Article Text
PDF
Abstract
Difficulties relating to supportive and palliative care research are often reported. However, studies have highlighted that people near the end of life are happy to participate in research and want their voices heard. Thus, one may raise a twofold question: are we limiting the freewill of people who are seriously ill? And are we missing important data, which probably cannot be obtained from other sources? In light of this landscape, a new opportunity has emerged: the use of social media (SM). This paper provides a comprehensive summary of SM, including its theoretical underpinnings, and recent examples of successful uses of SM in healthcare research. It also outlines the opportunities (wider reach, direct access, the potential of Big Data, readiness of research data, empowered participants) and challenges (anonymity of participants, digital divide, sample bias, screening and ‘saying no’ to participants, data analysis) of using SM in end-of-life care research. Finally, it describes the practical steps that a researcher could follow to recruit patients using SM. Implications for palliative care clinicians, researchers and policymakers are also discussed, with a focus on the need to facilitate patient-centred care through the use of SM. The need for relevant and updated guidelines in this new, emerging field is highlighted.
- Ethics
- Methodological research
- Social media
- Internet
- Palliative care
Statistics from Altmetric.com
Introduction
Difficulties in supportive and palliative care research
Difficulties relating to supportive and palliative care research are well documented.1 ,2 The topic under investigation, namely the process of dying, is often considered to be the main challenge. Psychosocial research in end-of-life care usually involves interviews with patients where a number of topics are explored. Asking patients to tell their story is often approached, and rightly so, with caution, to ensure that patients only disclose as much as they are comfortable with.3 Dickson-Swift et al4–6 also suggested that when thinking about what sensitive research entails, we should take into consideration not only the topic under investigation, but also its impact on all people involved in it. Healthcare professionals may feel that discussing participation in a study could be difficult during a short consultation. They may also act as ‘gatekeepers’ by only approaching patients who they deem would be well enough to take part in the research. A recent review of palliative care research revealed that 39% of studies excluded patients that were considered as too ill by the healthcare professionals.7 This may be linked to the patients’ physical abilities to take part, but also to the notion of adding to the patients’ burden. Therefore, in palliative care research, both the focus on patients’ experiences and their involvement, as well as that of those caring for them (both relatives and healthcare staff), may be considered as sensitive.
A number of studies have discussed strategies to maximise recruitment of seriously ill patients into healthcare research. Ryan et al7 suggested that close collaboration between medical staff and researchers could facilitate the recruitment process: medical staff by monitoring the physical status of the patient and identifying times when patients feel better, and researchers by being more flexible in their recruitment procedures. However, these strategies may be too time-consuming for healthcare professionals to adopt in the demanding day-to-day clinical practice. Agar et al2 recently provided an overview of mechanisms to ensure that ethical principles are met when obtaining informed consent in palliative care research. Giving potential participants the choice to opt out rather than opt into the study, for both patients and healthcare professionals, has been used with success.7 However, the use of these techniques is largely dependent on regulations by ethics committees, to ensure the well-being of and respect towards those involved.
On the other hand, some authors have highlighted that seriously ill patients are happy to be asked whether they are willing to take part in research and engage as participants.8 In fact, some have expressed concerns about having the right to make decisions removed from them, hence ‘being treated as already dead’.9 One could also argue that distressed patients may also benefit from having their voices heard in the context of research. The recent Cancer Patient Experience Survey10 revealed that only 37% of patients had been asked whether they wanted to take part in cancer research. There were great variations across trusts, ranging from 14% to 62%. More importantly, 92% of patients were happy to have been approached about taking part in research.10 A recent review11 exploring the views of patients, caregivers, health professionals and researchers on research at the end of life, highlighted how not only participating in research showed positive outcomes for both lay carers and patients, but also how it was only a minority of patients who experienced distress. While it might be the case that some patients would not find taking part in the research always beneficial, not asking them puts the matter in a different place altogether. Thus, one may raise a twofold question: are we limiting the freewill of people who are seriously ill? And are we missing important data, data which probably cannot be obtained from other sources?
In light of this landscape, a new opportunity has emerged: the use of social media (SM). Despite it being a relatively new concept, a number of definitions have been already proposed. The General Medical Council (GMC) has defined SM as ‘web-based applications that allow people to create and exchange content’, such as ‘blogs and micro-blogs (such as Twitter), internet forums (such as http://www.doctors.net.uk), content communities (such as YouTube and Flickr) and social networking sites (such as Facebook and LinkedIn)’.12 Other SM platforms allow users to do collaborative projects, participate in content communities, and interact in virtual social worlds.13
Aim of this article
The purpose of this paper is threefold. First, we aim to give a brief summary of SM, including its theoretical underpinnings and some examples, so that researchers and clinicians can familiarise themselves with this rich, complex aspect of the Web. Second, we will outline the opportunities and challenges of using SM in palliative care research, and its clinical, research and policy implications. Third, we will describe the practical steps that could be used to recruit patients using SM, as an aid or as the only recruitment procedure.
What is SM?
SM is embedded within the paradigm of the Web 2.0. The term Web 2.0 was coined in the early 2000s,14 and is used to designate the current stage of the Web, where the focus is on the content created by the users (as opposed to data created by institutions or organisations). The ‘2.0’ in Web 2.0 refers to the second conceptual stage of the Web. Before that, the Web was unidirectional: the Web 1.0. During the 90 s, the internet became popular but content was created only by some, and internet users were not able to change it or interact through it (please see section ‘What is SM’. below for the distinction between SM user and internet user). In the Web 1.0, most users were only consumers of content, not creators. The internet was to be read, but not to be written.
Content, created by users in the shape of text (eg, blogs), images (eg, Flickr) or videos (eg, YouTube), is what powers the Web 2.0. The more people that use this content, the richer this data gets.14 In other words, SM cannot exist without users. For instance, the value of Facebook lies in the number of updates people post on it; if no one posts updates, Facebook becomes empty. The same principle applies to all SM: it is the central aspect of Web 2.0. There are hundreds of SM platforms, each with their own features and target population. The best known examples of SM in 2013 in the Western world are probably Twitter and Facebook. The rapid growth of SM is evident in the commercial sector. Although this expansion has not been mirrored in healthcare research to the same extent, we are seeing more and more examples of the use of internet and SM in this context. The extent of utilisation of internet including SM in health-related studies varies from using it as one of many ways of recruitment, through seeing internet-based recruitment as a parallel method to traditional methods.15 There seems to be a growing interest in it in the field of palliative care. A recent paper16 summarised the SM landscape in regards to palliative care, focusing on the advantages and disadvantages that its use has for both patients and clinicians. Granger17 reflected on how the use of SM by palliative care practitioners and patients like herself may have an impact on the improvement of healthcare. A growing number of researchers are already using SM to disseminate their work and collaborate with others by using platforms such as LinkedIn and Researchgate. Key organisations in the field, such as the European Association for Palliative Care (EAPC), maintain blogs aimed at providing a platform for researchers and service users to share the ideas and recent developments (see http://eapcnet.wordpress.com/). Nicholson18 suggested a number of professional Twitter accounts related to palliative care that clinicians and researchers may find of interest.
In summary, within the Web 2.0 paradigm, recruitment for research can be operationalised as a continuum. At one end, there are the traditional recruitment methods, in which no SM procedures are used. At the opposite end of the continuum, there is recruitment using exclusively SM, with no traditional methods being used at all. Finally, between these two extremes, any given study may use any amount of each method. For instance, some researchers might choose to use SM alongside traditional recruitment methods, or use traditional recruitment methods with the aid of SM at a later stage.
Information and communication technologies (ICT), including smartphones and websites, and more recently, SM, have shown promising results in other healthcare research fields.15 ,19 ,20 Indeed, a recent study reported recruiting 529 participants solely via Twitter.21 Close et al22 demonstrated that, in fact, the use of ICT including Facebook, can be the most successful method in recruitment of children with genetic conditions into clinical trials, with a recruitment rate of 91% for internet-based recruitment approaches and 9% for the traditional approaches respectively.
SM users
It is commonly asserted that SM is just for young people, or that SM demands a considerable amount of time and skill. As controversial as these statements are, they may not be true anymore—if they ever were. People of all ages surf the Web and use SM. In fact, a term has been created for older people who have the skills to use ICT effectively: ‘silver surfers’.23 Studies have reported how older people are able and keen to use ICT for health purposes.23 ,24 In regards to time investment and level of skills, in the past decade there has been a huge focus on the usability of the Web, with a focus on trying to make ICT as user-friendly as possible. This has partially contributed to the success of these technologies with people of all ages, and has allowed the development of SM.
It is worth noting that not all internet users are SM users. Not everyone that goes online participates on SM: a portion of people choose to read websites, send or receive emails or purchase products. None of these activities are SM. A recent US-based report on internet use25 reported that 72% of adult internet users used a social network platform when they go online. While this is a majority, it still leaves over 1 in 4 people not using SM when online. On a typical day, 48% of adults who go online use such platforms, 28% watch videos on YouTube and 18% send instant messages.25
Using SM in supportive and palliative care research: challenges and opportunities
Research around end-of-life care has always been dotted with challenges, and the use of any new method poses new concerns. SM is not an exception to this.
We will describe in this section the main challenges and opportunities associated to the use of SM in supportive and palliative care research. A summary is provided in table 1.
Summary of challenges and opportunities
Opportunities
There are several advantages regarding the use of SM in end-of-life research. We are mindful that some of the opportunities below come with challenges as well, which we will address below.
Wider reach
Traditional recruitment approaches are greatly limited in terms of time and space. In regard to time, traditional recruitment is limited by the working hours of the people who recruit for us: clinicians, nurses and so on. It is also limited to the availability of the potential participant: normally, the times when she or he can attend the healthcare centre. On the other hand, SM allows the information to be available online around the clock. Once a researcher has posted the information about the research project and/or recruitment, this information stays online all day, every day, so that the potential participant can access it at his or her convenience. Similarly, SM can override space barriers. Traditional recruitment methods are highly limited in terms of space: usually to the healthcare centres or professionals that the researcher has made links with. Instead, SM can reach anywhere in the world that has an internet connection. This is especially relevant when researching rare conditions, or rare types of cancer. In summary, SM allows the researcher to reach a much wider potential participant spectrum.
Direct access
In traditional recruitment strategies, access to potential research participants is often challenging in terms of time and finances, and almost always there are mediators. Such mediators may be ethical bodies, clinicians or administrative staff, among others. These often receive the informal name of ‘gatekeepers’, as it is how they are perceived by researchers. On the contrary, the use of SM allows a much more direct, fluid communication with the population of interest. One of the main characteristics of SM is its responsiveness and agility. It seems likely that a person who is SM active will be proactive in communication.
Big Data
The potential of Big Data is related to the vast amount of data available online, and that requires new ways of processing and structuring.26 It can be found in any field, from meteorology to technology, and also in healthcare. The three key features of Big Data are volume, velocity and variety. Volume refers to the vast amount of data that is generated and can be stored (eg, all the personal data from the 1.15 billion Facebook users worldwide). Velocity is the speed at which this data is created, and also the speed with which it changes and evolves. Finally, variety refers not only to the different kinds of data (medical, sociodemographic, financial and virtually any field), but also to the variety of ways in which Big Data can be processed. Unsurprisingly, SM generates enormous amounts of data. As technology develops and makes it possible to store, process and analyse this new kind of data, researchers in all areas are only now starting to realise the potential of Big Data for healthcare research. Big Data is already being used in cancer genomics research, where processing information from huge numbers of cases has provided very specific data on how to treat patients individually.27 SM provides access to Big Data in research too: huge amounts of data are already available online. The main challenge with Big Data nowadays is being able to make sense of it (‘data mining’). A number of initiatives (collaborations between universities and industry) and instruments (computing languages) are rapidly emerging to address this.28 Finally, researchers should be aware that, Big Data mining and analysis will incur financial cost.
Readiness of the data
In some occasions, data may already be there thanks to SM: for instance, people who leave video testimonials on websites such as YouTube talking about their process of disease, or patients who write about their condition on their own blogs. This can, in some specific cases, eliminate the need to recruit, since the data is already there. For instance, a growing number of studies are using SM audiovisual platforms as the content for their research.29 ,30
Empowered patients
Many studies have reported an association between the use of SM and increased levels of empowerment in patients, including feeling more in control and improved self-efficacy.31 The use of SM may also facilitate patient empowerment in the context of research. Hanley et al32 identified three levels of involving service users in research including: consultation, collaboration and user control. While consultation involves asking service users for their views on a particular issue or stage of the research, collaboration shifts control to the patient by inviting them to be part of the committee or undertaking the research. User control enables service users to lead the project and oversee its progress. While a number of suggestions have been made on how to implement different levels of service user involvement in traditional settings,33 little attention has been paid to how this could be implemented in the context of SM. SM could be used to empower patients in research in the following ways: asking patients about research ideas, having them comment on the design of a given project or encouraging them to disseminate research findings via their networks. Equally, it seems reasonable to expect that participants recruited in this manner will be more proactive and prone to share their experiences. It has benefits not only for the researchers, but also for the participants, who might feel empowered by having their views listened to.31
Financial
Despite some studies having used paid advertisement in SM to recruit participants, in principle the only cost of SM is the necessary connectivity. This feature makes it especially attractive to projects with a limited budget.
Challenges
Privacy and anonymity
When using SM to recruit in supportive and palliative care research, confidentiality and anonymity are paramount, as they are in using traditional recruitment methods. This is especially relevant when considering the use of internet forums for recruitment. There is a subtle difference between internet forums and other SM platforms, with regards to ethical concerns. For platforms such as YouTube or Twitter, the user is aware that his or her data is in the public domain: when someone uploads a video to YouTube, they know that it is going to be made available to everyone wanting to watch it. The same does not apply to internet forums. On such platforms, users chose to create a login, provide more or less information about themselves, and the content that they post (text, pictures, videos) is in principle only intended to be accessed via the forums. In other words, it is not intended to be made available at the same level as it is in platforms like YouTube or Twitter. This must be taken into account when designing the SM recruitment strategy, and researchers need to make a decision during the early stages regarding the level of public access, to ensure that privacy of potential participants is respected.
Digital divide
While the use of the internet and SM has become mainstream in many high-income countries during the past decade, there are still millions of people that cannot access this technology. This may be due to lack of internet connection or equipment, or to a lack of skills, and has been linked to age and level of education.34 This inequality in terms of internet access is known as the digital divide. The increasing availability of mobile applications (widely known as apps) for SM means that people in lower income countries are often able to access the internet, even in situations where getting online via a computer is not possible. While there are many initiatives aimed at reducing the digital divide, there is still a long way to go.
Ownership of content
A myriad of legal issues must be taken into account, since data posted in SM is stored on the servers of private companies and corporations. For example, videos and pictures posted on Facebook are still copyrighted by the creator of the images (the user), but can be used, modified, shared and sold in any way by Facebook. Users are often not aware of data protection policies, although they are explained in the Terms and Conditions of these platforms. This is especially relevant for Facebook and Twitter. Legal and ethical concerns may arise as part of the research process and must be considered before making a decision to use them.
Sample bias
There may be concerns whether participants recruited via SM are different from those recruited via traditional methods.35 A study by Reed15 on metastatic breast cancer highlighted that although the participants recruited online and in a traditional way were comparable on a number of demographic variables, the participants recruited online were younger and represented the lowest level of deprivation compared to that of those recruited via traditional methods. Participants recruited online also scored significantly lower on social well-being and reported lower satisfaction with the services.15
Screening and ‘saying no’ to participants
Terminology in end of life care is complex and might be a bit confusing for patients with an advanced disease. A recent survey conducted by National Cancer Survivorship Initiative revealed that the language spoken by clinicians is not always understood by patients.36 Another study highlighted that staging of the disease was especially confusing for patients.37 This may have implications for the recruitment of patients via SM. When utilising the traditional methods of recruitment, screening of patients based on the inclusion criteria is usually undertaken by the medical team. For researchers based in the academic context, not only the identification of the patients but also the first contact with the potential participants is usually performed by the relevant clinicians. Recruitment in SM is based on self-referral process and its success depending on the participants’ accurate understanding of the clinical information given to them by the clinical team.
Reed 15 compared face-to-face and online recruitment strategies into a survey-based study with women with metastatic breast cancer. One of the challenges they have come across was that some patients who considered themselves as having metastatic cancer have in fact either local cancer recurrence or primary breast cancer. In those cases, such questionnaires were not included in the analysis, without challenging the participants’ understanding. For qualitative studies which may recruit via SM but undertake data collection face to face, the issue may need to be considered in a different way. The level of information available to patients about their conditions needs to be considered, especially with regards to prognosis and severity, as researchers may inadvertently reveal information related to their condition. While this is true for any interview study, the issue may be magnified when using SM to recruit, when the initial screening is performed in fact by the participant. The researchers need to consider the implications of this when designing the project.
Data analysis
The issue of potential bias in a sample will also have an impact when analysing the data. The decision on whether data obtained via SM and traditional methods can be analysed together or separate will mainly be guided by the type of methodology employed. In qualitative studies, sample homogeneity will be more important for specific approaches, such as Interpretative Phenomenological Analysis,38 while the ability to compare data coming from different sources may be an advantage, for instance in Grounded Theory.39 For quantitative studies, it may be necessary to compare the characteristics of the sample recruited via traditional methods and SM to gain a full picture of the data obtained.
Conclusions
Implications for research: guidelines on ethical procedures
When using SM as a way of recruitment, a study must follow the same ethical procedures that would be complied with when using traditional methods of recruitment. The GMC and the British Medical Association (BMA) have recently issued comprehensive guidelines for healthcare professionals on the use of SM and its impact on the patient-doctor relationship.12 ,40 However, its use in a research context has been largely overlooked. Both the British Psychological Society (BPS) and the American Psychological Association (APA)41 have recently started addressing the need for guidelines on internet-mediated research, with BPS having published the comprehensive Ethics Guidelines on Internet mediated research.42 In it, they highlighted a number of challenges which need to be addressed when designing such studies including data protection, opportunities for debrief or obtaining informed consent.42 As rightly pointed out, these new guidelines ‘should be considered as supplemental and subordinate to the Society's Code of Human Research Ethics and the overarching Code of Ethics and Conduct’.42 The responsibility for considering these issues rests with the ethics committees and the researchers.42 It is crucial that local ethics committees are aware of these guidelines and facilitate discussion involving not only researchers, but also service users.
The guidelines are based on the assumption that ‘the design normally involves acquisition of data from or about individuals in the absence of face-to-face co-presence’.42 There might be instances, however, where the individual is recruited via SM but the data collection is carried out face to face. In those instances, some of the challenges mentioned earlier such as screening of participants may introduce additional complexities. Researchers need to use their judgement and think critically about the potential impact of their research, not only directly on the potential participants, but also on the wider Web community, and should be guided by four principles of the BPS Code of Human Research Ethics,43 which include respect for the autonomy and dignity of persons; scientific value; social responsibility; and maximising benefits and minimising harm.
Research methods using SM are by nature faced with a dynamic and ever changing landscape, which makes it difficult to predict its future. The speed of these changes may not always be immediately acknowledged in the appropriate guidelines42 and it is essential that researchers engage in the debate about the participant's safety on a regular basis. Box 1 provides a summary of 10 simple steps which researchers can follow when using SM for research.
10 simple steps to maximise participant recruitment using social media (SM) in 2014
Create a Facebook page for your research project.
Create a Twitter account for your research project.
Update the above regularly. SM is an ongoing conversation: you need to update your SM platforms as often as you can, to keep your project visible.
Establish online links with other researchers and relevant bodies in your community: be proactive and start following relevant accounts.
Keep it professional. Many people use SM for leisure, and you might do too, but the purpose of the accounts that you will create for your project is professional. For many potential participants, this will be the first time they will learn from you or your study, so you want to set the tone from the beginning.
Strictly follow the ethical guidelines established by the relevant bodies in your field. Respect and ensuring the participants’ wellbeing are as important in a virtual environment as they are in a face-to-face environment.
Maximise the visibility of your research project in search engines (such as Google or Yahoo) by making sure that other relevant websites link to it. If you can, create a web page in your institution's website (NHS, university…). Link to your SM from there and also link to that from your SM.
Create a short video (between 3 and 5 min) explaining your project in lay terms and update it to YouTube and/or Vimeo. Again, don't forget to add the links to your other SM.
If you already have a blog, post about your project. As in step 3, remember to post brief updates. Always be aware of potential copyright issues.
Make sure you stay active and visible. Exchange knowledge, opinions and links. It is a good idea to keep a log (maybe a simple Excel file) of how many participants you are getting per SM platform. This will help you identify patterns in the recruitment of participants, and may be relevant during the data analysis stage.
Implications for clinicians: paradigm shift
The irruption of the Web 2.0 and SM has involved a change of paradigm in the medical profession. Until the early 2000s, health information was only available to professionals, and was of course only created by professionals as well. This has radically changed: anyone with access to a device connected to the internet can create and publish health information online. This very often happens through SM. It can be on a larger scale, as is the case when individuals write blogs or record series of videos where they reflect on their conditions, treatments and experiences. It can also take place on a smaller scale, as is the case when people use microblogging platforms (such as Twitter) for the health purposes. This has very recently come to the attention of the medical community, with two high-profile cases: Kate Granger, a British doctor diagnosed with incurable breast cancer, and Lisa Adams, a US citizen living with stage IV breast cancer. These two people have been openly sharing their experiences with advanced cancer and their treatments online, using Twitter and blogging. Some have questioned the ethical implications of this,44 while others have highlighted the paradigm shift that this controversy illustrates.45
As rightly noted by Coiera,46 SM provides also a new way of engaging with public. SM has favoured a conversation between stakeholders (patients, clinicians, researchers) in a way that had never been seen before. This has of course meant a shift from the biomedical model towards the biopsychosocial model, where patients are empowered, proactive and truly the central element of their healthcare.47 Palliative and supportive care as a discipline has always been known for being patient-centred. However, this transition to a more holistic vision where the patient is the centre has brought a change in the dynamics between healthcare professionals and their patients. For better or for worse, people now have access to a myriad of information about health online. There has been much debate about the benefits and dangers of such access.44 Some clinicians have been, and still are, rather reluctant to lose their position of information-holders and share information with patients. On the other hand, the benefits of health-related information being freely available on the internet have been proven in a number of conditions.48 ,49
Recommendations for policymakers
In a recent report,50 SM users themselves expressed concerns regarding consent, anonymity and avoiding undue harm when using SM for research. The same report also highlighted that even among users of SM there is confusion ‘about what the term ‘research using SM’ meant’.50 The regulatory bodies have been relatively slow in responding with guidelines to mirror the increasing interest of people in use of SM. As pointed out before, there are some guidelines which provide researchers with a number of issues to be considered when designing and conducting internet-mediated research. These issues should not discourage researchers from using this innovative way of engaging with participants but rather strengthen projects on both scientific and ethical level. Guidelines must be monitored and updated on a regular basis to mirror the current developments of technologies.
General conclusions
People near the end of life are often hard to reach populations in healthcare research. However, research has shown that a vast majority of seriously ill people would be happy to be asked to participate in research. Furthermore, policymakers may lack information from dying people, simply because they are considerably difficult to recruit. In the light of this landscape, SM is rapidly rising as a potentially useful tool to reach and recruit in palliative care research. This paper has offered a summary of the theoretical underpinnings of SM, and described how it can be used as a tool in recruitment for palliative care research. It has also highlighted the main challenges and opportunities associated with its use. Using SM to recruit participants might be tempting, but it must be carried out in a regulated manner. Even more so in palliative care, where patients’ cognitive and physical abilities may be compromised, thus limiting their use of SM. This is not different to other methods of recruitment, where patients declining health may prevent them from participating in research. The aim of this paper was to offer a comprehensive and objective overview of the challenges and opportunities of SM in palliative care research, so that researchers are more able to make an informed decision and design their research protocols accordingly. The particular use of SM discussed in this paper requires a set of skills and information. While there has already been a considerable effort to create guidelines, researchers and policymakers need to work closely together to develop guidelines which respond to the changing landscape of SM.
We have illustrated the difficulties researchers in palliative care experience to access potential participants. Unfortunately, their voices often go unheard due to the combination of factors described in this paper. On the other hand, we have highlighted how crucial it is to take their views into account, especially in service planning and development. While public health services have shifted towards a patient-centred approach, and user involvement seems to be nowadays a crucial aspect of health research, there is still much to do until we can talk about participant-centred research. SM has the potential to bring patients to the centre stage in research.
Acknowledgments
The authors would like to thank Professor Eila Watson and Professor Mary Boulton for their support in writing this paper.
References
Footnotes
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.