Inclusion under the Mental Capacity Act (2005): A review of research policy guidance and governance structures in England and Wales

Abstract Objective To investigate how people with communication and understanding difficulties, associated with conditions such as dementia, autism and intellectual disability, are represented in research guidance supplementary to the Mental Capacity Act (MCA: 2005) in England and Wales. Methods A documentary survey was conducted. The sample comprised the MCA Code of Practice (CoP: 2007) and 14 multi‐authored advisory documents that were publicly available on the Health Research Authority website. Textual review of key words was conducted followed by summative content analysis. Results Representation of people with communication and understanding difficulties was confined to procedural information and position statements that focused mainly on risk management and protection. Whilst a need to engage potential participants was recognized, guidance provided was imprecise. Conclusions Tensions exist between the protection versus empowerment of people with communication and understanding difficulties in research. The development of structured, evidence‐based guidance is indicated. Patient or public contribution People with communication and understanding difficulties and carers participated in a working group to explore, discuss and interpret the findings.

difficulties in research is widespread. 3,4 This may affect our understanding of the needs of these groups, and the efficacy of new treatments and interventions. The Research Governance Framework for Health and Social Care in England and Wales 5 requires that research participants reflect the diversity of the wider population and advises against the routine exclusion of under-researched groups, including those with disabilities. 6 Researchers are responsible for ensuring that their research complies with the requirements of the MCA (2005), and that people with capacity-affecting conditions are included appropriately. 7 Gaining consent is a fundamental prerequisite for involving human beings in research. 8 Founded on the principle of respect for autonomy, [9][10][11] it formally recognizes people's interest in making decisions, acting voluntarily, and understanding and processing appropriate information relating to these decisions. However, certain conditions, such as dementia, autism and intellectual disability, may affect the individual's decision-making capacity. 12,13 Another common characteristic of these conditions is communication and understanding difficulties, which may pose challenges and complicate assessments of capacity. 14,15 People with capacity-affecting conditions make up a significant proportion of the UK population. The number of people with dementia is projected to rise to 1.6 million by 2040 16 ; 1 in 100 people have autism 17 ; there are over 1.2 million stroke survivors 18 ; and a further 1.5 million people with intellectual disabilities. 19 The Code of Practice (CoP: 2007) accompanies the MCA (2005) providing guidance for interpreting the legislation. 20,21 For example, paragraph 11.29 of the CoP (2007) stipulates that the patient's wishes and feelings must be considered regardless of their capacity for deciding about their own research participation. In the context of research, the MCA (2005) has been criticized for lacking an appropriate balance between protection from exploitation and empowerment, with emphasis placed on the former. 22 Indeed, the MCA post-legislative scrutiny criticized the implementation of the Act in care contexts, noting a culture of protection and paternalism amongst professionals working with people who may lack capacity. 23 This appears to be in opposition to the culture of empowerment that has grown in recent years within health and social services in the UK for different care groups, including people with intellectual disabilities, 24 autism and 25 dementia. 26 There is an obvious tension between managing any risks associated with research participation and ensuring individual rights are supported. 5,27 Assessment of participant capacity (the ability to understand the information relevant to the decision, to retain, to use or weigh it up and to communicate the decision) is a primary requirement in ethical research under the MCA (2005). 1,28,29 A person's decision is viewed as their own and therefore sacrosanct. However, where a person is lacks capacity, additional requirements of the MCA (2005) need to be satisfied to authorize their participation in research. In such cases, the appointment of a consultee-a person who is able to advise on the individual's likely wishes and desires regarding participation in research-is recommended. The involvement of another person might seem diametrically opposed to any notion of empowerment and might prioritize the consultee's personal views over those of the individual. Furthermore, Jackson 30 observed that capacity and incapacity do not have clear boundaries and should be viewed on a continuum. 31 The CoP (2007) 21 attempts to redress the imbalance between protection and empowerment by recommending consideration of the views of potential participants who lack capacity. This is consistent with empowerment theory, which highlights strengths and capabilities, rather than cataloguing risk factors. [32][33][34] As a construct, empowerment theory examines interaction of individual competencies, systemic support or facilitation, and proactive behaviour affecting policy development and the social change process. 35,36 Joining efforts with others to gain access to resources and to achieve goals are considered critical to empowerment. Zimmerman and colleagues identified three components of empowerment: intrapersonal (how people think about their capacity to influence others); interactional (the transactions between people and environments); and behavioural (what is done to influence change in the environment). 35 Thus, empowerment may be viewed as a complementary process to the protection of rights, but also one that invites support from others. 37 43,44 ; and 'Active Support', 45 are not formally recognized practices. The national statement on ethical conduct in research recommends that information is presented for optimal accessibility by the person. 46 For example, simplified language in large print with pictorial support has been used with people who have intellectual disabilities (termed Easy Read) 47 ; aphasia-friendly resources for people post-stroke, 48,49 and digital computer technology with people with autism. 50 According to relevance theory, we naturally engage with information (spoken, written or symbolic) that requires the least cognitive effort for the most successful understanding. 51 This requires deliberate address of critical aspects of language including its form (syntax and grammar), content (semantics or meanings and vocabulary) and use (intended purpose of the message). 52 The current study was part of an investigation into the ethico-legal landscape for the development of an assent-based process for the inclusion of adults with communication and understanding difficulties in ethically sound research in England and Wales. The aim was to investigate how people with communication and understanding difficulties are considered in the research guidance in England and Wales. This is an international concern for legal systems across the globe. 53

| The CoP (2007)
The CoP (2007) was retrieved from the UK Government website.

| Word referent frequency
In order to establish how adults with communication and understanding difficulties are represented in research guidance, a survey of surface-level vocabulary was conducted. Firstly, a list of key words (termed referents) was generated that related to the following domains of interest: communication and understanding difficulties; decision-making and capacity; and inclusion in research. A first iteration was compiled by each of two researchers independently.
Secondly, the researchers met with the Chief Investigator and the two lists were compared. This involved grouping similar words together (eg communication, language and speech) and identifying any differences between the two lists of referents. Once consensus had been reached, a combined list was generated. Thirdly, the list was used to locate word referents and their frequency of occurrence in an electronic search of each document using NVivo-12 software.

| Summative content analysis
Summative content analysis 55 was carried out on the entire dataset by two researchers working independently. Latent content analysis was conducted to explore meanings. 55 Based on their homogeneity, meanings were grouped, resulting in a first level of nodes, termed 'organising themes'. This gave rise to a second level of nodes, termed 'sub-themes'. The Chief Investigator (last author) facilitated two meetings at which the assignment of excerpts to organizing themes and sub-themes was reviewed, until team consensus was achieved.
Next, associations and dissociations occurring amongst thematic content were identified. Finally, a visual representation of organizing and sub-themes was generated. This analysis procedure was then applied to Chapter 11 of the CoP (2007).  Table 1 summarizes word referent frequency in the three domains of interest: people with communication and understanding difficulties; decision-making and capacity; and inclusion in research.

| Word referent frequency
The total word referents per domain and the number of sources in which they occurred are shown (see Appendix A for document title). Each domain contains word referents as indicated in the central column. A '+' immediately after the core referent indicates that it is a 'multi-stem' phrase, to include related terms. For example, autism + included a variety of different words used to describe autism, including autistic spectrum disorder/condition/ ASD/Asperger's syndrome/AS. Referents that did not include related terms are referred to in this paper as 'single-stem' phrases.
Referent frequency is rank ordered from highest to lowest within each category.

| Summative content analysis
Initially, research guidance was defined by two levels of relevance: general (sources: 11, references: 327); and specific (sources: 14, references: 479). The first level, general, was excluded because of its focus on operational principles, rules and regulations associ-  The safeguarding of participants was emphasized: 'If someone is unable to provide consent for themselves due to a lack of mental capacity, the next step to consider is whether the legal requirements and '…nothing must be done to or in relation to the person who lacks capacity which is unduly invasive or restrictive'.
[CoP: 11.12] and that the research benefits to the person who lacks capacity, be proportionate to any burden resulting from participation: 'The research must have some chance of benefiting the person who lacks capacity…the benefit must be in proportion to any burden caused by taking part…' [CoP: 11.12]

| MCA (2005) & Decision-making
The 'Mental capacity is considered to be lacking if, in a specific circumstance, a person is unable to make a decision for him or herself because of an impairment or a disturbance in the functioning of their mind or brain'.
[ Source 11] 'Researchers should assume that a person has capacity, unless there is proof that they lack capacity to make a specific decision'.

| Accommodations
The Accommodations organizing theme featured in 10 research

| Triangulation
The organizing themes of the research guidance and the CoP (2007  Conversely, empowerment was addressed infrequently within the research guidance and the CoP (2007), which is consistent with comments from the House of Lords select committee. 2 In accordance with the legal test for capacity, the research guidance adopted a binary approach to mental capacity assessment, whereby a person was either assessed to have capacity or not. 21,22,29,30 However, fluctuations in capacity and how these affect an individual's eligibility to participate in research requires consideration. 21,22,29,30 Although the need for accommodations was recognized, detail into how they might serve the principles of determining and supporting an individual's capacity and communication was not provided. 22 Thus, the idea of joining efforts with others in order to achieve understanding, which is consistent with empowerment theory, 35 was largely neglected. The inclusion of incapacitous participants in research was linked to procedural aspects of the consultee process, with limited practical strategies for facilitating engagement and decision-making that is evidenced in the emerging use of assent in some areas of research, for example dementia. 6 In conclusion, the CoP (2007)

ACK N OWLED G EM ENT
We would like to acknowledge Yvonne Plenderleith's contribution to the content analysis of the HRA documents.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests. All authors listed on the manuscript contributed to the study design, data collection, data analysis and interpretation of the findings, and drafting and reviewing the final manuscript.

AUTH O R CO NTR I B UTI O N S
HR was involved in data retrieval, analysis, interpretation and visualisation; writing of the original draft and review and revision of subsequent drafts. RH was a consultant on all legal aspects of the work and review of the original draft and contributing to subsequent drafts. OJ, AK, PEL and CS were involved in review of the original draft and contributing to subsequent drafts. KB was involved in the design, data analysis and interpretation; writing of the original draft and review and revision of subsequent drafts.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data addressed in this paper are publicly available via the UK Health Research Authority website (www.hra.nhs.uk).