1 How might patient and public involvement (PPI) improve recruitment and retention in surgical trials? A qualitative study exploring the views of trial staff and PPI contributors Joanna Crocker (Lead Researcher) OxBRC Research Fellow, University of Oxford 10th May 2017 Website:
2 Project team Jennifer Bostock (Lay Partner) Louise Bowman (MRC CTSU Hub for Trials Methodology Research, University of Oxford) Richard Bulbulia (MRC CTSU Hub for Trials Methodology Research, University of Oxford) Alan Chant (Patient Partner) Jonathan Cook (University of Oxford & MRC ConDuCT-II Hub for Trials Methodology Research, University of Bristol) Nicola Farrar (University of Oxford) Jenny Hislop (University of Oxford) Louise Locock (University of Oxford) Sophie Petit-Zeman NIHR Oxford Biomedical Research Centre & Unit) Keira Pratt-Boyden (University of Oxford) Sian Rees (University of Oxford) Shaun Treweek (University of Aberdeen) Kerry Woolfall (MRC North West Hub for Trials Methodology Research, University of Liverpool) Visit study website for further information and contact details.
3 Definition of PPI Grant application Patients Trial Steering CommitteeTrial Management Group Advisory panel Consultation exercise Patients Carers Service users Public “PPI contributors” may be part of e.g… NOT researchers recruiting people to be participants in trial, or researchers disseminating information about trial to patients or public consulted by or work alongside researchers in e.g. Focus groups Surveys Interviews Choosing topic Designing Planning Doing Communicating findings Talk through definition used in this study In reality, PPI is often implemented with little planning Research process
4 PIRRIST project www.phc.ox.ac.uk/pirristPIRRIST team are developing a PPI intervention aimed at improving recruitment and retention in surgical trials – mixed methods project with 4 stages Qualitative study (stage 2) explored views and experiences of PPI, recruitment and retention in surgical trials
5 Aims of qualitative studyTo explore: 1) Needs and challenges PPI in surgical trials Recruitment & retention of participants in surgical trials 2) Views about impact of PPI on recruitment & retention 3) Ideas for PIRRIST intervention Number 2 is focus of this talk
6 Data collection & analysis31 surgical trial staff 21 PPI contributors 2 PPI coordinators 6 focus groups 7 interviews 54 total participants 6 focus groups 4 with surgical trial staff at Aberdeen, Birmingham, Bristol & Oxford universities 2 with PPI contributors at Birmingham Library Verbatim transcripts & textual data analysed thematically (3 researchers) to identify emerging themes. PPI contributors unable to attend focus groups were offered a one-to-one interview in person or by telephone. All participants as well as those unable to attend focus groups were invited to submit additional comments in writing. 11 written contributions 8
7 Results: Impact of PPI on recruitmentinforming trial design / recruitment methods assessing patients’ willingness to take part directly recruiting participants improving relevance of research question publicly endorsing trial Drawing on their experiences, participants proposed several ways in which PPI contributors could improve recruitment to trials… Informing trial design e.g: benefits and burdens for participants recruitment process (where, when, who) patient information materials Colour code: Purple = from PPI contributors Turquoise = from trial staff Black = from both PPI contributors and trial staff 11
8 Results: Impact of PPI on recruitmentRecruitment methods They [PPI contributors] had the idea of using social media as a possible avenue to approach patients because of the type of patients in the trial – they were younger and they're more inclined to use Twitter and Facebook... And with their input we started to develop entries for Facebook and to use on Twitter, and our recruitment virtually tripled as a result of using that. (Staff FG2) I would love to show you several quotes but due to limited time I have selected one to give you a flavour…
9 Results: Impact of PPI on retentionassessing burden/acceptability of follow-up methods making data collection tools more patient-friendly suggesting retention strategies communicating with participants during trial changing which/how outcomes collected challenging regulatory barriers to adopting new data collection methods Suggested ways in which PPI contributors could improve retention in trials included… Communicating with participants during trial - e.g. newsletter updates, explaining why important to stay in trial Regulatory barriers: “Now, [um] for a lot of [er] these patients it is quite difficult to get them to come back to hospitals because it's expensive and they might not be able to get there etc., so an obvious answer would be to use things like mobile apps and things, which is fine. But when you try to do that, so many hurdles are thrown up from regulatory perspectives and it becomes almost impossible to adopt new technologies, and I think it's a case of…as a trials unit we're seen as having a vested interest on making our lives easy and things. Right, whereas it's patient groups saying, 'Look, we want to participate in these trials, but we're being prevented from doing so by these ridiculous sort of rules etc., stopping us to adopt simple, easy mechanisms.' […]…[…]And the patients get it but you can't get the people in charge of trials to get it …. PPI is not being exploited enough in that area.” (staff) 12
10 Results: Impact of PPI on retentionData collection tools One of the tasks that I had to do was actually to review the patient diary… and say, you know, "Does this make sense to the average patient?“ […] It was a big task but then in some cases there was no consistency from one page to another, you know, and it wasn’t severe but, you know there were gaps and what I found particularly refreshing is when I went back with my comments, and the changes were made. (PPI FG6)
11 Results: Impact of PPI on recruitment/retentionPPI unhelpful / less helpful if: Involved too late Different from trial target population Literacy level too high (PILs) Bigger barriers are operating Involved too late - e.g. at PIL stage Different from target population - e.g. condition, background) Bigger barriers are operating e.g: - Clinicians acting as gate-keepers (recruitment) - Patients already reluctant to attend follow-up clinics (retention) 14
12 PPI unhelpful/less helpful if…Literacy level too high [PPI contributors] are very good and they're commenting on… all aspects of the trial and, you know. But perhaps when you send the patient the information sheet to look at, they might… be reading it at a higher level than a lot of the people you might be contacting. (Staff FG4) There were several discussions about PPI contributors not being representative of the target population – being more literate, more intelligent, enthusiastic about research, not having lived experience of the condition being study, etc.
13 PPI can have a negative impact…One trial we paid for translations of information sheets which, well I thought was good. But we also paid for recordings of the translations as well which we were told by a PPI person, and by one of the research nurses involved in the trials management group as well, was going to be a very good thing, and… it's the costliest PPI thing we've ever done… All the recordings are available online, so we know how much they're used, and we can also compare when, you know translations were distributed or recordings were distributed and see if there's any change in recruitment, and there wasn’t. (Staff FG2) There were 2 examples of PPI having a negative impact in terms of being a waste of time and money (not reducing the number of participants recruited). In both of these cases, only one PPI contributor was mentioned. This perhaps highlights the importance of seeking more than one patient perspective before spending large amounts of money or time on a PPI initiative?
14 Conclusions Potential for PPI to improve recruitment & retention in various ways Potential for PPI to be unhelpful or harmful Plan carefully in advance Involve as early as possible Include patients with target condition Seek diverse perspectives Diverse perspectives – reduce likelihood of doing something unhelpful or harmful, and increases “representativeness” of target population
15 Some final words of caution…I see [PPI] more as improving the research process, to be honest, rather than just trying to get people into trials… Because, from my perspective, there’s no point in trying to get people into trials that are not going to be worthwhile. (PPI I50) From one PPI contributor
16 Email: [email protected] Website: www.phc.ox.ac.uk/pirristThank you All participants for their time and generosity Project team Collaborators and advisors (expertise in PPI, trial methodology and surgical research) Funders: NIHR Oxford Biomedical Research Centre & Network of MRC Hubs for Trials Methodology Research Visit study website for further information and contact details. Website:
17 Developing the PPI interventionStage 0: What is known? (Systematic review) Stage 1: PPI current practice in UK surgical trials (Online survey) Stage 2: PPI & recruitment/retention challenges and needs; explore possible components of intervention (Focus groups) Stage 4: Selecting an intervention and evaluation strategy (Consensus workshop) Stage 3: Rating possible components of a PPI intervention and barriers to recruitment & retention (Online survey) Systematic review of impact of PPI on recruitment and retention in clinical trials Preliminary results presented as a poster (number P57). Stage 1 online survey – complete and will present some of the key findings today Stage 2 focus groups with key stakeholders - underway
18 Recruitment methods Stage 1 survey participantsAdverts ( & Twitter) distributed to and cascaded by regional/national PPI and surgical networks/groups Article published in RCS Bulletin Professional contacts
19 Data collection 6 focus groups:4 with surgical trial staff at Aberdeen, Birmingham, Bristol & Oxford universities 15 trial managers/coordinators, 7 investigators, 7 research nurses, 1 clinical trial administrator and 1 research associate 2 with PPI contributors at Birmingham Library 7 interviews with PPI contributors (2 face-to-face, 5 telephone) 11 written contributions (4 trial staff, 7 PPI contributors) PPI contributors unable to attend focus groups were offered a one-to-one interview in person or by telephone. All participants as well as those unable to attend focus groups were invited to submit additional comments in writing. Focus groups all had between 6 and 9 participants PPI contributors who took part in focus groups: N=14 Written contributions: 2 of the trial staff had taken part in focus groups; 2 had not. All 7 PPI contributors took part in focus groups. 8
20 Data Analysis: Coded deductively by 3 researchersTranscript checked & anonymised Coded deductively by 3 researchers Focus group transcribed (verbatim) Coding framework agreed Coding reports further coded inductively by 3 researchers Agreed codes transferred to Nvivo
21 Directly recruiting participants Participant information sheets:Results: Impact of PPI on recruitment Directly recruiting participants Clinicians are wonderful about doing the research but they're not all of them wonderful about communicating with… patients. So, recruit somebody into the trial, into the team rather, to say, "Look, I can sell this idea to patients," and let them do the recruiting and use the same skills that that person has got to keep them on-board. (PPI FG5) Participant information sheets: They [PPI contributors] did actually pick up a mistake … which nobody else had done considering it had been through REC and peer review.. The facts should have been right (Staff FG1) While some PPI contributors supported the idea of directly recruiting participants, one felt that the idea of employing someone to ‘sell’ the trial to patients was unethical. Other examples: Staff FG2: they said that, 'It was a waste of time,' quote, to have [um] patient information sheets to be given to children under the age of eight, which is what our ethical regulations require us to do. [um] So, at that time we…we then took that back to ethics and it was accepted, and that was removed for every single children's trial – bear in mind we had a hundred plus going PPI FG5: So, when it comes to getting people engaged into taking part in a clinical trial, what I think needs to…attention needs to be paid to, is the language and the personal skills to the…that somebody can sell them the idea. Clinicians are wonderful about doing the research but they're not all of them wonderful about communicating either with their juniors, or other clinical staff, or patients. So, recruit somebody into the trial, into the team rather, to say, "Look, I can sell this idea to patients," and let them do the recruiting and use the same skills that that person has got to keep them on-board. PPI FG: it's about having awareness of the culture of the people you're trying to recruit and the area they come from etc. etc., not only just age and other demographics but the way people perceive things, because automatically we [Indian communities in City 14] think, 'what's genetics about? Will they be taking blood samples and will they be coming and cutting our bodies all this sort of stuff?'
22 Results: Impact of PPI on retentionRetention strategies they [patient focus group] suggested that certain strategies you might want to employ [er] to encourage patients to continue to stay in the study. So, the use of, you know financial payments or putting people into a lottery or shortening the questionnaire, or because it's a young population, you know collecting the data online, when we were doing everything through the post. (Staff FG2) Assessing acceptability of follow-up methods “I think for this trial we are asking too much… I think a PPI group would have said, ‘It’s unrealistic’… we could show them some questions and say ‘Would you mind filling those in? Just imagine yourself in this situation. When are you going to lose interest?” (Staff FG1) FG1: In this example, a trial staff member speculated on how PPI might have improved the follow up of one of their trials after they lost 60% of their participants,
23 PPI unhelpful/less helpful if…Different from target population Just because you’ve got PPI doesn’t mean to say that the end result is going to be, you know, people are going to want to join your study or things. Because, obviously, they’re self-selected, enthusiastic people. And that’s not the population you’re recruiting from usually. (PPI I50) Involved too late So this entire [computer] programme somebody spent ages making it [device] all pretty to encourage a stroke patient to be able to move both hands to make a cup of tea. And the patients just turned round and says, "Everyone can make a cup of tea one handed, try doing something like buttering toast," and the guy was just [clicks fingers] deflated like that. I thought, 'You should have asked patients a year ago.' (Staff FG4)
24 PPI can have a negative impact[PPI] generated a huge amount of workload, because there were so many patients who thought they were eligible, phoning up and ing and contacting, who then had to be screened and weren’t eligible. (Staff FG3) In this example, the PPI was having a patient tell their story about the intervention and trial process in a public newspaper article.