Background
Research drawing upon participants’ perspectives and experiences is increasingly being undertaken as part of clinical trials to help inform recruitment, trial delivery and improve the conduct of future trials [
1]. To date, the overwhelming majority of studies have focused on patients’ understandings and experiences, and how these may affect recruitment, retention and adherence to the treatments investigated during the trial [
2‐
10]. However, the benefits of including staff perspectives have also been highlighted, with studies showing how these can be used to improve recruitment [
11], adherence to trial protocols [
12] and the support given to individuals working on future trials [
13]. Most research undertaken with staff has focused on, and drawn attention to, barriers to recruitment [
14,
15], such as busy clinics, lack of resources [
16,
17], insufficient consultation with clinical staff [
17,
18] or there being fewer patients meeting inclusion criteria than originally anticipated [
19]. Studies have also highlighted how recruitment and trial delivery can be affected by poor understanding amongst staff about what a trial is trying to achieve [
16,
20], perceived pressures to meet trial recruitment targets [
13] and funding systems and financial considerations [
21].
Complementing these studies, a growing body of literature has highlighted the ethical issues which can arise for trial staff due to their experiencing conflicts between research and clinical roles [
22‐
25]. This literature has also explored how trial staff may attempt to address these role conflicts, for instance, by separating research and clinical activities and/or by inserting individualised care into the trial [
23‐
27]; in some instances undermining adherence to trial protocols [
18,
25]. More recently, the emotional impact of trial work on staff has begun to be raised and discussed, this issue having erstwhile been implicit in the trials literature [
22,
28,
29]. Notably, in a recently published synthesis of findings from interviews with staff involved in recruiting to six pragmatic randomised controlled trials (RCTs), Donovan
et al. document what they term ‘hidden challenges’ to recruitment [
30]. Specifically, they highlight the emotion and discomfort staff can experience when they are confronted with conflicts between research and clinical roles and/or when they struggle with the concept of equipoise [
30,
31]. In an interview study with staff involved in recruiting children to clinical trials, Shilling
et al. likewise uncovered emotional issues that arose for staff [
32]. Specifically, they found that staff could worry and feel anxious about burdening families and overwhelming them with information.
In this paper, we seek to advance the literature further by bringing the emotional aspects of trial work further to the fore. To do this, we report findings from interviews undertaken with staff involved in an RCT for people with type 1 diabetes who were responsible for recruitment and delivering the interventions investigated during the trial. The aim of these interviews was to identify issues which arose for staff during the trial which could be used to help explain trial findings. We also sought to inform recommendations for the rollout of the treatments investigated and the conduct of future trials. However, like Donovan
et al. [
30], our interviews with staff not only revealed issues which are already well recognised in the literature, such as organisational barriers to recruitment, they also uncovered unexpected and hitherto undocumented emotional challenges. Specifically, as we report in this paper, staff not only highlighted the emotional impact on themselves of undertaking work during the trial, they also described having to undertake emotion work on patients and their colleagues. In doing so, we will argue that emotional labour, a concept which is now well recognised and widely used in the nursing and midwifery literatures, has been surprisingly overlooked in the literature on trials. As a precursor to presenting our findings and discussing their implications, we will begin by briefly introducing the concept of emotional labour, the trial which formed the focus of our study and our qualitative study design.
Emotional labour
The concept of emotional labour first came to the fore thanks to a ground-breaking study of flight attendants undertaken by Hochschild in the 1980s [
33]. Defined as, ‘the induction or suppression of feeling in order to sustain an outward appearance that produces in others a sense of being cared for in a convivial and safe place’ (page 7, [
33]), Hochschild showed emotional labour to be a key aspect of flight attendants’ work, one which, like members of other service sector industries, they were explicitly trained and paid to undertake [
33]. Subsequent to this, this concept has been extended to understand the work of healthcare providers, principally those working in nursing and midwifery disciplines [
34‐
36]. However, unlike Hochschild’s flight attendants, and because of heavy emphasis placed on physical labour in nursing and midwifery work [
34], researchers have suggested that emotional labour remains an ostensibly invisible and undervalued aspect of the work of nurses and other healthcare providers. Hence, a key agenda of this literature has been to raise awareness of the importance and skilled nature of this aspect of healthcare providers’ roles, and to call for better training and support to be offered to staff in the future [
34‐
36]. Such an agenda is also taken forward in this paper; albeit, in this instance, to offer better and tailored support to staff working on trials, such as the one we now go on to describe.
The Relative Effectiveness of Pumps Over MDI and Structured Education (REPOSE) trial
The Relative Effectiveness of Pumps Over MDI and Structured Education (REPOSE) trial is a two-arm, multicentre RCT comparing two types of insulin treatment delivery for adults with type 1 diabetes provided with high-quality education - pump therapy versus multiple daily injections (MDI) [
37]. At present, most patients with type 1 diabetes use an MDI regimen and the trial was developed to determine whether pumps might be a better and more cost-effective option. Trial participants were required to attend a week-long structured education course called Dose Adjustment For Normal Eating (DAFNE), so they could receive comprehensive instruction on how to accurately count carbohydrates and adjust their insulin doses accordingly [
38,
39]. As DAFNE courses were originally developed for patients using an MDI regimen, an adapted course was used for those randomised to the pump arm of the trial so that they could also receive training in the use of pump technology. To be eligible for the trial, patients could not have previously used an insulin pump, attended a DAFNE course (hence patients on DAFNE course waiting lists were targeted for recruitment into the trial), or have a stated preference for one type of treatment over the other. Checklists were used to help ensure patients met all trial eligibility criteria. Patients were provided with an information sheet prior to deciding whether to take part and those taking consent were instructed to discuss key aspects of the trial to help foster patients’ understanding, including knowledge of pump therapy and the randomisation process. Consenting patients were allocated a place on one of two DAFNE courses. These courses were then randomly allocated to either pump or MDI treatment at least four weeks in advance of the first course date in the pair. After attending their pump or MDI courses, patients’ clinical care was returned to their routine healthcare providers which, in practice, were often the same staff who were responsible for trial recruitment and delivery. Clinical data for the trial are being collected at six, 12 and 24 months [
37].
The trial is being conducted in eight centres in Scotland and England, most of which ran three pairs of DAFNE pump and MDI courses. The courses were normally delivered by two DAFNE educators, usually a diabetes specialist nurse and a dietitian. As well as being responsible for delivering the courses, one or more of these educators also took responsibility for recruitment in each centre, supported by a central Clinical Trials Unit (CTU). As well as undertaking a health economic evaluation and statistical analysis of the final trial findings, the CTU was responsible for overall management of the trial, including: periodic updating of the trial protocol and operating procedures; data management, randomisation and trial monitoring; providing centres with detailed reminders of trial timelines and recruitment deadlines; facilitating educator teleconferences and providing regular updates on the trial’s progress, including recruitment numbers achieved in the different trial centres.
Discussion
This study explored the experiences and views of staff involved in a clinical trial investigating two methods of insulin treatment delivery. What staff’s accounts have served to highlight is that, alongside the physical and logistical challenges involved in trial delivery [
14‐
16], and the ethical challenges which can arise from staff having to balance research roles with clinical responsibilities [
23‐
25], trial work can also involve attendant emotional challenges [
30,
31] and require staff to undertake emotion work. Such work may not only require staff to pre-empt and manage patients’ emotions, such as their potential anger or disappointment on hearing the outcome of randomisation, it may also entail staff managing their own emotions and those of colleagues. Like the emotional labour uncovered by James in her study of hospice nurses [
34] and first reported by Hochschild in a study of flight attendants [
33], this emotion work is arguably an invisible and hitherto underreported aspect of trial work. It is, however, a kind of work which deserves more consideration and attention to help ensure staff are appropriately supported to undertake trial work in the future.
Since the early work undertaken by James [
34] and others [
35,
36], there has been a growing recognition of the importance, yet undervalued nature, of emotional labour in nursing and midwifery work. Indeed, a now extensive and diverse range of studies have shown this kind of labour to be an integral aspect of nursing and midwifery work, spanning a wide range of disciples and sub-specialties, including: mental health nursing [
43], end of life nursing [
34,
44], dementia care [
45] and gynaecology [
46,
47]. As this body of work, when brought together, has also served to highlight, and in keeping with the findings reported in this paper, emotional labour can be a complex and multi-faceted task. Specifically, it can entail staff resourcefully employing empathy and ‘emotional intelligence’ [
48] to understand patients’ perspectives and provide them with emotional support [
46‐
48], as well as offering similar support to colleagues [
49]. It can also require staff to emotionally manage themselves in order to present an appropriate emotional response and persona to others [
46,
50], and it can extend from public and workplace settings into domestic and private spheres [
45]. As some studies have further served to highlight, the emotional labour undertaken by staff can also be affected by organisational features and constraints, such as ward routines and requirements to follow strict protocols [
34,
51]. The impact of organisational features was also apparent in our study in which we found that staff members’ experiences of stress and the attendant emotional work undertaken could be affected by whether their centre was a new or established DAFNE centre with a long or short patient waiting list.
As indicated earlier in this paper, a common agenda of the nursing and midwifery research has been to raise awareness of the importance and highly skilled nature of the emotion work undertaken by staff. It has also been to highlight the inadequacy of current staff preparation and training [
47,
48], which has mostly been informal and learnt on the job [
34,
47,
52]. Hence, these studies have called for better and more formalised training and support to be offered to staff to undertake effective emotion work [
34,
43,
44,
48]. Such a call, it could be argued, should be extended to staff working on clinical trials [
30,
31]. However, it should also be kept in mind that, despite the rapid growth of scholarly activity within the field of emotions, it has had disappointingly little impact on everyday nursing and midwifery practice [
49,
53]. This situation has been attributed to emotional work being marginalised by the dominant emphasis on clinical outcomes, cost containment and evidenced-based practices [
53,
54], and has led to calls for pragmatic and realistic approaches to be developed [
54].
In developing pragmatic and realistic approaches which could be used in trial work, a number of potential avenues could be considered. One would be to check that sites interested in participating in a clinical trial have the capacity to deliver the requirements of the study and have in place an infrastructure to support implementation. Some of the staff we interviewed indicated that they had felt ill-prepared for the work involved in undertaking recruitment and would have liked to have been more involved in decisions about taking part in the trial. Hence, we would recommend that site checks should entail discussion with all front-line staff who would be involved in trial recruitment and delivery, as well as with principal investigators.
Another potential avenue to consider would be to develop and use effective approaches for pre-empting and preventing potential disappointment about treatment allocation (and, hence, the need for staff to undertake emotion work on patients and themselves). This could potentially be achieved by improving patient and staff understandings of trial designs, in line with others’ recommendations [
8,
20]. However, as qualitative research undertaken with patients and/or their proxies (for example, the parents of newborn infants) has served to highlight, this may not necessary be an easy task [
7‐
9]. This is because, even when people are given clear and accessible information (as happened in the REPOSE trial) and are able to accurately recall it, they can still find it very difficult to grasp and understand concepts such as equipoise and randomisation. Hence, provision of information to patients in isolation is unlikely to be sufficient [
8]. To address this problem, promising work has been undertaken by Donovan
et al. [
55] who used a complex intervention, informed by ongoing qualitative findings, to improve training given to recruiting staff (for example, through individualised feedback following review of audio-recorded recruitment appointments and through updating of training materials throughout the trial). This intervention was shown to have a positive impact on patients’ understandings of the trial, informed consent and, hence, acceptance of treatment allocation [
55]. The benefits of staff training have also been highlighted in another study [
12]. In this, staff described how they had been able to undertake trial work in an unbiased way after receiving training to raise their awareness of their own treatment preferences and how these might impact on trial delivery practices. Hence, this has led to a recommendation that similar, proactive training be incorporated into future trials [
12].
In addition, we would recommend, like others, that trial staff would benefit from increased resourcing and administrative support [
15]. Staff might also benefit from greater provisioning of experienced research nurses to undertake recruitment. Indeed, use of such staff might not only lessen trial staff’s workloads, but also the ethical issues and attendant emotional challenges which can arise when conflicts occur between clinical and research roles [
25]. However, given these kinds of interventions can be expensive, other potentially more cost-effective and easy to implement alternatives may need to be considered. To this end, and drawing upon the ideas, suggestions and positive experiences reported by staff who took part in the REPOSE trial, future trials could include regular sessions for frontline staff, perhaps led by the principal investigator or other senior team member, with local psychological support. These could be used to enable and encourage staff to discuss emotionally challenging issues, seek input and emotional support from colleagues and to critically reflect on experiences [
47]. Depending on the size of the trial team within particular centres, these sessions could be centre specific or held between sites. In addition, staff could benefit from being able to access a psychologist or counsellor on a one-to-one basis, when required, to discuss complex emotional and ethical issues, seek advice on how these could be addressed and receive targeted support for themselves. We would also recommend that staff are provided with training to help them to develop and use effective strategies for dealing with potential patient disappointment or anger following notification of treatment allocation. This could comprise role-play of hypothetical scenarios before the trial begins and in anticipation of problems and challenges arising, and use of actual cases occurring during the trial to learn from ongoing experiences and share good practice (see also Donovan
et al. [
55]). To deliver these kinds of interventions, we would also recommend that appropriate funding is included in future grant applications. In addition, we would recommend that such interventions should be evaluated for effectiveness and cost-effectiveness.
In making these recommendations, it is important to consider that they are informed by findings from an exploratory study of a relatively small number of staff working on one particular trial, all of whom were women and all of whom were either nurses or dietitians. This may limit the generalisability of the findings, not least given Donovan
et al.’s observation that different ethical issues and attendant emotional challenges may arise for different categories of trial staff; in their case, doctors versus nurses [
30]. It should also be taken into account that emotion work is stereotypically considered women’s work [
34] and, hence, one which female staff members are thought to be particularly well placed to undertake. Therefore further research involving different categories of staff working on different kinds of trials would be beneficial; this could also include observational as well as interview methodologies. These reflections notwithstanding, we would suggest that our findings and recommendations do have a potential resonance and relevance for staff working on other types of clinical trials, especially those which, like REPOSE, are non-blinded and, hence, where there is potential for patient and staff disappointment about randomisation outcomes. It is also pertinent to consider that REPOSE is a relatively uncontroversial trial in so far as, while it deals with a chronic health condition, it does not involve patients at the end of their lives or those who might be motivated to use a trial to access potentially lifesaving treatments. Hence, it is possible that the emotional issues highlighted in this paper could be even greater in trials involving more controversial treatments or with greater morbidity and/or mortality implications.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
JL conceived of the design of the study, analysed the data, drafted the manuscript and is the guarantor. JK contributed to the design of study, collected and analysed the data and helped draft the manuscript. DW contributed to the interpretation of the data and helped draft the manuscript. DR contributed to the design of study and helped draft the manuscript. CC contributed to the interpretation of the data and helped draft the manuscript. SH contributed to the design of the study and helped draft the manuscript. All authors have read and approved the final manuscript.