Introduction
This paper draws from my research on service user involvement and older people and discusses how by involving older people, knowledge creation becomes a personal and social process. I argue that by learning from, and understanding the worlds of others, we can change the way researchers think about what they are doing and thereby create better quality research. I am indebted to a number of people who have contributed to my thinking through our work together. Our common interest has been in developing a conceptual approach and evidence-base for understanding of the values, approaches and methods of involving patients and the public in nursing and health services research. We have been fortunate to work throughout with Sally Brearley, who as a lay representative and applicant on most of the studies has provided fresh insights into involvement and by her perceptive questions challenges us to think differently.

User involvement broadly encompasses different approaches to direct participation or representation (Beresford 2003, Smith et al 2008) – either as a one off occurrence or an ongoing relationship. The development and maintenance of this relationship of involvement can vary considerably; see for example Arnstein’s (1969) levels of involvement through consultation, collaboration and control.

In this paper I am interested in the personal and interpersonal relationships between researchers and service users and what that means for changing how we think about research and the politics of knowledge creation. I think of myself as first and foremost a nurse and then an academic and now a manager. It is my experience as a district nurse that motivated me to ask questions and do research. I therefore want to start with looking over my shoulder at my PhD study, which I started part-time while working as a district nurse.

My PhD focused on how doctors and nurses collaborate on medication management for older people. It evaluated a patient-held drug record in terms of how it improved information and doctor/nurse agreements about what medication their older patients were taking. It was a before and after study and it showed that, at both time points, the professionals did not report the same information about medication their patients were taking, and indeed relied on the older people themselves to tell them. This highlighted for me, what we know intuitively, that the patient knows best. Furthermore it challenges the conventional understanding of compliance, turning it on its head to consider the issue of the professionals’ non compliance rather than the patients’ (Ross 1991). This raised the whole issue, which I go on to discuss next, how we can find ways to listen more effectively to what our patients, in this case older people tell us so we can use this information to improve practice.

Learning from older people involved in a study on falls
This study was one of 11 commissioned in a basket of studies to develop the evidence around primary care in London. With considerable foresight, the commissioners required all studies to have consumer involvement and to support this they provided additional funding for training for both the researchers and the service users. We invited older people with a common interest in preventing falls to be part of a consumer panel. They worked alongside us and contributed in a number of ways. Total membership was 21, ages ranged from 60-80 and the average attendance at meetings was 13. The role of the consumer panel encompassed:

1. Advice on the interview schedule,
2. Participation in the pilot study,
3. Insight to the first level of analysis from the interview data and, using the anonymous data from the interviews, development of stories / vignettes. A vignette was then used to structure the interview with the health and social care professionals,
4. Contributing to local dissemination of the results e.g. by speaking at community meetings, and influencing strategic change on falls prevention through committee roles on the PCT.

At the end of the study we asked the consumer panel for some reflections. Not surprisingly they regarded their involvement as essential and not “tokenistic” or an optional extra. They cautioned the research team on use of language, requesting we express ourselves simply and avoiding the professional obsession with acronyms. They asked to be treated as equals and not as one participant described a “Faberge egg” (Ross et al 2005).

One of the powerful bits of learning for the research team was the important contribution the consumer panel made to the quality of the research, how relationships changed as the study progressed and the way the service users challenged our priorities. Their role in disseminating knowledge was also important, because they were readily given recognition by service partners, both the PCT and the Local Authority who were keen to access the expertise of the consumer panel to inform a user view on their falls prevention strategies.  

Involvement of service users with long term conditions in a study of organisational governance in primary care
This study was funded by NIHR SDO to explore how changing organisational governance and incentives in health care has an impact on the professional experience working with people with long term conditions (Ross et al 2009). We carried it out in 3 case study sites (PCTs) – one in London, Surrey and the South West. Here I am going to explain how we worked with service users. I am going to pick out two interconnected issues to illustrate how we have developed our understanding of, and got better at the process of involvement.

Firstly we also had a national/overarching service user advisory group and three local service user reference groups (SURGs), one drawn from each case study site). All participants had experience of a long term condition – either a non psychotic mental health problem or a physical condition e.g. stroke or COPD. At the beginning of the study we worked with the SURGs to explore their views on good and not so good care. These views were collected to ground the professional interviews in the experience of service users. Establishing these groups, ensuring engagement throughout the study and enabling opportunities for the service users to not only understand the nature of the research, but also to feel they were influencing its methods and outcomes was important. All of this required considerable personal work and relationship building.   

Secondly, there was the issue of researchers own learning about ways to achieve success in meaningfully engaging service users. As researchers were working separately in three sites, but working to a common approach, we had to find ways to train and support them. Perhaps not surprisingly we had to overcome a level of anxiety. Although they were all highly skilled and experienced researchers with a social science background they expressed a lack of confidence about entering into new, somewhat uncharted territory of building relationships with service users in this way. They needed support and reassurance to think about forging and maintaining relationships to develop the real opportunities for learning whereby the experience of long term conditions expressed through the views of service users could be used to inform and influence the conceptual framework of the study and the focus of the interviews with managers and professionals. We addressed this by providing training facilitated by one of the co-applicants who is also an expert in representing service user views, through informal learning and sharing in team meetings and email support.

Different domains for learning
The examples described above are selective and in the interests of brevity, may appear oversimplified, but I hope they illustrate some points about the how service users may become involved and the opportunities and challenges this poses for researchers in sometimes having to rethink relationships.

The framework below brings together the different types of involvement of service users. It is derived from a review of the literature and evidence of service user involvement in nursing, health visiting and midwifery research (Smith 2006) and attempts to conceptualize the different stages at which service users may be involved in research. It illustrates the multiple domains for involvement, either as part of researcher-led or user-led research or part of a partnership approach. However, it is constructed in terms of requirements of research, rather than thinking about what service users may gain from participating, also it is not able in this form to describe the personal and interpersonal relationships that need to develop between professional researchers and service users. For that we need something more sophisticated and dynamic, which is a challenge for future research. 

There is no doubt that user involvement can be a challenge for professionals. It takes time, resources and different sorts of skills. Sometimes it means changing plans, modifying research designs and shifting priorities. Success, I believe comes from attention to “personal” work, meaning researchers responding to the small things that matter to service users for example follow up after meetings, making individualised arrangements for people to attend meetings, identifying and responding to the preferred way of receiving information etc. 

In conclusion then what should we doing as researchers? I believe if we are to be able to use research to change practice then shifting the way we think about research, the questions we ask and the way we go about it has to be grounded in the authentic experience of service users. To shift the balance of power in research we need to give some thought to the types of relationships we have and how we develop them with key participants, whether they are working in collaborative relationships or the more traditional researcher/participant model. In this context, I believe we can improve the quality of what we do by being more open in our relationships with service users and listening to their challenges, reflections, sense checking and critique.

Acknowledgements
This paper is based on a keynote given at the British Society of Gerontology conference in July 2010. I am grateful to many colleagues who have worked with me, in particular Elizabeth Morrow nee Smith, Sheila Donovan, Sally Brearley, Sara Christian and Ursula Hawgood.

References
Arnstein A. (1969) A ladder of citizen participation. Journal of the American Institute of Planners (AIP Journal), 35(4) 216-24.

Beresford P. (2003) User involvement in research: exploring the challenges. Nursing Times Research, 8(1) 36-46.

Ross F.M. (1991) Patient compliance – whose responsibility? Social Science and Medicine, 32(1) 89-94.

Ross F. Donovan S.,  Victor C., Brearley S., Cottee M., Clarke E., Crowther P. (2005) Involving older people in research: methodological issues. Health and Social Care in the Community, 13(3) 268-275.

Ross F. et al (2009) The professional experience of governance and incentives: meeting the needs of individuals with complex conditions in primary care. London: NIHR SDO (available online: http://www.library.nhs.uk/Improvement/ViewResource.aspx?resID=343611).

Smith E. (2006) Involving service users in nursing, midwifery and health visiting research. Briefing Paper. London: NIHR SDO (available online: http://www.sdo.nihr.ac.uk/projdetails.php?ref=08-1305-069).

Smith E. et al (2006). User involvement in the design and undertaking of nursing, midwifery and health visiting research: a review of evidence and practice. International Journal of Nursing Studies, 45(2), 298-315.