There has been increased recognition of the potential benefits of patient and public involvement in research, and researchers are increasingly being asked to demonstrate patient involvement to funding bodies [1]. Suggested benefits include improved quality of the results, more direct applicability of research to patients and improved translation into clinical practice [2]. Lay involvement has also been seen as ethically mandatory [2].

However, there is also concern that current patient and public involvement is often tokenistic and aimed at ‘ticking boxes for funding applications’, rather than placing a true value on patient input and involvement [1, 2]. Domecq et al. [2] carried out a systematic review of studies which described involving patients in the conduct, design and/or dissemination of research. They found that lay people were mostly involved in agenda setting and protocol development, and it was much less common for lay people to be involved in execution or translation of the research; only a very small number of studies involved lay people in data collection. Evaluation of different approaches to involving patients and the public is required [1, 2].

The extent to which lay members can carry out these more extensive research roles has been debated. For example, Ives [3] and Martin and Finn [4] have discussed whether it is possible to give lay people the training required to conduct research without turning them into ‘professional’ researchers and losing their unique lay perspective. There has also been discussion about the best way of achieving representation in terms of lay involvement [5].

We have recent experience of involving lay people in a health services research study that has seen their roles extend beyond agenda setting and protocol development and into data collection. The study explored how inpatients can be involved in medication safety while in hospital and the potential role of electronic prescribing systems as a barrier or facilitator (the Inpatient Medication, Patient Relationships and Electronic SystemS (IMPRESS) Study [6]). The first phase of this study involved carrying out observations of doctors’ ward rounds, pharmacists’ ward visits and nurses’ drug administration rounds in two English National Health Service (NHS) hospital trusts (organisations), to explore if and how healthcare professionals facilitate inpatient involvement in medication safety and how this may differ between paper and electronic systems. The project had a patient and clinical engagement group that met quarterly and originally included three lay members and relevant healthcare professionals. At the first meeting of this group, one of the lay members expressed the view that our proposed patient and public involvement was not sufficient to truly include lay perspectives in the research. The suggestions were put forward that additional lay people should be recruited to the patient and clinical engagement group and that lay people should be involved in conducting some of the observations; these suggestions were adopted.

The aim of the present paper is to describe our experiences of lay involvement in conducting research, from both the lay observers’ and researchers’ perspectives, in order to inform the future role of lay people in carrying out health services research. We describe experiences of the preparation for the lay involvement, the lay observations themselves, the value added by involving lay observers in data collection and the challenges that arose.

The three lay members (representing both genders) and four researchers all agreed to take part and were interviewed. The duration of the interviews ranged from 10 to 45 min. Verbatim quotes are used to illustrate themes.

Benefits of including lay observations

Positive experience for lay observers and their role in research

All three lay observers described conducting the observations as an enjoyable and interesting experience. One of the lay observers also expressed the view that the lay observers had an equal relationship with the researchers. Three of the researchers identified benefits to the research that subsequently resulted from these positive experiences. Two of these researchers expressed the view that being involved in the data collection had resulted in increased lay observers’ motivation and involvement with the project. The third was of the opinion that having the lay people play such an integral part in the research may also benefit any other research projects they were involved in, as they would have an increased understanding of the research process.

The bottom line is, as I say, I really enjoyed it and I found it really interesting and quite eye opening. I think it’s a really valuable experience. (Lay observer 1)

It was at corporate level playing field there is no differences basically a group of people who want to produce some improvement in a particular service area, healthcare and who want to join forces and a very co-operative way and the relationship between the healthcare professionals and the patient representatives is really excellent. (Lay observer 3)

It has… probably contributed to how involved that they’ll be later in the project and discussing our findings and their findings … has meant that their involvement isn’t token. It’s probably increased their motivation and interest as well so that will be useful. (Researcher 2)

Different perspectives

The two researchers who were most involved in conducting the observations and working with the lay observers were of the opinion that the lay observers had greatly contributed to the research findings by bringing different perspectives to those of the researchers; all three lay observers also expressed this view. While the researchers reported being focused on discussions related to medication, the lay observers brought a wider perspective of general communication and engagement management and the impact that this may have on medication-related discussions between healthcare professionals and patients.

Because I am far less knowledgeable about medication and prescribing than the investigators I was working with, I concentrated much more on the human relationship between the inpatient, the patient in the bed and the consultant … when it was a consultant led ward round, or the pharmacist on a pharmacist ward round or the nurse/nursing student on a nurse ward round, the interaction, the body language, the relationship, the physical interaction, physical location and placement [of] the healthcare professional in relation to the patient in the bed. (Lay observer 2)

I felt the lay observers were picking up more about the overall conversation, about how the patient was involved, and not specific to medication-related issues, it was more of a general “how’s the patient being involved in decision making and in the conversation”, that kind of thing. So they definitely brought a different perspective, so it was really positive, it’s a really positive thing and it almost made me reflect how I was looking at the situation as well. (Researcher 1)

Secondary analysis of the IMPRESS data demonstrated that there were additional codes added to the coding framework specifically as a result of the lay observers’ involvement. These comprised length of consultation, consultation structure, communication style, the level of engagement of the patient and the lay perception of the patient situation. The secondary analysis also suggested that lay observers of different backgrounds brought different perspectives. For example, the lay observer who had a background in communication and customer management brought a fresh perspective on the effect that the consultation structure and communication style of the healthcare professional team might have on patient involvement in medication decisions and the lay observer who had his/her own experience as an inpatient documented observations based on empathy with the inpatients’ perspectives.

Overall, interviews and secondary analysis suggested that combining the lay and researcher perspectives together gave a richer output than either could have given alone (Fig. 1). For example, the researchers reported observing that some discussions about medication that happen on doctors’ ward rounds did not involve the patient. The lay observers noted that ward round consultations were structured so that there was a ‘technical phase’ where the healthcare professional discussed the patient, in which the patient was about ‘as involved as a coffee pot’ followed by a ‘summary phase’ to the patient (coded as consultation structure). Taking the lay and researcher perspectives together, it became clear that discussions about medicines did not involve patients when they took place in the ‘technical’ phase and were then not included in the summary to the patient.

Impact on study design

Finally, the process of making arrangements with potential lay observers also shaped the overall direction of the project. One of the potential lay observers suggested that he would particularly like to observe rounds where carers would be present, and so the research team took the decision to carry out some observations of evening drug administration rounds, which coincided with visiting hours to the ward.

This study has demonstrated that involving lay people in data collection is one way of overcoming the problem of tokenistic patient involvement in research. Both the lay observers and the researchers working closely with them were of the opinion that they had contributed and added real value to the research. This perhaps provides a way forward in terms of breaking out of the self-fulfilling cycle of tokenism identified by Snape et al. [1]. In their systematic review, Domecq et al. [2] described the reported benefits of involving the lay public in research were that it improves patient enrolment in research studies and decreases subsequent attrition and aids in dissemination. These are all important roles, but the current study has demonstrated that the lay contribution can go far beyond this to adding a different perspective to the data collected and the ensuing research findings.

However, this increased involvement also brought challenges, such as maintaining the correct balance of the insider/outsider status of lay observers. Martin and Finn [4] have suggested that there may be a tension between achieving a true partnership rather than a consultative relationship between researchers and lay people and still keeping the lay input distinct from that of the researchers. Similarly, Ives et al. [3] discussed the paradox of the contribution of lay people needing to being outsiders to the system but needing to be trained to conduct research and become insiders. Ives et al. [3] suggest that this paradox may mean that it is not possible to give lay people the training they need to conduct research without compromising their ability to bringing a fresh perspective as an outsider and that their involvement should be limited to being consultative. The current study findings suggest that lay observers were involved in data collection in true partnership with the researchers, while still bringing their unique perspective to the research. The mechanism for this is unclear, but our researchers specifically sought to give the lay observers the minimum training necessary to fulfil their role. Yet, the findings also suggest that while lay members were given training on research governance procedures, they may not have been able to follow these fully without taking on a researcher/healthcare professional perspective of research ethics. It may be possible to find a way of maintaining the right balance of keeping the unique perspective of lay observers while still ensuring appropriate research governance; this is a subject for future research.

The need to recruit a diverse and representative range of lay people has been recognised [5]. Martin [5] discussed two models of representation, one of having statistical diversity in terms of demographic characteristics and the other embodied in the ability of the lay people to represent the needs of a diverse group of people, rather than being a diverse group in itself. In this study, it was easier to establish the latter as our lay observers included an experienced former inpatient who was therefore able to represent the patient experience and a former communications manager who had experience in making sure the needs of a diverse population were met. Researchers also need to take into account that lay people who are willing and available to take part in data collection are also likely to be a limited resource,

We have also identified the need for infrastructure to be developed to support this type of patient and public involvement In terms of training. INVOLVE [8] has developed resources for many lay involvement roles, but this does not extend to the specific training needed for data collection.

Some of these challenges have previously been described with relation to patient and public involvement in research, including patient frustration with lengthy processes of training and meetings, time restraints of researchers, extra time needed to complete research and additional funding needed [2]. However, introducing a greater level of involvement, as described here, intensifies these challenges and calls for the development of systems to minimise them. While additional time and resources may have alleviated these problems to some extent, we feel that an agreed system for negotiating access is needed.

A theoretical concern of patient and public involvement discussed in the literature is ‘scope creep’, i.e. that involving patients in the research may lead to irrelevant concerns and issues being suggested which would make the research unfeasible [2]. This was not a concern raised by the researchers or lay people involved in the present study. While the lay observers had a different emphasis in their observations, this only served to strengthen the findings.

On the basis of the positive results of lay involvement in data collection, we also subsequently took the decision to involve lay members in data analysis in a later part of the IMPRESS study, and all lay members are authors on the resulting publication (submitted for publication).

Lay members of the public can add value by being involved in data collection for health services research projects by bringing their unique perspectives to the research. Maintaining the correct balance of not ‘professionalising’ lay people and giving them sufficient training to conduct research is a challenge. There is a need to address this further in future studies and to build infrastructure to support lay involvement in conducting research.