Background
Clinical practice guidelines and health technology assessments have emerged as a source of support to individuals wanting to use research evidence in their healthcare decisions. The science and methodology behind clinical practice guidelines has developed enormously over the last 20 years [
1]. Good quality guidelines are now widely recognised as useful authoritative statements of best practice across many clinical areas [
2,
3]. Guidelines are typically produced for health care providers but there is an increasing interest in developing derivative products for the public and many organisations now produce patient versions of guidelines. In the UK, for example, the National Institute for Health and Care Excellence (NICE) and the Scottish Intercollegiate Guidelines Network (SIGN) produce freely accessible patient versions. The Finnish Medical Society, Duodecim, publishes patient versions of national Current Care guidelines and a comprehensive collection of guideline-based patient information in Duodecim’s Health Library. Professional groups are also producing patient versions of their guidelines, for example, the Netherlands Association of Post-traumatic Dystrophy.
For patient versions of guidelines to be useful, members of the public and patients need to know about them (or their health professionals need to know about them), the information within them needs to be pertinent to their healthcare decisions and that information has to be understandable. This is not always the case, as our earlier literature review found [
4]. Others have also highlighted challenges, for example, that public and patients may perceive guidelines negatively as a way to ration access to medications [
5], or that research information is often general when what people want is information that is very specific to them [
6]. Indeed, even for guidelines directed at health professionals, alternative presentation methods, especially presenting information in layers, with the most important coming first, makes it easier for guideline to be used [
7]. In other words, it is important to know what your users need, and in what format.
The aim of this study was to explore what patients and the public understand about the purpose and production of clinical guidelines, and what they want from clinical guidelines to support their healthcare decisions. It complements our earlier systematic review [
4] and is part of the DECIDE project [
8].
Methods
The study involved focus groups with a heterogeneous purposive sample selected to represent a range of the likely users of patient versions of guidelines. This included individuals with selected health conditions (diabetes and depression), members of the public without experience of the topic of the guideline under discussion, a group of health communication professionals, and a group of young people (18 to 25). Diabetes and depression were selected based on the availability of voluntary organisations in the SIGN network that were able to facilitate recruitment and because they had been selected by the DECIDE project management group as being relevant to all countries within DECIDE. Nine focus groups were undertaken in total.
Participants were asked about their awareness and understanding of clinical guidelines and were presented with scenario recommendations, or draft materials from patient guidelines to prompt discussion (see Additional file
1). The questions and materials shown to the groups followed an iterative process of development and were revised for use with the different groups sampled. All focus groups took place between May and November 2012, in eight in locations across Scotland, and one in England. All were 30 to 90 min long. All focus groups were recorded and transcribed prior to analysis. Each group was facilitated by one or two researchers and hand-written notes were taken to capture relevant non-verbal communication.
The focus groups were supplemented with four semi-structured interviews with people experiencing homelessness, using draft patient guideline materials to prompt discussion. Interviews were conducted with this group because they provided a more flexible format that seemed to work better for the people experiencing homelessness. All interviews were audio recorded and transcribed prior to analysis. Focus groups were conducted by several members of the working group (MC, KG, KL, & ST).
Participants
Participants for the focus groups were recruited through the Scottish Health Research Register (SHARE) [
9], the Scottish Health Council (SHC), the Scottish Intercollegiate Guidelines Network (SIGN) network of voluntary organisations and the University of Dundee. The communication professionals were recruited with the help of Sense About Science [
10]. The communication professionals were included as part of a heterogeneous sample of the public. They were purposively selected because communication professionals provide much of the material patients and the public receive. We were interested in their views on what the public and patients (i.e. their audiences) knew about guidelines and what opinions they had on how guideline producers could make their own health communication jobs easier. The communication professionals included journalists/multimedia managers and editors from UK based voluntary organisations, newspapers, journals and broadcasting companies. They all used guidelines as the basis for some of the material they produced for the public.
A total of 62 people participated in the focus groups; each group was composed of between four and 11 participants (see Table
1). Participants were recruited from a range of geographical locations across Scotland, with the exception of the focus group with communication professionals which took place in England. This group was held in London, England because it was a more convenient location for the participants. Communication professionals in England may often access different resources and guidelines than those based in Scotland but their general requirements are likely to be similar.
Table 1
Details of focus groups
G1 | No medical condition. Recruited from Dundee University. | 7 | 7 female | 30s – 60s | Dundee, Scotland |
G2 | People with a diagnosis of diabetes. Recruited from Diabetes UK, via the SIGN network of voluntary organisations. | 7 | 4 male, 3 female | 40 - 75 | Glasgow, Scotland |
G3 | No medical condition. Recruited from the SIGN network of patient organisations. | 4 | 1 male, 3 female | 45 - 75 | Inverness, Scotland |
G4 | People with a diagnosis of depression. Recruited via the SIGN network of voluntary organisations. | 7 | 3 male, 4 female | 30 - 75 | Edinburgh, Scotland |
G5 | No medical condition Recruited through Scottish Health Research Register (SHARE) and the University of Dundee. | 7 | 6 male, 1 female | 41 – 70 | Dundee, Scotland |
G6 | Young people. Recruited by the Scottish Health Council. | 5 | 2 male, 3 female | 18-25 | Glasgow, Scotland |
G7 | Health communications professionals including representatives from the voluntary sector, media, and academic journals. | 11 | 3 male, 8 female | not recorded | London, England |
G8 | No medical condition. Recruited through Scottish Health Research Register (SHARE) and the University of Dundee. | 9 | 2 male, 7 female | 40s-60s | Dundee, Scotland |
G9 | No medical condition. Recruited through Scottish Health Research Register (SHARE) | 5 | 1 male, 4 female | 40s – 70s | Dundee, Scotland |
The four people experiencing homelessness, one woman and three men, were recruited via National Health Service (NHS) Tayside Health Board.
Ethics
Ethics approval for work done by KL and ST was given by the University of Dundee, where both worked at the time. Other aspects of this research were undertaken as part of SIGNs service evaluation. Information on the planned data collection for the DECIDE project was provided to the East of Scotland Research Ethics Service and they confirmed that it did not require full ethics committee review under the terms of the Governance Arrangement for Research Ethics Committee (GAfREC) in the UK. All participants were provided with study information, assured of confidentiality, and informed that they could withdraw consent at any time. All participants provided written consent.
Data analysis
All analysis was carried out using QSR NVivo10®. NVivo® is software that supports qualitative and mixed methods research. The data were analysed using the framework analysis technique described by Ritchie & Spencer [
11]. This method involves a number of separate but interconnected stages of rigorous and methodical qualitative analysis (Table
2 contains a description of the stages of framework analysis). It allows for the use of a priori themes linked to the study objectives, and for emergent themes to arise from the data. A number of themes were identified from the interview schedules and the previous work of DECIDE [
4], however the majority of themes emerged directly from the data.
Table 2
Steps in the framework approach [
12]
• Familiarisation – All data sources are read and sorted with reference to the study objective. |
• Identify a thematic framework – Key issues concepts and themes are identified with reference to the objectives of the study and the work of the DECIDE project to date. |
• Indexing – The thematic framework is applied to all the data sources. |
• Charting – A matrix of findings for each theme is developed, allowing the comparison of different groups. Each cell of the matrix contains a summary of each group or individuals contributions to the theme, with a reference that can be directly linked to the original text. |
• Mapping and interpretation – All charts and notes were compared to search for patterns and connections in the data. The process allows for the exploration of consensus and disagreement both within and across cases. |
Two researchers (NF & JK) examined all the transcripts independently, and met to discuss and agree an initial thematic framework. Subsequently they carried out the indexing and charting stages independently and met to review and agree all findings derived from the data. The researchers who conducted the focus groups/interviews also examined the findings of the analysis, to ensure the themes identified were an accurate reflection of the material.
Discussion
The main aim of this study was to explore what patients and the public understand about the purpose and development of clinical guidelines, and what they want from clinical guidelines to support their healthcare decisions. The individuals who contributed to the study represent a range of people who may access patient versions of guidelines, including people with long-term conditions, younger people, homeless people, and other members of the public, as well as healthcare communication professionals who use guidelines as the basis for the materials they produce for the public.
The main message from this study is that those charged with the development and implementation of clinical guidelines have a considerable amount of work to do to increase awareness in the first instance and make patient versions of guidelines acceptable and useful to patients. Five themes emerged: Access and awareness, What patients want to know, Properties of the guideline, Presenting the evidence, and Format. These overlap with the themes coming from our systematic review (with which NF and JK, who led the analysis in the current study, were not involved): Applicability of guidelines, Purpose of guidelines for patient, Purpose of guidelines for healthcare system and physician, and Properties of guideline [
4]. Again we conclude that these themes need to be incorporated into the design of patient versions of guidelines if they are to be useful to the public and patients.
The public is generally unaware of the existence of guidelines, though people are enthusiastic about them once they are made aware of them. It is unlikely that a guideline website is the first port of call for the majority of people looking for health information, meaning signposting by health professionals (or their organisations, such as the NHS) is important. The professional health communicators group mentioned the importance of making guidelines easy for charities (several members of this group worked for charities) to access and understand so that they can incorporate guideline information into their materials for the public. The relevance of this advice is easily demonstrated by a Google search of ‘diabetes treatment’, which, ignoring paid-for ads, puts the charity Diabetes UK as the #1 search result, with NICE at #17 on the second page of results. NHS Choices was #2, emphasising the importance of using brands that the public may be more familiar with and, as our study found, are already trusted.
Participants emphasised the need for information that enables them to choose between treatment options, which includes information on harms. Moreover, they would like help with this from healthcare professionals. Treatment harms will always be of concern to patients [
16,
17] and healthcare professionals need to be aware that health information they provide may be competing with anecdotal information. Making the trade-offs between harms and benefits, especially when they are close, is an area where guideline-based tools to support shared decision making could have an important role, a point made by others linked to the DECIDE project [
18]. Members of same group suggests that production of these tools can be semi-automated as demonstrated in the MAGIC guideline project for antithrombotic therapy [
19]. What members of the public do not want is to be handed a lot of material and then left to make their own decision, which is a narrow view of shared decision making [
20]. The public wants input from professionals.
Presenting risk information in patient versions of guidelines is challenging [
4]. Participants in this study often differed in their support for numerical information and graphs, and suggested that there is no one-size-fits all approach. This has led DECIDE to suggest layering of information in guidelines [
7] with the most important information (generally the recommendation) being presented first and then users can access more detailed information if they want it, down to the full evidence profile. Layering is possible even within a paper document using colour or boxes, as done by the Scottish Dental Clinical Effectiveness Programme in their patient guideline on preventing gum disease [
21] and SIGN in their glaucoma guideline patient version [
22], which attracts readers to the recommendations first.
Strengths and limitations of this study
A strength of the study is that it involved a wide range of participants from many backgrounds and from across Scotland. The extent to which the findings are transferable to other populations, particularly non-UK populations, is uncertain. Qualitative research does not seek to gather a representative sample of a population but rather to purposively gather a heterogeneous sample and explore the transferability of the results beyond the sample gathered. The findings presented here do concur with those from our earlier systematic review [
4], which was global in scope. We are confident about the wider relevance of the five themes that emerged from the focus groups and interviews.
A rigorous method of qualitative analysis was used to analyse the data, which was led by two researchers (NF and JK) working independently of each other and who were not involved in the data collection, or the earlier review. This has ensured that the themes emerged from the data, rather than being imposed upon it by the rest of the team. Finally, we have made our findings tables available as a supplemental document so that others may review it and potentially use if for their own work.
The sample was purposively selected and heterogeneous in nature and therefore the materials and the exact questions posed to the participants varied somewhat between the groups. The materials and questions were tailored to the groups and developed as the research progressed. Each focus group/interview focused around the topic of what each group understood about and wanted from clinical guidelines and was open in its structure. This did not hinder the analysis. Given the focus of the research question and the range of groups included, the focus of the analysis was on similarities across groups and where discussion moved away from guidelines this material was excluded from the analysis. The analysis process specifically sought to identify differences between groups at the charting stage (see Table
2), and where differences were found they have been highlighted.
There are issues of trust that need to be addressed alongside efforts to clearly convey the balance of risks and benefits associated with particular treatments or courses of action and the importance of evidence when making potentially life changing decisions. The GRADE process [
23] has been instrumental in clarifying what that balance might be for specific interventions, and includes the values and preferences of patients and healthcare professionals. DECIDE has built on that work to improve the presentation of that information in ways that are meaningful to different audiences, including patients.
Acknowledgements
Dr Ilkka Kukanammo and Dr Helena Liira, of the School of Medicine, University of Tampere, Finland, provided input on this research for which we are very grateful.
Competing interests
No conflicts of interest have been declared.
Authors’ contributions
NF carried out data analysis and drafted sections of the manuscript. JK carried out data analysis. MC recruited participants, conducted focus groups and semi-structured interviews and carried out preliminary analysis of the data. KG conducted focus groups. KL conducted focus groups, commented on and edited sections of the manuscript. RH contributed to the design and coordination of the study. NS commented and edited sections of the manuscript. EM commented on and edited sections of the manuscript. JT commented on and edited sections of the manuscript. ST contributed to the conception and design of the study and drafted sections of the manuscript. All authors read and approved the final manuscript.