Extent, quality and impact of patient and public involvement in antimicrobial drug development research: A systematic review

Abstract Background Patient and public involvement (PPI) is increasingly recognized as bringing a range of benefits to clinical and health services research. Recent systematic reviews have identified and synthesized many benefits (eg higher recruitment rates) and some costs (eg extra time need). Much of the literature focuses on PPI in long‐term conditions rather than more acute health care in which the majority of microbiological research is undertaken. Objectives The aim was to identify the extent, quality and impact of PPI in antimicrobial drug development research. Objectives were to identify any relevant reporting of PPI in antimicrobial research; appraise the quality of reporting on PPI using recognized PPI reporting and critical appraisal tools; and extract and synthesize data on the impact of PPI. Search strategy A systematic review was undertaken with a search strategy based on four word groups (PPI, patients, antimicrobial drug development and outcomes). Eight online databases were searched. Inclusion criteria English language publication, publication between 1996 and 2016 and studies describing PPI in antimicrobial drug development research. Main results No studies were found through online searching that met the search strategy and inclusion criteria. One relevant protocol paper with a brief mention of PPI was identified through expert recommendation. Commentary papers recommending PPI were identified through website searching and expert opinion. Discussion and conclusions Despite strong policy guidance encouraging PPI at the international and national levels, and anecdotal accounts of PPI taking place, evidence for the extent, quality and impact of PPI in antimicrobial drug development research has not yet appeared in the peer‐reviewed literature.


| Background
Patient and public involvement (PPI) is increasingly recognized as bringing a range of benefits to health services research. Recent structured 1 and systematic reviews 2,3 have identified and synthesized many benefits (eg improved study design, better participant information materials, higher recruitment rates) and some costs and challenges (eg extra time need, risk of tokenism). Public involvement in research has been defined as "an active partnership between the public and researchers in the research process, rather than the use of people as 'subjects' of research." 1 Much of the literature focuses on PPI in long-term conditions (eg mental health, rheumatology) rather than more acute and laboratory settings such as those in which much microbiological research is undertaken. There is increasing concern in the field that PPI in research is often either not reported or reported inadequately. 4 Notably, the bibliography of PPI in research studies published by the UK National Institute for Health Research (NIHR) advisory group INVOLVE 5 identifies no studies with a focus on PPI in antimicrobial drug development research (ie phase I, II and III studies carried out prior to regulatory agency approval or post-marketing studies).
Over the last decade, there has been increasing interest in PPI in drug development more generally, with a number of recent calls for action. 6,7 More in-depth studies have included Duckenfield and Rangnekar's report on patient group involvement in drug development for chronic and progressive diseases, in particular two case studies on the involvement of patient groups in research on muscular sclerosis and Parkinson's disease. 8 Houyez discussed patient group involvement in drug development, with particular reference to AIDS patient organizations and activists. 9 Patient advocacy groups are identified by Smits and Boon as important actors in pharmaceutical innovation. 10 New PPI forums and initiatives have been established, such as the US Food and Drug Administration (FDA), Patient-Focused Drug Development initiative 6

and the European Patients' Academy on
Therapeutic Innovation (EUPATI) launched in 2012 as a patient-led Innovative Medicines Initiative (IMI) project. 11 EUPATI has published a useful paper of case reports on patient involvement in industry-led medicines development, 12 but these all related to drugs for long-term conditions (including HIV/AIDS) rather than to antimicrobial drug development for acute infections.
This distinction between PPI in long-term conditions and in antimicrobial drug development is potentially crucial as the opportunities for involvement and impact may vary considerably in the different settings. With many long-term conditions, there are well-established international, national and local patient groups that researchers can involve from the beginning of the research journey. In addition, clinical researchers in long-term conditions may have regular contact over a long period with relevant patients and carers, giving them ample opportunity to involve them early or at any subsequent stage of the research process. By contrast, there are few if any established patient groups for those with bacterial infections. Patients often experience acute bacterial infection as a one-off experience which is either successfully treated with antibiotics or may be fatal. If recovered, they may thus have less experiential knowledge of their condition. Moreover, such patients are unlikely to have any on-going identification as an "acute infection" patient, although they may identify with a chronic underlying condition which predisposes to infection, such as cystic fibrosis or diabetes. Thus, the opportunities for involving patients in research on certain types of infection may be limited. What was unknown at the beginning of this study was the extent to which researchers were able to overcome these barriers and successfully involve patients in antimicrobial drug development research. If and when these patients are involved in research, it was also not known whether their impact was similar or different to that reported in previous reviews on the involvement of patients with long-term conditions in the research process.
As part of IMI-funded COMBACTE-MAGNET, 13 a large programme of research to develop new antimicrobial agents with a strong commitment to developing PPI, there was a requirement to establish the existing evidence base on relevant approaches to PPI that could be built on for the programme. Therefore, this rapid systematic review was conducted to provide an evidence base for the COMBACTE-MAGNET PPI initiative, in parallel with a rapid qualitative mapping of approaches to PPI in microbiological research which will be reported elsewhere.

| Aim and research questions
The aim of the research was to systematically review the microbiol-

| Design and search strategy
The design and search strategy followed expert guidance 14 adapted the approach successfully used by one of the authors in a recent systematic review of PPI within surgical research 15 Four relevant word groups were identified to capture terms relating to PPI, patients, antimicrobial drug development research and outcomes ( Table 1). The search strategy sought to reflect the diversity in international terminology relating both to PPI (eg involvement, engagement, participation) and to patients (eg citizens, public, users), but we recognized that at the time there was no consensus on appropriate terminology either in research or in the PPI field more generally. The outcome column relates to impacts on the research rather than to the ultimate health outcomes for patients which are unlikely to be identifiable in PPI research.

| Ethical considerations
Although formal university ethical review was not required as this study did not involve collecting data from human participants, consideration was given as to whether it raised any significant ethical issues.
The subject matter (public involvement in research) was not a sensitive issue, and authors of studies reporting microbiological research were unlikely to be vulnerable. PPI in research does not require review and approval, but ethical issues may arise, so ethical considerations as a data extraction field were included.

| Quality assessment
Previous reviews have commented on the difficulty of assessing the quality of articles reporting PPI in research. 1,2 Only a relatively small number of papers report primary research on PPI; more usually, PPI is reported by researchers in relation to studies where the substantive focus is on a health issue, and reporting on the PPI in the primary study is a secondary or subsequent concern. Given that the extensive INVOLVE bibliography and the references included in previous structured and systematic reviews on PPI do not include any studies of PPI in microbiological research, 1-3 it was deemed unlikely that there would be many eligible articles identified in this review, and that it would be unwise to set too demanding a quality threshold.
We therefore followed the approach in Brett et al. 2 and planned to include any study with a clear statement of aims, methods and reported results. It was agreed that eligible studies would be quality appraised for PPI reporting according to an adapted version of the GRIPP checklist 4 and the critical appraisal guidelines developed by Wright et al. 16 . All included studies were to be independently assessed by EB and DE, and any cases of disagreement were to be resolved through discussion.

| Data extraction
Data were to be extracted under the following categories: publication details (author, year, title, journal, volume, number, pages); study details (aims, design, ethics, participants, results); PPI (aims, conceptualization/terminology, number involved, ethical considerations, methods, results, impacts, strengths and limitations identified by authors); and strengths and limitations identified/other comments by reviewer.

| Data synthesis
We anticipated that most or all of any reports of PPI in microbiological research would be reflective sections within main study reports,

| Patient and public involvement
As part of the wider literature on the potential benefits and impact of PPI in research, there are an increasing number of published accounts of PPI in systematic review research. 18,19 In principle, it would have been preferable to have had PPI in this review from the beginning. However, this rapid review was undertaken on a short timescale, while a microbiology patient panel was being established as part of our wider PPI initiative for the COMBACTE-MAGNET research programme. A pragmatic approach was therefore adopted. The key finding of this review is that there were no papers identified which focused on PPI in antimicrobial drug development research and only one in which there was a brief discussion of PPI, secondary to reporting on the main study. Moreover, this one paper was a protocol rather than a report of findings, and it was concerned with a new use of the drug rather than reporting pre-authorization or post-marketing Due to the brevity of the PPI section in our one identified paper, we also followed up with queries to the lead author of the paper. He reported that eight to ten members of the SURF group were involved in initial discussions of the design of the trial, and one PPI member was recruited to the trial steering committee. She wrote the PPI section of the grant application and has advised and helped throughout the trial.
Her role included writing the participant information sheets and when the study experienced some issues around consent; she wrote and undertook a questionnaire survey and a small number of interviews with participants. She has presented with the lead author at a PPI trials conference and will be drafting a specific report on PPI and the consent issues as well as the PPI section of the final report to NIHR (Thwaites, personal communication).

| DISCUSSION
It might be considered that finding only one relevant study briefly reporting PPI in antimicrobial drug development research is a negative and therefore a disappointing finding; however, negative findings of systematic reviews are important to publish as the failure to can lead to serious bias or omissions in the literature. 22  However, as we also searched full texts, it is very unlikely that any papers with a substantive focus on PPI have been missed. There is also the possibility that the search terms used were insufficient to identify relevant literature, given the documented inconsistencies in PPI terminology and reporting. 4,16 For example, our search did not include the term "consumer," a term identified in a recent article describing the development and testing of a search filter for identifying PPI in health research. 25 However, the proposed search filter was unpublished and therefore unavailable at the time this review was conducted. Furthermore, consultations with experts in both the fields of PPI and antimicrobial research only uncovered one additional paper, suggesting that few, if any, have been missed.  26 Interestingly, the authors note that this article was produced specifically in response to a comment by Catherine Stihler MEP at a European workshop on the need for researchers to communicate the value of meaningful patient involvement in EU projects. They therefore report their experience of patient involvement, but offer up key principles for success of patient involvement in other research: "involve early, involve deeply, have patients feedback on project progress, include patients in dissemination and help patients convey their own story." 26 In addition, they provide a comprehensive list of barriers and difficulties for meaningful involvement, and suggestions from their experience on how to overcome these. Taken together, these principles and guidelines provide a template for other drug development researchers. It is also notable that the principles and guidance are very similar to other advice coming from other IMI projects 27 and more generic PPI reviews. [1][2][3]5 In discussing the systematic review findings with our microbiol- phasized the need for researchers to make panel members feel comfortable and safe to ask questions or make points; they need to feel there are no "wrong" or "silly" questions.
A recent study found pharmaceutical industry professionals to be positive but uncertain about PPI in medicines research and development. 28