Basket | Login


Part 1: Ethical Challenges Linked to Public Involvement

2019: Overcoming ethical challenges affecting the involvement of people with dementia in research

Part 1: Ethical Challenges Linked to Public Involvement 

What is meant by Public Involvement?

People with dementia can contribute towards research in different ways. The organisation ‘INVOLVE’[1] makes a useful distinction between three possible ways.


This consists of the active involvement of people in research projects and in research organisations other than as research participants.


This consists of people taking part in a research study as a research participant (formerly often referred to as a research subject).


This consists of people receiving information and being informed about a research study.


The term ‘public’ is understood as including patients and potential patients, informal (unpaid) carers, parents and guardians, people who use, or have used, health and social care services, people with disabilities, but not people who are employed as health or social care professionals, or academics (INVOLVE 2017).

In Part 1 of this discussion paper, we focus on the contribution of people with dementia to research in the context of Public Involvement and in Part 2 on their contribution as research participants (i.e. providing researchers with data for their studies).

The concept of Public Involvement

The term ‘Public Involvement’ (PI) is usually understood as meaning carrying out research and developing policies with or by members of the public and patients rather than on or for them. This is how the term is used in much of the literature on this topic. Although the term covers a broad range of people, the scope of this discussion paper is limited to the involvement of people with dementia in PI. Consequently, discussions surrounding PI in this paper relate to the involvement of people with dementia. Please see the sub-section ‘a word about terminology’ for more details.

PI is not about merely raising awareness or providing information about ongoing or completed research (sometimes referred to as patient engagement) or about being a research participant. Rather, it is about creating a partnership between researchers and patients, whereby all contribute collaboratively in varying degrees towards the research process or the research output. It is not a specific method but an approach to involving people in research other than as research participants. PI could be considered as an overarching term which groups together a wide range of approaches and methods (see Figure 1).

F“I share this disease with millions of other people - so why should I be ashamed of it? On the contrary, I want to speak out about how it feels to live with dementia. You might all have a long history of research in the field of dementia but we, who live with dementia, can tell you much better about our demands, needs and wishes. We are the real experts of our condition! Don’t use me but involve me!”  (Angela, person with dementia, Austria).


PI is not rigidly defined according to the extent of the contribution to a particular research project or to the nature of the task. The nature and extent of the involvement may differ from one research project to the next. It can occur along a continuum from involvement in an isolated task, through involvement at several or all stages of the research process up to full involvement as a core member or leader of the research team. Furthermore, different people may be involved at different stages of the research or in different tasks.

Figure 1: Extent of involvement in PI and issues affecting involvement

Methods used in PI may sometimes be used which resemble those used in qualitative research such as interviews and focus group discussions. The use of specific methods does not determine whether something is PI or research. Rather, it is the reason for using a particular method and how the information is used that are the determining factors. If the information is used as data to answer a research question, then it is research; if the information is used to develop or improve the research process, or if involvement promotes democracy, the right to voice, justice and fairness, then it is PI.  Hoddinott et al. (2018, p. 3) summarise this as follows:

“Discussion with patients at a workshop can seem similar to collecting data in a focus group, because both involve listening to patients’ perspectives, but the context and outcomes from listening differ. PPI means that researchers are in a continuing and reciprocal relationship with patients and make decisions with them about the research. In qualitative research, researchers listen to patients in order to improve their understanding of a topic”.

The following provides a brief overview of some of the key areas in which people with dementia might typically be involved in PI. These activities are ideally planned and organised in such a way that they are accessible to a broad range of people and do not require prior knowledge of research. This is not an exhaustive list.

  • Providing insight into the experience of living with dementia (in the context of a consultancy role, not as research data),
  • Monitoring plans and the conduct of research in order to help prevent the exploitation and maltreatment of people with dementia,
  • Influencing decisions about which studies to fund,
  • Helping to shape and define research questions so that they reflect the needs, priorities and interests of people with dementia (e.g. with individual researchers or through priority setting initiatives such as the James Lind Alliance in the UK mentioned in the introduction),[2]
  • Sitting on ethics panels and project steering/advisory committees,
  • Improving recruitment (especially of people from groups which appear difficult to reach),
  • Ensuring transparency and accountability of the research process,
  • Offering the perspective of a person with dementia about the topic, research question or methods,
  • Encouraging reflexivity amongst researchers (see Part 3 of this discussion paper),
  • Challenging set views and restrictive paradigms (either explicitly or simply by their presence),
  • Contributing towards the design of methods and tools which are more appropriate and accessible to people with dementia,
  • Contributing towards the collection of data and the interpretation of findings,
  • Helping ensure that outcomes will be relevant to people with dementia (e.g. that services will be appropriate and hence used),
  • Contributing towards accessible findings,
  • Helping disseminate accessible information about the study.


How can researchers best describe what PI is about and attract people with no prior experience or knowledge about research to it?

The rationale for Public Involvement

The above list of areas in which people with dementia contribute towards research through PI provide some indication as to why PI is important for research and researchers. The rationale for PI is broadly based on two key arguments (Ives, Damery and Redwood 2013, Gradinger et al. 2015) which also highlight its importance for people with dementia, members of the public and society in general.

  • The first, based on normative arguments, is sometimes described as ideological or process orientated. It emphasises ethical, social and political concerns. PI is portrayed as ‘an end in itself’, linking involvement to democracy (e.g. democratic decision making, public accountability, legitimisation and transparency), people having rights (e.g. a right to voice, a right to be involved in research relevant to one’s own condition) and to ethical principles of justice and fairness. It has also been argued that research ‘done to’ rather than ‘with’ people could be considered a breach of autonomy even though research participants provide informed consent to participate in research and those involved in PI are seldom also participants in the same study (Ives et al. 2013).
  • The second, based on substantive arguments, is sometimes described as pragmatic, portraying PI as a ’means to an end’, linked to attempts to improve the quality, validity, relevance and/or utility of research (both from a research and user perspective). It can also be considered as instrumental in providing knowledge that might otherwise be missing. This includes, for example, highlighting issues and asking questions about things that researchers have perhaps not considered. Tritter and McCallum (2006) suggest that key contributions often arise from personal experience and a non-medical or technical frame of reference.


Do values such as ‘democracy’ and ‘a right to voice’ have different meanings for different people?

Some researchers argue that the rationale for involving ‘patients’ is different to that for involving the public (McCoy et al. 2018). They suggest that people who have experience of a particular disease or condition are well suited to being involved in health research or policy related to their own treatment or care. The public, or so they argue, should be involved in decisions related to research or policy with wider implications (e.g. the allocation of scarce resources or for matters which require the ability to make trade-offs between competing values and interests) for which a degree of neutrality, with no established perspective and a willingness to consider competing perspectives. They associate public involvement with the goals of democratic representation, accountability and transparency (similar to some of the goals described in the first argument above) and patient involvement with some of the arguments described in the second argument above (e.g. sharing their unique experience of a condition) (McCoy et al. 2018).

However, the two arguments/rationales described above, in the context of involving people with dementia in PI, are equally valid. It may be important to involve a broader set of people (i.e. not restricted to those with a certain condition) in research for some issues (which are of relevance to a broader range of people) but this does not detract from the goal of involving people with dementia in all kinds of research as a means to ensure transparency, legitimacy and accountability. Similarly, it would be wrong to assume that people with dementia (or any other condition) are incapable of considering competing perspectives fairly just because they have a condition which is particularly relevant to the debate.

In the past, it was largely assumed that people with dementia were unable to express their opinions or share their experience. This often resulted in them being silenced and in some cases represented by informal or professional carers who do not always have the same perspectives or fully understand their views and experience. The modern disability movement, which started in the 1960s in America, brought people with disabilities together to fight for a common cause. The people with disabilities, mainly physical disabilities and sensory impairments at the start, argued that they were best placed to determine their needs and voice their interests. They coined the phrase ‘nothing about us without us’. People with dementia are now part of this movement and it is important that they have a seat at the table and are listened to.

PI represents a step towards recognition of the importance of hearing the voices of people with dementia and addressing what Flicker describes as ‘epistemic injustice’. This consists of ‘testimonial injustice’ whereby less credibility is granted to people on the basis of social prejudices (e.g. consisting of unjustified beliefs that they are incompetent) and ‘hermeneutic injustice’ whereby marginalised groups are deprived of the opportunity to contribute their social experiences to a comprehensive understanding of a particular event or condition and are harmed as a consequence (Jongsma, Spaeth and Schicktanz 2017). The harm arising from epistemic injustice in the context of research is that when research is not inclusive of people with dementia (particularly from minority and less heard groups), those people are silenced and thus discriminated against. It is now increasingly accepted that people with dementia have an important contribution to make to research and PI is increasingly required for research funding or ethics approval in some countries. However, whilst the concept is gradually becoming more common in the field of dementia research, the practice is not yet widespread across the whole of Europe.  

How Public Involvement differs to other approaches to involvement in the research process

We mentioned earlier possible confusion about the difference between PI and research in relation to the various methods used. There is also sometimes confusion about how PI differs from other related concepts such as co-production, co-design and co-creation which are specific approaches to PI.

To complicate matters, the term ‘engagement’ is used by some organisations and individuals to describe what others would call ‘involvement’ and vice versa. In the United States of America and Canada, and in Europe, for example in research linked to the Innovative Medicines Initiative (IMI), the term ‘engagement’ is typically used to denote what we describe in this paper as ‘involvement’.

There are some similarities between participatory research and PI, such as seeking to include and give a voice to people who are typically marginalised. A major difference is that whereas participatory approaches typically seek to involveallrelevant stakeholders from a particular community, both as advisorsandas participants with the aim  of bringing about social change (Knowles, Voorhees and Planner 2015), PI focuses on people with a direct experience of a particular condition as distinct from people being research participants. Knowles et al (2015) nevertheless suggest that researchers might benefit from looking at areas where one approach could learn from the other (e.g. PI could perhaps learn more about going out into relevant communities and meeting people in their own spaces, doing things their way and on their terms and conditions).


Researchers often experience difficulty recruiting a sufficient number of participants for their studies and PI means seeking to involve an even greater number of people in research. It could be argued that there is a need to consider the sustainability of PI and the possible synergetic burden. In other words, does the goal of involving people in research through PI and as research participants risk placing too many demands on people with dementia and would people with dementia who might otherwise have become research participants opt for PI instead? Whilst this issue is worthy of attention, it should not be assumed that the same people are attracted to either role in research or that involvement in the context of PI would rule out being a research participant at another point in time (or vice versa).

On the other hand, PI is a relatively new concept, not only for people with dementia but also for many researchers. Researchers therefore need to be clear about their aims and needs in relation to their research and to make it clear to people with dementia what these are and what it might bring those people to contribute to research either through PI or as a research participant. Some people with dementia will be more drawn towards participation in research involving little responsibility, interaction or reflection beyond what is asked of them as research participants, and would not be interested in getting involved in discussions with researchers, for example, about research design and procedures. Some might simply be interested in the topic or the goal of a particular study (e.g. to improve care or test a new drug) or agree to be involved in either PI or as participants based on having been asked by a respected person or gatekeeper. Others might find PI activities more rewarding or motivating but may assume that they need to have experience in or knowledge of research (which should not be necessary). It is therefore important for sustainability to ensure that people with dementia fully understand that there are different ways of contributing towards research, that they have options. The co-existence of both approaches should not be seen as competition for a limited set of people but rather as an opportunity to enable more people to contribute towards research in a way that they consider meaningful and personally rewarding.

“Six months ago, I was diagnosed with dementia and really wanted to do something constructive. I wanted to play a role in finding a solution to this terrible disease. I volunteered for a clinical trial but unfortunately had a bad reaction against the drug and eventually withdrew from the study. Then I heard they were looking for people with dementia to do ‘PI’ for another study. This was completely new to me but I decided to give it a try. Now, I really feel I’m doing something useful and am pleased to be contributing towards research in this way” (Juan, person with dementia, Spain).

A word about terminology

The term ‘involvement’ is sometimes preceded by the term ‘public’, ‘public and patient’ or just ‘patient’. The reference to patients is increasingly considered as problematic in that a lot of people with dementia do not see themselves as patients and are voicing their concerns about being ‘positioned’ in this way. Everyone who consults a medical doctor is at that moment a patient but this is not the sum total of a person’s identity. A ‘person first’ approach to the concept would suggest emphasising the person rather than the fact that they have a medical condition. This, however, would include the entire population, whereas Public Involvement, and variations of this term, is about involving people either because they have lived experience of a particular condition and/or because they are members of the public (i.e. not health or social care professionals or academics).

In 2017, when Alzheimer Europe published its position paper on this topic, co-authored by INTERDEM[3] and the EWGPWD, the term ‘public and patient involvement’ was preferred. The topic of terminology was raised in a parallel session at Alzheimer Europe’s annual conference in The Hague in 2019, which was chaired by the Vice Chair of the EWGPWD and included presentations from people with dementia about working groups of people with dementia involved in PI. Presenters and members of the audience, including representatives of some Alzheimer Associations, emphasised the desire to move away from ‘patient-focused’ language. We are therefore using the term Public Involvement in this discussion paper. The fact that public involvement is a term which covers quite a broad range of people should not be interpreted as implying that the involvement of one set of people is interchangeable with that of another. Our focus is on the involvement of people with dementia.

Recommendations for researchers

  • Be clear with people with dementia what PI is about, why it is important and how it differs from being a research participant.
  • Bear in mind that different levels of involvement are possible and that the level and type of involvement should be dependent on a range of requirements linked to the research itself and to the needs, wishes and interests of the people with dementia involved. See section on ‘Levels of involvement and the issue of power’.
  • Be clear about which groups of people are involved in PI (e.g. just people with dementia or also other people) and about the contribution of each (which should be specified in any subsequent reporting).


Recommendations for research ethics committees

  • Be clear about the role, purpose and value of PI in research and how this differs from research participation.
  • Seek clarification if necessary about how PI is embedded in the study and how those involved will be supported.
  • Suggest the implementation of PI activities at different stages of research, including the design of the study.
  • Include a person with dementia as an external expert if not already an ordinary member of a research ethics committee.
  • Consult a person with dementia when reviewing a protocol and the informational materials.


Recommendations for funders 

  • Invite people with dementia to have a say in prioritising research proposals.
  • Integrate patient relevant research questions or outcomes in open calls.
  • Assign a score to PI activities when selecting projects for funding.
  • Provide a dedicated budget for PI activities.
  • Recognise in funding applications the range of roles that PI may play.
  • Ensure that PI is explained in enough detail to determine meaningful involvement.



Public Involvement (PI) is about involving people with dementia in the research process, but not as participants. It means doing research with rather than on or to people with dementia. This is based on their right to voice their needs and perspectives and to democratic processes such as legitimisation and transparency. It is also a means to an end in that it enables researchers to benefit from the lived experience and perspectives of people with dementia, thereby fulfilling the criteria for good and hence ethical research. There is not one ‘right’ way to do PI. Indeed, it must be a flexible process which responds to the needs, possibilities and interests of a diverse set of people with dementia. In this discussion paper, we are using the term ‘Public Involvement’ (albeit with a unique focus on the involvement of people with dementia) as a move away from ‘patient-focused’ language.

Promoting and valuing diversity

Capturing diversity versus seeking representativeness  

Through PI, people with dementia can contribute towards research by sharing their unique insight into living with dementia and how this relates to the research topic and to the methodology with researchers. However, concerns about representativeness may block attempts to begin to involve them (Tritter and McCallum 2006) and call into question the value of their contribution. In a synthesis consisting of a mapping of values found in diverse literature on PI in health and social care research, frequent concerns were detected about whether people engaged in PI were representative of the community being studied because of selective recruitment or difficulty reaching certain groups of people (Gradinger et al. 2015).

A systematic review of PI in research[4] revealed that most of the 142 studies included had used convenience sampling[5] and the authors suggested that this was acceptable, adding that random sampling is the least biased approach but can fail for very small numbers (Domecq et al. 2014). The authors commented that despite the risk of convenience sampling and volunteering resulting in “a sample of patients that are not truly representative of the target population”, researchers should use the approach best suited to the availability of ‘subjects’ and based on the research topic. The use of terminology related to quantitative research when referring to people engaged in PI may contribute towards confusion and lack of clarity about their role.

People engaged in PI should not be described in terms of their ‘representativeness’ in the statistical sense (i.e. with associated assumptions that their views and experience can be generalised to the wider population of people with dementia). People with dementia have knowledge and experience of their condition and should not be considered as representing everyone with dementia. Any information they share about their experience is valuable and it is important to involve a sufficiently broad range of people in order to obtain a sufficiently comprehensive and nuanced understanding of the experience of dementia. However, for some issues just a few people might be sufficient. PI should not be confused with qualitative research for which it is often important to reach a point of saturation (see section on sampling in Part 2 of this paper). In PI, each person is expected to share his/her expertise and unique perspective with the research team (albeit it with appropriate encouragement and support). The perspectives, thoughts, feelings and experience of people living with dementia are valuable. These may be quite different from one person to the next and this does not make them any less valuable. Rather it enables research to benefit from a greater understanding of the diversity of people’s experience of living with dementia and of the complex nature of the condition.   

On the other hand, in keeping with the principle of equity/justice, everyone should have an equal opportunity to be involved in PI. Consequently, if people with very similar characteristics are involved to a much greater extent than others, this suggests that a large proportion of society is being excluded and that PI is not capturing the diversity of experience of people with dementia across Europe. Researchers are sometimes criticised for involving people who are the easiest to find. Indeed, the term ‘hard to reach’ is increasingly being recognised as serving to blame people from minority groups for low levels of involvement (e.g. reflecting assumptions that they have made themselves inaccessible or are unwilling to get involved) (INVOLVE 2012). Clearly, some forms of PI are more intensive and demanding than others, which requires a trade-off between inclusiveness and demandingness (Jongsma and Friesen 2019). There is a risk that some people with dementia, especially those who are often marginalised, may find the kind of participation proposed too challenging and hence not get involved. Some symptoms of dementia may also result in some people needing extra support or not being able to take part in certain activities (e.g. because of restlessness). Attention must be paid to ensure that different methods and opportunities for inclusion are appropriate and adapted to the needs of a broad range of people with dementia.

Some people have more opportunities to engage in PI than others because they are asked more. Possible reasons for being more frequently solicited include, for example, having connections with researchers or organisations, having certain skills and moving in circles that are familiar to many researchers. The contributions of people who are frequently involved are no less valuable for that reason. Nevertheless, people who share certain characteristics (e.g. linked to level of education, ethnicity and socio-economic status) and are frequently involved in PI may sometimes find themselves referred to in a disparaging manner (INVOLVE 2012) (e.g.  as ‘the usual suspects’ or as ‘super patients’) (Black et al. 2018, Beresford 2019). Such people are often more experienced, confident and assertive and more likely to make points which the researchers might not want to hear (Beresford 2019). Criticisms have also come from other people engaged in PI that researchers are perhaps too dependent on ‘super patients’, take their perspectives into account more often and miss the perceptions of a more diverse set of people, including people from marginalised groups (Black et al. 2018). This may also lead to concerns from other people with dementia. Jongsma, Spaeth and Schicktanz (2017) report criticism from some people with dementia of a lady who was very vocal in PI, but whose experience, whilst acknowledged as being valuable, was not considered by other people with dementia as reflecting their experience. Whilst the views and experience of people engaged in PI are not expected to represent those of other people, if one or a few people dominate discussions, their views may be the only ones heard.

The problem, however, lies not in the people with dementia who are currently involved in PI but in the fact that researchers are often only drawing on people from a narrow section of the population with a valuable but limited range of experience and perspectives, and failing to be inclusive (e.g. overlooking the insights and experience of other people who are also very much affected by the topic of the research and may have a lot to offer) (INVOLVE 2012). This is an issue of relevance to the principle of fairness and to the ethical requirement to conduct good quality research. It should, however, be noted that diversity is not always visible. Assumptions are sometimes made based on ignorance and stereotypical beliefs about what people from minority groups ‘look like’ without really knowing about those involved (e.g. their ethnicity, life experience, religious beliefs, level of education and gender identities). Moreover, it should not be assumed that people with the same condition or from a particular sub-group have by definition/default the same opinions and interests.

Equality and equity

There are also several factors which may make it more difficult, and hence less likely, for some people to contribute to PI, some of which are addressed in the next section on ‘Understanding multiple and intersecting identities’. An understanding of the difference and relationship between equality and equity is important in this context. In 2012, Craig Froehle produced the following graphic (Figure 2) to illustrate the point that “equal opportunity alone wasn’t a satisfactory goal and that we should somehow take into consideration equality of outcomes (i.e. fairness or equity)” (Froehle 2016).

Figure 2: The relationship between equality and equity


Gillon (1994) emphasises that justice is more than mere equality in that people are sometimes treated unjustly even if they are treated equally. He argues that it is important to treat equals equally and unequals unequally in proportion to morally relevant inequalities. An approach based on equality involves treating everybody in the same way without taking into account differences between people, which may be inherent, linked to circumstances or structurally determined. In the context of PI, this might involve organising PI sessions on a holy day when people from a particular faith would be unlikely to participate, discussing highly technical documents which require a level of literacy that some people do not have or having meetings on the 5th floor of a building with no lift, which would exclude people with mobility problems. These are all examples of equality, which when coupled with lack of sensitivity for individual needs and preferences, are basically unfair and not conducive to attracting and involving a diverse set of people with dementia in PI. Equity is therefore not about simply providing everyone with same opportunities but about fairness and equality in outcomes. Issues related to inequity therefore need to be addressed when striving for equal opportunities and outcomes.

Perceptions of the value of Public Involvement

Ensuring meaningful and valuable PI touches on respect for the individual in many ways and if carried out properly contributes towards good quality research of relevance to people with dementia. If approached in a tokenistic way, it involves deception, contempt and exploitation, perhaps providing a seat at the table, but failing to take into consideration people’s contribution. Many researchers carry out PI on the basis of ethical, pragmatic and methodological motives (Ball et al. 2019). However, as recognition of the importance of involving people with dementia in research increases, so too does the risk of tokenism (Brett et al. 2014, Hardavella et al. 2015). In the context of mental health research, Patterson, Trite and Weaver (2014) point out that incorporating lived experience into the research process may improve the quality, relevance, acceptability and ethical status of research but is by no means universally accepted. Critics have claimed that conducting PI is time-consuming, challenging and often tokenistic. Researchers sometimes plan for PI simply because it is a necessary requirement for funding or in some cases to obtain ethical approval. It is not uncommon for organisations working with people with dementia to be contacted, often at the last minute shortly before submission of a proposal, by researchers wishing to involve people with dementia in their research without any clear plan or goal, and without appropriate funding. This would most likely result in a half-hearted, tokenistic involvement of people with dementia in PI as these researchers may not be intrinsically motivated to conduct PI or may underestimate the need for the planning of such activities.

How could funders encourage researchers to conduct meaningful PI?

People with dementia, particularly people from various minority groups, may sometimes feel that their involvement is simply ‘to tick the box’ or tokenistic in the sense that they are, for example, the ‘representative’ Black person or Sikh person or person with an intellectual disability in the PI team. In other words, they feel that they are only there so that researchers can say that someone from their group was involved in the research and that what they have to say is perhaps of little or no importance to the researchers. Researchers should try to ensure that people are not made to feel this way as a result of their treatment or attitude because it is a form of harm and injustice, and may also affect people’s willingness to share valuable information about their experience of dementia.  This may sometimes be because researchers do not yet have sufficient cultural awareness or skills to reach out to a more diverse set of people in an appropriate manner so as to make each person feel valued as an individual. As more people with a broader range of characteristics start to take part in PI, their complex identities will hopefully be recognised and their contributions no longer reduced to a narrow, stereotypical understanding of a particular category into which they might otherwise have been placed.

Some researchers may have practical concerns about the time or skills needed to conduct PI properly or about specific challenges linked to the involvement of people with dementia. Some may be unclear about the concept of PI and what it entails. These are valid concerns which need to be addressed for PI to be effectively and meaningfully included in a research project. Lack of enthusiasm or resistance to the concept of PI by some members of the research team, however, not only risks a constant struggle to legitimise and promote the importance of PI in research but also jeopardises its effectiveness and risks tokenism.

Negative attitudes about PI and lack of responsiveness by researchers may also be detrimental to the wellbeing of the people with dementia involved and a major source of frustration. In some cases, people have experienced stigma, prejudice and discrimination. Examples from people involved in PI for mental health research include excessive scrutiny or, conversely, over-valuation of their contribution, undermining of their validity and status as an expert by experience if a research activity is performed too well, being considered as one of ‘the usual suspects’ (as mentioned in the previous section) or as ‘non-representative’ and experiencing resistance from researchers wishing to retain traditional research hierarchies (Patterson et al. 2014). There are also reports of positive experiences of PI and of mixed feelings within the same project. Some people with dementia contributing to PI in brain donation projects, for example, have reported feeling valued and involved whereas others, at different sites, described their involvement as tokenistic and frustrating (Hayes et al. 2018). 

Tacit assumptions and expectations of people with dementia, carers, researchers, funders and members of research ethics committees about PI need to be addressed and a way forward negotiated (Poland et al. 2019).  More research into the impact of PI would also be helpful in convincing some researchers about its value. In addition, it is important to consider what makes people feel valued and respected in some studies or at some research sites and not in others. It may be linked not only to attitudes but also to practices and procedures which do or do not contribute towards a respectful, equal partnership and opportunities to contribute meaningfully to research through PI.

Recommendations for researchers

  • When involved in specific studies, discuss how you and your co-researchers feel about PI, the concept of lived experience and the goal of creating equal partnerships. Acknowledge and discuss different perspectives to agree on how to present PI to the people with dementia who may eventually contribute to your research in this way.
  • Communicate your perception of the role and value of PI in research in the most appropriate way to the people with dementia involved and listen to their perspectives.
  • Consider how invitations to people with dementia from marginalised or minority groups are interpreted by those concerned and try to ensure that they feel valued as individuals (and not simply ‘to tick the box’ or as a token gesture).
  • Try to ensure that activities are not overly demanding and are inclusive of a variety of perspectives of people with dementia.
  • Ask people with dementia directly (e.g. in invitation letters or pictorial adverts for PI) to let you know if there is anything that would make it easier for them to be involved.
    • Include the concept of PI in training on research methods.


Recommendations for research ethics committees

  • Assess whether researchers have a specific training/the required expertise to carry out PI activities.
  • Require a detailed description (and not just general statements) of PI activities in research protocols.
  • Assess whether methods, times and opportunities for PI are appropriate to promote equity in PI.


Recommendations for funders 

  • Encourage researchers to allocate part of their budget and adequate time to attempts to support ongoing involvement of a more diverse set of people with dementia.
  • Take into consideration that more inclusive PI activities may require a higher budget.
  • Take into consideration that more inclusive PI activities will increase the social value of research.



Promoting and valuing diversity is about seeking a diverse group of people with dementia for Public Involvement (PI) activities. It is not about representativeness, at least not in the statistical sense (e.g. people with dementia think X, Y or Z) but about listening to the perspectives and learning from the lived experience of very different people. It is too easy to categorise people as being ‘hard to reach’. We need to be looking at why people from some groups within society or with certain characteristics are being excluded and at how practices, procedures and structures within society contribute to such exclusion. If we want to promote inclusive research, it is important first to address the issue of inequity.

Hierarchies and power

Levels of involvement and the issue of power

Research is about producing worthwhile knowledge, yet knowledge is often associated with power[6]. This perhaps partly explains the popularity of Arnstein’s (1971) ladder of citizen participation which is frequently evoked as a means to evaluate and in many cases to judge different levels of PI in research, with some forms of PI being considered as better than others. The typology of the ladder was initially proposed by Arnstein in 1971 to be deliberately provocative with each rung of the ladder corresponding to the alleged extent of citizens’ power in determining three US federal social programmes, namely urban renewal, antipoverty and Model Cities. Nevertheless, it is frequently reproduced, slightly adapted or used uncritically (in terms of its original purpose, focus and historical context). As a result, consultation (labelled in Arnstein’s article as tokenistic) has become almost a dirty word. However, Arnstein only specified that consultation should be considered tokenistic if proffered by power holders as the total extent of participation whereby citizens hear and are heard but lack the power to ensure that their views are heeded “by the powerful”. Moreover, as suggested by Maier, “the process of increasing participation should not be simplified to a one-dimensional parallel of climbing a ladder” (2001, p.716) and it should not be assumed that people with dementia engaged in PI are all preoccupied with power.

People with dementia can engage in PI in different ways and to different degrees such as by being involved in consultations about various issues, sharing perspectives and providing advice, collaborating with researchers more directly on specific tasks or by carrying out research. Co-production is one approach to PI which particularly emphasises the sharing of power and responsibility from the start to the end of the project, including the generation of knowledge (INVOLVE 2018). This tends to be perceived in terms of joint ownership of projects and working together in the pursuit of a joint understanding. The term ‘empowerment’, which is also often associated with co-production, reportedly challenges embedded knowledge hierarchies ‘of the expert versus the lay subject’, and recognises that communication is not ‘a one way transfer from a knowing subject to a supposedly ignorant one’ (Porter, 2010) and that there are ‘experts by training’ and ‘experts by experience’ (Cheffey, Hill, McCullough and McCullough 2017). Tritter and McCallum describe this as being,

“not a hierarchy of knowledge – relevant professional versus irrelevant lay – but rather a complementarity between forms of knowing, set within a willingness to acknowledge differences” (2006, p.164).

However, whilst certain aspects of involvement are perhaps more central or emphasised in co-production, they should also be present in any form of PI. Key principles such as sharing power, embracing diversity, striving for inclusivity, respecting and valuing the knowledge of all those involved, reciprocity, and building and maintaining relationships (INVOLVE 2018) should be reflected in all PI, without power necessarily being the focal point, the sole measure of participation or the criterion for evaluation of the value of a particular contribution.

As Tritter and McCallum (2006, p.164) suggest in relation to Arnstein’s ladder, involvement should not be perceived as a “contest between two parties wrestling for control of a finite amount of power”, thereby excluding opportunities for collaboration and shared decision making. People with dementia can contribute towards research in different ways according to their interests, experience and abilities. Not everyone is interested in the same level of involvement and capable of the same level or kind of participation in decision making. One person may wish to contribute to a small part or particular aspect of the research, whereas another might prefer to play a much more active role.

At the level of society, it has been suggested that the very goal of working together in a positive and collaborative manner with some marginalised groups is idealistic and naïve. This is because of assumptions and philosophies underlying much research, particularly in the field of medicine and psychiatry, and because people with mental disorders can still be deprived of their freedom, lose certain rights, be subjected to coercive measures and not be taken seriously. Rose and Kalathil (2019)[7] state that most current research in the fields of medicine and psychiatry still has remnants of Enlightenment thought and Eurocentricity, with a focus on rationality, the thinking, reasoning subject and the alleged ‘racial’ and cultural superiority of White Europeans and that this undergirds much of health research. For Rose and Kalathil, such a situation is not conducive to the goals of co-production. They suggest that a change in thinking is needed before this can be achieved.

Equality and clarity of roles

People engaged in PI and researchers tend to be perceived and defined in dichotomous terms. This is reflected in the social meanings attributed to members of each group, in expectations about who will conduct various tasks, in legal and financial responsibilities (irrespective of the extent to which people with dementia contribute towards key decisions) and in the use of language and titles (Morrow et al. 2010). The constant positioning of the various people contributing to research through language and other means serves as a constant reminder of differences and may raise questions about equality and the role of each.

People with dementia are usually invited by researchers to engage in PI. In many cases, the researchers determine the timing, duration, frequency and nature of involvement of the latter. Flexibility and sensitivity to the needs and interests of people with dementia (e.g. in terms of the timing and preferred mode of involvement) are important but it is often inevitable that PI corresponds to the development/progress made in a particular research project. For this reason, unless people with dementia have an extremely high level of involvement in the research (e.g. a co-researchers), the researchers will largely determine the timing of PI and the kind of involvement required. Researchers also usually manage the budgets necessary to pay for meeting rooms, travel, accommodation, materials and any out-of-pocket expenses. Academic discussions often revolve around whether PI is valued by the researchers and in many cases, researchers take the lead and have final responsibility for certain tasks (e.g. linked to accounting, safety issues, applications for ethical approval, reporting obligations and statistical analysis), however much shared decision-making there may be. It is therefore important to be aware that terms such as PI, co-researcher and researcher (to name but a few) reflect acquired roles which are actively negotiated and sustained within social contexts and that various constraints may affect how people do and do not act (Morrow et al. 2010).

Respect for equality should not be interpreted as the need to deny differences in people’s roles and responsibilities. At the same time, attention should be paid to how people’s roles and responsibilities are portrayed and communicated so as to avoid social positioning and the creation of unnecessary friction between people who are all investing their time, skills, energy and experience in the research. Terms such as ‘expert by experience’ and ‘expert by training’ may reflect different types of expertise but if each type of expertise is truly valued on an equal basis, this has to be demonstrated through actions not just words. Moreover, the emphasis of these two terms is on expertise, albeit of different kinds. This confirms the belief that different kinds of knowledge can exist side by side but in making this distinction there is also a risk that meanings come to be associated with each and that these do not necessarily reflect an equal value.

Tailoring Public Involvement to the interests and abilities of all people with dementia

Ethical involvement of people with dementia in PI calls for mutual respect, different opportunities for involvement possibilities for shared decision making, which correspond to individual interests and abilities, and clarity about the roles and responsibilities of everyone involved. Every effort should be made to ensure that all people with dementia with an interest and ability to contribute to PI have an equal opportunity to do. With regard to ability, efforts should be made to address inequity and to promote inclusion. The concept of ‘reasonable accommodation’, mentioned earlier, is relevant in this respect. This means that reasonable adjustments/adaptations should be made to enable people with dementia to contribute to research.

Measures to promote inclusiveness need to be balanced against the need for PI to be effective (in terms of contributing towards improving research and giving a broad range of people a voice), the need to promote the wellbeing of those involved and to work within the context of certain constraints (e.g. linked to available funds, group dynamics, progress with a specific study and academic, ethical and administrative formalities).  This requires a certain degree of flexibility (in terms of what is considered possible), acceptance of different needs and differences, a readiness to consider various forms of adaptation/adjustment as ‘reasonable’, and above all, a desire to promote diversity within research[8].

Researchers should have a clear idea about why and how they wish to involve people with dementia in their research in the context of PI, but a discussion between researchers and the people providing PI about the goals and expectations of each at the start of the project would be helpful (Black et al. 2018). It may also be helpful to consider different types and levels of involvement of people with dementia in PI as occurring along a continuum (see Figure 1 on page 15), perhaps with different people intervening at different points and on different tasks. This would ensure the involvement of people with dementia, in accordance with each person’s interests and what they are able to contribute, at every stage of the research. It is not helpful to denigrate anybody’s contribution on the grounds that it does not represent a certain type or level of involvement. Furthermore, if a person with dementia wishes to contribute to a certain part of a study and has valuable experience to share, he or she should be encouraged to do so. For some people, participation is the goal of their involvement and that is what they find rewarding and empowering. Others may have a keen desire to take on responsibility and to play a role in decision making.

It should not be assumed that everyone is capable of or interested in the same kind or level of involvement. Parveen et al. (2018) propose a person-centred and culturally sensitive approach to working with minority ethnic communities in PI (which could be equally valuable to the involvement of people with dementia in general). This approach involves providing people who are potentially interested in engaging in PI with information about the study and opportunities for involvement and encouraging them to reflect on their possible level of involvement and the aspects of the study they would like to be involved in. This helps avoid judgemental and value-laden reflections about different levels of involvement and implicit power relations. This more tailored, person-centred and cultural approach would help ensure that people with the appropriate skills and interest (not necessarily the same ones) are involved throughout the whole research process in a continuous manner. This also creates more of a balance in that the researchers seek the support of people with dementia and people with dementia decide in what way and to what extent they would be willing and able to provide that support.

This was exemplified in a recent IMI-funded research project for which Alzheimer Europe asked members of the European Working Group of People with Dementia (EWGPWD) if they would like to be part of a Patient Advisory Board (PAB). Some members of the group expressed an interest in contributing towards certain tasks periodically and others were keen on a greater involvement (more regularly, on a wider range of tasks). The latter had a particular interest in the topic of the study. The EWGPWD proposed having a general PAB and a Core PAB. The members of the Core PAB agreed to be contacted at relatively short notice and to provide feedback rapidly by various means (e.g. email, teleconference or through additional meetings). This has enabled a more interactive approach between people with dementia and the researchers within tight timeframes reflecting the demands of the study. This approach has not undermined the continued work of the general PAB on issues which can be planned more in advance and for which more support can be provided to the people with dementia who might need it.

Recommendations for researchers

  • Formulate a tailored, person-centred and cultural approach to PI.
  • Discuss the goals and expectations of PI with people with dementia at the beginning of the process of involvement and revisit as needed.
  • Be clear about roles and responsibilities of people engaged in PI and researchers, while avoiding devaluing the contribution of people with dementia.
  • Ensure the involvement of people with dementia in accordance with each person’s interests and what they are able to contribute.
    • Have clarity from the outset about how PI will be included in the study and build this into a project timeline.
    • Regularly update people involved in PI on progress with the research and plans for their ongoing involvement.


Recommendations for research ethics committees

  • Assess whether research protocols clearly describe roles and responsibilities of people engaged in PI and researchers.
  • Assess whether research protocols envisage different possibilities of involvement at the different stages of the research process to ensure that the persons’ interests and possibilities of contributing are respected.


Public Involvement (PI) needs to embrace diversity, strive for inclusivity and be based on respect for everyone’s involvement (irrespective of the type and level of involvement of each person). Flexibility and a readiness to make reasonable adjustments/adaptations are essential to enable a broad range of people with dementia to engage in PI activities. It is important to build and maintain mutually respectful relationships between people with dementia and researchers. An over-emphasis on power and the portrayal of involvement in the form of a hierarchy devalue the meaningful and valuable involvement of many people with dementia. Decision making should be shared wherever possible but the roles and responsibilities of researchers and people with dementia may differ depending on the type and level of involvement of each person in a particular study. It is important to ensure that this does not impact on the perceived value of the contribution of each.

Understanding multiple and intersecting characteristics/identities

There are numerous factors which interfere with the willingness and ability of people with dementia to contribute towards PI. In this section of the report, we look at some of these factors. Most challenges are experienced to a greater or lesser extent by many people but some people face considerable challenges and often more than one. Different people, depending on their history, personal situation and the level of support available to them, are better or worse placed to address various challenges they may face. For example, the challenges and experience of taking part in PI in a foreign language of a white, 55-year old, highly-educated, man with dementia are likely to be different to those encountered by an Asian woman of 85 who is illiterate in her own language. Similarly, people who are living in care homes, who have an intellectual disability or who have problems with substance abuse, to name just a few examples, may be affected differently by various challenges which need to be addressed in order to include them in PI activities. Sometimes, symptoms of dementia, inappropriate support and unmet needs may result in people with dementia acting in ways that other people find challenging and this may also represent a barrier to their possible involvement. In this section of the paper, we first examine the concept of intersectionality and then look at some concrete examples of related issues.  

About intersectionality

People with dementia form a diverse heterogeneous group made up of people with different characteristics such as different ethnicities, ages, genders, disabilities, levels of education and socio-economic backgrounds etc. There is not ‘one dementia’ and not just one experience of it. If researchers want to involve a more diverse group in PI (and as participants in research), it is important to consider these differences and how they impact on involvement in research or, more specifically, on exclusion which is a form of discrimination[9]. However, focusing on difference often involves categorising people into what come to be perceived as neatly defined, separate groups. This process of categorisation when combined with drawing a boundary between ‘we’ (usually members of what has come to be considered as the majority group, often consciously or unconsciously seen as the norm) and ‘them’ (usually members of what has come to be considered as the minority group) may serve as a mechanism for constructing ‘otherness’ (Torres 2015). Through dialogue and social interaction, meanings are constructed around the different categories which reinforce and perpetuate perceptions of normality and deviance, and contribute towards power imbalances and discrimination[10].  

Whilst it is important to identify characteristics which are commonly associated with discrimination, it is also important to look at ways in which socially constructed categories intertwine and may create new forms of discrimination which are not immediately obvious (Angelucci 2017). This is often described as intersectionality, a term coined by Kimberlé Crenshaw in 1989 to describe the different lines of oppression and marginalisation in society that can affect a person, based on their ethnicity, age, gender, ability, sexual orientation, class or the intersection of two of more of those aspects. It is a lens to explain the underlying mechanisms of the multidimensionality of marginalised subjects’ lived experiences (Nash 2008). Intersectionality therefore provides a lens to address complexities that different identities entail and the position of these identities, as an advantage or disadvantage, on a structural level. 

Intersectionality is about exploring the relationships between socio-cultural categories and identities. Knudsen (2006) describes it as something that can be used to analyse the production of power and of social and cultural hierarchies within different discourses and institutions, whereby some people come to be positioned not only as different or ‘other’ but as troublesome. These categories can be considered as socially constructed because, as Pickering (2001, p.72) points out, “conceptions of the Other and the structures of differences and similarity which they mobilize do not exist in any natural form at all… the location of the Other is primarily in language”. In other words, minority groups ‘exist’ because societies have chosen to focus on specific attributes (or features or experiences), name them and label people as belonging to groups reflecting those attributes. The task then becomes one of seeking to understand their difference (the assumption being that these differences reside in ‘them’) and “the logic behind their peculiarities” (Torres 2015, p.941). Locating ‘the problem’ in the individual detracts attention from the way that structures, organisations, procedures and systems create problems and blames individuals and groups for not being able to adapt (e.g. to Western culture, language and traditions). There is a risk of people considered as ‘other’ being blamed, for example, for not being able to adapt to Western culture or to ‘age successfully’, and of serious forms of discrimination (e.g. racism and ageism) being attributed to bias and prejudice alone, and overlooking the role of institutions and systems in creating and promoting oppression, discrimination and marginalisation (Torres 2015). 

It has been argued that failing to consider intersecting identities may even render certain characteristics or vulnerabilities invisible. Writing about the intersection of ‘race’ and ‘sex’, Crenshaw (1989) warned against thinking about discrimination as occurring along a single categorical axis. She argued, taking the example of sex discrimination in the work place, that discrimination against a white woman and discrimination against a black woman is not comparable and that failing to recognise the impact of the combined discrimination against black women by reducing the issue to the category of sex and taking the situation of white women as the norm renders the impact of ‘race’ invisible. This contrasts with identity politics, which encourages mobilisation around a single axis of human traits (such as the illness as identity, or gender as identity). Similarly, Torres (2015) points out that fairness and non-discrimination does not necessarily result from treating men and women exactly the same because it amounts to treating a woman as if she were a man (and vice versa) which is not the case. The risk in putting pressure on people to classify a particular axis as their single defining feature is that they may see themselves as complex individuals who cannot be represented so selectively or reductively. People may, however, identify more strongly with some personal characteristics than others at certain times or in certain situations. The important point is to recognise the possible impact of how different identities intersect. 

Unquestioning acceptance of certain ‘norms’ can also contribute towards intersectional invisibility. Purdie-Vaughns and Eibach (2008) have described the impact of androcentrism (the tendency to consider male experience as the norm), ethnocentrism (the tendency for members of the socially dominant ethnic group as the norm) and heterocentrism (the tendency to consider heterosexuality as natural/the norm and other forms as deviant or lifestyle choices) on people with intersecting identities. This contributes towards people being defined as non-prototypical members of marginalised groups. The impact of where or how a person is positioned is also affected by the context:

“Capitalist women would be identified as oppressed when situated in patriarchy and as oppressors when situated in capitalism. Working class men would be identified as oppressed when situated in capitalism and as oppressors when situated in patriarchy. The situation would obviously get more complicated once we added race to the schema of oppression” (Smiley 2008, p.99).

What does this mean in relation to the involvement of people with dementia in PI? Firstly, it suggests the need to consider social and cultural categories as dynamic and fluid, in a permanent process of construction, deconstruction and renegotiation of meanings affecting people’s identities, social roles and relationships, and with implications for marginalisation and potential discrimination. This does not mean that categoriesper seshould be denied or rejected but rather that broad and sweeping categorisations (bordering on stereotypes) should be avoided (Knudsen 2006). Secondly, it necessitates a focus on “the complexity of relationships among multiple social groups and across analytical categories and not on complexities within single social groups, single categories, or both” (McCall 2005, p. 1786). Marginalised intersectional identities amongst people with dementia can be taken as the starting point to explore the complexity of the lived experience of people with dementia (through participation in research) and for that diverse experience to influence the research process in the context of PI.  Finally, it highlights the importance of reflexivity on the part of researchers as a means to increase awareness of their own assumptions, possible restricted or biased world views and potential intersectional blindness at the individual and structural level (please see the section on reflexivity and positionality in Part 3 for more details).

The categorisation of people into sub-groups on the basis of perceived difference along a unique axis is one way of making sense of the world and of taking short cuts so as to make the vast array of information available more manageable but it has social consequences. Torres (2015) encourages researchers to reflect on the implications of using broad social categories such as class, gender, sexuality, ethnicity and age which typically result in assumptions. She asks whether and if so why a particular category matters, what assumptions are attached to it, whether these are helpful and what this says about individual responsibility versus structural factors.


Which social categories do you notice most?

Do you make assumptions about people based on these (e.g. this is a woman or this is a person from a traveller community, so this probably means X, Y or Z)?

What do you think the advantages and disadvantages are (for the people who are so categorised) of doing this?

What is your intersectional societal position? How does it impact on your research (research question, methodology, and writing)? How does it impact on your collaboration with people with dementia with their intersectional societal position?


Language and communication difficulties

Difficulties with language and communication are frequently experienced by people with dementia. This may, for example, involve having difficulty finding words, pulling ideas together, keeping track of conversations, remembering what has already been said and formulating sentences. Documents may become hard to read and understand. Such difficulties often result in people feeling uneasy, embarrassed or awkward when interacting with other people, and having problems at work. This may lead to loss of confidence and self-esteem, to confusion and misunderstandings and even to withdrawal from situations which require literacy and communication skills. These difficulties with language and communication tend to increase as dementia becomes more advanced. Consequently, people with more advanced dementia may find tasks which rely heavily on language and communication more challenging. This may result in them being excluded from PI, either by researchers or as a result of self-exclusion.

“Two years ago, I was diagnosed with dementia and have been participating in many surveys and studies.  I think that patient involvement is essential for a good study and I am happy to answer. But it is getting difficult for me now to read and write, although I was an educated engineer and communicated in both English and Flemish without any problems” (Geert, person with dementia, Belgium).    

Lack of fluency in the main language used in a particular country can also be a major barrier to involvement in PI. Many people from minority ethnic groups have difficulties understanding and communicating in the national language of the country in which they live. Impairments affecting communication (e.g. linked to hearing, vision and speaking) may also result in the exclusion of people with dementia.

Most PI work is conducted in the national language of a particular country and involves reading, writing and speaking at some point. The consequence of conducting PI in this way is that large numbers of people are prevented from contributing to research. It is estimated, for example, that 60% of first-generation immigrants in the Netherlands lack proficiency in the Dutch language (Uysal-Bozkir, Parlevliet and de Rooij 2013). In the South Asian community in the United Kingdom, only 35% of older people over 65 years of age can speak English and only 21% can read and write English, often communicating in their first language, which for many is Urdu (Blakemore et al. 2018).

In some minority ethnic groups, there is also a greater likelihood of older people being illiterate in their primary language (i.e. their mother tongue), especially older women (Bhattacharyya and Benbow 2013, Nielsen and Jørgensen 2013) and women in some communities have fewer opportunities than men to learn the language of the country (Beresford 2019). To complicate matters, in some languages, there is no word for dementia. This is the case in many Asian languages including, for example, the Punjabi language (Lawrence et al. 2011, Mohammed 2017, Uppal and Bonas 2013, Sagbakken, Spilker and Ingebretsen 2018). Similarly, it cannot be assumed that there is a clear understanding about dementia within each country's Deaf community because of the lack of access to information in their own language. Some countries have a sign for dementia in their own signed language, but this is dependent on their understanding about dementia (Young, Ferguson-Coleman and Keady 2018). It has also been suggested that being bilingual or multilingual may delay the onset of dementia by a few years. However, as dementia progresses many people revert to their primary language or experience ‘intrusion’ which consists of mixing the two languages (Khan 2011).

Some people with dementia have additional difficulties because they do not have adequate reading and writing skills in the national language of the country in which they live. These difficulties are not limited solely to people from minority ethnic groups. According to UNESCO, “A person is literate who can, with understanding, both read and write a short statement on his or her everyday life,” whereas UNICEF emphasises the ability “to use reading, writing and numeracy skills for effective functioning and development of the individual and the community” (UNESCO 2006).[11] However, a person may lack literacy skills but be able to communicate fluently or be literate but communicate non-verbally.

Digital literacy

People who are not ‘digitally literate’ (able to use computers, tablets and smart phones well) face similar difficulties to people who have difficulties reading and writing (Beresford 2019). This may include using email, Word, Excel or social media and searching for information on the Internet. Figure 3 below, from Eurostat (2018), shows the Internet use of adults (aged 16-74) in the European Union in 2016 in percentages. This clearly demonstrates the digital divide by age and educational attainment. Moreover, some groups of people are likely to have less access to Internet than others such as care home residents and some people know how to use it but are wary of it (Beresford 2019). In the UK, the Office for National Statistics reported that in 2014 only 37% of adults aged 75 years and over had ever used the Internet (compared to almost 99% of 16 to 24-year-olds). There were only small differences in Internet use between ethnic groups. The use of computers and the Internet for webinars or teleconferences may be a way to involve people who are not able to attend meetings (e.g. who have limited mobility or live in rural areas) but can also exclude people who are not able to use them or do not have access to them. It should also be borne in mind that people with dementia may develop difficulties writing and using computers.

Figure 3: Percentages of Individuals who used the internet on average at least once a week, by age group and level of formal education (Source Eurostat 2016).

In many cases, people who have problems with language and literacy are also living in socially vulnerable situations (e.g. in poverty and in poor housing conditions etc.). They have different needs and are at greater risk of health disparities. It is therefore important that their views are taken into consideration by researchers when designing and conducting research.

Years of education

Socio-economic status, education, language ability, health literacy, geographical location and being a member of the main ethnic group are all factors which may affect whether a person with dementia knows about research and how to get involved in a particular study in the context of PI or as a participant. According to Wilson (2000), people with higher levels of education sometimes have a greater awareness of societal problems and may be more likely to be asked to volunteer. This could be because they are often already involved in organisations, have high literacy levels and may have developed certain useful skills such as running meetings. They may therefore also feel more confident about their ability to contribute.

Older people on average tend to have had fewer years of formal, compulsory education compared to their younger counterparts. In the past, children could leave school at a relatively earlier age than the children of today and were sometimes allowed to leave school to take up employment, either permanently or periodically. Some went into apprenticeships and some had their schooling interrupted by world wars. Nowadays, in member states of the European Union (EU-28), the leaving age is between 15 and 18[12], with this age sometimes being dependent on certain conditions being fulfilled (European Commission 2016). People from some minority ethnic groups have had little or no formal education (Berdai Chaouni and De Donder 2018). Historically, and even today in some parts of the world, children with intellectual disabilities are excluded from the general education system, whether or not through choice, and placed in ‘special schools’ where the quality of educational attainment is often less than that provided in mainstream education (Right to Education 2018). People from traveller communities also tend to have low literacy and educational levels (Liégeois 2007, Condon et al. 2019). A 2011 census of England and Wales further revealed that only 40% of people from traveller communities had any formal qualifications.

The issues of self-confidence and how PI is portrayed and conducted therefore need to be addressed. People with lower levels of education may be reluctant to take part in activities which are perceived as highly intellectual or academic (e.g. working alongside researchers, helping develop research materials and being advisors in the research process). Approaches are needed which do not focus on written texts and preparatory reading, which take into account different abilities and ways of communicating and which are culturally sensitive and appropriate for the people involved.  

English as the working language for Public Involvement

At the European or international level, researchers often have a single working language and this is often English. In some studies, PI involving people with dementia is carried out at different levels. For example, there may be local/national groups in various countries and a broader group which brings representatives of these smaller groups together. There may, in addition, be one or two people who sit on a project steering committee. The people who contribute to PI in such projects other than at local/national level, and in many cases have more direct contact with the leaders of various aspects of the study, tend to be either native English speakers or people who have acquired a high level of competency in English as a second language. Some researchers and organisations go so far as to set a minimum level of English (e.g. B2) needed to join a patient PI pool or Patient Advisory Group, even stating a preference for people with a scientific or research background. B2 is the ‘vantage’ or ‘upper intermediate level’ of language competency described in The Common European Framework of Reference for Languages (CEF or CEFR). It applies to a person who is able to:

  • “understand the main ideas of complex text on both concrete and abstract topics, including technical discussions in their field of specialization.
  • interact with a degree of fluency and spontaneity that makes regular interaction with native speakers quite possible without strain for either party.
  • produce clear, detailed text on a wide range of subjects and explain a viewpoint on a topical issue giving the advantages and disadvantages of various options”.

Some researchers, in addition, have a preference for people who already have knowledge or experience of research, or experience interacting with different relevant stakeholders (e.g. in the domain of drug development, health technology assessment (HTA) bodies, regulators or payers).

As mentioned earlier, many older people did not have as many years’ formal education as their younger counterparts and in some countries, English is not the foreign language prioritised in schools. Moreover, a survey commissioned by the European Commission found that 62% of people in Europe over the age of 55 do not speak any language other than their mother tongue well enough to have a conversation (Eurobarometer 2012). In Hungary, Italy, the United Kingdom and Portugal, over 60% do not speak any foreign language and in Spain, Romania, Bulgaria, the Czech Republic and Poland at least 50% do not. This way of organising PI results in large numbers of people being prevented from taking part in research.

Recommendations for researchers

  • Try to create an atmosphere of trust and acceptance (favour approaches which encourage continuing involvement so that this atmosphere can be built up over time but do not discourage people who prefer more limited involvement).
  • Consider going out into the community in order to build rapport and trust with people who might feel intimidated about meeting researchers and attending meetings at a university.
  • Ask people in advance what they would like to know or feel they might need in order to make a decision about getting involved.
  • Be ready to propose possible support and information if people are not sufficiently aware of PI to know what might be helpful.
  • Recognise the value of augmented and alternative communication to supplement or replace speech or writing in order to facilitate PI.
  • Be proactive in trying to find people from different backgrounds rather than immediately approaching whoever is available. This may require extra time and funds (e.g. for translated materials or support with communication), which should therefore be planned for.
  • Plan for resources and time for finding people to involve in PI.
  • Avoid, where possible, the unnecessary use of academic titles (e.g. in German, Frau Professor Doktor Schmidt) in joint meetings between people engaged in PI and researchers (whilst respecting cultural differences in approaches to the use of titles and customary levels of formality).
  • Strive for flexibility in the promotion and organisation of PI so as to accommodate as far as possible the interests, needs and priorities of a diverse group of people with dementia.  This may involve some of the following:
    • Ensure that a high level of education, language or literacy is not needed to engage in PI by avoiding overly focusing on texts (e.g. develop interactive activities and small group discussions), bearing in mind that some people may find it easier to express themselves in writing.
    • Encourage people with dementia who are not highly educated and who may have problems with language and literacy to engage in PI (e.g. emphasising that high levels of education are not necessary and the importance to the researchers of their contribution, stating the wish not to exclude them on the basis of difficulties with language and literacy and asking what kind of support they might find helpful). 
    • Aim for accessible documentation and materials (e.g. use plain language, encourage people to ask if anything is unclear, and avoid jargon, abbreviations and unnecessarily complicated language).
    • Provide interpretation and encourage people to bring along a supporter. Bear in mind that some people may need more than one supporter (and budget for this).
    • Develop together with one or more people with dementia accessible materials to be used in PI. Discuss also planned activities and clarity of instructions or questions.
    • Test/pilot materials on a few people to ensure that they are fact accessible, regardless of people’s literacy, before using them with a PI group.
    • Adapt existing documents to make them more accessible such as providing clear explanations about what certain terms mean, about the topic of the research or about research goals and methods and using a relatively low level of language competence (e.g. B1 level).
    • Send relevant documentation well in advance and be available for questions or to meet in advance to discuss agenda items.
    • Plan for sufficient time during meetings to present the key issues in the documentation and for any questions.
    • Try to ensure that methods of communication are suited to the people involved. Where appropriate, for example, use verbal methods of communication rather than written or computerised techniques (e.g. in small, informal group discussions or one-to-one exchanges rather than written feedback of computerised documents by means of track changes).
    • Do not assume that everyone has access to Internet or computers and knows how to use them.
    • Recognise the added value of visual or pictorial communication where appropriate in the agenda, action points or minutes (e.g. through use of photosymbols for people with intellectual disability), bearing in mind that some people with dementia have difficulties understanding abstract images.
    • For large-scale European research projects, for which the working language is often English, take measures to increase possibilities for people who are not fluent in English to contribute to PI.
    • Challenge the reliance on English. Seek alternative approaches which do not rely on people engaging in PI having high levels of fluency in English.
    • Bear in mind the possible impact of involving people who are unable to contribute to discussions directly (in terms of group dynamics, maximising the contribution of everyone involved and the wellbeing of the individuals concerned and the other people involved).  Look for approaches and methods of support which are beneficial to the group, the individuals concerned and the PI process.
    • Seek feedback about people’s experience of PI and how it could be improved.
    • Consider the value of seeking feedback via pictorial or non-verbal communication methods in addition to verbal or written ones.


Recommendations for research ethics committees

  • Assess whether developed methods and materials have the potential to include people with different characteristics such as different ethnicities, ages, genders, disabilities, levels of education and socio-economic backgrounds.
  • Ensure that PI methods are appropriate to the study and will facilitate appropriate involvement of people with dementia.

Recommendations for funders

  • Ensure that PI methods are appropriate to the study and will facilitate the involvement of people with dementia and promote diversity.


People with dementia are often considered as being members of a homogeneous group but people experience dementia differently. Recognising such difference is important but categorising people with dementia into different sub-groups risks over-emphasising difference, drawing a boundary between ‘us’ and ‘them’ and thereby constructing ‘otherness’, which may be disempowering and lead to discrimination. In keeping with the concept of intersectionality, it is important to recognise the complexity of people’s identities and the dynamic and fluid nature of social and cultural categories so as to avoid make broad and sweeping categorisations.

Such categorisations may reflect certain personal characteristics but they also result from the way society is organised, often unfairly. Careful attention, consideration and appropriate methods are needed to prevent inequity in the context of PI. This also calls for researchers to reflect on their assumptions and possible restricted world view in order to gain insight into interpersonal and structural factors which might, in this case, hinder the promotion of diversity and inclusive research.

Ageism (public and internalised)

Older people are sometimes actively sought as participants for a particular study because of their age which actually puts them in the target population, but in many cases older people (especially the oldest old – see below) are excluded from research, both as participants and in the context of PI. Whereas there may be some challenges to involving older people in PI, a systematic review of older people’s patient and public involvement in health and social care research found that the benefits to research and for the older adults themselves outweighed the challenges (Baldwin et al. 2018). The prevalence of dementia increases with age so it is particularly important to involve a broad range of older people with dementia in PI. However, there is a lack of consensus or awareness about who counts as ‘older’ (i.e. older than what?).

In many countries, people aged 65 and over are defined as older people (Ouchi et al. 2017[13]), whereas the United Nations considers people over the age of 60 or 65 as older (or elderly persons) and those over the age of 80 as the ‘oldest old’[14]. AGE Platform Europe defines the 50+ age group as older[15]. Targeted measures to encourage greater involvement of older people with dementia in PI need to be clear about the target group. 50+ is perhaps too broad a target category to be effective in attracting a balanced range of people from 50 to over a 100. It would cover over 200 million people in the European Union extended over at least five decades[16].

The exclusion of older people from research, including PI, may be partly a reflection of ageism (i.e. defined by the World Health Organisation as stereotyping, prejudice and discrimination against people on the basis of their age[17]). Ageism, like racism and sexism, reinforces social inequalities by legitimising and sustaining inequalities between groups, and is more pronounced towards women, people with a low socio-economic status and people with dementia (Ayalon and Tesch-Römer 2017, WHO 2019).

The internalisation of ageist attitudes, whereby older people themselves see age in a negative light and adhere to negative stereotypes of older people, may affect their self-confidence and willingness to take part in PI activities. Beresford (2019) highlighted the influence of confidence and self-esteem for people of all ages involved in PI and this can be further hampered by negative stereotyping. The internalisation of various forms of negative stereotyping has been shown to affect people’s confidence in and actual performance on certain tasks. Barber (2017), points out, for example, that when older people are placed in situations which they fear might confirm the stereotype that older people are not cognitively capable, their level of performance often falls below their own potential. It is not known to what extent people with dementia are affected by possible internalised stereotypes such as the stereotype that they are not capable of expressing their views (i.e. simply because they have dementia).

Training and support

In some cases, people with dementia engaging in PI have prior or even current experience of conducting research. When this is not the case, some researchers feel that people engaged in PI should be offered training so that they can better understand the topic, the issues and implications of adopting various methods and what different tests and procedures involve, and thereby contribute more meaningfully to discussions. Brett et al. (2014) argue that training in research methodology would empower people to contribute to discussions surrounding the research design and to ask questions about the study rather than limiting their contribution to accounts of their lived experience.

By way of example, a study which set out to evaluate the impact of involvement of older people with dementia and age-related hearing and/or vision impairment in a multi-site European research study provided six hours of training (reduced from a six-day programme) broken up into manageable chunks of one hour “to support their input to the research” (Miah et al. 2018). The training covered research awareness, understanding the research process (including making sense of published papers), qualitative and quantitative methods, developing and evaluating interventions, health economics, ethics and governance.

Some researchers have found the Participatory Learning and Action research approach helpful in situations where they have no familiarity with the languages or culture of the target population, where the targeted minority groups lack trust in the researchers who are perceived as representing ‘the establishment’ or dominant cultural group, and where some of the potential participants have low levels of language and literacy (O’Reilly-de-Brún et al. 2016). However, this may necessitate fairly extensive capacity building skills and training, and a high level of involvement which would not correspond to the wishes and abilities of every person interested in PI.

There may be considerable differences regarding the content, duration and type of training offered to people with dementia engaged in PI. Such training may be helpful and beneficial if it is the right level for the people involved but if perceived in terms of training people with dementia ‘to bring them up to the level of researchers’, it could be construed as failing to value the knowledge that they already bring to the research process.

One of the key aims of PI is for researchers to benefit from the experience and insight that people with dementia have in relation to the research topic and the conduct of their study. Training is sometimes perceived as a threat to the input from people with dementia. Hayes et al. (2018), for example, highlight the threat of lay representatives over time becoming so highly informed that they identify with researchers to the extent that they become distanced from what would normally have been their own viewpoint. Goffman (1963) mentioned a similar situation in his seminal work on stigma in which he described situations in which people with a particular attribute (albeit a discrediting one), have a new career thrust on them, that of representing their ‘category’ (i.e. people who share the same attribute). In so doing, he suggests they break out of a relatively closed circle and start to have dealings with people from other categories, gradually ceasing to be representative of the people they represent (1963, p.39). Caron-Flinterman et al. (2007) use the term proto-professionalism to refer to the internalisation by patients of biomedical scientific language and principles. They point out that in the context of biomedical research, these people come to be regarded and treated as ‘real partners’ but are considered by some as having lost sight of the patient perspective.

As mentioned earlier, people with dementia are not expected to be representative (in the statistical sense) when contributing to PI. One of the aims in involving them in research in this way is to hear their views and learn about their experiences and perspectives as a person with dementia (in relation to all areas and stages of a particular study), not as a professional researcher. Ives, Damery and Redwood (2013) call this the ‘professionalisation paradox’ which they claim represents a fundamental tension between the moral and pragmatic motivations of PI in health research. They emphasise the importance of PI providing researchers with access to the lay perspective and suggest that even if researchers also had that perspective personally (i.e. themselves having the condition which was the topic of research), they would lack ‘outsider’ status and the ability to be a critical observer, free from the concerns and pressures of publication, generating income and building up an impressive CV.

Ives and colleagues also ask with regard to PI, “Why would we want to put unskilled lay people in a position where they can direct a process about which they know and understand little?” (2013, p.183). They go on to reflect on the benefits of training in empowering people engaged in PI and ensuring that they can legitimately enjoy an equal position of power alongside professional researchers but conclude that this jeopardises their ability to act as lay representatives. They do not claim that training causes people engaged in PI to lose their experience of illness or disability but that they become less able to authentically reflect the lay perspective and that their contributions become tamed and more in line with those of the researchers.  

As a counter argument, Staley (2013) points out that PI is a complex activity involving many different kinds of involvement, not all of which require specific training. For some forms of involvement, such as advising on how to reach or better motivate people to take part in research or how to make documents more accessible, a person who is ‘closer to the ground’ may be highly effective and would not necessarily benefit from formal training in research (Staley 2013). She further remarks that even when training is provided, in most cases, the aim is not to equip people to do research but rather to provide them with sufficient knowledge to understand the basics of research and to be confident that they can contribute to the study effectively.

A possible exception is where people engaged in PI actually ‘do’ research (e.g. help recruit participants, gain informed consent and collect and analyse data). This is in keeping with the concept of PI occurring along a continuum and with the possibility of people with dementia being co-researchers but they must have the necessary scientific rigour to conduct high quality research. In addition, researchers may need to obtain informed consent from research participants for their anonymised data to be shared with people engaged in PI and supervision would be needed (Hoddinott et al. 2018). It may also be necessary to address certain governance issues (e.g. in relation to terms of employment, legal issues and the right to payment (Hoddinott et al. 2018). Such involvement would need to be supervised because a full grasp of all the relevant issues and assimilation of knowledge required to conduct high quality research would normally take several months or years to acquire, not just several hours.

“Training in photovoice and time to get to know others in the research team made all the difference.  I liked that it was at a slower pace to start with. There was lots of time to understand it before we started using it” (Co-researcher with intellectual disability in Scotland).

It could be argued that training should be mainly about enabling people to contributemoreto the research process than solely accounts of their personal experience (Brett et al. 2014) and about helping them to communicate their opinions and perspectives more effectively to researchers (Morrow et al. 2010). In many cases, it would be helpful to reframe training so as to focus on the issue of providing support, which could include providing accessible information and helping people to develop skills that they might find helpful for PI. If researchers present PI as something for which training is necessary, this may give the wrong message to people who might otherwise be interested in it. It may be perceived as indicating that PI is only for highly educated people or for people who are good at and willing to study. People who do not fall into that category may feel discouraged or disempowered. The term ‘co-learning’ suggests a more constructive approach. This term is sometimes used to describe collaborative learning for which the goal is not to turn people into researchers but to help them to understand the research process and at the same time enable researchers to learn about a person-centred approach and PI (Hoddinott et al. 2018).

Could it be that the need for training to do PI indicates the lack of ability of researchers to make PI accessible?

Should the concept of training for PI be replaced by that of appropriate support and the opportunity to acquire useful skills?

Recommendations for researchers

  • Make it clear when looking for people to engage in PI activities that there is no age limit on involvement and that everyone’s perspectives and experiences are important (e.g. those of younger and older people, those of people with mild, moderate and advanced dementia etc.).  
  • Where training is a requirement for PI (for example in a co-researcher role), ensure this is done appropriately over an extended timeframe if required and with ongoing support and reinforcement of learning.
  • Avoid making training a requirement for others forms of PI.
  • Ensure that any training offered, is in keeping with the nature of the proposed activities, the requirements of the study and last but not least, the abilities and desire of the people involved to receive it, taking into consideration also the challenges that many people with dementia face.
    • Recognise that the suggested need for training may be off-putting to some people, especially those with lower levels of education or who lack confidence in their ability to learn, who have intellectual disabilities or who have difficulties with language and literacy.
    • Create the conditions in which people who wish to share their experience of dementia with researchers can do so without having to do any training.
    • Clarify the roles and responsibilities of the key researchers involved in the study as well as those of people engaged in PI (and how this corresponds to the possible need for support).
    • Learn how to make PI accessible to people who do not have prior experience or knowledge about research or who are new to PI (e.g. offer support to develop certain skills or confidence to express views to researchers or ask questions).
    • Bear in mind what different people who are interested in contributing to PI can realistically be expected to learn in the time frame available and in keeping with their individual capacities and interests. 
    • Consider the possibility of offering, and seeking support for, appropriate and accessible support (including explanatory information) in another language.

Recommendations for research ethics committees

  • Assess how relevant support is organised in the context of PI and whether it is suited to the needs and abilities of people with dementia.
  • Consider the role of people engaged with PI and ensure they are fully and appropriately supported, including for any essential training required.

Recommendations for funders

  • Offer the possibility to apply for costs to cover the promotion and provision of appropriate skills and information materials for people engaged in PI activities where appropriate.
  • Do not make training compulsory for most PI activities.


The prevalence of dementia increases with age so it is particularly important to involve older people with dementia in Public Involvement (PI) as well as younger people. The continued exclusion of older people from research, including PI, may reflect ageism which, like racism and sexism, reinforces social inequalities by legitimising and sustaining inequalities between groups. The internalisation of ageist attitudes, based on negative stereotyping and devaluing older people, may have a negative impact on confidence and self-esteem and interfere with readiness to be take part in PI.

PI enables researchers to benefit from the experience and insight that people with dementia have in relation to the research topic and the conduct of their study. It is sometimes suggested that people with dementia should be provided with training so as to have a sufficient understanding of research to be able to contribute meaningfully to discussions. A counter argument is that training may result in an over-identification with the position of researchers, make people with dementia less able to authentically reflect the lay perspective and jeopardise their unique outsider status and position as critical observers. We are not aware of evidence that this occurs. We suggest that the key issue is that PI work should be organised and presented in such a way that training is not necessary because it may be off-putting to some people and make them feel inadequate. Opportunities to develop relevant skills and to obtain clarification about relevant issues should be provided if and when requested.

Acting on input from PI activities

Managing different perspectives

A key principle of ethical research is that it needs to be carried out in accordance with rigorous scientific standards, but not everyone engaged in PI will necessarily be familiar with or fully understand those standards. This could be considered a limitation to the involvement of people with dementia in certain PI activities. It could also be a strength in that they can look at various aspects of the research with a critical eye, without subjecting themselves to the kind of self-censorship that might come with a more in-depth knowledge of research methods and methodologies. This brings us back to the issue of whether training is appropriate.

In their systematic review of the impact of PI on health and social care research, Brett et al. (2014) report two examples of researchers adapting their research design (i.e. not having a placebo arm) in response to PI feedback about what constitutes a more ethical study design from the lay/user perspective. They state,

“While this may have impacted on the robustness of the study design for researchers, it provided a more ethical study design for users. (…/…) This can provide an important challenge for researchers and highlight potential tensions between academic criteria of good quality research compared with the user perspective of what constitutes appropriate research. It is the researchers’ role to ensure any study is of high quality scientifically and to explain and justify the research design and negotiate changes that make the study more acceptable to service users without compromising its robustness or validity” (2012, p.641).

Drawing on the CIOMS guidelines for biomedical research involving human subjects[18], Emanuel et al. (2000) remind researchers that unsound research (i.e. not conducted in a methodologically rigorous manner) is unethical as it exposes people “to risks or inconvenience to no purpose”. Researchers therefore need to be able to balance the desire to respect and include input from the public and patients in their research against ethical concerns about the scientific validity of the research. Procedures are needed to manage such situations in a collaborative way that values the perspectives of all involved and, where possible, leads to a decision that is acceptable to all, but someone has to have ultimate responsibility. In most cases, that someone is the Principal Investigator although he or she may seek advice from an internal steering committee and people with dementia may be sitting on that committee.

“Lay people need to know enough about the research process to appreciate which aspects cannot be altered without harming scientific quality. But ultimately it is always the researchers’ responsibility to draw on their technical expertise to ensure the robustness of the research” (Staley 2013, p.1).

Ensuring involvement at a sufficiently early stage  

Commissioning structures can be very complex and bureaucratic with technically defined subject areas and methods which may preclude meaningful involvement, and often a lack of awareness and resistance from professionals (Tembo, Morrow, Worswick and Lennard et al. 2019). Not surprisingly, people with dementia are rarely involved in research commissioning (e.g. priority setting and developing calls for funding). According to Tembo et al. 2019), significant changes are needed regarding communication, practices, systems, structures and cultures which all hinder meaningful contribution of patients and the public to research commissioning. They point out that concerns have been reported, for example, about the lack of impartiality of members of the public, suggesting that the public lack objectivity, may be biased and may be influenced by personal interests (Tembo et al. 2019), thus calling into question the ‘rationality’ of involving the public in decisions about the allocation of funds for research. While Tembo and colleagues raise an interesting point, the accusation seems to be misplaced and overstretched. Surely other stakeholders and the researchers themselves are not impartial; this is not only a concern about the public. It thereby fails to balance these concerns against others linked to the potential bias and self-interest of researchers (e.g. linked to scientific curiosity, rivalry for scarce funds, the desire for peer recognition and the need to publish for job security), which might contribute towards tokenistic involvement of people with dementia.

People with dementia are also often absent at the stage of development of research proposals. In many cases, their involvement starts when the project starts because funding is not yet available for PI until funding has been awarded. Consequently, it may be difficult for them to shape the nature of their involvement and certain aspects of the research design if these have already been described in detail in the proposal and the accompanying description of activities, and for which specific funds have been allocated in the budget. If and when it becomes apparent that certain decisions have already been made, people engaged in PI may be reluctant to raise questions or suggest changes, anticipating that their input will be opposed, ignored or not considered a realistic option (Morrow et al. 2010). Some people with dementia who contribute in the very early stages of research may find that they are unable to continue once the research officially starts because their health or circumstances may have changed in the time between the application for funding and the start of project (which can be quite lengthy e.g. up to a year).

Reflecting on the issue of payment


In the vast majority of cases, people contribute to PI (and/or become participants in research) on a volunteer basis. According to Wilson (2000, p. 215- 216):

“Volunteering is any activity in which time is given freely to benefit another person, group or cause. Volunteering is part of a cluster of helping behaviors, entailing more commitment than spontaneous assistance but narrower in scope than the care provided to family and friends. (…/…) Some think the desire to help others is constitutive of volunteering. Others subscribe to the view that volunteering means acting to produce a “public” good: no reference to motive is necessary.”

Volunteering is common and somewhat of a tradition in some countries. A longitudinal survey of volunteering amongst some 30,000 people over the age of 50 in 11 Continental European countries found that it was most common in Sweden, Denmark, Germany, the Netherlands, Belgium, France and Switzerland (ranging from 14.4% to 21%) and less so in Spain, Greece, Italy and Austria (2.9% to 8.3%) (Hank and Erlinghagen 2010). People who volunteer to be involved in research do so for a variety of reasons. Some of these reasons may result in personal benefit such as increased self-esteem, social contact or learning new skills, without this necessarily being the driving force for such activities.

There is a complex relationship between volunteering and a range of factors such as the number of hours that people spend in paid employment, overall income, employment status and other responsibilities, such as caring (Wilson 2000). One European study found that people with higher levels of education and those not in paid employment were more likely to take up volunteering (Hank and Erlinghagen 2010) but again this is complex, not always linear and may be linked to certain types of volunteering. Data is lacking on how people with dementia hear about and come to be involved in PI and how they can be facilitated (i.e. what do they need to be on board, which methods of involvement are appropriate?). As the concept and recognised value of PI in research becomes more common, it will be important to reflect on how to attract people with dementia from diverse backgrounds and from countries where volunteering is not common, and to explore the challenges and barriers which hinder the participation of some groups of people.

Payment for Public Involvement

In the last few years, there has been a movement towards paying ‘patient representatives’ for their expertise, time and effort based on concerns about fairness and equality. An argument in favour of payment is that the expertise provided by people with dementia (‘experts by experience’) is of equal importance to that provided by experts in other domains (‘experts by training’). In this way, payment for involvement (as opposed to merely covering out-of-pocket expenses) is seen as formal acknowledgement of the equal value of the expertise provided by people with dementia and people with expertise in other domains.

Payment for PI is gaining ground particularly in the area of research into the development of new drugs as witnessed by policies and guidelines being developed by the EFPIA[19] (European Federation of Pharmaceutical Industries and Associations), as well as in the ‘Patient Pool’ currently being set up by the Innovative Medicines Initiative (IMI) which funds a lot of dementia research in Europe[20]. Aside from the issue of equally valuing the expertise provided by people with dementia, it is perhaps particularly important for pharmaceutical companies, who invest millions in dementia research and if successful make considerable financial profits, to demonstrate fairness in properly compensating everyone who contributes towards making the clinical trial a success.

An accessible report explaining to the general public and patients what they might want to know about payment for PI National Health Service-related, public health or social care research was produced in the United Kingdom in 2016 by INVOLVE[21].  The authors emphasise the importance of payment for PI in terms of recognition of time, skills and expertise, equality and helping ensure that a wide range of people have the opportunity to influence the research process. 

“It is good practice for organisations to pay you for your involvement in research. This is one way in which you can be rewarded for the time, skills and expertise that you contribute to the research process. Paying people for their involvement in research helps to support more equal partnerships between researchers and members of the public. It helps to support the inclusion of people who might not otherwise be able to get involved, whether for financial or other reasons relating to access. Consequently, it widens the potential pool of people who might influence the course of research” (INVOLVE 2016, p.8).

Payment of out-of-pocket expenses (e.g. for travel, accommodation and meals), including swift and easy reimbursement of costs, may make it possible for people on a low income to engage in PI and without this affecting their wellbeing. Many researchers/funders also cover the costs of supporters to enable people with dementia who need assistance for travel or during meetings to engage in PI.

Challenges linked to payment

There are mixed views in the literature and amongst those directly involved about payment for PI (Black et al. 2018). Many people with dementia have limited opportunities to earn money and people with limited financial resources might appreciate the opportunity to be paid for their contribution to research. Payment might also encourage people from more diverse backgrounds to engage in PI (i.e. people for whom payment would be an incentive, people who are perhaps less motivated by an awareness of social and health related issues and/or people would not normally do volunteer work).

There is also the issue of payment being interpreted as a sign of respect, quality and value. This has also been reported in relation to participation in qualitative research (Head 2009). However, offering payment for PI can also be perceived as devaluing volunteering. In an article on incentives for research (including financial payments), Grant and Sugarman state,

“Incentives may induce people to do the right thing, but for the wrong reason, and thus undermine responsibility, altruism and other important values” (2004, p. 722).

Moreover, payment does not automatically convey or guarantee respect for the person receiving it and their contribution. Likewise, voluntary involvement does not preclude respect for a person’s contribution or value.

It is important consider whether payment represents a financial incentive to motivate people to contribute towards research, thereby making the invitation more attractive, or amounts to coercion (an exercise of power resulting in people do things against their principles or better judgement) (Grant and Sugarman 2004). In the context of research, payment is sometimes considered as a potential means of coercion. Payment may be considered as involving ‘undue influence’ and thereby undermining consent, but this would not apply to PI which does not normally require informed consent as the people involved are not research participants. Moreover, as PI does not tend to involve the same risks as participation in research, especially in clinical trials, Millum and Garnett (2019) argue that payment for research participation is not a form of coercion that undermines consent, which necessarily involves threat, but that payment constitutes coercion as subjection[22]. As the issue of informed consent is not relevant to PI, it should not be necessary to be as strict about payment for PI as for participation in research.

In PI, payment tends to be understood within an economic paradigm, almost as a form of trade or payment in exchange for a service. Were it to become an obligation to involve people with dementia in PI, financial incentives might start to be used primarily to attract them to PI rather than in recognition of a fair exchange. Some similar issues might then need to be considered as for  participation in research. However, the issue of risk and burden is not comparable. Payment might nevertheless also run counter to the principle of justice in that more people with dementia who would welcome payment might eventually be involved in PI, their willingness to be involved exploited and a further imbalance/lack of diversity created. Nevertheless, it is probable that financial incentives to engage in PI would not consist of inducing people to do something to which they are adverse but more to do something that they might otherwise have had no particular interest in.

Moreover, not everyone who contributes to PI wishes to be paid. Payment suggests a commercial exchange of goods or services, accompanied by obligations to provide what the other party wants in return for the payment. People with low self-esteem may be wary of getting involved if payment is offered as they may worry about whether they will be able to provide what is required (Beresford 2019).

With regard to people who are on a low income or receiving benefits from the State and who might benefit from payment for their contribution, such payment may be problematic. It may, for example, jeopardise their rights to benefits, result in them having to make complicated or costly tax declarations or involve a risk of being accused of fraud (e.g. based on travelling to and contributing towards meetings and discussions despite having been declared unfit to work on the grounds of disability).

In smaller projects such as those conducted by PhD students or researchers in institutions with limited funds, the obligation to pay people for PI activities may be problematic and hinder potentially valuable research. Payment may result in competition, with people interested in engaging in PI being attracted to projects offering a better rate of payment. Competition can be positive but may not necessarily attract people with dementia interested in PI to the most worthwhile projects. Researchers involved in small-scale research projects who can perhaps only offer EUR 20 to EUR 50 per person per meeting may find it difficult to compete with offers of EUR 200-500 for involvement in a certain task. If PI is conducted, as generally recommended, throughout the whole research process, even small payments can add up and necessitate sizeable funds.

The significance of the amount paid also needs to be considered. Small payments may be considered insulting, especially when people are working alongside professionals who are perceived as being highly paid. The amount paid may also depend on who is conducting the research, the amount of funds available and the scale of the project. In its position paper on PI, drafted in collaboration with INTERDEM and the EWGPWD, Alzheimer Europe argued that if funds are available for the payment of external experts (e.g. fees to attend a meeting or daily allowances), the same money should be offered to people with dementia engaged in PI on an equal basis (Gove et al. 2017).

Input from people with dementia would be helpful on this issue to determine, for example, whether and if so on what basis people with dementia should be paid for PI, what amount they would find acceptable and reasonable, whether they would prefer a gift voucher or a donation to be made to a good cause in their name). It is unlikely that a decision could be made for every group of people with dementia involved in PI or for every research project but people with dementia need to have their say in the matter. When trying to decide what would be fair, this is sometimes phrased in terms of compensation according to ‘fair market value’. Some organisations have clear guidelines. The American Patient-Centred Outcomes Research Institute (PCORI) (2015, p.1), for example, recommends that payments for people engaged in PI[23] should reflect “the level of expertise, commitment, responsibility, the type of work involved, and the degree of participation contemplated” and should also take into account a range of associated factors such as the effort involved, a person’s experience, skill level, time for preparation, comparable levels of pay and responsibility of participating professionals, local wages and national restrictions and that this should be informed by principles of equity.

Recommendations for researchers

  • Involve people with dementia in PI as early in the process as possible (e.g. during the development of the research proposal and protocol).
  • Give people with dementia involved in PI prior to the official start of a research project an estimate of how long it may take before the research starts. Also, inform them of the risk that the project may not receive funding and hence eventually not go ahead. 
  • Explain at the start of their involvement how input from people with dementia will contribute towards the research and about any restrictions which might already exist (e.g. linked to funding, ethics approval, available resources etc.).
  • Explain that all ideas are welcome, that the researchers welcome criticism and suggestions, that not every suggestion will necessarily be acted on but that there will be transparency about how decisions are made and that people with dementia engaged in PI are key players in the decision making process. 
  • Avoid making global assumptions about whether or not people want to receive payment.
  • Clearly state whether PI for a particular project is on a volunteer basis or in return for payment (and in the case of the latter, if it is possible to decline payment).
  • Irrespective of whether participation in PI is on a voluntary or paid basis, provide a clear and understandable explanation of what costs can be reimbursed and/or what payment people are entitled to receive, how to go about obtaining such reimbursements or payments, and ensure that reimbursement or payments are made promptly. 
  • Develop a clear and transparent payment policy (if payment is envisaged) in collaboration with people with dementia.
  • If payments are made to experts by training for the same kinds of tasks as those being carried out by experts by experience, offer the same payment to each.
  • Consult relevant guidance on fair remuneration.
  • Ask people involved in PI, who were offered payment, whether they considered the amount fair and take this into consideration when discussing any future possible payments.
  • If payment is offered for PI, also provide information about the possible impact this may have with regard to taxes, benefits and certain rights or entitlements that a person might have (e.g. linked to disability, mobility or tax exemptions)[24].
  • Try to ensure that people engaged in PI can have available payments transferred directly from the research funds to a charity or worthwhile cause of their choice, without this necessitating any money passing through their own accounts.


Recommendations for research ethics committees

  • Ask researchers to provide details of pre-study involvement of people with dementia (e.g. in the topic selection, plans for PI and possible suitability of various methods for involving people with dementia).
  • Assess whether the dedicated budget and criteria for the reimbursement of people involved in PI are fair.


Recommendations for funders

  • Involve people with dementia in discussions about funding and commissioning.
  • Assign a score to the involvement of people with dementia in the writing of the research protocols when selecting projects for funding.
  • Provide a dedicated budget for the reimbursement of people involved in PI.


PI is carried out within the context of research which must fulfil certain criteria (e.g. for social value, safety, rigour and scientific validity). However, all aspects of research, including the PI component, must be ethically sound and this requires some degree of flexibility from all involved. Researchers nevertheless stand accountable for their research and need to be able to justify their decisions. PI needs to start from the very beginning, before key decisions have been taken about the study. People get involved in PI for different reasons. Some people with dementia are interested in doing PI work on a voluntary basis and others would appreciate being paid. In some countries, volunteer work is more common than in others but volunteering is an individual choice. Arguments in favour of paying for PI are often focused on concerns about fairness, equality and respect but this should not be taken to imply that people who do PI on a voluntary basis are being used or disrespected or that payment automatically translates into respect. Ideally, people should have a choice and any policy for payment (or reimbursement of costs) should be clear, understandable and transparent.

Monitoring and reporting 

Planning PI right at the start of a research project is important but does not in itself guarantee success. As PI is still in its early days, more work is needed to evaluate its effectiveness, justification, the appropriateness of methods used and its conceptual aspects, including ethical implications. With regard to the issue of involvement being meaningful, some form of monitoring and evaluation is needed (e.g. to check whether people with dementia feel that their voices are being heard, to check whether researchers are acting on the input they receive and to evaluate in what way and to what extent PI is having a positive impact on the research process).

Researchers are increasingly taking measures to report on PI and to evaluate it. However, whilst there may be some pressure on researchers to engage in PI, there are often little or no requirements for detailed reporting beyond the initial plans. In 2011, Staniszewska and colleagues developed the GRIPP (Guidance for Reporting Involvement of Patients and the Public) checklist, which represented the first international attempt to strengthen the quality of PI in research. In 2017, this was revisited and resulted in GRIPP2, a short form and long version of a reporting checklist for reporting of the context, process, and impact of PI in health and social care research (Staniszewska et al. 2017). They highlight the ethical imperatives of reporting research in such a way that it is understandable and possible for other researchers to learn from and build on the experience acquired by their peers.

Staniszewska et al. (2017) suggest that failure to report PI in study reports may be seen as a form of misreporting and misrepresentation. With regard to the ethical involvement of people with dementia in research in the context of PI, they also emphasise the need to develop a patient or service user version of GRIPP2. This would not only enable people with dementia engaged in PI to play a more active role in committees responsible for reviewing research reports but also to develop assessment criteria which reflect their priorities and concerns.

Ethics approval and legal agreements

The issue of ethical approval for PI is often unclear. Hardavella et al. (2015) raise the issue as to whether PI raises any ethical concerns for those involved. They state:

“They are actually acting as specialist advisers, providing valuable knowledge based on their experience of a health condition or public health concern; therefore ethical approval is not needed for the active involvement in research, e.g. helping to develop a protocol, questionnaire or information sheet, or for being a member of an advisory group or co-application” (2015, p.224). 

Many researchers have reported being obliged to obtain ethical approval for PI activities. Some have voiced the opinion that it would be necessary to seek ethical approval if discussions were audio-recorded or if they would like to publish an article about the PI activity in a peer-reviewed journal. There is even anecdotal evidence of ethics approval being requested for a person with dementia to co-author an article. Fortunately, there are also good examples of research ethics committees having reviewed outlines for planned PI work and confirmed that this did not constitute research and did not require ethical approval (Poland et al. 2014).

In an article by Di Lorito et al. (2017), a small number of studies were reviewed on the topic of ‘peer research’ and the European Working Group of People with Dementia (EWGPWD) contributed towards the interpretation of the findings. The members of the EWGPWD had no objection to meetings and discussions being audio-recorded and felt that it was important to contribute towards articles in peer-reviewed journals if and when appropriate. Their focus was more orientated towards empowerment than protection from harm or methodological issues linked to actual participation in research (as opposed to PI). They emphasised the importance of:

“human rights (e.g. autonomy and respect), justice and equity (e.g. opportunities for people with dementia to be invited to participate in research) and tokenism (e.g. relevance of involving people with dementia from the beginning of the research and not just in the process but also in the decision making e.g. around which areas of research should be prioritised)” (Di Lorito et al. 2017, p. 66).

Researchers who submit applications for ethical approval usually have to demonstrate that they have reflected on any issues which pose a potential threat to their own wellbeing (e.g. psychological, physical or other) as well as that of research participants. The involvement of people with dementia in research in the context of PI should in principle also not result in harm to themselves. Consequently, Principal Investigators responsible for seeking ethical approval should also reflect on the wellbeing and protection from harm of people engaged in PI in the project. This should not be taken to imply that ethics approval should be obtained for their involvement but that their wellbeing and safety should be equally considered.

As mentioned earlier in the sub-section on training, people with dementia can contribute to PI in different ways. In situations where people with dementia are directly involved in research tasks (e.g. conducting interviews, moderating focus groups or accessing data), they should have the necessary and appropriate skills and should be able to conform to the same ethical requirements as other researchers, namely with regard to the conduct of good science and the promotion/protection of the wellbeing of research participants.

Some researchers and organisations ask people who take part in PI activities to sign contracts covering issues such as confidentiality, intellectual property, data protection, financial compensation and travel expenses. This is particularly common in the field of biomedical research in which pharmaceutical companies are involved. These documents may be difficult to understand and quite daunting in general, but even more so for people who have cognitive and other impairments or with low levels of language and literacy.  In addition, some contain clauses about potential conflicts of interest and propose exclusivity clauses. Exclusivity clauses, especially when based on a narrow definition of conflict of interests, may seriously limit opportunities for PI and the independence of people with dementia involved. Guidelines on this issue were developed in 2018 by Myeloma Patients Europe, WECAN, PFMD and experts of pharmaceutical companies.[25] Some of these have been incorporated into the recommendations below.


Recommendations for researchers

  • Include details in the research protocol about how PI will be conducted (e.g. when, in relation to which aspects of the research, methods for involvement, funding required for relevant activities etc.).
  • When reporting on studies and in peer-reviewed articles, include details of how PI was conducted, with whom and what the impact of it was.
  • Be specific, when reporting, about the origin of various input from mixed groups  (e.g. was a particular point which led to an important change or decision made by a person with dementia, a carer or a member of the general public? Did different stakeholders disagree on particular issues or have different perspectives?).
  • Mark documents as ‘confidential’ if they are and remind people with dementia at the beginning and end of discussions if these were confidential (in addition to the usual requirement not to repeat issues raised in the context of PI outside of that context).
  • Offer or at least be prepared to talk through any documentation that people with dementia may be asked to sign and to address any concerns they may have.
  • Summarise key requirements (e.g. related to codes of conduct or practice) in any contracts that people with dementia are asked to sign, and refer the signatory to a readily available full document so as to keep the contract as concise as possible.
  • Plan and budget for pre-meetings if required. For example, PI involving people with an intellectual disability can include a meeting in advance of steering/advisory group meetings to talk through the agenda and plan involvement, resulting in increased confidence and active participation.

Recommendations for research ethics committees

  • Make it clear to researchers that although ethical research should include PI and that PI should be performed in an ethical manner, PI does not require ethical approval from a research ethics committeeper se(and explain any exceptions to this rule).
  • Ask in feedback how researchers will involve people with dementia in PI prior to the start of studies if this is not stated in applications.
  • Ask in feedback how researchers will involve people with dementia in PI prior to the start of studies if this is not stated in applications.
  • Ask researchers to include in applications details of PI which was conducted prior to review by the research ethics committee.

Recommendations for funders

  • Make detailed reporting of PI a requirement for funding (e.g. link it to a deliverable).
  • Do not request or expect exclusivity from people with dementia for PI activities.


It is important that Public Involvement (PI) is rigorously monitored and evaluated.  Often, there are no formal requirements for this. People with dementia need to be involved in this under-developed aspect of PI work (i.e. in the development of monitoring, evaluation and reporting tools). However, the concept of PI is not yet widely understood across Europe. Ethical approval is not required for PI because people engaged in PI activities are not research participants and are not providing data. People with dementia engaged in PI should nevertheless not come to any harm as a result of their involvement. Whilst ethics approval should not be required, their wellbeing and safety, as well as that of the researchers, should be given due consideration.

Contracts covering issues such as confidentiality, intellectual property, data protection, financial compensation and travel expenses are often long, complicated and difficult to understand in general and especially for people with cognitive difficulties and/or difficulties with language, literacy and education. Some contracts contain exclusivity clauses which limit opportunities for PI. Short accessible summaries and necessary support should be provided, and exclusivity clauses avoided for PI activities.


[1] Please see the INVOLVE website which contains a wealth of information on this topic:

[2] More more information and a guidebook, please see:

[3] Insert INTERDEM website link here.

[4] The review was of Patient Engagement but the authors’ definition corresponded to that of PI adopted in this paper.

[5] The issue of sampling for participation in research is addressed in Part 2 of this report.

[6] The phrase “knowledge is power” can be traced as far back as 599-661 CE to Imam Ali

[7] The context of Rose and Kalathil’s argument is co-production involving researchers, policy makers and non-White service users with mental disorders, the latter described as the “mad” and the “racialized mad”.

[8] Also with regard to the trade-off between inclusiveness and demandingness mentioned earlier (Jongsma and Friesen 2019).

[9] e.g in the form of epistemic injustice (see p.18).

[10] This has some similarities to Link and Phelan’s (2001 and 2006) conceptualisation of stigma involving the labelling of socially salient attributes shared by a group and a process of cognitive separation (i.e. making a distinction between ‘us’ and ‘them’), whereby those with the attribute are considered as in some way deviant.

[11] For further information about literacy and illiteracy, including definitions from several countries, please see Chapter 6 of the 2006 report commissioned by UNESCO:

[12] In 5 of the German ‘Länder’, the leaving age is 19. For a comparative table, including additional countries in Europe, please see:

[13] In Japan, proposals have been made to redefine old age, suggesting the term ‘pre old age’ for people aged 65 to 74 and old age for people aged 75+

[14] Please see

[15] Please see

[16] According to EUROSTAT figures from 2018 (see Age Platform Europe above)

[17] For more informaiton, see WHO section on website about ageism in healthcare

[18] CIOMS stands for the Council for International Organisations of Medical Sciences (1993)

[19] For more information about EFPIA, see:

[20] For more information about the IMI, see:

[21] INVOLVE is a national advisory group funded by the National Institute for Health Research (NIHR) to support public involvement in NHS, public health and social care research.

[22] the state of being dominated, controlled or influenced

[23] Described as engaged research partners in their document. See also “Working together with patients: principles for remunerating patients, patient organisation representatives & carers for work undertaken with the pharmaceutical industry” produced by EFPIA (June 2019).

[24] For an example of a document providing advice on payment to people in receipt of welfare benefits (INVOLVE 2018), please see:

[25] The guidelines are available at:



Last Updated: Friday 13 November 2020