Introduction
-
What barriers do research teams perceive to enhancing the recruitment of under-served groups into trials and what are the current approaches used? This was examined using a survey and a semi-structured online interview.
-
What are researcher views on the usability and value of the INCLUDE Ethnicity Framework in supporting engagement and recruitment of under-served groups into trials? Examined via completion of the INCLUDE Ethnicity Framework and semi-structured online interview.
-
Are there areas of future development and resourcing that may improve engagement and recruitment of people from under-served groups into research? Examined using a survey and a semi-structured online interview.
-
What are the optimal ways of engaging diverse patient, public and community advisory groups (often described as PPI, PPIE or PCIE groups) and effectively enabling them to engage with the INCLUDE Ethnicity Framework? Examined via a consultation process with public contributor advisory group.
Methods
-
Survey: a survey of researchers working on trials for NIHR, a UK funder
-
Qualitative interviews with trialists: INCLUDE Ethnicity Frameworks were completed by a subsample of researchers identified via the survey, and then all these researchers completed a qualitative interview to explore perceptions of the Ethnicity Framework and its acceptability
-
Consultation process with public contributor advisory group: feedback from multiple patient, public and community advisory groups across two locations on a brief guidance document for researchers regarding involving patient, public and community involvement and engagement (PCIE) contributors in the INCLUDE Ethnicity Framework
Survey
Qualitative interviews with trialists
Consultation process with public contributor advisory group
Results
Participant flow and characteristics (see Additional file 1 for the participant flow diagram)
Survey findings (see Table 1)
Question n = answered within this section | Response option | % (n) |
---|---|---|
Under-served groups that researchers identified as important for their particular trial n = 115 | Ethnic minorities | 44.3% (51) |
LGBTQ+ | 7.0% (8) | |
People with cognitive impairments | 22.6% (26) | |
Socio-economic disadvantage/low-income | 53.9% (62) | |
Male/female gender (depending on context) | 20.9% (24) | |
Age extremes (e.g. under 18 and over 75) | 40.9% (47) | |
People living in remote areas | 20.9% (24) | |
Religious minorities | 10.4% (12) | |
Other (e.g. people with physical disabilities; with complex or severe mental health needs; substance users; carers) | 26.1% (30) | |
The previous/planned involvement of members of this under-served population within the relevant trial n = 115 | Review of funding application | 40.0% (46) |
Patient and public involvement (PPI) representatives from under-served groups will be/are advisors | 76.5% (88) | |
PPI/service user researcher(s) from under-served population | 36.5% (42) | |
PPI from under-served groups to co-create intervention or other aspects of study design | 41.7% (48) | |
None | 3.5% (4) | |
Other (e.g. recruiting more participants from under-served populations into the trial) | 11.3% (13) | |
How researchers identified the under-served groups that were relevant to their trial n = 115 | Previous experience | 85.2% (98) |
Research literature | 35.7% (41) | |
Toolkit or set of guidelines (e.g. INVOLVE and PROGRESS-Plus) | 1.7% (2) | |
Other (e.g. review of the clinical epidemiology of the target illness; support groups; PPI) | 17.4% (20) | |
Approaches to increasing recruitment of under-served groups considered by trial teams n = 125 | Patient and public involvement | 81.6% (102) |
Staff training | 43.2% (54) | |
Recruiting from community organisations | 32.8% (41) | |
Cultural adaptations | 23.2% (29) | |
Use of toolkit to identify under-served groups | 8.8% (11) | |
Other (e.g. design of recruitment materials; recruiting from deprived areas) | 23.2% (29) | |
Researcher views on funders mandating recruitment and inclusion of under-served groups n = 173 | It would be difficult to have a quota for all groups | 61.8% (107) |
Having a quota is a good idea | 8.7% (15) | |
Funding would be required to increase inclusion | 45.1% (78) | |
Mandating inclusion is not relevant to all trials | 54.3% (94) | |
Other | 4.6% (8) |
Qualitative findings
Superordinate theme | Subordinate theme |
---|---|
Current barriers and strategies | Structural barriers |
Researcher barriers | |
Current strategies | |
Enhancing engagement and recruitment of under-served groups | Resources suggested |
Structural changes | |
Usability and value of the framework | Usefulness of the Framework and general impressions |
Barriers to implementing the framework | |
Usability issues with the Framework |
Current barriers and strategies (to increasing engagement and recruitment of under-served groups in trials)
Structural barriers
“So I suppose the barrier first of all is if people aren’t accessing those services, then they may not access the trial.” ID12
“So people from deprived and minority ethnic groups will struggle to get access to services in the first place. There’s quite a lot of hesitancy around medical interventions for mental health disorders themselves and we’re basically asking people to take an additional medication.” ID15
“One of our issues is that the pool we are recruiting from is already not representative and how you fix all those barriers to women’s health and accessing care.” ID11
“I think that they’ll (new, diverse sites) be hard to recruit in because they’re not as research active… We really need to persist with them, with those sites, because we’ll get very different participants. And also acknowledging that I think it will make it harder.” ID12
Researcher barriers
“It would be difficult to know in advance which ones [ethnic groups] might be more or less receptive to research, and, I mean, I don’t know which cultures those might be.” ID7
“It got me thinking about the people who we actually don’t include into the study…. It’s various factors, like not having the study documentation in different languages.” ID10
Current strategies
“This [the Ethnicity Framework] would be so helpful to have during the design phase and then we could appoint our PPI representatives more appropriately perhaps.”
“It’s mainly white middle-class at the moment…it’s a challenge to open it out, it’s easier with ethnic minority groups… but I think it’s more difficult with people from socioeconomic deprived backgrounds to get representation in the PPI group from those groups.” ID13
“Most of our PPI groups all across the country are not necessarily representative of the population.” ID1
“I really don’t want it to be an exclusively white middle class university educated panel.” ID11
“Those (questionnaires) are all standardised so it’s not like we could get them translated easily.” ID8
“We are recruiting from clinical teams with high numbers of Black and South Asian patient populations, in areas of economic deprivation.” ID15
Enhancing engagement and recruitment of under-served groups
Resources suggested
“Some central mechanism for translation and production of patient information sheets that was in multiple languages, that would have been useful.” ID7
“Whether the NIHR could provide someone to input into studies and whether there could be maybe someone that you could go to and have that point of contact.” ID10
Structural changes
“I wondered if this was something that in phase two of the grant applications you should have a very abbreviated version of the form and then actually, I think it will impact on the trial design.” ID11
“I’m surprised that NIHR, for example, when then they’re… considering funding, don’t say, look… why haven’t you included budget for things like accessible formats, large print, and audio, and things like that for your patient facing materials.” ID8
“Even if it was part of some sort of IRAS [NHS ethics] application or something as well.” ID10
“Of course the regulatory system does have a means of influence in that. So for drug trials it would be the MHRA and the HRA… And the UK Medicines Agency. So they all now could be saying actually we want to show that diversity is present, so that you can be sure that the medicines will work in the full range of population.” ID3
Usability and value of the framework
Usefulness of the Framework and general impressions
“I think it was really useful to actually complete, and it would have been really useful to have that at the beginning of every study so it does get that initial thought-provoking ideas and stuff in place.” ID10
“Overall I generally felt it was a very comprehensive questionnaire and it did what it was supposed to do, put it that way.” ID1
“It’s a very useful, sort of, checklist to go through, to make sure that you're not making any obvious blunders when you set it up. So I think I found it helpful from that point. But, as I say, I'd struggle to say it was definitely this that we changed as a result of it.” ID3
“Prompts would be good because the, the questions…I mean, the questions are very open ended.” ID8
“(If) you answered the question and then there were some actions that you wanted to take, I think that would be really helpful…. There was a bit at the end after the worksheet that was actually about what you were going to do about it, wasn’t there…and I had other meetings and I just had to kind of leave that.” ID12
“But I think to expect the researchers to have the answers to those questions before doing the trial is quite a stretch. But I think they should set up the systems before the trial so that they can answer the questions before and during the trial.” ID2
Barriers to implementing the framework
“It was very clear, and I think it’s a really good exercise if you’re at the point of designing your trial.” ID8
“If I, kind of, had this information when I was developing the documentation, it could have been a case of, actually, do we want to think of X, Y and Z to have this or…and when they were, kind of, developing the protocol for the study and obviously applying for the grant, thinking about those aspects and stuff as well, to try and include as many people as possible into the study.” ID14
Usability issues with the INCLUDE Ethnicity Framework
“I did the first part, I was like, yay…and then there were like three more parts, and I was just like, actually, I can’t…I think it’s unrealistic to think that researchers (have time).” ID12
“Almost like a list of the social graces, just to be like, you could think about class, you could think about geography, you could think about sexuality, just…or something like that.” ID5
“I found that I had to copy and paste the prompts just below each question so the way the form is designed at the moment it just has the top heading numbered with each question. I had to copy and paste the prompts just below the question to help me answer the question. Because I just found that they were very vague.” ID6
Advisory group consultation
Initial meetings (Manchester and East Midlands separate groups): key questions
When and how to involve patient, public and community contributors in the INCLUDE Ethnicity Framework?
Who should make the first contact and how should this be made?
What should researchers say or ask about clinical trials?
Meetings 2 and 3 (Manchester and East Midlands separate groups)
Why involve patient, public and community contributors in a specific trial?
Why use this INCLUDE Ethnicity Framework?
Is the fact that the trial might not get funded an issue?
Final meeting (combined Manchester and East Midlands groups)
Discussion and conclusion
Statement of principal findings
Strengths and weaknesses of the study
Meaning of the study
Role of PCIE/PPI
Key recommendations
-
Minor changes to the INCLUDE Ethnicity Framework. For example, adding more specific prompts, encouraging researchers to detail specific actions they could take, and providing additional resources (where available)
-
Use the INCLUDE Ethnicity Framework and related resources when designing clinical trials, and then reviewing during the trial (e.g. once recruitment has commenced to ensure strategies are effective)
-
More detailed and robust evidence that examines the use of different strategies to improve inclusion
-
Researchers to consider the funding and time resources required for diverse and inclusive recruitment
-
Funders to publicly signal that they support this time and resource when they are essential in the pursuit of better, more representative and generalisable research
-
Consider diversity within PCIE groups and take proactive steps to involve relevant under-served groups