research. The motion for the debate was:
This house believes that the use of social media enhances dissemination and engagement in clinical research.
As a frequent user of social media (including Academia.edu, Facebook, ResearchGate, LinkedIn and Twitter) I have often argued for the greater use of social media in clinical research and other academic activities. So my heart sank a little when I realised that Kelly’s invitation was to speak against the motion. However, as I began to explore the issues and to develop an argument against the motion I started to convince myself that greater care should be taken when seeking to use social media in clinical research.
During the debate I based my argument on the following five points:
- There are still too few patients, especially those over the age of 65, who engage with the Internet and social media. While there are a growing number of e-patients and e-carers who are willing to engage with the internet, there are still significant numbers who do not use the internet or make use of social media. Any social media recruitment and retention strategy can only be used as part of a broader strategy, but as the number of internet and social media users continues to grow the use of social media in clinical research should also increase.
- Using social media could cause sampling bias and have consequences for the quality of the research. One of the commonest criticisms of clinical research is that study populations do not reflect the patient population who might eventually use the new drug or intervention. Using social media to aid recruitment risks contributing to sampling bias by only accessing that part of the population that can be reached through social media.
- Researchers are not yet ready to use social media. Social media is still a relatively new technology and many researchers do not themselves yet engage with it. Until they do, social media can only be used as part of a much broader recruitment and retention strategy.
- Online patient communities are not always representative of patient populations. Online patient communities have a number of useful functions but they are not always established in such a way that they are usefully representative of patient groups. Focusing social media strategies on such groups risks wasting researcher time and could, again, have consequences for the quality of the research.
- Social media is not as secure as it needs to be. For those who know what they are doing, it is not too challenging to hack into social media accounts, to access personal information or to send malicious posts. There is also a danger of disclosure by those using social media. There is research demonstrating how frequently healthcare professionals risk identifying patients through post on social media.
I do not believe that the above points should prevent researchers from using social media but they do raise a number of issues that researchers should consider. Research is all about creating evidence so it is interesting that there is so little evidence about the use of social media as part of a recruitment and retention strategy in clinical research. We need to create that evidence to inform researchers.
As the Chair of an NHS Research Ethics Committee I know that one of the commonest protocol amendments to come across my desk is the result of poor recruitment, with researchers seeking to change their recruitment strategy or to extend the duration of their study. I have no doubt that researchers could use social media more effectively to support recruitment and retention of research participants. The point I made during the debate is that researchers should use social media to support recruitment and retention with great care.← Research nurse moves beyond clinical research delivery to design and dissemination 5 minutes with Denise O’Donnell, Senior Hepatology Research Nurse in Oxford →