Paper summarized: Elberse JE, Pittens CA, de Cock Buning T, Broerse JE. Patient involvement in a scientific advisory process: setting the research agenda for medical products. Health Policy. 2012 Oct;107(2-3):231-42. doi: 10.1016/j.healthpol.2012.05.014. Epub 2012 Jun 25. PubMed PMID: 22739128.

Summary of the Paper by Jill Pooler. Thanks to Carina A.C.M. Pittens and Janneke Elberse for their inpunt on clarifying some of the details.

Introduction

The authors argue that the involvement of patients in health sector decision-making is a growing phenomenon (Ref). It is believed that involving patients in decision-making results better acceptance (Ref). Scientific advisory bodies are also engaging with this trend (Ref). However there is concern that patient involvement may result in loss of scientific authority as patient contributions are considered to be subjective (Ref). In addition to this there are perceived challenges when scope of patient involvement is restricted by topic (Ref). This paper seeks to address these issues in relation to how researchers can involve patients in setting the agenda for health research. In particular it examines how the needs of a broad range of patients can be articulated, in such a way that they can be used in a scientific process. This is accomplished by describing a case study whereby patient groups were involved in setting a research agenda for the development of medical products (drugs, medical devices related to diagnosis and care, and tissue replacement products).

The case concerns a request from the Dutch Minister of Health to the Health Council of the Netherlands (GR) for a research agenda for medical products. The GR comprised 12 members of which only one was a patient representative. Only having one patient representative on their panel was considered too arbitrary and not representative. Therefor a research team was tasked with consulting patients from a wide range of disease domains (15 disease domains). To this end, twenty nine patient groups across fifteen disease domains were recruited to contribute to this advisory process. Taking the position that participation is a dialogic process, the Dialogue Model (DM) for patient participation was used to structure the process of collaboration between groups. The underlying principle is that all participants in the groups have a ‘unique and relevant’ perspective and that dialogue will result in shared understanding and agendas.

Method

The DM comprises four phases:

1. Exploration, during which disease specific patient groups were contacted and interviewed to assess i) their willingness to participate ii) the required social conditions for participation to occur iii) and their ability to invite 15 patients to take part (there was no inclusion or exclusion criteria). In addition a literature review was conducted.
2. Consultation and prioritisation, during which a distinction was made between the six disease domains with an already identified research agenda and the nine without. Of the six disease domains with a research agenda, 23 in-depth semi-structured interviews were conducted. . The focus of the interview was to explore in more detail the data in the already established research agenda. The interviewees were selected because they participated in setting the research agenda or because the collaborating patient organisation(s) felt that the person could add new information to the existing agenda, relevant for future medical products. For the nine disease domains without a research agenda 15 focus groups were conducted. Some (patient) organisations sent an email to their entire database, others selected patients which were also patient experts because the organisation expected that these persons could represent a larger group of patients. One or two members of the GR committee and staff also attended the focus groups.In addition, in the disease domains of dementia and cerebro vascular accident (CVA) a focus group of informal carers was conducted. Four exercises were used to assist patients identify needs regarding medical products (as defined earlier). These included asking ‘what?’ and ‘why’ questions and the rationale for decisions explored. Some patient organisations requested some complimentary interviews. If this phase did not yield three top priorities agreed by the participants, the project team analysed the data and proposed three priorities which were then sent to each patient for their views. The priorities were not ranked. 
3. Integration, the aim of which was to analyse the data and integrate the priorities identified. A report of the findings of the consultation exercise was compiled for the GR. 
4. Follow-up, which involved gaining some insight into the usefulness of the process to the GR by interviewing key members; patient experience derived from informal conversations and feedback from them, and the extent to which patient views were incorporated into the GR’s recommendation. 

Complimentary to theses phases, six key principles were taken into account: (1) active engagement,

(2) creation of good social conditions,

(3) respect for experiential knowledge,

(4) mutual learning,

(5) emergent and flexible research design, and (6) neutral process facilitation.

Data collection is summarised in Table 1:

Table 1. Consultation overview

The ones indicated with an* are considered patient representatives 

In total 211 respondents were consulted through interviews and focus groups.

Analysis

Interviews and focus groups were audio-taped and transcribed. Two researchers coded the transcripts separately. The codes were discussed by the research team who developed a coding structure to analyse the data which reflected the topics relevant for answering the study questions. These were described as:
Analysis sought to answer the following questions:

1. how did patients articulate their needs during the process,
2. what medical products did patients prioritise, and
3. what effects did the three criteria of the GR committee (specified focus, highly specified, limited to three priorities) have on the process?

Results
The results of the exploration phase revealed that patients or their representatives were willing to be involved in the exercise, with just one patient group withdrawing as they did not consider the topic to be relevant. In addition it was problematic for patients to stay focused on needs relating to medical products with discussions drifting more towards societal issues around the disease.
The results of the consultation phase suggests that those groups with an already established research agenda struggled to identify or prioritise needs which could lead to medical product development without reference to the already established research agenda as a prompt. Those groups without an already established research agenda also leaned towards discussion about problems not amenable to the development of medical products and needed the discussion to be steered back to it. Over the course of the focus group exercises whereby patients were asked about the medical products they were already aware of in relation to the disease, they began to suggest ideas for product improvement. However it was problematic prioritising future development of medical products.
The result of phase 3 found that it was problematic categorising the prioritised medical products into drugs, medical devices related to diagnosis or care, and tissue replacement products because it de-contextualised the need. Thus data were re-categorised to five research directions indicating the purpose of the medical product: i) tailor-made care, ii) general improvement of therapy, iii) self-management, iv) early and correct diagnoses, and v) delay of the progression of the disease and/or recovery.
There was little overlap amongst the disease specific domains as medical product development was disease specific. However there was overlap in the research directions.
The result of the follow-up phase showed that the GR valued the exercise and reported to the Minister the three highest ranked medical products per disease domains as articulated by the patients.
Limitations of the study included the representativeness of patient participants/representatives of the disease domain. In addition there was a dilemma in maintaining the focus of the patients on the development of medical products and allowing sufficient scope for them to contribute their experiential knowledge.
The study concludes that that patient involvement in scientific advisory processes does not need to compromise scientific authority of an advisory committee. Indeed it is possible for advisory committees to determine the focus, set criteria, and keep the mandate for decision-making whilst patients can make a valuable contribution.

Other articles on this model:

Elberse J, Laan D, de Cock Buning T, Teunissen T, Broerse J, de Boer W. Patient involvement in agenda setting for respiratory research in The Netherlands. Eur Respir J. 2012 Aug;40(2):508-10. doi: 10.1183/09031936.00018812. PubMed PMID: 22855474.

Elberse JE, Caron-Flinterman JF, Broerse JE. Patient-expert partnerships in research: how to stimulate inclusion of patient perspectives. Health Expect. 2011 Sep;14(3):225-39. doi: 10.1111/j.1369-7625.2010.00647.x. Epub 2010 Dec 22. PubMed PMID: 21176013.

de Wit MP, Elberse JE, Broerse JE, Abma TA. Do not forget the professional - the value of the FIRST model for guiding the structural involvement of patients in rheumatology research. Health Expect. 2013 Jan 31. doi: 10.1111/hex.12048. [Epub ahead of print] PubMed PMID: 23363240.

Broerse JE, Zweekhorst MB, van Rensen AJ, de Haan MJ. Involving burn survivors in agenda setting on burn research: an added value? Burns. 2010 Mar;36(2):217-31. doi: 10.1016/j.burns.2009.04.004. Epub 2009 Jul 4. PubMed PMID: 19577849.