Evaluating patient and public involvement in researchBMJ 2018; 363 doi: https://doi.org/10.1136/bmj.k5147 (Published 06 December 2018) Cite this as: BMJ 2018;363:k5147
- Antoine Boivin, co-director1,
- Tessa Richards, senior editor/patient partnership2,
- Laura Forsythe, director of evaluation and analysis3,
- Alexandre Grégoire, patient partner1,
- Audrey L’Espérance, senior adviser1,
- Julia Abelson, professor4,
- Kristin L Carman, director of public and patient engagement3
- 1Center of Excellence for Partnership with Patients and the Public, University of Montreal Hospital Research Center and Faculty of Medicine, Montreal, Canada
- 2BMJ, London, UK
- 3Patient-Centered Outcomes Research Institute, Washington, DC, USA
- 4Department of Health Research Methods, Evidence, and Impact, Centre for Health Economics and Policy Analysis, McMaster University, Hamilton, ON, Canada
- Correspondence to: A Boivin
Research funders increasingly recommend and require patient and public involvement (PPI) in the design, conduct, and dissemination of health and social care research.123 In the literature and policy discourse, PPI is justified by two lines of argument: one on the basis of ethical principles, the other on the assumption that it may improve the quality, relevance, and uptake of research.4
The scientific community holds polarised views on involvement, but all are calling for stronger evidence.5 Those critical of PPI want more evidence on the costs, benefits, and risks before they undertake anything more than a tick box approach to obtaining grants. For advocates already engaging with patients and the public, evaluation is necessary to understand how best to do PPI and fully reap the benefits of working together.
Over the past 10 years, the international literature evaluating PPI has more than tripled.6789 The recent systematic review by Cocker and colleagues focusing on PPI’s effect on enrolment and retention in clinical trials (doi:10.1136/bmj.k4738)10 is an important addition. Their review suggests that PPI can improve rates of enrolment in clinical trials, thus bringing the most robust evidence to an association that has long been hypothesised.8 The finding is important because recruitment difficulties can reduce trial validity, add costs, and increase the risk of studies being abandoned and not reported.11
Although PPI significantly improved recruitment, the effect size was modest. This might reflect the fact that all forms of PPI were pooled, from simply having a patient on an advisory committee to full patient partnership in research governance, design, and peer recruitment. It also suggests that PPI alone cannot be expected to solve all recruitment problems.
A second important finding is that the effectiveness of PPI is strongest when people with lived experience of the condition being studied are involved as research partners. This supports the view of patients and the public as experience based experts who contribute knowledge that is complementary to that of scientists and professionals.12
Moving forward, PPI evaluation should be embedded in the broader evaluation of research quality, relevance, and effect. PPI evaluation should be grounded in key principles (box 1).
Key principles for PPI evaluation
The language, definitions, and goals of PPI vary between stakeholders, cultures, and countries. Clarity of goals and definitions is a precondition for evaluation. This starts with characterisation of PPI—its core features, activities, and mechanisms—and being explicit about which goals are being pursued by each person involved.13 We may not reach perfect consensus, but if we hope to support those who view involvement as an ethical imperative and those who want proof of effectiveness to proceed, we need to attend to the information needs of both.
Evaluators need to be reflective about unspoken values and power imbalance underpinning patient and public involvement and its evaluation.1415 For example, calls for evaluating the effect of PPI raise further questions about “effect on what and for whom”? While some researchers may involve patients and the public to increase study recruitment rates, patients and citizens may be more interested in ensuring that research answers questions and includes outcomes that matter to them and their communities.
Strengthening the science underpinning PPI is key to ensuring that it becomes an integral, robustly conducted, and well resourced component of research, not a last minute add on.16 Progress has been made in defining and measuring high quality PPI in different contexts,17181920 how to report it,21 and how to assess its effect.1622 International collaborations are also helping to foster involvement science and methodological rigour.23
Greater transparency, including proper scrutiny of the involvement process and support structures, is essential to interpreting whether an “absence of effect” is because the PPI is ineffective (theory failure) or because of inappropriate resources and suboptimal implementation (practice failure).24 Funders and publishers can help open the black box by requiring more detailed information about PPI in grants and publications.
We must move forward pragmatically, to ensure that evaluation efforts are not paralysed by the misguided perception that PPI is too controversial or complex to be studied. Evaluation need not be complicated to be useful. Simple feedback between patients and researchers can improve the involvement process, spur mutual learning, and change researchers’ mindsets and future practice.2526
Finally, as we move toward co-production models of research (focusing on co-leadership, mutual learning, and shared benefits between science and society), reciprocity becomes a core principle of PPI and its evaluation. Involvement has to become a truly joint enterprise where patients and citizens are empowered to act as full partners in research and its evaluation, on an equal footing with other stakeholders.27
A vast amount of public money and human capital is invested in health research. Since PPI is increasingly seen as pivotal to improving the value and relevance of research, we need to get serious about how it is done and equally serious about how it is evaluated.
We thank Susan Hrisos (international conference on evaluation of patient and public involvement in research, Newcastle University), Sophie Staniszewska (University of Warwick) and Valerie Lahaie (Centre hospitalier de l’Universitéde Montréal) for their contribution.
Competing interests: We have read and understood BMJ policy on declaration of interests and declare AB holds the Canada research chair on patient and public partnership and is assistant professor at the University of Montreal’s department of family medicine and department of management, evaluation, and health policy. AG is funded by the Quebec SPOR-SUPPORT Unit patient and public partnership strategy.
Provenance and peer review: Commissioned; not externally peer reviewed.