Easy read and accessible information for people with intellectual disabilities: Is it worth it? A meta‐narrative literature review

Abstract Background The proliferation of “accessible information” for people with intellectual disabilities in UK health care has accelerated in recent years, underpinned by policy guidance alongside the recent introduction of mandatory standards. However, questions have been raised as to the impact of such resources as a means of enhancing involvement in health care and addressing health inequalities. Objective To review and synthesize the evidence from different approaches used to evaluate the impact of accessible information for people with intellectual disabilities using a meta‐narrative approach. Search strategy Literature searches were iterative and incorporated formal databases, grey literature and hand searches alongside more intuitive and opportunistic methods. Inclusion criteria Included English language papers published before December 2015 described the design and evaluation of written information adapted for adults with intellectual disabilities. Data extraction and synthesis We organized the papers into five groups according to similarity in authors’ writing styles and presentation, epistemology and theoretical foundations, aims and methodologies, professional and organizational identities. Main results The 42 included papers in the five groupings occupied diverse positions on (i) public communication vs individualized materials, (ii) literacy as decontextualized skills vs social practices and (iii) the expertise of patients vs professionals. There was limited evidence for the impact of accessible health information, notwithstanding the potential benefits associated with their creation. Conclusions Individually tailored information is more likely to meet personalized health information needs for people with intellectual disabilities. The emergence of different social formations in the creation of accessible information has potential for advancing engagement of diverse groups.


| INTRODUCTION
The provision of "accessible information" to people with intellectual disabilities is often characterized as the answer to some "wicked" (persistent, resistant to solution) problems 1 relating to equal access to health and well-being. Policy documents and statements by politicians, practitioners, researchers and self-advocates suggest that accessible information will allow people with intellectual disabilities to become more self-determining, minimize health inequalities, promote active citizenship and bring about the empowerment of a social group whose voice is often excluded and ignored. 2,3 Consequently, the availability of accessible information has proliferated greatly in recent years, especially in the health field. There are a number of guidance documents available that supply instructions for creating accessible resources 4,5 and an ever widening range of information resources have been published in an accessible information format (a recent UK web search turned up 24 different resources relating to blood tests alone). The requirement for service providers to make "reasonable adjustments" to their provision, including information they offer about their services, is now enshrined in the 2010 Equalities Act. 6 The establishment of the NHS England Accessible Information Standard 7 is a recent attempt to operationalize the adjustments that are expected from all health and social care agencies.
There have been some more sceptical voices, raising questions about the real impact of such resources and whether they are in fact fit for purpose, conveying all the necessary information in a clear and simple way. 8,9 Some have queried whether the import of accessible information is more symbolic as a marker of ideological commitment to inclusion, rather than a practical means of enhancing the knowledge of people with intellectual disabilities. 8,10 Others have highlighted the risk that accessible information will be handed to people with intellectual disabilities without appropriate support or attention to their individual communication needs. 11 To date, there is little systematically reviewed evidence that supports this policy-level commitment to adapting health information for people with intellectual disabilities by confirming its effectiveness.
This review is therefore an attempt to fill this gap and moreover to include a wide range of perspectives and approaches to exploring the value of accessible information.

| METHODOLOGY
Greenhalgh and colleagues 12,13 suggest "meta-narrative" as an approach to literature reviewing that does not seek to iron out major theoretical and methodological differences between studies, but collects similar studies into defined research traditions and paradigms.
Each of these traditions brings to bear its own epistemology and methodology in addressing the research question. The reviewers' job is to highlight the insights from each tradition, comment on the agreements and disagreements between them compare and come up with higher order themes that encompass these. This orientation is explicitly constructivist; the focus is on systems of meaning making associated with different paradigms, rather than determining any one underlying truth.
More recently, the principles of meta-narrative review have been used in smaller scale studies 14,15 with a more focused scope than the original, wide ranging reviews. In similar vein, we adopted the key strategies of meta-narrative review (see Figure 1), keeping in mind the six guiding principles of this approach (pragmatism, pluralism, historicity, contestation, reflexivity and peer review) and using recent guidance 16,17 as a benchmark to evaluate the quality of our study design, execution and writing up. Our research ques- • Incorporating intuition, informal networking and browsing to map the diversity of perspective and approaches • Identifying the prevailing conceptual, theoretical, methodological elements characterizing different approaches and the language and imagery used to 'tell the story' of the work. • Extracting and collating the key results, grouping comparable studies together • Identifying all the key dimensions of the problem that have been researched and giving a narrative account of the contribution from each approach • Addressing the agreements and conϐlicting ϐindings through the generation of higher order themes.

3.
Comparing the different groupings, ○ Can commonalities be found between the assumptions, approaches, findings and conclusions of the different groups?
○ How do agreements and disagreements among the groupings suggest higher order insights?

4.
What further research is indicated?

| Definitions
The definition of "accessible information" for people with intellectual disabilities raised challenges. There is no single generally accepted conceptualization, which means it is often ill-defined within contemporary policy documents. Within health-care settings, understandings are skewed towards easy-to-read resources, 5 although others have insisted that the mode of delivery of information 18 or the involvement of people with intellectual disabilities in creating accessible information resources 2 is key.
For instance, "easy read" is a term that has come into common use in the UK to describe information designed specifically for people with intellectual disabilities as a group with particular literacy needs. It has supplanted other terms ("easy to read," "easier information") and is found on many adapted public documents. This term is not so common in other English-speaking countries, although is gaining ground in Australasia.

| Scoping
The initial scoping phase of the review involved reflecting on our prior knowledge as researchers and clinicians. We contacted others engaged in similar work, tracked citations from the reference lists of guidance documents and opinion pieces and embraced serendipitous discoveries of different areas of work. A combination of Medical Sub-Headings (MeSH) terms supplemented with free-text words relating to "accessible information," "easy read"/"easy to read," and terms for intellectual disabilities was drawn up to identify potentially relevant literature in searchable bibliographic databases.

| Searching
We Next, an iterative process was initiated whereby we identified additional references by hand searching the reference lists and citations of relevant papers. We included "grey literature" such as reports published by public bodies found through Internet searching.

| Study selection
Study titles and abstracts were screened for eligibility, allowing for discussion to resolve any uncertainties. For studies not excluded on title and abstract, we obtained the full paper and assessed it in more detail. We established inclusion and exclusion criteria in order to come up with a more manageable number of papers, whilst avoiding an overly restrictive limitation of the studies to be included ( Figure 2).
Although our primary interest was in health information, we did not restrict our search to that field; a few papers were included that F I G U R E 2 Inclusion and exclusion criteria Included: • Published between 1990 and December 2015 (including online ahead of print) • Described not only the design and production of written information adapted for people with intellectual disabilities, but also had an element of evaluation of a single or multiple resources or of the design process. • The materials were aimed at adults with intellectual disabilities • Written in English.

Excluded:
• Materials aimed at children • Guidelines on creating adapted information • Studies of accessible information designed for patient groups other than people with intellectual disabilities. • Editorials, commentaries and opinion pieces • Description and evaluation of software, websites, apps or other computer or digital resources • Not written in English looked at information relating to voting, 19 consent to participation in research 20 or the criminal justice system. 21 We did not include video or digital resources. We recognize that these formats offer many exciting possibilities for people with intellectual disabilities; however, their creation and use require access to relatively specialist and expensive resources, compared to print media or audio recordings.

| Data extraction
We tabulated key bibliographic features of the studies. Reading, rereading and discussing our selected studies with an eye on this table helped us to identify groupings of papers where there were commonalities in styles of writing and presentation; aims, methodologies and study design; professional or organizational identities and ideological affiliations. This was not a one-off process; our categorization underwent considerable adjustment and refinement. We did not find a great deal of cross-referencing between studies and struggled to identify "seminal" or foundational studies that received most citations and themselves stimulated development of what might be identified as a "research tradition." In some cases, we went further than the authors themselves in associating them with wider research fields or epistemological frameworks as these aspects of the papers were largely implicit.
Inductive coding using NVivo 11 (QSR, 2015) of one group of papers led us to create a coding framework that we then applied to the other groupings. This allowed us to contrast and compare across the different groupings. To enhance consistency of data extraction, we coded the same two papers individually and then compared the data extracted and code headings. Following discussion and agreement on the code headings, we divided the remaining papers and coded them.

| Quality appraisal
Is often not appropriate for a meta-narrative review to apply the same quality criteria to all the studies under review. Where distinct approaches to the research topic have been identified, Wong et al. 16 recommend that "studies in these separate traditions should be appraised using the quality criteria that a competent peer-reviewer in that tradition would choose to use (p19)" rather than following a single predetermined protocol to evaluate quality. Therefore, considerations of study quality are addressed within the summaries of each research grouping in the results section below. Nevertheless, it is important to note that only the papers published in practitioner or academic journals, as distinct from grey literature and reports, were subjected to a transparent peer-review process.

| RESULTS
Forty-two papers were included in the review (see Figure 3), which we collated into 5 groups ( Table 1). The most recent print versions of papers are listed. In this section, we present accounts of each group before offering a synthesis of findings and inferences across the groups.

| Group 1: Practitioner accounts
The authors of these articles were mainly clinicians who created their own resources for distribution or use by a specified group of people with intellectual disabilities. These articles were generally published in journals accessed by practitioners that provided opportunities for authors to share information and experiences and publicize their work and raise the profile of the service they worked in, rather than ad-

| Group 2: People with intellectual disabilities as resource evaluators
These authors of the nine papers in this group described projects where the opinions of people with intellectual disabilities were foregrounded as reviewers of accessible information resources (one or many) that had been created elsewhere. In four cases, authors were part of independent research teams who had been commissioned to review a specific resource. 21,[30][31][32] Participants were in many cases associated with self-advocacy organizations 31,33 or those that worked with people with intellectual disabilities as coresearchers. 34,35 The authors of these papers were more likely to align explicitly with the social model of disability. 36 The importance of people with intellectual disabilities taking an equal role in designing accessible resources and also having control over when and how the resources were used was emphasized. Authors drew attention to power imbalances between people with intellectual disabilities and others (carers and paid staff). They suggest that staff and supporters themselves can present barriers to disabled people in terms of distribution of resources and their use to promote free and unbiased decision making by people with intellectual disabilities. 32 The findings of this group were more ambivalent than Group 1.
Many participants in these studies were appreciative of efforts to make written information more accessible. However, in most of these papers, limitations in the accessible information were highlighted, particularly ambiguous visual images 32,33 and wordings, 37 having too much information, being too difficult to read, even for "competent readers". 37,38 Some authors pointed out how simplifying a visual image or written information made its meaning less, rather than more clear.
Perhaps because of their background in disability studies and self-advocacy, these authors offered a more complex understanding of access for people with intellectual disabilities 39 that encompassed not only the existence of adapted resources, but also their distribution and availability. Many of their participants mentioned problems in getting hold of resources 30,31,36,38 ; these constituted further barriers to access and might be caused by funding for a resource drying up 34,35 or professionals not handing them on to people with intellectual disabilities. 30

| Group 3: People with intellectual disabilities reflecting on process of creating resources
The three studies in this area were papers authored or co-authored by people with intellectual disabilities, or referred to work authored by someone with intellectual disabilities. They appeared in British specialist intellectual disability journals that are welcoming of papers in easier to read formats, 40 and all included illustrations. The stated aim of these studies was to highlight the capabilities of people with intellectual disabilities as producers of accessible information. In many ways, they depart from what is usually considered acceptable as a research report, with little reference to background academic literature or impersonal appraisal of data. Their value is judged to reside in their close alignment with the principles of inclusive/participatory research as reflecting the agency and voice of people with intellectual disabilities. 41 There is an awareness of how creation and use of written information engage not only the cognitive aspects of the self, but also sociality and emotion. The self-advocate authors mentioned how their role/job gives them social identity and enhances social bonds and self-esteem. 42  Their conclusions were mixed. In contexts where the information under discussion had particular valency or significance (either emotional or professional) for the supporters, for instance when it was part of a health promotion consultation, 18 or a discussion about cancer, the supporters' own views and interpretations of the materials appeared to take precedence. 46 This might mean that supporters used the materials to emphasize normative behaviour regarding health promotion (rather than outlining choices) or avoided aspects of the resource they found difficult or upsetting. Relying on reports of carers or professionals, rather than using observational methods, perhaps gave a more positive view of the impact of accessible information in real-life settings, such as visiting the doctor. 47 A drawback for these papers is the very small numbers of people involved in the studies with ten or fewer participants with intellectual disabilities in all cases.  56 was the only one to explore the impact of presenting text in an audio rather than video format. This variety does pose some difficulties in drawing overall conclusions from the studies' findings.

| Group 5: Evaluation of effects of accessible information
These studies recruited larger groups of participants (between 13 and 85 people), although tended to do so "opportunistically." Most were described as having "mild" or "moderate" intellectual disabilities and potential participants with very restricted or no verbal language were excluded.
Nevertheless, the impact of using accessible information was disappointing. Where understanding of the texts was assessed as an outcome measure, and adapted and non-adapted texts were compared, 19,20,[52][53][54][55][56] only one of these studies 53 found an advantage for groups who had been given information that had been simplified linguistically or otherwise adapted to make it easier to understand.
On the whole, the cognitively more able, or those who did better on reading tests, seemed to do a better job of decoding. For those who struggled more, adapting the text in various ways did not tend to help.

| SYNTHESIS
In this section, we identify higher order themes through reflecting on the conceptual, methodological and empirical differences between our groupings. These are represented as tensions between diverse positions with implications for practice, policy and future research; we reflect on these in the final discussion section.

| Public or individualized resources?
Creators of accessible resources are faced with a dilemma regarding whether they address potential consumers of their texts as a group

| Literacy skills or literacy practices?
Use of reading assessments and readability metrics among

| People with intellectual disabilities and professionals: whose expertise?
The different groups of studies adopted diverse positions on the role of people with intellectual disabilities in the initiation, design and evaluation of accessible information along a continuum of involvement. 63 At one end of the continuum, judgments about the readability of a text were calculated through computer algorithms 50 -people with intellectual disabilities were not involved at all. In other Group 5 studies, people with intellectual disabilities were involved as research "subjects" usually recruited through service settings that they attended, where it might be argued they were something of a captive audience. As participants in these studies, people with intellectual disabilities did not always have significant involvement in the design of the materials they were given to test.
At the other end of the continuum of involvement, people with intellectual disabilities were described as the originators of ideas for accessible materials, 43,64 authors and editorial advisers, 35,65 or coresearchers 31,34 into the impact of accessible information.
Between these two positions, people with intellectual disabilities were engaged as consultants at the beginning and quality checkers and approvers towards the end of the design and production process.
The importance of involving people with intellectual disabilities was

| Limitations of this study
Meta-narrative review regards intuition, personal and professional knowledge and networks, and serendipity as resources available to reviewers, although at the expense of the replicability of the review. We therefore acknowledge that this review is itself a crafted narrative, albeit one that we have tried to support throughout with reference to our primary sources. Other reviewers might well identify different groupings of studies and highlight different themes.
The relatively discrete focus of our review and our limited resources in undertaking it meant adherence to the guiding principles of historicity (unfolding of research traditions over time) and peer review were a challenge. We refer throughout to "groupings" rather than "research traditions" as we found limited mutual awareness and intergroup citation, even between very similar studies, meaning that it was not easy to get a sense of later work building on and elaborating earlier work. Our engagement with peer review only extended as far as seeking feedback from peers and mentors; ideally, we would have undertaken the review with support from a reference group that included different stakeholders, including practitioners, policymakers and people with intellectual disabilities.

| DIRECTIONS FOR FUTURE RESEARCH
In selecting our papers for review, we found that we had to omit literature that was potentially very relevant and valuable, because the authors had not explicitly included people with intellectual disabilities in their research samples, or had in some cases excluded them. 70 In the wider literature relating to written health communication, simplification of health messages and health literacy we found widely cited review studies, that explored, for instance the usefulness of using visual images in health communication 71 or compared different strategies for simplifying health information. 72 We also discovered a separate literature on making information accessible for people experiencing aphasia after a stroke. 73,74 Further research is needed to clarify whether there are common strategies that are likely to improve the accessibility for a range of groups who need communicative support, or whether different groups have different requirements. Some of the authors of our reviewed papers suggested that resources developed for people with intellectual disabilities offer many benefits to the wider population. This claim needs further empirical exploration.
The review revealed a lack of information about how people with intellectual disabilities and the people with support them use accessible information in their everyday lives. Unfortunately, observational research on the communicative experiences of people with intellectual disabilities is still scarce 75 ; this is even more the case regarding literacy practices. 76,77 If these resources are being used in real-life situations, how can we monitor their impact and specify appropriate outcomes? Different bringing their own resources to bear on interpretation of a text. This suggests a direction for future research that seeks to deconstruct accessible health information texts to discern how people with intellectual disabilities, health professionals and biomedical institutions are constructed within them and also map how texts are taken up, recontextualized and transformed in use. 82