Do social media uploads, comments, blogs and videos provide rich opportunities for research or should we stand back on moral grounds? Answer: we need a framework
By Dr Dafydd Brooks , Dr Mark Taubert, clinicians at Velindre Cancer Centre and Dr Annmarie Nelson and Dr Noreen Hopewell-Kelly, researchers at the Marie Curie Palliative Care Research Centre, Wales Cancer Trials Unit, Cardiff University, UK
Patient and carer peer support on social media sites:
Many people with health problems and their carers use the internet to seek information about their condition. The rise of social media has only served to expand the sharing of information, with micro blog sites like Facebook hosting pages for groups to discuss both common and rare conditions alike.
This form of communication makes patient and carer interaction across the globe with similarly affected individuals very easy, and directly and indirectly impacts on heathcare services. There is also huge potential for research into what people are saying and writing. (1, 2, 3)
Research on peer support via social media:
We looked at research articles focussing in particular on social media hosted peer-support mechanisms for people with chronic conditions. Numerous research papers are emerging looking at online behaviours of patients with conditions such as diabetes and mental health diagnoses. Focusing on peer-support mechanisms, there is a trend for openly available online interactions to be analysed by researchers using both quantitative and qualitative methodologies.
For instance, one such paper, a recent research article in PLOS ONE (4), analysed 3,044 comments on 19 Youtube videos created by people who self identified as having severe mental illness, focussing on how they gained peer support through sharing advice or describing their own health experiences.
Analysis of data showed that peer support occurred across four themes:
– minimizing a sense of isolation and providing hope;
– finding support through peer exchange and reciprocity;
– sharing strategies for coping with day-to-day challenges of severe mental illness;
– learning from shared experiences of medication use and seeking mental health care.
Lack of anonymity:
These broad themes were consistent with already accepted notions of peer support in mental illness, consisting of a voluntary process aimed at inclusion and mutual advancement through shared experience and also developing a sense of community. Data from this study also suggested that the lack of anonymity and associated risks of being identified as an individual with severe mental illness on YouTube appeared to be overlooked or felt not important by those who uploaded videos and wrote comments. This lack of data awareness carries with it potential concerns for researchers interested in analysing and publishing such data.
Metamorphosis: from casual upload to full blown research data
An important area for consideration in this type of research paper is the moral dilemma of analysing data that, while being in the public domain, was unlikely to have been intended for research analysis by the persons uploading them. This is an area where informed consent for the uploaders’ views to be included in a particular research study is absent, and more generally, inclusion in any kind of research consent is non-existent unless specifically detailed in the social media sites terms and conditions, (which most users are unlikely to read). It is difficult to quantify to what degree users of social media might object, if they found out that their online uploads were also to be found in a research article, having been anonymised, analysed and dissected with a specific question in mind.
Also, there is the issue about where such research information is stored; a Youtube clip that has been analysed would presumably be stored and saved in the researchers’ place of work, as with any other form of audiotapes from qualitative interviews. A subject’s personal views, in addition to being in their intended place online, are also now in print or on file elsewhere, e.g. in a scientific journal where they can be viewed by a different audience, or in a research centre’s storing facility.
Publicly available, therefore fair ‘research game’?
Issues surrounding (informed) consent, so integral to biomedical ethics including Good Clinical Practice, are going to be increasingly challenged by this interesting, very accessible new form of research. Both the General Medical Council and the British Medical Association have issued guidance on social media use for doctors, mainly focussing on conduct and confidentiality issues. But what is also needed is detailed guidance on how this effects the research domain, for instance from the Medical Research Council within their Good Clinical Practice recommendations.
On the other hand, this openly available data can provide potentially valuable and very raw insights into patient and carer experience and some researchers may feel that this publicly available information is ‘fair research game’.
Metamorphosis: from caterpillar to butterfly and from tweet to research quote
Our palliative care research centre initiated a quality improvement project, exploring associated ethical dilemmas. The research centre has no definitive set of ethical or practical guidelines to inform work with social media, so the centre initiated two consensus days to construct a workable, ethical framework for future research in the field. The consensus days were constructed to discuss and debate the complex ethical challenges embedded in social media research. The ethics of using social media, specifically in relation to privacy and anonymity and public/private domains were debated. Researchers within the centre (n=8), including experienced post doctorate social scientists, clinical academics and two external experts in online research met to focus on constructing a set of ethical guidelines. Issues of confidentiality, anonymity and informed consent were all discussed.
A framework for social media research was proposed with specific criteria relating to recruitments of participants, information giving, and consent, and will form part of a forthcoming discussion paper in a learned journal. The consensus was that using historical text is problematic and therefore not encouraged. Closed forums created specifically for research purposes are considered to be the most appropriate way to conduct online research.
The view of participants was that social media provide a unique opportunity for palliative care research. What is of paramount importance though, is that ethical principles in relation to this medium are clearly defined and that guidelines are in place prior to the commencement of any such research activity. We feel these need to be in place and addressed within the published research paper in order for the informed caterpillar to become an ethical butterfly.
(1) Chou W-YS, Hunt YM, Beckjord EB, Moser RP, Hesse BW (2009) Social media use in the United States: implications for health communication. Journal of Medical Internet Research 11: e48. doi: 10.2196/jmir.1249
(2) Taubert M, Watts G, Boland J, Radbruch L Palliative social media. BMJ Support Palliat Care 2014;4:13–18. [Abstract/FREE Full text]
(3) Comabella C, Wanat M Using social media in supportive and palliative care research BMJ Support Palliat Care doi:10.1136/bmjspcare-2014-000708
(4) Naslund JA, Grande SW, Aschbrenner KA, Elwyn G (2014) Naturally Occurring Peer Support through Social Media: The Experiences of Individuals with Severe Mental Illness Using YouTube. PLoS ONE 9(10): e110171. doi:10.1371/journal.pone.0110171