Background
Clinical trials are considered the ‘gold standard’ method for assessing the efficacy and safety of pharmaceutical treatments and other health-care interventions. It is common practice for qualitative research to be undertaken with patients and staff who are involved in clinical trials.161,162 This research usually takes place during a trial’s pilot or early phases to improve recruitment, patients’ understanding of trial processes and the solicitation of informed consent.163–168 Qualitative research has also been undertaken during trial delivery to explore adherence to trial protocols and treatments, and aid interpretation of trial findings.169,170 Although the closeout of a trial potentially presents challenges for both patients and health professionals, especially when patients may be required to stop using the treatment(s) under investigation, this aspect of trial participation and delivery remains surprisingly under-researched. The limited work undertaken to date suggests that patients may experience a form of trial bereavement on closeout,171 and some may wish to continue using trial treatment(s) despite the trial failing to show clinical benefit.172 How staff address closeout issues with patients, and what their own information and support needs are, remain unknown.
Closing out REPOSE
Recruitment to the REPOSE study commenced in November 2011 and the first trial centre began to close out patients [i.e. commenced 2-year (final) follow-up appointments] in April 2014, with the final centre closeout appointment in June 2015. The insulin pumps used during the REPOSE study were provided free of charge by Medtronic, with a warranty that covered only the trial’s 2-year duration. After extensive negotiations, pump consumables were funded at a local or national level (e.g. by the DH, Chief Scientist Office or a primary care trust) for the duration of the trial and on the understanding that pump therapy would be withdrawn post trial unless a clinical benefit could be demonstrated for individual patients and local funding provided. This was communicated to potential trial participants in the patient information sheet for the trial (see Appendix 8). On closeout, each REPOSE centre was advised (as per the trial’s SOP for closeout; see Appendix 15) to make their own clinical decisions about which patients should remain on a pump and who should revert to a MDI regimen, with centres having to find local funding for patients who remained on pump therapy.86 It was also agreed (as formalised in the SOP) that the patients would not be told whether or not they would keep their pump until after their data had been collected at the final 2-year appointment because of concerns that this knowledge might influence how they completed their questionnaires.
Early reports from trial staff and ongoing review of trial data indicated a large variation in closeout practices between the REPOSE centres (listed in Table 11). Although, in some centres, most or all patients remained on pump therapy, in others, the majority of patients had pump therapy withdrawn and were reverted to a MDI regimen. Early anecdotal reports from staff also indicated that patients’ emotional reactions to withdrawal of pump therapy were variable, with some presenting major challenges to staff. Specifically, some staff expressed concerns about the lack of guidelines, procedures and support structures for themselves to manage and support patients effectively at closeout when withdrawing pump therapy. In light of these reports, it was decided to systematically evaluate staff experiences of closeout to generate insights and recommendations to support the conduct and closeout of future trials, especially those where an expensive health technology is being tested that may be withdrawn at the end of the trial period. A case was made to the funder to undertake this additional piece of qualitative work using some underspend within the grant. Approval from the funder for this substudy was given on 13 April 2015.
Aims
This qualitative study drew on the experiences, understandings and views of health professionals who were involved in closeout of the REPOSE Trial in order to:
better understand variations in practices between trial centres on closeout and establish whether or not, and to what extent, these arise from local clinical guidelines and practices, individual physician/health professional beliefs and/or other factors and considerations
inform guidance and support for staff involved in the closeout of future clinical trials, particularly those in which investigated treatment(s)/device(s) may be withdrawn.
Research questions
What are health professionals’ experiences of closing out the REPOSE Trial? What (if any) practical/ethical/other issues arose for staff, and how did they attempt to address these?
What factors and considerations informed health professionals’ decisions to continue or discontinue pump treatment in individual patients?
What processes and procedures do staff think should be put in place to support patients and staff involved in the closeout of future clinical trials, especially those where expensive health technologies are being investigated and may be withdrawn?
Overview
The qualitative work was completed to plan and on schedule, enabling a comprehensive investigation of staff members’ experiences of closeout in the seven main REPOSE centres. Although it had originally also been our intention to include patients’ views, we were unable to involve this group because of the limited time available to gain NHS research ethics and R&D approvals at the REPOSE centres, and undertake the data collection and analysis. One journal article has been accepted for publication, which reports key findings from the following analysis (Lawton et al.173).
Study design and methods
In-depth interviews were used to collect data about staff experiences of study closeout, as these afforded the flexibility needed for participants to raise and discuss issues that they perceived as being salient, including those unforeseen at the study’s outset.142,174 The use of one-to-one interviews also afforded privacy, allowing participants to share their views about the processes and procedures for closeout at their study centre. The study was informed by the principles of Grounded Theory140 and entailed simultaneous data collection; this allowed the areas explored in the later interviews to be revised in light of emerging findings.
Recruitment and sample
Working closely with the CTRU to identify relevant individuals, we targeted all staff members (physicians, diabetes specialist nurses and dietitians) in the REPOSE centres, who were thought to have been actively involved in closeout appointments. Staff were recruited from seven of the eight participating centres. The eighth centre was not included because it was a reserve centre that was added at the end of the trial to deliver two courses only, and this centre had only one patient using a pump at the end of the trial. Staff were recruited via written (e-mail) invitations accompanied by information sheets and opt-in forms. When staff had opted in, NH contacted them to arrange an interview.
Data collection and analysis
The University of Edinburgh’s Centre for Population Health Sciences, Ethics Review Group granted ethics approval for this study in June 2015. The interviews took place between June and August 2015. Participants were offered the choice of a telephone or face-to-face interview at a time/place most convenient to them; only six (29%) staff members requested a face-to-face interview.
Interviews were informed by a topic guide that was developed in light of literature reviews and findings from qualitative research conducted earlier in the trial,147,155 and were revised in light of emergent findings from the early interviews. The final version of the topic guide is appended to this final report (see Appendix 16). Interviews lasted for ≈60–90 minutes. The key areas in the topic guide were covered and explored in depth in all interviews. Interviews were digitally recorded (with consent) and transcribed in full for in-depth analysis. By the time recruitment and interviewing had stopped, data saturation had been achieved, that is, no new findings or themes were identified in new data collected.
The interviews were analysed thematically by NH and JL using the method of constant comparison.148 Individual interviews were read through repeatedly to look at differences and similarities in individuals’ perspectives and experiences before being cross-compared to identify common issues and experiences across and within study centres. NH and JL wrote separate reports before meeting (both during and after data collection) to discuss and reach agreement on key themes, identify emerging findings requiring more detailed exploration and develop a coding frame. The qualitative analysis software package NVivo10 (QSR International, Warrington, UK) was used to facilitate data coding and retrieval. Coded data sets were subjected to further, in-depth analysis to identify subthemes and illustrative quotations.
The findings presented below are structured under our original research questions. To safeguard participants’ confidentiality, pseudonyms for individuals, Dr X (diabetes specialist) or EDX (DAFNE educator – diabetes specialist nurse/dietitian), and centres (A–G) are used throughout this report, and all identifying information has been removed or deliberately altered.
Findings
Participants
Twenty-four staff members were invited to participate. In one case a staff member said that they had no direct experience of closeout/end-of-trial consultations. Two others opted in, but an interview could not be arranged at a convenient time, hence 21 (87.5%) staff members were interviewed. Full details of the final sample are provided in Appendix 17.
As can be seen from Appendix 17, we achieved good representation of different types of staff: clinical diabetes specialists, diabetes specialist nurses and dietitians. Between two and five (mode three) members of staff were interviewed at each centre. With the exception of centre A, at least one DAFNE educator and one diabetes specialist was interviewed from each centre.
Staff experience of delivering DAFNE varied from 5 to 17 years (mean 10 years). There was also variability with regard to individuals’ experience of pump therapy, ranging from 2 to 37 years (mean 10 years). It should be noted that the majority of staff interviewed at centres D and E had relatively little experience of pump therapy prior to delivering REPOSE. Although many staff had previous experiences of working on clinical trials, few had been involved with studies that had required new technologies to be withdrawn at the end of the trial. The main exceptions were those staff members (n = 5) who belonged to the three study centres that had been involved in the delivery of a DAFNE pump pilot study. This 12-month study27 – 6 months’ recruitment and 6 months’ follow-up – had taken place between 2009 and 2010.
In order to understand staff members’ experiences of closing out REPOSE, it is necessary to provide an account of what happened at the end of the trial in the various centres. Thus, prior to answering the research questions, we will describe the variations in closeout practices that staff described in the different centres.
Background: closeout practices in REPOSE centres
As noted above, the CTRU issued a SOP for closeout (see Appendix 15), which outlined what was to happen up to the point at which all trial procedures were completed (i.e. final blood samples were taken, QoL measures collected and data from the pump downloaded). What happened to trial participants afterwards – whether or not they remained on MDI/pump, whether or not they had pump therapy withdrawn or initiated – was a clinical decision, taken by staff at the individual centres. In other words, the decision to leave patients on, start or terminate pump therapy was not a trial decision. However, many of the staff involved in delivering REPOSE experienced these post-trial treatment decisions and, more specifically, patients’ reactions to them, as part of their trial experience. Thus, for the purpose of this chapter, we will talk about post-trial treatment decisions as part of the closeout process because this is how the staff perceived and interpreted them.
To ensure that resources (i.e. pumps) were allocated appropriately and fairly at the end of the trial, most centres put site-specific procedures in place for closeout (i.e. what would occur after the final downloads had been logged). Some centres adopted very formalised operational procedures for decision-making about post-trial treatment. In these centres, decisions about individual participants were made at a multidisciplinary team (MDT) meeting, involving all of the research team and other staff members, and which took place a couple of weeks before the closeout of each of the groups. The MDTs’ decisions were governed by strict NICE/Scottish Intercollegiate Guidelines Network (SIGN) criteria (see Research question 2) and (normally) documented. Each patient then attended a post-trial consultation with a clinician/educator after the final data collection session to discuss his/her treatment plan.
Other centres took a less formal approach to closing out their patients. Dr H described what happened in one such centre (centre B):
I guess it wasn’t a formal MDT. But yeah it was just a chat with the educators and myself about each individual patient as they were coming up for the end of the study, about who/what the best way forward was for them.
In some of these centres, the whole research team met in advance of the post-trial appointments to discuss what might happen in individual cases; in others, the educators briefly spoke to the clinician after the patient had provided their final download and before they went in for their post-trial consultation. In these centres, although some, or all, trial team members had some input into post-trial treatment decisions, it was individual clinicians who made the final decision during the post-trial consultation, often taking the patient’s views into account:
We started a fairly neutral conversation about how it been and what would they want to do if the option were that they could keep it. And then if they said well you know, they’d really like to stay on pump therapy then I said ‘OK, well you know, let’s have a look at how you’ve got on with it’ and obviously I’d got a feel for that already. So before they came in [educator] and I sat down and looked through and looked at how they’d got on, and what had happened to hypoglycaemia frequency, what had happened to HbA1c, and then obviously they came through and told us how they felt in terms of, the impact on quality of life and things.
Dr G
In all of the centres, the clinical appointment to discuss post-trial treatment occurred after the final appointment to collect trial data. In some cases, patients were seen on the same day on which they came in for their final trial appointment; in others, this clinical appointment occurred a couple of weeks later.
Research question 1
What are health professionals’ experiences of closing out the REPOSE Trial?
Staff who had been involved in follow-up appointments during the trial, primarily the educators, said that they had become increasingly aware that withdrawal of pump therapy at closeout/the end of the trial would be difficult. This was a result of their observations (see ED7 below) that patients were becoming increasingly emotionally attached to their pumps as the trial progressed, an issue which became particularly apparent from the 12-month follow-up appointment onwards.
. . . at the kind of the routine REPOSE follow-ups when we asked them how they were feeling about the pump, they all reported that they loved the pump, that they felt it was making their life so much easier, and that they couldn’t imagine going back to having to inject multiple times a day . . . So they were all very vocal that they really wanted to stay on their pump. And that they would be prepared to fight for it, if needs be.
ED7
For this reason, some staff reported worries and concerns about how patients might react to the withdrawal of the pump at the end of the trial: ‘I knew it was going to be difficult and I wasn’t looking forward to it’ (ED11).
Dealing with stressful situations
The ways staff experienced these post-trial consultations varied, and was related to whether or not patients were able to remain on their preferred therapy and, as will be described later [see What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?], whether or not staff had put pre-emptive measures in place to manage and prevent problems arising from the withdrawal of pump therapy. In some cases, when patients who wanted to remain on a pump were told they would have to revert to MDI, staff members, including Dr C, described situations that had been stressful and difficult to manage because patients had become upset and/or angry:
I had trouble in the course, because one lady when she came to the end of her trial, her HbA1c was appalling. I mean it was appalling. There was no way you could justify leaving her on pump, because she was getting no benefit from it biomedically. What she needed was a complete change in how she managed her life. And she was very upset to have the pump removed. But what was fascinating was she did not say that at the closeout interview . . . . Next thing I know she’s written streams of letters of complaints to all and everybody, because we removed the pump from her, and refused to give her any supplies after 3 months.
Dr C
Later in the interview, Dr C reflected on how this and other similar experiences had ‘. . . kind of tainted the whole study for me, because it was really quite difficult for a little while. It was very uncomfortable’. Dr B reported a similarly stressful encounter with a patient:
The one locally that really didn’t go well, was a gentleman whose control had got worse on the pump. And I was explaining that in fact on balance it was actually more dangerous for him to remain on pump. And he was the one that walked out. He didn’t shout or give me any indication. He just stood up and said ‘OK’ and walked out.
Dr B
Like Dr C, Dr B had been taken aback by this experience: they had been ill prepared for it, primarily because, like the other clinicians in the study, they had been less involved in the trial follow-up visits, which had been mainly carried out by educators.
Smooth transitions
However, the withdrawal of pump therapy was not always experienced as generating such negative reactions; indeed, a small group of patients, across the centres, were described as having been ‘happy’ to revert to MDI, with some requesting this transition at the end of the trial. Moreover, in another centre (centre A) at which pump therapy had been withdrawn from the majority of patients, staff said that closeout had been relatively straightforward and non-confrontational. As will be described further later [see What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?], this appeared to be due to staff having put procedures in place to pre-empt, prevent and manage disappointment among those patients.
In approximately half of the centres, patients received the treatment that they wanted at the end of the trial, and closeout, as a result, was experienced as raising few issues for staff. This was particularly the case in Scottish centres where, in 2012, the Scottish Government had made funding available for pump therapy, with a target of ≈5% of patients with T1DM to be using pump therapy between 2013 and 2015. As Dr E reflected, because of the Scottish Government’s largesse, closeout was very straightforward in that centre because the majority of patients were able to remain on pump therapy if they wished to do so:
Our closeout has probably been less complex than most places. And that’s because of this impetus to increase the number of people with pumps . . . Happily for us, because the timing was just perfect, so that the end of the trial was within this expansion up, we were actually able to fairly straightforwardly continue with pumps on a routine NHS way for all of the patients who wished to.
Dr E
ED3, from another resource-rich centre, similarly said ‘I think it [closeout] went well. There was nothing certainly from our side in [site D]. I don’t think there were any issues for us’.
Staff at the Scottish centres did comment that, had government funding not been put in place during the trial, closeout would have been more challenging and problematic:
Well I guess we would have been in the same situation as other centres where there was no funding stream to continue patients. And we would have had to say: ‘sorry. We don’t have any money for you to continue on this’. And I think it would be very difficult. I mean obviously I would imagine in other places it’s caused a bit of damage to the doctor or health-care professional relationship . . . I guess people having invested a lot of time in it over the course of the study you’d feel a bit let down if somebody’s told that there’s no money. Sorry, give it back.
Dr H
Although staff at such centres did not generally have to manage patients’ reactions to the withdrawal of the pump, they did have other issues with which to contend. First, as Dr E noted, they had problems providing timely training for all MDI patients who were offered, and accepted, pump therapy at the end of the trial: ‘Most of our control patients were really quite keen to go on pumps, afterwards. And the, you know, there’s a degree of work just dealing with that’. Second, even though patients usually received the therapy they wanted after closeout, the staff said they still had to reassure and ‘calm down’ patients when they came in for their final downloads because they were anxious about losing their pumps:
On the day of their final visit I think they were all extremely heightened, they were very worried I think most of the patients who came in. We kind of had to almost calm folk down a little bit. We had quite a few who were walking in the door at that final visit very, very scared because they knew it was the end of the trial and they didn’t know what was going to happen now.
ED6
As ED6 commented, dealing with patients’ anxiety throughout the trial was particularly difficult in their centre, as although staff realised that most people would have their pump therapy funded after the trial, they still had to follow the trial SOP, which required staff to be more circumspect when patients asked about post-trial treatment during follow-up visits.
Differences of opinion within multidisciplinary teams
Finally, with regard to their closeout experiences, some staff indicated a lack of consensus within some research teams regarding the decision to keep particular individuals on pump therapy at the end of the trial:
And I think there was a difference in how some of the team viewed it as well, in that some seemed to say: well, it’s a trial for 2 years. And then they come off the pump and we see how they do. And then we may put them back on the pump. Whereas others are saying: well no there’s been significant improvement, So we’ll keep them on the pump. So I had kind of extremes.
ED12
I think others (in our site) were much more, I . . . I think that they thought that they were going to stick to the letter of the law and they’d take pumps away and that would be tricky . . . but then that’s their individual practice it’s not for me to tell colleagues particularly consultants how they should practice, and the practice is very different it’s such a personal thing.
Dr A
This particularly applied to those centres that had adopted less formalised closeout procedures, specifically where final decisions were taken by clinicians alone. In such centres, not only were disparities in decision-making between different team members noted, but also some team members described having not always agreed with their colleagues’ decision to keep individual patients on pump therapy. Several staff commented that they were not always convinced that patients were benefiting over and above what could be achieved using a MDI regimen and DAFNE education. Indeed, some such staff indicated that they would rather have used stricter guidelines for pump allocation at the end of the trial to ensure that NHS resources were distributed in a fair and transparent way in their centre (see Research question 3).
In summary, the interviews confirmed differences in staff experiences of closeout across the study centres. First, in centres at which pump therapy was routinely withdrawn from all but a few patients, staff had needed to manage some of the patients’ negative emotional reactions and some had felt ill prepared for this experience. Second, there was evidence that some centres had managed patient expectations about post-trial treatment more successfully than others, thereby pre-empting patients’ disappointment at having pump therapy withdrawn (see Research question 2). Finally, in centres where ample funding for pump therapy was available, the issues arising at trial closeout focused on calming anxious patients before final data collection and providing timely training for MDI patients commencing pump therapy.
What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?
Staff identified a couple of issues that may have affected their own and patients’ experience of the trial and closeout; these included the length of the trial and the ethical challenges arising from withdrawal of the pump. Although some of these issues had been identified and addressed prior to, or during, the main trial, others emerged only during the interviews, as staff reflected upon their trial experiences.
Length of trial
Some staff, as already indicated, reported that they did not really start picking up on patients’ anxiety about the removal of pump therapy until they attended their 12-month follow-up. Thus, the length of the trial, or, specifically the length of time spent on pump therapy, was identified as an issue by a number of staff who questioned that this may have affected patients’ emotional reactions at closeout:
I have been involved with trials where the treatment has been withdrawn, but it’s been a shorter period of time. I think 2 years is quite a long time and people get very used to things, don’t they. And then they do start to think that the pump’s theirs. So I think that’s more difficult. When I’ve been involved with other trials of equipment it’s been more like a few weeks, 6 weeks or something like that. And so patients are very aware that it’s just for that trial period.
ED11
Although ED11, like others, saw the length of time spent on the pump as affecting patients’ reactions at closeout, ED1 regarded the trial’s relatively long duration as indirectly influencing some of clinicians’ decisions to continue pump therapy for certain individuals in their centre:
The REPOSE SOP for the end of the study was based on what we did for the pilot. And the only thing I could say, I hadn’t really thought about that until just this morning. And whether there was something to do with the length of the trial, the duration, which made it more difficult to make that decision [to remove pump therapy] at the end.
ED1
Learning from experience gained during the pilot study
Three of the centres in the main trial had been involved in the pilot,27 and staff who had taken part in the pilot talked about how these earlier experiences had influenced the ways that they approached the main trial. These staff described how they had entered the trial with some, but perhaps not enough, awareness that terminating pump therapy at closeout might be problematic, and how this had led them to putting some pre-emptive measures in place to prepare patients for removal of their pumps:
When we did the pilot . . . some people were really devastated that they couldn’t keep the pump, even though we told them. So we were much clearer we think, this time round with: you’re not, you know – although everything was signed – with the fact that they needed to give the pump back. And I think we were before. But I think we just reiterated it throughout the process more.
ED10
Indeed, in one such centre (centre A) staff designed a clear formal protocol for ending the trial, which not only set out criteria for who was to stay on pump therapy (see Research question 2), but also helped them to manage patients’ expectations throughout the trial and their emotions at closeout. This centre had reverted the majority of patients to MDI at closeout and they had followed strict procedures for this including explaining why pumps were being removed, what removal meant and how reversion to MDI might be a temporary state of affairs, which could be revisited in the future. This centre also provided patients with spare consumables so that they could continue to use their pumps in the immediate short term before reverting to MDI at a convenient time, thereby giving them a chance to adjust psychologically and practically to the transition. These strategies appeared to work, for although this centre had withdrawn pump therapy from most patients, the staff reported that this had gone reasonably smoothly:
So that we didn’t switch them there and then on that 24-month visit. We reminded them, we had a few people were quite upset and grumpy about it. And we said: look, how can we? We have some kit we can give you that can tide you over for another month or 6 weeks, while we sort out your pens and getting you back – to switch you back onto MDI and doing it in a supportive as way as possible. We didn’t rip the pump off them at that appointment and say: there’s your pens back, off you go. And so having that discussion at the meetings [MDT] before for all of them just helped us come up with a kind of individual plan to sort of, damage limitation really.
ED14
Leaving the door open to revisit patients’ eligibility for a pump
Although centre A was the only centre to consistently allow patients a lead-in period to revert back to MDI, staff in other centres described how they had tried to manage anxiety and disappointment by making patients aware that they could make a case for them to have the pump reinstated in the future if they struggled to manage their diabetes using DAFNE + MDI:
And we did say to him, as we said to others. This is does not mean that pump therapy is completely closed to you. You know, what you need to do now is go back on injections, really apply DAFNE. You know, monitor, keep records, make adjustments, and you know, down the line, if you’re still not managing to achieve an HbA1c or you’re getting hypos, then we can consider a pump again. But you need to put the work in.
ED1
Staff in all of the centres, but especially those in pilot centres, also talked about how they had tried to manage patients’ expectations about closeout throughout the trial, particularly the likelihood that they may not continue on the pump:
Cause we did – had done the pilot as well we’d sort of expected – we knew what to expect, cause you know, we’d done the pilot before REPOSE. So we’d been involved and the same sort of thing had happened: people you know, of course if they liked the pump, they like the pump and want to keep it. So it was really about just reminding people of the rules and we tried to do that each time we met them as well, just to remind them that this was about the trial, this was about seeing if the pump was effective and if they didn’t meet NICE criteria the pump would go back. So we tried to talk about that at each meeting time as well, not just leave it to the end.
ED11
In a couple of centres, in addition to raising the issue of withdrawal of the pump during trial visits, patients in the pump arm were given encouragement several weeks before closeout to use the remaining period of the trial to demonstrate that they could use their pumps more effectively by the time of their final download or, as Dr H described, to prepare patients for closeout and ease their disappointment if a clinical benefit could not be evidenced:
I think we just felt better that we’d given them the opportunity. You know if you’re pre-warned that there’s going to be an exam[ination] result in another 3 months kind of thing, then if you don’t do so well in it, you think: oh well, at least they told me kind of thing. I think we were just thinking that a warning shot is quite a good idea. And might make the that’s all, no you can’t have a pump any more discussions easier.
Dr H
The ethical challenges of withdrawing pump therapy
For some staff closing out the REPOSE Trial was seen as throwing up distinctive ethical challenges not only because the patients had time to get used to pump treatment, but also because of the nature of the treatments involved. As Dr H noted, unlike drug trials through which treatments might be replaced by seemingly similar forms of therapy, REPOSE required the withdrawn treatment to be replaced by a very different therapeutic option, which, as they noted, may be seen by some patients as not really an option at all:
I mean it doesn’t really have parallels to other studies. I mean I guess if you’re on a new tablet for x, y or z at the end of the study you might not be able to continue it, but there’s usually an alternative. And it’s you know tablet versus tablet instead of you know, pump versus another way of giving insulin which is very different . . . as I say it’s not like this is trying one pump versus another pump, and you at the end of the study you go back to the old pump. But you take away the new fancy one. This is like something, getting something versus getting nothing.
Dr H
Dr A raised further ethical issues regarding the withdrawal of pump therapy. This clinician, like others, argued that if individuals were benefiting, or even perceived themselves as benefiting, in ways that went beyond the clinical criteria outlined in the NICE/SIGN guidelines then it would not be right to remove pump therapy at the end of the trial period, not least because these individuals had given their time to take part in a research project. In other words, as long as patients were using their pump safely then they had the right to keep it after the trial had finished:
Just because we’re doing a research project doesn’t mean you don’t continue to have a therapeutic relationship with people and I mean you can call me a softy, but I think we owe it to our patients who participate in research to do the best by them, and as I said at the beginning you can get a pump for anybody if you want, and I just think making a judgement that they don’t benefit therefore they should stop. If they think they’re benefiting, then I’m not comfortable saying I know better than you.
Dr A
Dr A, like others, also stressed that removing pump therapy at the end of the trial could potentially undermine an ongoing therapeutic relationship with a patient – especially as was the case for this doctor, when health professionals delivering the trial were also responsible for providing patients’ routine diabetes care.
Although Dr A was based in a centre at which access to funding for pump therapy was limited and some patients had pump therapy removed at the end of REPOSE, ED5’s centre, in contrast, had plenty of funding available and the majority of patients had stayed on the pump following closeout. ED5, however, was acutely aware of the different funding situations across the trial centres and commented that the removal of pump therapy at the end of the trial in some centres and not others was just further evidence of the existence of what they regarded as an unethical ‘postcode lottery’:
That’s the state of the NHS that really at the end of the day if the patient’s benefiting then I feel it’s quite sad that someone can remove something from someone that they’re benefiting from. And I think that it just highlights in the NHS a bit of a postcode lottery really regarding pumps, and that hopefully in the future that’s going to be more standardised. Because your care really wherever you are should be equitable.
ED5
In summary, staff in all of the centres anticipated that closeout and the withdrawal of pump therapy might be an issue for patients and, hence, had developed a range of potential solutions to address these, including developing strict protocols for managing expectations and emotions, and reminding patients that pump therapy was a research intervention whenever they attended trial visits. In addition, staff identified a couple of ethical issues, such as the problem of withdrawing treatment from patients who perceived themselves as benefiting from it, potentially compromising an ongoing therapeutic relationship, and the inequity of the postcode lottery for funding treatments in the UK.
Research question 2
Variability between centres
The number of patients staying on pumps after closeout varied markedly between centres and, as noted in Research question 1, it was clear that the staff in the different centres, including ED10, were aware of this:
ED10:
And there are always going to be clinical judgement and exceptions. But it feels a bit like people [sites] have done things slightly differently at the end.
I:
What did you do at the end?
ED10:
We said to everybody, you have to give it back.
Ultimately, it was the availability of resources, specifically the availability of funding to keep/move patients on to pumps in routine clinical practice, which determined what happened to individual patients at the end of the trial. In centres E, B, D and F, for which generous funding was available, the majority of REPOSE patients stayed on/commenced pump therapy if they wanted to:
So we were very fortunate in that sort of financially there wasn’t going to be any problem here about asking patients for the pump back at the end of the study. It was agreed that it would be daft to do that and then restart them again on a pump 3 months later or something. So although the patients didn’t know and obviously we wouldn’t say to them, because that wasn’t, that wouldn’t have been good. You know within the group it was realised that there was a sort of secure funding stream to continue those that were benefiting from the pumps at the end of the study.
Dr H
In centres A, C and G, for which funding was scarce, staff were acutely aware that pump therapy needed to be rationed and restricted to those patients who demonstrated a clinical need or benefit, independent of the patient’s wishes:
[Dr] was quite cut and dried about it. Unless there was a medical reason or unless they met NICE [criteria] already from a hypo[glycaemia] point of view they had to come off. And you know if there was any, you know, trouble, they would come down and talk to the patient themselves if necessary.
ED14
Different interpretations of National Institute for Health and Care Excellence/Scottish Intercollegiate Guidelines Network criteria
The interviews suggested that staff in resource-rich and resource-limited centres tended to use different criteria when making decisions about individual patient’s post-trial therapy. Two resource-limited centres adopted very strict criteria for allocating pump therapy at the end of the trial so that, in general, only those patients with a clinical need who met NICE/SIGN criteria,13,175 as tightly defined (namely, HbA1c > 8.5%, attempts to reach target with MDI resulting in disabling hypoglycaemia), continued using pump therapy following closeout. The remaining patients in these centres were informed that they would revert to MDI:
My view was that if they had shown significant benefit in terms of HbA1c and, and/or reduction in hypoglycaemia frequency, then we would continue them on pump therapy. And that’s effectively what we did . . . . they had to effectively fulfil what NICE would expect. So the NICE guidance is based on an expectation of a 0.9% reduction in HbA1c, and I felt that was what they should be achieving for us to say that they should continue pump therapy.
Dr G
There were some that we knew had done really well on the pump. We knew that they’d really enjoyed being on the pump, that we knew that because they had never had, kind of from a NICE guidance point of view, a period of time having had what we would consider a, you know having done DAFNE and seeing if DAFNE works first, before putting them on a pump, and had never seen that, we couldn’t justify it from a hypo[glycaemia] point of view. They had done really well no doubt. But we couldn’t justify it from NICE to keep them on the pump.
ED14
This approach contrasts with that adopted by resource-rich centres that applied a much looser or subjective interpretation of the NICE/SIGN criteria when determining who remained on pump therapy. In one of these centres, nearly all of the patients in the pump group were allowed to remain on the pump when the trial finished, with some individuals, such as Dr F, justifying their decision by referring to the ambiguity inherent in the NICE/SIGN criteria:
Yeah. I think it was difficult to remove something that somebody’s doing well with, and wants to continue. If you know you have that funding available. And as I say SIGN and NICE are very vague. So you know, you, I felt I could justify it.
Dr F
Using quality-of-life criteria to inform decisions
These centres frequently took QoL issues, as well as biomedical criteria, into account when deciding who should remain on pump therapy. Dr F, for example, commented that they took into consideration how ‘well’ people were doing on pump therapy when making treatment allocation decisions in their centre:
I:
What do you mean by doing well?
Dr F:
It’s interesting isn’t it. So doing well might be having good blood glucose values. But doing well might just be engaging with their diabetes better than they did before. So we had a couple of quite chaotic people who don’t have perfect glycaemic control, but they’re testing, they’re entering information and they’re keeping in touch with us in a way that before they weren’t. So I guess you know, doing well can be something over and above what their blood sugar’s telling us. And certainly their control, it’s not perfect, it’s better and safer than it was before. So I think that’s what I would sort of class as doing well.
Likewise, Dr H, from another resource-rich centre, described how decisions about post-trial therapy at their centre were governed by the team’s ‘global impressions’ about how individuals had coped on pump therapy:
We didn’t have any sort of hard criteria. It was going to be more just a sort of global impression taking into account of all the team’s views. You know, for instance this guy . . . early on in the study I think everybody would have said if he ever makes it to the end of the study, when he gets there he shouldn’t be on a pump. But he eventually got there with using it. So the people kind of relaxed a bit more about it. But I think he was the only person potentially that we would have taken off.
Dr H
Ensuring patient safety
Finally, independently of the availability of resources to fund pump therapy, decisions around individuals’ continuation on pump therapy following the trial were primarily affected by consideration of safety issues. As ED3, who was based in a well-resourced centre, indicated:
So as a team we reviewed all the people on pumps and made the decision about whether we felt, based on the information that we had and their downloads etc. they were using the pump first of all safely, cause that’s the key priority really is, the safeness and then whether they were getting any benefit from it.
ED3
Indeed, in resource-rich centres, safety appears to have been the only reason for removing people from pump therapy at the end of the trial, unless patients requested to come off the pump:
Oh it was definitely individual, definitely. I mean if we had funding but thought that person wasn’t safe. It wouldn’t have mattered if the funding was in place.
ED4
In summary, post-trial treatment decisions in all centres were influenced by assessments of patient safety and efficacy plus the availability of funding for pump therapy. Access to resources ultimately dictated the decision-making strategy that was adopted by the different centres; in resource-limited centres individual treatment decisions were NICE/SIGN-guideline driven and based on strict, objective efficacy criteria, whereas in resource-rich centres, decisions about individuals were based on looser, subjective views of efficacy or patient benefit and a desire to safeguard an ongoing therapeutic relationship.
Research question 3
Strategies used in REPOSE
As already outlined above (see Research question 1), staff had developed some strategies either proactively or during REPOSE to manage and prepare patients for potential withdrawal of the pump. In the main, staff saw these strategies as having been helpful, effective and appropriate, and said they would use them (and recommend them to others for use) in future trials of a similar nature to REPOSE. Such strategies included preparing patients for closeout by reminding them, at the outset, that pump therapy was only funded for the duration of the trial:
. . . it’s about the expectations those people had from the start. And I do think that if you’re very clear from the outset, if people’s expectations are at a certain level, then those conversions (post trial) are much easier. But it’s about being very clear from the outset.
ED14
In addition, staff recommended that patients be given reassurance that they will be monitored to determine whether or not they needed a pump in the future so that a case could be made for them to access one; separating (ideally in space and time) the clinical appointment to discuss post-trial treatment from the final trial appointment and, if possible [see What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?] giving patients a window of time after closeout to adapt before therapy was removed:
You know maybe there should have been a wash-out period or something afterwards, you know like this is the end of the trial maybe you’ll have 2 or 3 months or something to discuss with your team the way forward or whatever rather than people thinking right on the day it finishes and that’s it, it’s very difficult to just whip something off somebody and say here you are go back to your pen so I think that might have been the only thing, and that’s just feedback from the patients really.
ED6
Staff also identified two general areas in which they felt that their practice could have been improved and which could help support patients and staff involved in closing out future trials involving potential withdrawal of treatment. These were formalising post-trial procedures and improving communication between/within teams and with patients.
Formal post-trial procedures
Staff at a number of centres commented that the post-trial period is relatively neglected in trial planning compared with trial set-up and delivery. In light of their experiences of working on REPOSE, these staff members highlighted a need to acknowledge and prepare for the ending of a trial from the outset:
Maybe if I’d been in a trial where something had been taken away, we would have formalised this a bit more . . . But it’s difficult to envisage that when you’re writing a protocol so far in advance isn’t it? At that point the major thing is: can we get enough people into the study. That’s always the major hurdle. And in hindsight we probably ought to have sorted out the closeout in more detail once we were up and running. And set aside time to actually do that. With amendments or whatever it needed.
Dr B
There was widespread acknowledgement that thinking about closeout in advance and adopting a more detailed or formalised set of procedures for decision-making about ending/continuing trial therapy would have been helpful for staff managing this process. Some staff commented that appropriate costings/resources would also be required for this and to ensure staff had dedicated time to manage the closeout effectively rather than trying to fit it into already busy work schedules:
And that we had enough time for it – I think one of the issues as well is, because we’re such a busy clinical team and this was kind of fitted in as part of our clinical work as well. Although there was some backfill and things it was still a very busy time for us. So making sure that we did have the time and it was given to that
ED13
Staff in resource-limited centres, in particular, highlighted a need to develop a more formalised process with regard to decision-making about post-trial therapy, suggesting that this would enable staff to support each other when making difficult treatment allocation decisions and communicating them to patients:
But the final decision [post-trial treatment] was made by different people. In hindsight potentially I think all of the educators and the PI should have been probably together for all of them. And discussed them . . . I would have definitely met and gone through the SOP and gone through everything and checked that everyone was clear with what we were doing. And probably together supported each other and probably have continual meetings with those people particularly involved in the trial.
ED10
Local guidelines
As indicated earlier, staff also thought that having local guidelines in place was important to avoid inconsistent practices within centres and to help promote parity in decision-making, fair allocation of scarce resources (pumps) and also to help prevent potential disagreements and tensions within the team.
Staff in resource-rich centres also suggested that having more formalised procedures at the end of the trial could be useful and result in more transparent and accountable post-trial treatment decisions:
Would I have put something in at the end to kind of reassess, to kind of see whether or not there was, was it right to allow a participant who’d been given a pump to remain on a pump . . . So perhaps something that perhaps brought a bit more structure into that . . . But that perhaps would have been one thing to kind of do a fuller or a more structured assessment about whether or not it was the right thing to keep them on a pump.
ED8
The staff speculated that adopting more formalised procedures for closing out patients would ensure that staff with the requisite skills were available following closeout to train patients to use different technologies, if required, as well as provide emotional support:
When I saw the patients my team weren’t there. So, and that was a technical problem, because I had to teach people how to use MDI and how to use the bolus calculators . . . I struggled a bit, ’cause I’m not a trained educator . . . And it just wasn’t done properly. And that’s entirely my fault, because we didn’t set it up to do it. We didn’t think it through I don’t think. And they were the last patients. So we didn’t get the chance to improve it.
Dr C
Finally, some staff also indicated an explicit need for training/role play to deal with patients’ emotional reactions at closeout and suggested that this training could be usefully incorporated into the costings and design of future trials:
And I guess maybe yeah, just kind of sort of, kind of how to deal maybe with – if people are being – if something’s being withdrawn from the person as well, in terms of a sort of a therapy, how to kind of sort of handle that as well, and what kind of the, so maybe a little bit of training about kind of the best way to kind of present that to people.
ED13
Improving communication
Finally, nearly all of the interviewees talked about the need for better communication about closeout at the end of the trial. As Dr C said, ‘Most of our problems came from breakdowns in communication, I think’. First, many staff noted that better communication across trial centres would have been helpful in the REPOSE Trial, as this would have enabled staff in the different centres to prepare for, and alert others to, patients’ reactions and develop more consistent protocols and/or guidelines for good practice:
I think the only thing I might have done better is we might have had more of a discussion about the scenarios and shared the experiences so people got a more consistent message.
Dr A
Consensus and communication within centre teams was also seen as important for managing closeout effectively. Some staff noted that there had been a communication breakdown in their centre, with the result that some team members were not aware when pump therapy was scheduled to be withdrawn, and that this had caused problems for the staff and patients involved.
I think that was, that was, that was all not – we didn’t manage that very well, if I’m honest, because Dr did it on a day when I was on leave. And Dr didn’t – I didn’t know Dr was going to take the pumps off them there and then. So I would have, I would have liked to have seen them to have gone through their regimes on pens with them. And to have given them a bolus calculator meter, which would be like the bolus calculator on their pumps that they were used to. So we didn’t – I didn’t know they were going to walk into the consultation with a pump and leave without one.
ED12
Other staff members at this centre commented that better communication within the team would have enabled them to better support each other through the closeout process:
I think probably there should have been more of a team effort at the close. Because there was a lot of people involved in the team, but it was more or less left to you know, the educators and the dietitian. And then the consultant saw them later. But I think you know, if the whole team were involved there wouldn’t have been so much awkwardness at the close.
ED10
Finally, staff argued that closeout of these sorts of trials would be potentially easier if there was better communication with research participants. As indicated earlier, some advocated continually reminding participants that pump treatment was funded only for the trial’s duration [see What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?]. Others, who supported adopting more formalised end-of-trial processes, suggested that these could be explained to patients so they are made aware in advance of how decisions about their post-trial therapy would be made:
You need to let the patients know this, you know, you could have some fixed set criteria for whether they keep the pump or not. Or we say, at the end of the trial, you come off the pump for 3 months. And after the 3 months your diabetes control will be reviewed again to see if the pump therapy is suitable for you. So that they actually – and that might be the better way to do it – so that everybody knows they’re going to come off the pump for 3 months. And then they’ll get a review, rather than this.
ED12
As noted earlier [see What (if any) practical/ethical/other issues arose for staff and how did they attempt to address these?], staff made a related point when they argued that it might be useful to make patients aware during the trial when they were currently not reaching the criteria for post-trial treatment (pump) so that they were prepared for the possibility of their treatment being withdrawn:
If we’d had the same conversations all along: your HbA1c‘s no better. You only bolus twice a day. And you have that conversation three or four times, then the patient is going to come to the conclusion: yeah I’m not going to keep the pump. I can’t do this. They would have got to that point themselves.
Dr B
Dedicated trial clinics
Although all staff regarded communication between staff and patients as a crucial factor in facilitating trial closeout, some acknowledged that developing relationships with trial participants is difficult, particularly in the larger trial centres. To overcome this, one member of staff suggested that, in the future, trials should set up clinics for trial participants, which, they reflected, would have been helpful in the REPOSE Trial when communicating with patients, particularly when terminating pump therapy:
Whoever finished the trial with the patient and communicated to them the decision should have known the patient. Just so that it was a – it was much more of a more – the way you would clinically, . . . the person who did know the patient should have been there at the time [the end] to support the patient through the transition. I don’t think we really realised how the patient would perceive the difficulty of the transition . . . unlike other studies that I’ve done I personally didn’t feel that I was engaged with the study’s subjects, which probably is correct from the point of view of the outcomes, but it did make the ending of the study a bit more difficult.
Dr C
As Dr C further suggested, having dedicated trial clinics would result in continuity of care across the trial and thus make it easier at closeout because staff involved would be known to the patients and vice versa: ‘I think the person who is terminating the study should have been involved throughout it – that would have made all of the difference’.
Dr B similarly commented that involving educators who were known to the patients in the post-trial consultations was helpful when it came to communicating with, and managing, patients’ emotional reactions:
I think, she [educator] knew some of the patients better than I did because she’d done the course with some of them. And so she was warning which ones might be tricky. And you know, she is a good judge of character. So that was really helpful I guess. I think one of the other times we ran into a problem where it, the doctor that didn’t know the background to the patient. You know, so if you, I suppose not had the pre-warning, if we’ve not had that discussion beforehand, it would come across, or could be, come across really quite cold and so I think that was helpful.
Dr B
Role of the Clinical Trials Research Unit
Finally, one member of staff, ED11, suggested that CTRUs have a major role to play in communicating and co-ordinating information about trial closeout by offering/co-ordinating training, hosting meetings/teleconferences to allow staff to share experiences of closeout, offering examples of good closeout practices, making sure that there is adequate resourcing to do the closeout/post-trial appointments properly and reminding centres that closeout is approaching so that they can make preparations, particularly for potential negative reactions to the withdrawal of treatment:
I think everyone should have been advised to have those difficult conversation – you know had the conversation about not being able to keep the pump at the end, being advised to do that at every visit and every opportunity, so that people had their expectations managed. [And] . . . maybe a reminder that we needed to meet and discuss who was going to stay on the pumps and who wasn’t. You know, so just a reminder to say: have you had that conversation with your team? Has the patient been primed? Something like that would have been helpful.
ED11
During the trial there were opportunities for staff to share experiences during regular TMG meetings involving local PIs and some lead educators from all of the centres, as well as during regular educator teleconferences. However, these teleconferences primarily focused on issues relating to trial delivery. Although closeout was discussed in advance in both types of teleconference, the educator teleconferences were stopped just before the start of the closeout period, as it was thought that, by this point, further meetings would not be necessary (a decision that may have been partly due to a lack of awareness of the problems which would arise for some staff at closeout). Hence, there were limited opportunities for some staff members at the different centres to share and discuss the difficulties they went on to encounter when withdrawing treatment.
In summary, although most staff did not regard themselves as having needed support for the closeout of the REPOSE Trial, they outlined a number of ideas that they felt would facilitate closeout in trials when treatments are withdrawn. In addition to the strategies they had already developed during REPOSE, staff suggested that more formalised procedures for ending trials should be adopted; specifically, procedures for post-trial treatment decision-making and for transitioning research participants back into clinical care. Second, they advocated for improved communication among trial staff both between and within centres and with patients.
Key findings
This study has highlighted and explored differences in staff members’ experiences of closeout, both within and across the REPOSE centres. In most of the centres with limited funding for pump therapy, all but a few patients were reverted to MDI at the end of the trial. In some such cases, staff had had to manage patients’ negative emotional reactions to the withdrawal of pump therapy. In other centres at which funding for pumps was more readily available, all patients who were safely using the pump, who wished to continue using it and who were benefiting, as broadly defined from pump therapy, were allowed to continue this treatment following trial closeout. As patients in these centres were able to remain on the pump if they wanted to, closeout in these centres was perceived as less challenging.
Most staff, but particularly those involved in the pilot phase, anticipated that the withdrawal of pump therapy might be an issue for patients, and hence had developed a range of potential solutions to address this. These included developing strict protocols for managing patients’ expectations and pre-empting potential disappointment/anger by reminding patients that pump therapy was a research intervention that may terminate at the end of trial whenever they attended trial appointments.
All centres developed site-specific procedures for decision-making about post-trial treatment, although some were more formalised than others. These decisions were influenced by assessments of patient safety and efficacy plus the availability of funding for pump therapy. Access to resources ultimately dictated the decision-making strategy adopted by the different centres (and, in some cases, by different individuals within those centres); in resource-limited centres individual treatment decisions were NICE/SIGN-guideline driven and based on objective efficacy criteria, whereas in resource-rich centres the treatment decisions were based on more subjective views of efficacy or patient benefit and a desire to safeguard an ongoing therapeutic relationship.
Staff described a number of ethical questions and issues concerning the withdrawal of treatment, which they felt had emerged in closing out the REPOSE Trial. These included whether or not it was right to remove a therapy if patients were deriving some benefit, or perceived themselves as benefiting, from it; the fact that removal of therapy might undermine the trust and confidence in an ongoing therapeutic relationship; and the existence of inequity in funding for post-trial treatment.
Staff identified a number of things that they felt could facilitate closeout of future trials when treatments may be withdrawn. In addition to the particular strategies they had developed during the REPOSE Trial, staff suggested that more formalised procedures for ending trials should be adopted; specifically, procedures for post-trial treatment decision-making and for transitioning research participants back into clinical care. Second, they advocated for improved communication among trial staff, both between and within centres, and with patients. In addition, staff highlighted the potential value of having several team members involved in post-trial consultations, including staff who had contact with patients during the trial.
Key recommendations
Planning for closeout should begin at a trial’s inception. Closeout should be addressed in the risk assessment for the trial, and consideration given to whether or not there may be ethical and practical issues related to removing a trial treatment.
Ensuring that the necessary resources, training and protocols are in place will require that realistic costings for closeout (e.g. training for staff, making sure they have dedicated time for post-trial clinic appointments and MDTs) are included in the grant application or are negotiated with trusts during the planning stage.
Having formal closeout procedures for decision-making about post-trial treatment and transitioning patients back into clinical care/other therapies may increase accountability and transparency, and aid the communication of treatment decisions to patients.
Consensus and communication within centre teams is important for managing closeout effectively.
If closeout is staggered across/within centres then regular meetings/debriefs during the closeout period would allow staff to share and learn from each others’ experiences.
If a treatment may be withdrawn at the end of the trial, trial staff should communicate this to patients at every opportunity during the trial to prevent/pre-empt disappointment.
Information about the potential withdrawal of treatment should be included in formal trial materials (e.g. the patient information sheet, see
Appendix 8), as occurred in REPOSE, as well as informal trial communications (e.g. trial newsletters). Participants could also receive a separate (local) closeout information sheet before the end of the trial, which explains the timescales involved, the training/support provided and arrangements for future monitoring of (new) treatment. Consideration could also be given to whether or not a statement about the potential withdrawal of treatment should be included on the consent form.
Continuity of care across the trial could be encouraged; this could take the form of running dedicated clinics for trial participants or, at the very least, ensuring that staff closing out the trial are known to the patients.
Research appointments to collect trial data and clinical appointments to discuss post-trial therapy should be distinct; if possible, these should occur at different times and in different places.
Allowing patients a period of time after the trial is ended to continue on trial therapy and adjust to the idea of the withdrawal of treatment may be valuable. Funding for this period of adjustment may need to be included in the grant application.
Examples of good and bad practice at closeout could be documented and used to create scenarios for role play/training staff involved in the closeout of future trials involving potential withdrawal of treatments.
Strengths and limitations
This study had very high opt-in levels from staff, providing us with a sample size sufficient to achieve data saturation and allowing good representation of a diverse range of views. Recruitment from the seven main trial centres enabled us to identify a number of variations in practice in closing out the trial and the impact of contextual (e.g. availability of funding for pumps), as well as individual, factors (previous experience of delivering pilot, exposure/lack of exposure to patients during follow-ups, etc.) on closeout experiences and practices. This wide-ranging approach also enabled us to identify broader cross-cutting ethical issues and challenges, experienced in most/all of the centres.
There are, however, a couple of limitations that must be considered. First, in some cases there was a time lag between closeout and the interviews; hence, some of the accounts may have been subject to a recall bias. Second, the interviews required staff to reflect on what proved, for some, to be sensitive experiences. This may have impacted on staff willingness to discuss these issues in too much depth, although this was not evident in the interviews. Third, the fact that all of the staff interviewed come from a relatively small research community has affected the material that we are able to report because of our ethical mandate to safeguard confidentiality. Finally, one major limitation of this study is that we were unable to interview REPOSE patients about their closeout experiences because, as noted above, there was insufficient time available to secure ethical and other approvals and to collect the data.
In summary, the REPOSE Trial, presented an opportunity to undertake research on the experiences, views and information/support needs of staff members involved in the closeout of a trial, which potentially involved the withdrawal of trial treatments, thereby allowing us to provide data on a much neglected topic. All of the objectives of this qualitative study were achieved, and all of the original research questions answered. A peer-reviewed journal article is in press.173
