CIs spoke of how they had learnt as the trial went along, revealing that their

‘practice had evolved’ (CI 14) and their skills had “changed beyond recognition […] now we’re much better equipped […] but at the time when [trial] started we had very little idea at all about what PPI involved or how it would help or how it would work” (CI 2). In light of these challenges, CIs spoke of how in Calcitriol proliferation future they would involve more than one PPI contributor, in particular by using focus groups or panels of contributors rather than individual contributors, enlist the help of relevant charities, and conduct surveys or use social media when there was a ‘burning question’ (CI 55). Use of responsive PPI rather than individual contributors was described as

‘gold standard’ PPI (CI 14), as this avoided “the danger of having a single opinion” (CI 76), provided structure for all parties, and helped to enhance the confidence of individual contributors. I would certainly have more involvement and some kind of framework around it […] a small user group and set boundaries […] try to agree how often we should meet and what peoples’ roles and responsibilities are […] and provided more structure […] to make them feel that their views are important, and their involvement is very important, I think that would go a long way to easing the process. (CI 41) Many CIs indicated that they would extend PPI in future by asking contributors to lead in the dissemination of findings to relevant groups, help in the development of research questions, study design, and involve PPI contributors as co-investigators. CIs placed particular emphasis on how ‘crucial’ it was to have ‘early input’ (CI 14): The most useful things are […] the design stage […] RCTs you’ve got to plan ahead [...] after the development phase you shouldn’t really be changing anything […] it is during that development phase when decisions are being made. (CI 115) Early engagement and appreciation that their input into the

question is really important […] with retrospect and for the future studies […] more involvement at the front end, less in the middle and more at the end. (CI 2) Finally, CIs reflected on the importance of ‘thinking through’ plans and being clear Brefeldin_A about whether, what and why PPI is needed for individual trials: Be clear about the link between particular methods [of PPI] and particular benefits and challenges […] it’s not all the same, there are so many ways of doing it but you have to have good reasons for choosing how to do it. (CI 20) “I don’t think it should be automatic that there must be PPI involvement in every study, and different types of involvement are necessary for different parts of study. Having a core group is not necessarily the right thing because at different points there are different types of people and types of involvement that would be useful.