Basket | Login


Informed consent to dementia research

2011: Ethics of dementia research

The necessity to obtain informed consent

Legal obligation

Within the context of medical treatment in the European Union, medical practitioners have a legal obligation to obtain consent before treating a patient and may be accused of physical assault should they fail to do so (Alzheimer Europe, 2009). In some countries, this obligation extends to medical research and in some countries there are either specific laws on research which cover the issue of consent or paragraphs in more general laws addressing the issue of participation in research. This is the case, for example, in Austria, Bulgaria Cyprus, Denmark, Estonia, Finland, France and Germany but the emphasis is clearly on medical research such as clinical trials, experimental treatment in hospitals or accepting treatment by a student doctor.

The obligation to obtain consent for participation in research is, in addition, covered in numerous codes and declarations such as the Nuremberg Code (1946), the Declaration of Helsinki (1964-2008) and the Additional Protocol on the Convention of Human Rights and Biomedicine concerning Biomedical Research (2005) to name but a few.

Informed consent therefore represents an autonomous decision made by a competent person to authorize medical staff or researchers to involve him/her in a particular study, which for certain types of research (such as medical research) is also a legal requirement. Informed consent is increasingly becoming an ethical obligation for all types of research (e.g. for academic recognition of the validity of studies or for funding purposes).

Ethical motivation

Buchanan and Brock (1990) describe the two main values governing the need to obtain informed consent as being 1) to promote and protect the person’s wellbeing, 2) to respect the person’s self-determination.

The issue of personal wellbeing, which is also linked to personal benefit, will be discussed in section 5 of this report). With regard to self-determination, this term describes the desire to make important decisions about one’s own life oneself. Generally speaking, the two are linked as it could be presumed that people tend to think that they are best placed to make decisions likely to affect their own wellbeing. However, this fails to be the case if they lack the capacity to judge what might contribute towards their wellbeing.

The terms autonomy and self-determination are sometimes used interchangeably but are actually quite different. Atkinson (2007) describes the difference between these terms as follows:

Autonomy is the capacity for self-government and self-determination: the ability to choose for oneself. (…/…) Self-determination requires an individual to have the capacity to formulate and carry out plans, desires and policies of their own devising. Self-government further requires the individual to take account of their own rules and values in making these choices. (Atkinson, 2007, p.83)

She further specifies that for a person to be able to exercise autonomy, in addition to the mental and physical capacities needed, there must be a social and political environment which recognises and promotes the value of autonomy and allows it to flourish.

From an ethical and legal standpoint, it is necessary to be able to assess whether informed consent can or has been given and this includes assessing a person’s capacity to make a decision about participation in a particular study.

Determining whether a person has the capacity to give informed consent

The components of informed consent

Informed consent can be considered to have been given if a person (or his/her legal representative) after having been fully informed about the study, having understood the information provided and being capable of making a decision on the basis of that information  freely agrees to participate in the study. Petrini (2010) describes the basic components for informed consent as being:

  • The possession of competence
  • Voluntariness (not having been forced to make a particular decision).
  • The provision of information (clear and truthful information, covering amongst other things details of any potential risks, benefits and alternatives to participants).
  • Enrolment (freely deciding to participate in the study on the understanding that the participant can withdraw at any time without having to justify his/her decision or suffering any negative repercussions).

Competence refers to the legal capacity to make decisions in a certain realm and may be legally defined or determined in court. This is usually but not always linked to cognitive capacity. For example, many people under the age of 18 may have the cognitive capacity to make medical decisions but are not considered legally competent to do so (DuBois, 2008).

There are different degrees of lack of voluntariness ranging from being subjected to research without having had any choice in the matter, being forced to participate on the basis of some kind of threat or deciding to participate on the basis of undue influence such as the need for money which results in overlooking the risks involved (DuBois, 2008).

It is the responsibility of researchers to ensure that potential participants have received all the information they need to decide whether to participate. Such information typically covers benefits, risks, inconvenience, tasks, the purpose of the study, participants’ rights and the fact that the study departs from ordinary medical practice (Miller and Wilner, 1974). There may be additional information that is specific to a particular study or type of research such as the possibility of receiving a placebo or some degree of invasion of one’s privacy. Such information may be provided orally but it is common practice to provide participants with a written participant information sheet and accompanying consent form.

Cognitive status and the ability to give informed consent

Part of the process of obtaining informed consent involves determining whether a person has the necessary competence. A person’s cognitive status or score on an assessment of mental status (such as the MMSE) cannot be taken as sufficient proof of their ability to give informed consent (Orwig et al., 2011).  The MMSE has also been criticised for its emphasis on a person’s current ability to recall or manipulate facts and failure to address life experiences requiring long-term memory (Hellström et al., 2007).

Moreover, it is generally accepted that a person who is competent for the purposes of decision-making has the right to make a decision which other people may consider irrational. Competence does not equate with perfect rationality (DuBois, 2008). However, some people are not considered as having the capacity to make decisions at all, even foolish ones (Feinberg, 1986).

The criteria for competence

The assessment of competence involves paying attention to four main abilities (Appelbaum and Grisso, 1988; Mental Capacity Act, 2005[1]):

First, the person must have sufficient capacity to understand the information. If the study involves a considerable degree of risk, more information must be provided, particularly about possible risks and benefits, and the potential participant must be able to understand such information.

Second, the person must be able to retain, use and weigh up such information long enough to be able to make a decision. In addition, they must also be able to understand what the decision is about, why they are being asked to make it, and what the consequences of making or not making that decision might be.

Third, possible benefits, risks and inconvenience linked to participating in research must be understood and weighed against the person’s own values and goals, which means that the person must understand how participating might affect him/her personally (High, 1992 and Stanley et al., 1984 in Olde Rikkert et al., 1997).

Fourth, the person must have the ability to communicate his/her decision.

Giving informed consent to research is a complex process requiring cognitive skills which gradually deteriorate as a result of dementia, making such informed consent difficult and in some cases even impossible to obtain. Many people with dementia will, at some point, lack the capacities for reasoning and deliberation needed to fulfil some or all of these conditions for most medical decisions that need to be made, including those related to research. On the other hand, many people with dementia have sufficient capacity to express their desire to take part in research and to engage in various consent discussions (Hougham, 2005). When determining whether a person has the necessary capacity, it is not the final decision that is important but rather the process that the person goes through to make that decision (Sessums et al., 2011).  

Exception to the necessity to obtain informed consent

One of the only exceptions to the necessity to obtain informed consent for medical treatment is that of the emergency situation in which it is imperative to act immediately in order to save a patient from imminent danger or serious harm. This would rarely apply to the research situation in which interventions are carefully planned in advance and designed to provide some kind of new knowledge. However, in certain circumstances it might cover the use of experimental treatments or procedures in the acute care or clinical setting (or at the scene of an accident by paramedics) for people who are unable to consent (please see following section on ethical issues for more details).

Obtaining informed consent directly from people with dementia

The term “competence” describes the capacity to complete a certain task. It is always task-specific in that it relates to the capacity to perform a particular task, at a particular moment in time and under specified conditions (Buchanan and Brock, 1990). The area of competence of relevance to the involvement of people with dementia in research is that of decision-making capacity.

People may have varying degrees of capacity. In the case of dementia, it may even fluctuate not only because of dementia but also due to a range of psychosocial, situational, medical, psychiatric and neurological factors (Holzer et al., 1997) However, whereas capacity is not an all or nothing matter in that people may possess varying degrees of capacities (with regard to a particular task), competence is. There is a threshold for competence with regard to a particular task which someone either reaches or does not as the case may be (Buchanan and Brock, 1990). So a person is either competent or incompetent to make a decision about participating in a study. The importance of making such a distinction is to determine whether the potential participant should consent to participating in a particular study or whether somebody else should make that decision on his/her behalf.  It is therefore relevant to the issue of informed consent.

Obtaining informed consent indirectly through advance directives 

A person who fully lacks the capacity to consent to research could be for or against participating in research in general. It may be possible to determine what his/her wishes would have been by consulting his/her close family and friends or an advance directive for research.

Advance directives were developed in the 1960s in the United States of America. They were originally intended to be written by competent individuals who wanted to express in writing their wishes with regard to medical treatment they might eventually need in the event of an accident or illness which rendered them incapable of exercising self-determination (Vollman, 2001). In this way, a person’s autonomy could be extended into the future well beyond the point that they were actually able to exercise it. The concept of the advance directive has been further extended to the research situation in some countries but not in most. Amongst those  which do not accept advance directives, some may accept consent to research from a proxy decision-maker (previously appointed by the person lacking capacity to make such decisions on his/her behalf). Advance directives may be legally binding or simply advisory depending on their legal status in each country and sometimes on the nature of the decisions to be made. Even in cases, where they are not legally binding, they may take on legal authority insofar as there is an obligation to consider the presumed will or previously expressed wishes of a person now lacking capacity.

Proxy decision making

Proxy decision making is often discussed in the context of healthcare decision making (e.g. linked to treatment or care decisions) and may be covered by laws or deontological codes. Proxy decision makers may, in some countries, be able to consent to research on behalf of people with dementia. If such consent can only be given by a legally appointed proxy decision-maker, this may slow down the recruitment process.  

In some countries, the appointment of a proxy decision maker is made by means of an advance directive whereas in others, it is a separate process. The principle of appointing a proxy decision maker is to give that person authority to make decisions on one’s behalf at some time in the future. This is different to the principle of the advance directive which is to make a statement of one’s wishes in future situations (Atkinson, 2007).

However, apart from the fact that people may use the advance directive to appoint the proxy, some people actually combine the two decision-making possibilities. This means that they express their wishes in the advance directive and also appoint a proxy decision maker with authority to override those decisions. This gives people the reassurance of having somebody who can be consulted in cases where there is some doubt as to how the wishes expressed in the advance directive correspond to a particular situation. It can also be useful in situations where there have been new developments or changes which might result in the person’s previously expressed wishes not being applicable to the current situation.

[1] of England

Information and capacity

The provision of information

It is the responsibility of researchers to ensure that relevant and adequate information has been provided in a manner that is suitable for the particular person, and that it has been understood and retained (Fondation Médéric Alzheimer, 2011). The provision of information in a way that the person does not understand may result in what Sessums et al. (2011) call “pseudo incapacity”.

The common practice of providing a written participant information sheet and consent form can be particularly helpful for people with dementia who may forget what they were told by the time they are asked to sign the consent form. The participant information sheet can also serve as a reminder throughout the course of the research of what the study is about and what is involved.

Research ethics committees typically require a number of issues to be covered in the participant information sheet. This might, for example, include what the study is about, what participation would involve, what is hoped to be gained from such a study, the fact that the person can withdraw at any time and whether the participants are likely to benefit personally etc. Please see appendix 2 for a list of possible issues to be covered in a participant information sheet). An important issue, which is often overlooked, is that regardless of any personal benefit which might possibly result from participation, the purpose of research is not to benefit the participants. Such information about the difference between research and medical treatment is of paramount importance when obtaining informed consent to research from people with medical conditions, particularly those for which there is as yet no known cure.

The capacity to understand the information provided

People with dementia who participate in research do not always fulfil the criteria for consent despite it having been obtained because they have not understood the information provided. Agawar et al. (1996) carried out an interesting experiment involving people with dementia who had already consented to and were participating in a clinical trial, as well as their carers. Responses to a questionnaire revealed that out of 15 participants, only 3 realised that they might be receiving a placebo, 8 that they could withdraw at any time and 7 that they were participating in a research study. It could be that some of the participants had since forgotten the information they were given and upon which they based their decision (which nevertheless indicates the need for an ongoing consent process). The researchers conclude that if one were to apply the criteria for consent to treatment to consent to research, very few people would be able to take part but that at the same time, people with dementia represent a vulnerable group and need to be protected from exploitation.

In another study, the researchers emphasise the importance of the stage of dementia. 415 participants who had already given informed consent to a low risk, longitudinal study were tested for their understanding of the information they had previously been given. 250 participants had very mild, mild or moderate dementia and 165 did not have dementia. The responses from the participants suggest that those with very mild to mild dementia had understood the information but that in those with moderate dementia one third demonstrated poor understanding. The researchers found that presenting the information twice, improved the rate of understanding amongst people with mild dementia but concluded, based on their findings, that for people with moderate dementia, a carer should be involved in the process (Buckles et al., 2003).

The two above-mentioned studies suggest that people with moderate dementia might not have the capacity to understand the information provided as part of the informed consent process and that for some, the presence of a carer might be beneficial. They also raise questions about researchers’ ability to assess capacity. However, although issues linked to capacity and the assessment of capacity may sometimes make it difficult to involve people with dementia in research, they do not rule out the possibility of such involvement.

The assessment of capacity

Tools have been developed to measure the capacity to consent to treatment and to clinical research such as the MacArthur Competence Assessment Tool for Treatment (MacCAT-T) and the MacArthur Competence Assessment Tool for Clinical Research (MacCAT-CR), which are sometimes recommended (Appelbaum and Grisso, 2001; Dunn et al., 2006; Kim et al., 2001). The MacCAT-CR  covers understanding, appreciation, reasoning and expression of choice. However, it takes 15 to 20 minutes to complete and requires substantial training to administer it and to ensure its valid interpretation. Jeste et al. (2007) have developed a much shorter instrument for the assessment of decisional capacity for clinical research called the University of California, San Diego Brief Assessment of Capacity to Consent (UBACC). They describe this tool as being easy to use and to score, taking less than five minutes to administer. The tool also enables researchers to identify participants who need more comprehensive assessment.

The MMSE, which according to Molloy and Standish (1997) takes about ten minutes to administer, is not always an accurate predictor of the capacity to give informed consent. Sessums et al. (2011) suggest that it may be a useful indicator in the case of people with an extremely low score and for those with extremely high scores, but for those in the grey zone in between those two extremes, it may have limited usefulness for the assessment of the capacity to consent.

In their discussion about tools to assess the capacity to consent to research and care,  Moye and Marson (2007) point out that there may be differences in the type and rate of impairment of people with different conditions and that consequently, some tools may work less well in some condition than others. They also draw attention to the way that the assessment of capacity is becoming a field of study in its own right.

As pointed out by Kim et al. (2001), an appraisal of cognitive capacity is not sufficient to come to a categorical decision about a person’s capacity to consent to research. For such a decision, a score obtained by means of an assessment instrument is just one part of the process of obtaining consent, the other being clinical judgement. Researchers therefore need access to and expertise in the use of assessment tools for capacity as well as clinical experience.

However, the traditional competency-based approach fails to take adequate account of the situational aspect of capacity and the importance of interdependence and relationships (Dewing, 2007). It has been suggested, particularly by qualitative researchers, that the criteria for informed consent to research are heavily influenced by research ethics committees, which tend to focus on bio-ethical principles, and that this may actually lead to the exclusion of older people with dementia from research (Dewing, 2007). Dewing suggests the need to develop and propose to research ethics committees alternative methods of obtaining informed consent which are neither exclusionary nor paternalistic.

Preserved abilities

As mentioned earlier, the assessment of capacity must lead to a decision as to whether or not the person is competent to consent to participation in research. A person with dementia may lack the necessary capacity to consent to research but retain the capacity to decide about various aspects of research and express preferences that are reasonable (i.e. compared to decisions on the same issue made by people with capacity). Kim (2011) highlights a particular study he carried out with his colleagues in which it was found that the preferences expressed by people with AD concerning different research scenarios (i.e. involving blood draw, clinical trial, PET scan and brain surgery) were similar to people in the control group for three of the scenarios. People who had more advanced AD were not reckless with regard to risk taking but actually more cautious than their less impaired counterparts.

This highlights the need to involve people with dementia in decision-making linked to research, not just by informing them of decisions and obtaining their assent but also in a more active way based on their remaining abilities. There is a risk that lay people who are not familiar with the concept of capacity in relation to dementia will not think it appropriate to involve people with dementia in this way.

Capacity for decision-making about end-of-life research

People with end-stage dementia are likely to have limited capacity to make decisions about some forms of research and/or have difficulty expressing their wishes. However, expression of their wishes is not impossible as not all people dying with dementia have end-stage dementia (Cox and Cook, 2007). In addition, some may have made advance directives in which they expressed a desire to take part in research or appointed a trusted person to make such decisions on their behalf.

In some cases, there may be fluctuations in participants’ physical and mental condition. Perhaps one day, they feel able and motivated to take part but not the next day. Some may be receiving highly burdensome treatment for co-existing conditions, be experiencing extreme fatigue and experiencing distressing symptoms. Addington-Hall (2002) questions whether it is ethical to ask them to participate in research at such a time.

The timing and form of consent

Prior consent

Informed consent should be obtained before any research procedure is carried out. Depending on the type of research being carried out, informed consent may be a one-off procedure and a person may be asked to confirm having given consent in writing. This is also a legal matter in that obtaining informed consent may serve as a protection against possible prosecution for assault. In this respect, prior consent is a legal requirement for participation in research.

However, people with dementia may forget having consented to a study and not understand what is being asked of them or why. They may not realise that they are free to decide not to comply with the researchers’ requests or to withdraw from the study. Some may even be unaware that such requests are linked to research rather than treatment.

Experienced and ongoing consent

People do not always realise how they feel about a particular intervention or procedure until they have experienced it, particularly in the case of repeated invasive procedures like blood tests but also for procedures which some people might find disturbing (e.g. in-depth interviews on sensitive issues). People’s opinions as to whether research involves minimal risk and burden may differ in that certain vulnerable populations, such as older people and perhaps also people with dementia, may have a different perspective to that of the researchers (Annas et al., 1986).

It may be helpful for participants to experience what is involved before they give full consent. “Experienced consent” involves obtaining informed consent in the usual way before the intervention or procedure and then seeking it again, perhaps a week later, when then person has had time to decide whether s/he would be willing to repeat such an experience. Another approach is to seek consent at regular intervals throughout the whole study (Milton, 2000; Pratt and Wilkinson, 2001). This has the advantage of reminding people with dementia that they are participating in a study and that they are free to withdraw at any time.

Olde Rikkert et al. (1997) tested out experienced consent in a sample of frail older patients attending a geriatric hospital in the Netherlands and participating in biomedical research. They identified three advantages to its use. First, the actual capacity to consent improved (i.e. to understand information about the study and weigh the possible risks and inconveniences). Second, it enabled a true assessment of risk and convenience. Third, it gave participants time to reflect before being asked to sign a written consent form. The researchers point out that it may be easier for a person not to sign a consent form than to withdraw from a study after having given written consent. This might particularly apply to people for whom it is important that others do not consider them as unreasonable or making a fuss. It could also be helpful in situations where people wrongly presume some form of obligation based on their experience of signing documents and making contracts in everyday life.

However, Welie and Berghmans (2006) draw attention to the lack of a control group in the above-mentioned study, which means that the improved capacity may have been due to factors other than the try-out of the study, such as the passage of time or having received the information twice. They also draw attention to a possible catch-22 situation[1] in that capacity is also needed to consent to a try-out of the study and participants may not have such capacity. They state that whilst the standard for capacity to consent to try out part of a study is perhaps lower than the standard for capacity to consent to participate in the complete study, the rationale for participating in a try-out study may be more difficult for participants to grasp.

Layered consent

Sometimes, participants would like to take part in a particular study but have considerable concerns about some aspect of the procedure (e.g. giving blood, authorizing the researchers to obtain genetic information or being interviewed in their own home). In some cases, it may be possible for the researchers to exempt some of the participants from some of the procedures. If so, “layered consent” might be possible. This involves participants consenting to certain parts of the protocol and not others (White, 2000).

Situations where rapid consent is needed

For some types of research (e.g. research into delirium), rapid consent is needed but problematic to obtain. According to Holt et al. (2008), participants in studies into delirium are likely to be considered as vulnerable due to impaired capacity to consent as well as the possible underlying causes for delirium. One such cause is dementia. Yet, delirium has an acute onset and tends to be short-lived (e.g. just a few days). Sufficient lucidity to consent (for people who would normally have such capacity) tends to occur towards the end of the period of delirium. It is therefore necessary to recruit participants rapidly. If a proxy is needed to make the decision, that person must be found and contacted immediately. Attempts to protect participants may be in direct contradiction with their right to be involved in research which is appropriate to their condition. According to Adamis et al. (2005 and 2010), stringent testing of capacity may exclude many people with delirium from research into delirium. Those who do take part are unlikely to be truly representative of people with delirium and this has implications for the generalizability of the results. In their own study, in which they compared different approaches to obtaining consent to research from people with delirium in an accurate medical ward, they found that using stringent criteria resulted in younger, less cognitively impaired participants. This may also apply to other forms of dementia research but in the case of delirium studies (and other research carried out in emergency situations), there may be insufficient time to obtain proxy consent.

In England, there are special rules in the Code of Practice of the Mental Capacity Act (2005) governing the participation of people unable to consent in “urgent treatment research” (Holt et al., 2008). In such cases, the researcher can obtain the authorisation of a medical practitioner who is not linked to the study (but could be the patient’s doctor) or of the appropriate body for such decisions.  Holt et al. (2008) recommend this approach, also in relation to non-treatment protocols involving minimum risk such as observational studies.

Another approach would be to try to obtain consent in advance from potential participants who are at risk of needing a particular urgent medical intervention or experiencing a particular symptom for which a study is envisaged (Kalisvaart et al., 2006). This might still be possible for many people with dementia or their legal representatives but such an approach might cause undue anxiety about the probability of a specific medical condition occurring.

Form of and procedure for consent obtained

Nowadays it is increasingly common for researchers to obtain written informed consent and in some countries, such as the United Kingdom, they are obliged to do so in order to conform to the requirements of ethics committees. In the Clinical Trials Directive (2001/20/EC of the European Parliament and of the Council of 4 April 2001), it is stated that a decision to take part in a clinical trial must be written, dated and signed by the participant or his/her legal representative. It is further stated that people who are unable to write may, in exceptional circumstances, give oral consent in the presence of at least one witness. This process may contribute towards researchers treating the consent process as a one-off procedure or formality.

On the other hand, obtaining written consent, or at least making a detailed written record of consent given verbally, may help ensure that all relevant issues have been addressed and consented to. There is much more to consent than indicating whether one wishes to participate in a particular study. For example, consent may be needed to interview people, observe them or subject them to tests, to record interviews, to quote statements made by participants in future publications and to use data collected in case of subsequent loss of mental capacity or death. Consent forms also serve as proof that participants have been fully informed which is part of the process of obtaining informed consent. Participants can indicate on the form that they have been assured that data will be anonymised, that they are free to withdraw at any time and that they have been given a participant information sheet, had the opportunity to ask for any additional information they may require and that if so, such information has been provided. This structured and detailed approach to obtaining informed consent is generally considered as being in line with ethical requirements for research. Please see Appendix 1 for details of the kinds of topics covered in consent forms.

Nevertheless, it has been suggested that older people and people with dementia might be anxious and feel insecure or threatened when asked to sign a consent form even if they are interested in participating. Some researchers have proposed alternative methods such as verbal and behavioural consent (Bamford and Bruce, 2000; Bartlett and Martin, 2002; Cowdell, 2006; LaRue and Markee, 1995). Possibilities for alternative forms of consent must also be considered in the case of people with dementia with additional impairments (e.g. visual impairments and intellectual disabilities).

Research ethics committees sometimes require researchers to follow a set procedure to obtain informed consent which is clearly geared towards quantitative research and not adapted to many studies adopting a qualitative research methodology. For example in qualitative research, researchers can legitimately develop research tools, questions and even the consent process as the study progresses but in order to obtain ethical approval for the study, they may be required to submit final documents such as interview schedules, participant information sheets and consent forms and to state explicitly how participants will be selected and recruited (Allbutt and Masters, 2010).  At the same time, the conduct of good quality research requires researchers to adopt appropriate methodologies and to apply them effectively. It is possible that overly rigid and restricted approaches on the part of ethics committees might interfere with researchers’ ability to achieve this or alternatively, result in unauthentic/misleading declarations which amount to paying lip service to the requirements of the ethics committees. This would surely not be in the interests of research participants.

Possible factors influencing informed consent

State of well-being

The performance of people with dementia on tests of cognitive performance may be influenced by a range of factors such as the time of the day, level of concentration and motivation. Cognitive processing may also be affected by a person’s state of general well-being (Damasio, 2000). Consequently, tests which are carried out with minimal regard for the individual may affect the results obtained and lead to the unnecessary exclusion of some people with dementia from research due to low scores on various tests.

For this reason, Dewing (2007) advises researchers to find out how potential participants present themselves when in a state of wellbeing. She states that it may be necessary to find out about this from other people although clearly some people with dementia may object to being discussed in this way. Nevertheless, this may be easier to do in some settings than in others. For example, in a residential care home, it may be easier to adapt the process of obtaining consent to the individual but less so if the person needs to travel to an assessment or research centre.

Nevertheless, irrespective of the setting and even in the absence of personal knowledge about each potential participant, researchers can take measures to create a pleasant environment, be attentive to how the person with dementia is feeling and be as flexible as possible in organising the timing and place of the consent procedure.

Researchers must also be careful not to inadvertently disclose a diagnosis of which the potential participant is unaware whilst discussing possible participation in a dementia study (Blanchard, 2008). The disclosure of a diagnosis of dementia should be carefully planned and adapted to each individual.


In some studies, participants are not informed about the deception until after the study at which point they have the opportunity to consent to the data being retained and analysed for the purposes of the study. This is known aspost hocconsent and the process of informing them as “debriefing”. It implies that participants are asked to give informed consent either after having previously consented on the basis of ambiguous or misleading information or after having unwittingly participated in a study.

If participants do not know what the study is really about, it could be argued that they have not given informed consent (Dresser, 1981) and that the principle of veracity has not been respected (Fry and Veatch, 2006). They should therefore be informed about everything that might affect their decision with regard to possible participation in the study (Baumrind, 1976). Obtaining post hoc consent merely solves the issue of whether it is ethically and perhaps even legally permissible to use the data obtained from participant. It does not address other issues linked to the well-being of participants linked to their actual experience of participating in the study, to the violation of their trust or to disrespect for their personal values.  The ethical issues and justification for the use of deception in research is further discussed in sub-section 5.2.9 which addresses issues linked to protecting the well-being of people with dementia participating in research.  

Coercion, power and undue influence

Increasing difficulties to provide informed consent but also to communicate their wishes during a study make people with dementia potentially vulnerable and at risk of coercion or exploitation (Fondation Médéric Alzheimer, 2011). Coercion in the context of research could involve one person intentionally making an overt threat of harm to the potential participant in order to obtain consent to a study. This is not limited to physical violence. It could also involve deliberately leading the potential participant to believe that his/her condition will deteriorate or that s/he will die if s/he does not take part in the study (Fawcett and Garity, 2009). Exploitation would involve mistreating or unfairly using a person for the benefit of other people (e.g. solely for the researchers’ own personal goals).

Research ethics committees try to ensure that potentially vulnerable populations, such as children, people with incapacity and people in circumstances which severely restrict their liberty, are not exploited or forced to take part in studies. In some countries, this is explicitly stated in legislation on research or on involuntary internment (Alzheimer Europe, 2009). The issue of vulnerability is covered in more detail in section 5.2.7).

Researchers and doctors may be considered as figures of authority in their particular domain and hence accorded trust, which gives them a certain degree of power. Participants tend to go along with the rules and methods proposed by researchers which should reflect accepted practice within the research profession. However, whereas in the doctor-patient relationship, patients expect to be provided with medical expertise and services corresponding to their needs,  this is not the case with the researcher/participant relationship in which the emphasis is on the creation of knowledge and not necessarily or essentially for the benefit of individual participants. (Benham, 2008).  In the case of clinical research in which the researcher is a doctor and the participant is a patient, it may be unclear what reasonable expectations would be. Patients may become fully consenting research participants without understanding the difference between research and treatment. They may even be participating based on blind trust in their doctor. As a person with depression commented in a focus group on involving older people in clinical trials,

“If it was my doctor suggested it: “will you try this?” I’d say yes, but if anybody else asked me, I would probably say no.” (Bartlam, Lally and Crome, 2010).

This highlights the important role that doctors can play in facilitating recruitment but raises questions about the inappropriate use of power.

However, power may also be exerted on individuals who are not generally considered as belonging to a vulnerable group. For example, researchers may obtain approval from a hospital or residential care home for a study involving interviewing or observing staff. Ideally, each member of staff should consent to their participation in the study and such a process may actually take place. However, there may be cases where members of staff feel under pressure to consent based on insecurity about their jobs or about future professional advancement within the organisation. In some cases, where a whole department is involved, they may be afraid of standing out as a trouble maker or of the consequences of having to be moved to a different department during the study.

There may also be reasons why carers would feel under pressure to take part in a study or to give proxy consent (e.g. if they felt beholden to the people organising the study for the support or care of the person with dementia or feared reprisals against the person with dementia if the latter was in residential care).

Sometimes, the involvement or presence of carers during the consent procedure may result in carers making certain decisions on behalf of people with dementia and consequently those people with dementia not exercising independent will or making a free choice (Agawar et al., 1996). Whilst this is most unlikely to involve deliberate coercion, it raises concerns about voluntariness and the extent to which the carer helps as opposed to decides on behalf of the person with dementia, particularly as the carer might not have the legal authority to do so. Also, it cannot be presumed that the carer has the best interests of the person with dementia at heart.

Financial incentives

Sometimes researchers offer participants some form of financial remuneration. This might be intended as compensation (for their time, for expenses incurred or for risks or harms), as a mark of respect or gratitude or in order to motivate them to take part (DuBois, 2008).  The actual amount paid is important and affects whether it is considered as undue influence or justifiable. For example, a payment of EUR 40 to a person living below the poverty line with no opportunity to earn money might lead them to participate and to overlook the possible risks involved, whereas for a person on a good salary, it would probably not.  If the amount is too high, the participants may be inclined to try to please the researchers by providing the information they think the researchers want.

In some studies, participants are offered access to medical treatment or care that they would not be able to finance themselves but consider necessary. For a person with good health cover, this would not even be an incentive but to someone with no health insurance, it might be a deciding factor.

It could be argued that certain benefits such as money, treatment or care are clearly beneficial to those lacking them. However, it has been suggested that participation solely for financial or other related gain is an indication that the research is exploitive and that the participant is being used “merely as a means to somebody else’s end” (DuBois, 2008). Consequently, when deciding what might be a reasonable amount to offer a person, it is important to bear in mind that it is relative.  There are likely to be differences between and within different groups of participants. On the other hand, it would be equally unethical to offer different amounts of money or benefits to different participants in the same study.

In their report on the ethical issues linked to tissue donation, the Nuffield Council on Bioethics (2011) suggests that altruistic donation prevents the exploitation of the poor and may help ensure good quality tissue. They report mixed views on the issue of ownership of the body with some people feeling that they “own” their bodies and consequently can sell it as one might sell parts of a car, and others feeling that the recognition of ownership rights amounts to an inacceptable objectification or commodification of the body.

Researchers sometimes offer donations to established charities or gift vouchers instead of money. These options make the exchange of actual money unnecessary and may appeal to the altruism of participants or place the emphasis on something pleasurable as opposed to compensation.

The offer of money or other kinds of incentives might also be offered to doctors for their assistance in recruiting patients for studies. This might create a conflict of interests between a doctor’s willingness to help researchers and his/her duty to protect the best interests of his/her patients.

Doctors and possible ambiguity in their role in the research process

Where research is being carried out (or even just proposed to the patient) by medical practitioners there may be ambiguity between the role of doctor and that of researcher. In many countries, doctors are ethically bound by the Hippocratic Oath which states that treatment that is given should be in the best interests of the patient. As Berghmans and Ter Meulen (1995) point out, “The aims of conducting research are not primarily, nor exclusively to benefit the individual patient who is the research subject”. They further state, with reference mainly to biomedical research, that the knowledge obtained from such research may be “beneficial for future patients but that there is seldom or never a direct benefit for the patient” (1995, p. 648). However, people with dementia should have the opportunity to contribute towards such knowledge in the medical domain as well as in other areas.

After informed consent has been given

Withdrawal from the study

It is understandable that some participants give informed consent and later find certain aspects of the study more unpleasant or burdensome that they had envisaged. Whilst researchers prepare a protocol which contains a step by step plan of the various procedures to be carried out, it is difficult to predict how people will experience or react to those procedures. For some studies, various parts of the procedure (such as the actual questions to be asked in an interview) are developed or adapted along the way. This is part of the research method adopted.

Even if the questions are known in advance, a participant may find the level of probing too deep and not wish to continue. S/he may also be disturbed by some of the questions and in some cases may experience distress even some time after the interview. There is a possible conflict of interests in that the researcher needs to protect the wellbeing of the research participant but the withdrawal of the latter from the study may result in inconvenience to the researcher or have implications for the analysis and presentation of the data.

Older people are more likely than younger people to withdraw from a study (especially those involving trial drugs or the effects of existing drugs) due to worsening health, complications, the effects of polypharmacy, changing residence, going into hospital and dying (Orwig et al., 2011). As most people with dementia are quite old, this must affect a great deal of dementia research irrespective of the additional factors linked to dementia such as confusion, loss of mental capacity involving difficulties with communication and complying with instructions etc. In studies using quantitative methods of data analysis, withdrawal from the study may affect the generalizability of the results, especially if the participants who withdraw are in some way different as a group to those who remain in the study (Orwig et al., 2011). Also a reduction in the number of participants involved in the study may affect the statistical power of the study, making it difficult for the researchers to claim that the results they found were not simply due to chance. On the other hand, statistical methods are available to deal with drop out from studies by adjusting for the missing data (Orwig et al., 2011). Nevertheless, there is a potential conflict of interests and a necessity to ensure that the wellbeing of participants is respected with regard to their possible desire to withdraw from the study.

Blanchard (2008) points out that there may be consequences for the participant of withdrawal from a study such as undesirable side effects or financial issues, and suggests that patients who withdraw from a study should be followed-up.

When the participant loses capacity or dies before the end of the study

A person may agree to take part in a study and may have the capacity to consent but then lose that capacity during the study. This raises a few ethical questions such as: Should the person who has lost the capacity to consent continue to be involved in the study, who should decide on this matter and what should happen to data relating to that person that has already been collected? Similarly, a surrogate decision maker who consents to a study on behalf of a person with dementia may also lose capacity during the study and hence not be in a position to ensure that the rights of the person with dementia are being respected and that s/he is still willing to participate. Such problems should not normally arise if the person responsible for giving informed consent states his/her wishes with regard to this matter before the start of the study. Some research ethics committees specifically ask researchers to explain how they will address this issue when they initially apply for ethical approval for a study. A similar problem arises when a person dies before the end of a study in that it must be decided what to do with the data that has already been collected from that person.

Consent to data or human tissue being used for further studies

When researchers collect data or human tissue for the purpose of a study, once the study has ended that material is no longer used. In some countries, researchers are required by ethics committees to destroy it within an agreed period of time and in a specified way. However, the same data or tissue could often be used for additional research purposes. This would reduce unnecessary duplication, prevent wasting the time of additional participants and prevent exposing additional participants to unnecessary burden. However, researchers do not necessarily have the right to use data or tissue for additional research purposes.

Where this issue is not covered by law, an ethical dilemma exists. It could be argued that such data or tissue was given for a specific purpose which the participant felt was worthwhile and on that basis was willing to give consent to its use. An exception might be the post mortem donation of brain tissue for research. The participant might not be in agreement with the aims of the additional study or even object to other aspects of the study such as the researchers, other activities carried out by the research organisation or moral issues linked to the study.

Seeking additional consent for the reuse of the data or tissue is certainly not a mere formality and can be a major setback in longitudinal studies. It involves contacting every person who participated in the original study. In some cases, this might be thousands of people, many of whom will have moved house, not respond to requests for additional consent or no longer be motivated to contribute towards research. As many people with dementia are fairly old, depending on the timescale involved, some will have died. In addition, some may have had the capacity to consent at the time of the original study but may now lack that capacity. The researchers would then have to contact the person’s legal guardian who may not know anything about the original study and in some cases, not have the legal right to consent. Moreover, whereas the consent procedure may have been thorough initially, as the revised procedure consists of consent to the reuse of data or tissue rather than to actual continued participation, there is a risk that the procedure may be somewhat superficial.

In some cases, human tissue may have been collected in the course of clinical care and not for the purpose of research but could, nevertheless, be valuable to researchers. According to Emanuel and Menikoff (2011), research may be hampered by lack of clarity concerning the use of leftover biomedical specimens from research or from clinical use and a simple one-off written consent procedure in hospitals might overcome this problem. In the case of clinical care, some patients may have undergone particularly burdensome procedures for the removal of the human tissue. It would be unethical to submit an additional person to the same degree of burden merely for the purpose of research. This may render some types of research impossible yet, paradoxically, research using those existing samples might eventually benefit future generations with the same condition. 

The involvement of, or impact on, third parties

Involving carers in the consent process

Carers sometimes act as gatekeepers in that they facilitate or obstruct researchers’ access to people with dementia. In some cases, they may even refuse to let the person with dementia take part in a study even though the person with dementia is displaying a clear interest in the study (Hellström et al., 2007). Whether they have the right to do so is a legal issue but such gatekeeping may be quite subtle and hence difficult to detect.

From an ethical standpoint, such action can be interpreted in at least two ways. Some carers may, for example, be exercising power and disregarding the wishes of the person with dementia. This might also be expressed in the form of not informing the person with dementia of opportunities to participate in research which have been communicated to the carer. This would be a case of failing to respect the autonomy of the person with dementia, not respecting him/her and not acting in his/her best interests. However, some carers may be genuinely trying to protect the person with dementia from harm and feel that they have a better understanding of the risks involved based on their capacity to fully understand everything involved and their in-depth knowledge of the person with dementia. The person with dementia may be unaware of the probable impact that participation in a particular study might involve but nevertheless have consented to it. If s/he objects to the interference of the carer, it may be a difficult situation for all involved.

Disturbance to people not directly involved in the study

In some cases, people may suffer certain inconvenience or disturbance for a study for which they are not the actual participants. An example of this would be a study which involved interviewing a person with dementia in their own home as the spouse of the person being interviewed could feel under pressure to tidy the house, offer refreshments and allow a stranger into his/her home.

In the healthcare or residential care setting, a study involving the use of surveillance or monitoring techniques for people with dementia might also affect the right to privacy of staff employed in those facilities yet they might not have a realistic option not to be filmed or monitored. Possible solutions include the temporary assignment of members of staff who object to a different unit (Sifford and Bharucha, 2010) or considering that a person who continues to work under such conditions has implicitly consented to the procedure (Carlson, 2001). However, in the first case, the employee may find the temporary assignment to another post disagreeable or fear negative consequences as a result of his/her decision. In the second case, it must be borne in mind that people do not all have the financial security to leave their job or sufficient opportunity and relevant skills to change their employment. 

When the person with dementia cannot give informed consent

In the case of consent to research of no direct benefit to the participant

Whether or not people who are unable to consent should be involved in research is an ethical issue which was addressed in the Council of Europe’s Convention for the Protection of Human Rights and Dignity of the Human Being with regard to the Application of Biology and Medicine (1997). The convention differentiates between research that is beneficial to the participant and that which is not, stating that the latter should not normally be carried out on people who are unable to consent.

For example, article 17, 1, ii states that the results of research must have the potential to produce real and direct benefit to the participant’s health if s/he is unable to consent to the research. Article 2 of article 17 further specifies that in addition to certain standard requirements, exceptionally and under the protective conditions prescribed by law, research which does not have the potential to produce results of direct benefit to the health of the person concerned can nevertheless be carried out subject to further conditions being fulfilled:

  1. the research has the aim of contributing, through significant improvement in the scientific understanding of the individual's condition, disease or disorder, to the ultimate attainment of results capable of conferring benefit to the person concerned or to other persons in the same age category or afflicted with the same disease or disorder or having the same condition;
  2. the research entails only minimal risk and minimal burden for the individual concerned.

The above guidelines are linked to biomedical research and the benefits mentioned are perhaps understandably linked to participants’ medical condition or health. However, not all potential benefits are linked to health, even in the case of biomedical research. It has been suggested, for example, that there may be benefits linked to increased social interaction, human contact and other people taking an interest in one’s problems, and that taking part in research may even give people a sense of meaning (Cassel, 1985). Berghmans and Ter Meulen (1995) describe these as “secondary interests” and assert that although they may be important, they are not sufficient in themselves to justify participation in a particular research project. For some patient groups, the “direct benefit” might occur many years later if an effective drug results from the study but in the case of dementia, this period of time might be too long for this to be a valid argument.

In the past, the terms therapeutic and non-therapeutic research were used and these were eventually replaced by references to direct benefit. However, the term direct benefit, as it is currently used, is perhaps somewhat limited and unrealistic. Furthermore, as different research projects have different objectives, the possibility for benefit and the type of potential benefit may differ from one study to the next.

Advance directives for research

Ensuring that the future participant understands the nature of research

Backlar (1998) highlights the necessity to ensure that the person who makes the advance directive for research fully understands the difference between being a patient and being a research subject. However, as advance directives are not necessarily drawn up in the presence of witnesses, have a variety of forms and may in some cases only be used several years after loss of the relevant capacity, it would be difficult to assess whether or not this was the case.

Making decisions about an unknown future activity

Consenting to future research in an advance directive may be more problematic than consenting to treatment in advance. As pointed out by Berghmans (1998), it is difficult to give consent for a future experiment which has not yet been devised and which, by the nature of research, is likely to be innovative.

Buchanan and Brock (1990) draw attention to the fact that advance directives do not leave any leeway for discussion contrary to contemporaneous decisions. With regard to research, the possible risks may be considerable and perhaps the actual condition or situation of the person who made the advance directive is different to that which s/he envisaged when it was written.

On the other hand, advance directives are written by people with capacity who can be informed of any possible risks linked to the use of advance directives and then make an autonomous decision.

The involvement of third parties

Backlar (1998) suggests that the person’s treating doctor should not have any connection to future research in which the person having made the advance directive for research might participate. However, there may be occasions when it might be beneficial for the patient to be involved in research of which his/her treating doctor might be aware (e.g. in the case where a patient is not responding to existing treatment). It is therefore important to be clear about the kind of involvement on the part of the doctor that would be considered unacceptable and whether there could be exceptional circumstances.

An alternative to making an advance directive is to appoint in advance (e.g. by means of an advance directive) a proxy or trusted to person with responsibility for decisions relating to participation in research or to be consulted with regard to the interpretation of the advance directive for research (e.g. in case of ambiguity regarding the person’s expressed wishes or for particular situations which the person might not have considered).

Personhood – deciding on whose behalf and with regard to which interests?

There are also a few issues linked to personhood and personal identity which challenge the whole issue of using advance directives, including for research purposes. For example, some theorists claim that the person who previously wrote an advance directive is the same person now as when s/he wrote it but his/her interests may have changed. Others suggest that at some point in time, the person can no longer be considered as the same person and consequently question the validity of the advance directive.

Experiential and critical interests within stages of one complete life

Dworkin (1994) proposes theintegrityview of autonomy. This approach considers autonomy as a reflection of a person’s integrity as opposed to being based on concerns for their welfare. He distinguishes between two types of interests: experiential interests (the things that people do because they like the experience of doing them) and critical interests (the things that if not satisfied would make a person feel that they were worse off in some way or that life had been wasted such as having a close relationship or accomplishing a particular task or duty).

Dworkin claims that people with dementia in the later stages have no sense of a whole life with a past joined to a future and that they cannot have the projects or plans of the kind that leading a critical life requires. Furthermore, they are no longer able to act in a way that would make life more or less valuable. Consequently, although there may be a conflict between a person’s precedent autonomy and contemporary experiential interests, there is no conflict with their critical interests as they perceived them whilst still competent. Whilst his views about interests and quality of life involve some degree of stereotyping and should be considered in that light, his conclusion is worth considering, namely that even though experiential interests seem to take precedence over critical interests in advanced dementia, it is no reason to ignore the critical interests a person had when competent.

Dworkin’s view of a person with dementia seems to be one in which the person at a particular stage of their life has dementia, but this is just one stage in their complete life which has already involved different stages. As such, the stage they are now in, is affected by interests and concerns which transcend that stage and are important for their life as a whole. As such, the competent and incompetent selves are one and the same person. The critical interests, which previously gave meaning and coherence to life, are still important, even if at this particular moment in time, the experiential interests seem to be more in the foreground. Dworkin’s account would seem to suggest that advance directives should still be respected even when the person has lost capacity because they are to be viewed as expressions of the critical interests a person has, which are relevant to their whole life. Failure to respect them would, in his view, constitute “an unacceptable form of moral paternalism”.

Continuity between former and current selves – becoming a different person

Parfit (1984), on the other hand, describes a kind of psychological continuity between former and current selves which varies in terms of similarity to a greater or lesser degree. He argues that the two selves may become so radically different that they can no longer be considered as the same person and hence, the former self should not have decision-making power over the current self. Berghmans (1998) points out with regard to this theory that the degree of psychological continuity would probably be stronger in the case of early dementia compared to advanced dementia. However, this approach would justify the use of advance directives more in early stage than late stage dementia which is not something that people consider when they write such a directive. Also, such views would reduce the possibility of people consenting to end-of-life research which, to use Dworkin’s terms, may be a critical interest for some people.

Protecting the former interests of a future “non-person”

Buchanan (1988) takes the view that an advance directive serves to protect the interests of the person with capacity who wrote it and not the experiential interests of “the non-person that succeeds them.” The interests of the “former self” survive and extend to concerns about the future self (even though this is no longer the self). Again, a distinction is made between the person somebody once was and the person they have become, which sometimes sounds remarkably inhuman: 

“This interest survives not only his loss of the capacity for exercising a right of self-determination: it survives the loss of personal identity that results from the most severe and permanent dementia. Jones no longer exists yet his interest in avoiding the prolongation of the life of the body that was his survives him. It is the (surviving) interest of a (no longer existing) self in what happens to something (a living, sensing body) that is not that self. (Buchanan and Brock, 1990, pp 165-166)

Alzheimer Europe’s position on the use of advance directives from 2006

In 2006, Alzheimer Europe produced recommendations on the use of advance directives by people with dementia in which we stated our view that consent to research expressed in an advance directive should be accepted as a valid expression of a person’s wish to participate in such research (subject to certain conditions being fulfilled).

Values history

As part of future healthcare decision making, some people write what is known as a “values history”. This is usually a written document in which people describe what is important to them, covering, for example, their values, beliefs and preferences.  As such it is a very personal document. A person’s views about participation in research could be included in such a document in varying degrees of detail. A values history can be written alongside or instead of an advance directive. Whilst a values history does not have any legal basis, it can be very useful in helping other people understand what is important to a person, particularly in situations where somebody has no close friends or relatives but also in the residential care setting. Such documents may therefore contribute towards self-determination and beneficence, and help ensure more people with dementia have the opportunity to participate in research if they so desire, at the same time avoiding people who would rather not participate, doing so.

However, the issue arises as to who should have access to a values history. The person who wrote it presumably realized that it would be read by people who did not know them sufficiently well to know what they would have wanted. On the other hand, it is likely that they had in mind professional medical and care staff. The person may have trusted somebody (e.g. the family doctor or a trusted person) with the tasks of giving the document to the appropriate person or people when needed. If not, there is a risk that the document will not reach the appropriate people or that the person’s right to privacy will be violated.      Proxy decision making

Assessment of risk

To fulfill their role of proxy decision maker, a person should ideally have at least some understanding of dementia (Werner, 2001) as well as an understanding of the particular study being proposed and of the risks involved. In relation to an assessment of the risks of a particular treatment, Karlawish et al. (2000) found that proxy decision makers for people with dementia differed in their assessment of risk depending on their relationship to that person. Adult children of people with mild dementia were willing to accept a greater degree of risk than were the spouses of the people with dementia. Muncie et al. (1997) also found that proxy decision makers tended to be overprotective. This may prevent important studies from being carried out. Some proxies clearly overlook possible risk, based on the belief that the study would not be allowed if it were not safe. This casts doubt on the ability of proxy decision makers, presumably with full capacity, to make sense of the information they are provided with (Sugarman et al., 2001). Where the intervention was minimal, such as one extra blood sample, and of minimum risk, some proxies even reported having forgotten having consented and not being sure what was treatment and what was research during their visits to the clinic. In addition, the responses of some reflected a belief in the goodness of research and of the doctors and institutions involved which does not reflect a great deal of objectivity in assessing the possible risks involved in a particular study.

Basis for decision making

As proxies are expected to act in the best interests of the person with dementia, this would not normally include consenting to research involving burden unless some kind of benefit was likely for the person with dementia. Yet the person with dementia may well have been in favour of participating in research despite certain types or degrees of burden or risk. Indeed, the opinions of proxy decision makers may be different to those of the people they represent. Often, they do involve the person with dementia in the decision making process (Karlawish et al., 2001; Sugarman et al., 2001) but this does not necessarily mean that they always make decisions based on the perceived views of the latter, even though they clearly should.

Proxy decision makers are expected to make decision which reflect what the person would have preferred or, for issues which the person had perhaps never considered, which correspond to his/her values and interests, and to what makes life meaningful to him/her. They are expected to make authentic decisions (i.e. decisions which are congruent with the values of the person on whose behalf they are consenting) (Brudney, 2009; Sulmasy and Snyder, 2010). Kim (2011) emphasizes that an autonomous decision can only be made by the potential participant him/herself but that a proxy could make an authentic decision on that person’s behalf.

This requires a great deal of knowledge about the person and also the ability to make decisions on that basis rather than on the basis of one’s own values and wishes. Some people may be better at this than others. Proxy decision makers do not necessarily have such knowledge.

A study into the use of pressure relieving mattresses involving frail older people revealed that only 18% of those able to consent refused to take part in the study compared to a refusal rate of 45% from the proxies of those who were unable to consent (Mason et al., 2006). In a systematic review of end-of-life treatment decisions, inaccurate assessment of wishes by proxy decision makers was as high as 33% (Shalowitz et al., 2006). Several studies into proxy decision making in the context of healthcare involving chronically ill or older people have concluded that family members are not very good at predicting or making the same decisions as their relatives (Ditto et al., 2000; Pruchno et al., 2006; Seckler et al., 1991; Upadya et al., 2002 in Atkinson, 2007).

A small scale qualitative study involving interviews with 49 proxy/AD patient pairs revealed that proxies often make decisions based on what they believe to be in the best interests of the people with AD, with altruism as a secondary reason (Sugarman et al., 2001). The results of this study also suggest that proxies are considerably influenced in the decision-making process by the hope that the person will benefit in terms of more time or a better quality of life. Even though they seem to understand that this might not be the case, many express the view that it is better to try something than do nothing and even something experimental rather than current treatment. This is clearly not based on what they think the person would have decided and much more on what might be best for him/her. On the other hand, the study also suggests that they are influenced by similar factors as the people they represent such as hope, desperation, trust and altruism. However, Atkinson (2007) draws attention to a study in which more successful judgements were made by proxy decision makers when they were asked to make a substitute decision rather than one based on “best interests”.

Willingness to cede control to proxies

Many people with dementia are willing to cede future decision making to their proxies or families. In one study of 149 patient-proxy pairs, 82.9% of people with dementia (with capacity to consent to research) were willing to let their proxies decide on their behalf for future research instead of proceeding according to current directions of their own (Stocking et al., 2006). Amongst those who were not as keen on letting others decide on their behalf, some expressed their desire to maintain autonomy and some mentioned study-specific issues.

In another study involving telephone interviews with 246 people who had a close relative with dementia and who had previously taken part in clinical research, 81% stated a preference for giving advance instructions rather than letting someone decide on their behalf (Wendler et al., 2002). At the same time, 80% were willing to let their family enroll them in research with potential benefit even if this contradicted what they had written in an advance directive. Very few (13%) preferred to simply let their families decide about research in general yet of the 246 participants interviewed only 16% returned a completed research advance directive to a research centre after the interview.

These studies suggest that there are differences between people with dementia in their preferences regarding the involvement of proxies in future decision making about research. They also indicate an acceptance on the part of most people with dementia that their proxies or families might make decisions which are not similar to those they would (or even did) make themselves. This reveals great trust in proxies and low expectations about their ability to make decisions that they would have made themselves. However, as revealed in the second study, very few people wanted to let their families decide about research on their behalf. The fact that many people state an interest in expressing their wishes in advance with regard to research, yet so few write advance research directives, is a clear indication that more work is needed to promote the use of such documents.

A German survey-based study involving 100 people with mild cognitive impairment (MCI) or mild AD, 99 relatives and 93 doctors revealed that people with MCI or mild AD had, on the whole, high levels of confidence in their own ability to make decisions and a considerable preference to make decisions themselves (especially social decisions) (Hamann et al., 2011).  Those who were unsure of their decision-making capacity chose not to participate in decision making. The results also suggested that both relatives and doctors had difficulty predicting the preferences of people with mild AD or MCI, and that relatives wanted to play a greater role in decision making than required. This study was based on treatment decisions and social decisions (such as whether to stop driving and whether to go into a nursing home), and not on research. However, research and the reasons for wanting to take part in a particular study might be more medically or more socially orientated, and this might be reflected in the level of interest in participating in decision-making and the extent to which the involvement of relatives is desired by people with dementia.  

Autonomy and exclusion

There is a risk that a known diagnosis of dementia may result in family carers being asked to consent on behalf of a person with dementia without having the necessary legal authority to give proxy consent to research and without attempts having first been made to assess the capacity of the person with dementia to consent (Fisk et al., 2007). Researchers may have difficulty (or anticipate difficulty) in obtaining informed consent from people with dementia to participate in research. Consequently, some researchers may be inclined to approach people with dementia who have a proxy decision maker with authority to consent to research. Others may seek to involve proxy decision makers due to concerns about the level of risk involved and possible litigation (Bravo, Paquet and Dubois, 2003; Vass et al., 2003).

This could lead to a lack of equity in that people with dementia who lack the capacity to consent to research and have no proxy decision maker may be excluded from research. These people also represent a sub-group of people with dementia whose voices may not be heard.

Burden on proxies

Often the proxy decision maker is the person’s spouse but this is not always the case. Not all older people have a spouse or even relatives who could be the proxy decision maker. Some people may prefer to appoint a close friend as their proxy. Proxies may feel under pressure to take on this role if asked.

In a study carried out by Sugarman et al. (2001) (please see next sub-section for details), some proxies reported that decision making was burdensome and this depended to some extent on the degree of risk involved, the level of invasiveness and the extent to which the person with AD was able to participate meaningfully in decision making. This in turn was often linked to the stage of the dementia and the extent to which the proxy or the person with AD had come to terms with the diagnosis. The burden on the proxy is often overlooked but ethical research involves protecting the wellbeing of all involved.

The existence of an advance directive for research could perhaps assist proxies in their decision making. The results of a comparative study by Stocking et al. (2007) of the usefulness of research advance directives for people with dementia and their proxies suggest that they did not facilitate enrollment, ease decision making or make proxies feel more comfortable with their decision. However, the researchers recognise several drawbacks to their study and recommend further research. Perhaps, the usefulness of value statements should also be explored.


Even when somebody with the legal authorisation to do so consents to research on behalf of a person with dementia, the ongoing assent of the person with dementia is invariably needed (Fisk et al., 2007).  In the Clinical Trials Directive (2001), it is stated that the explicit wish of an adult who lacks the capacity to give informed legal consent, but is capable of forming an opinion and understanding, to refuse participation or to withdraw from a study should be considered by the researcher. However, expressions of assent may sometimes be misleading.

Sugarman et al. (2011) analysed interactions during a consent meeting between participants, companions and researchers in the context of a study of genetic markers for sporadic AD. They found that the vast majority of contributions within the discussion by the participants were positive statements indicating agreement and approval. Those of the companions were much more varied. The authors suggest that these affirming statements were perhaps simply a means of engagement and not a deliberate cognitive act  or indication of the person’s preferences or views about the proposed study. As such this does not constitute authentic assent. Whilst it is not clear from the study whether the companions were proxies, this study does have implications for how people with dementia who have proxies are involved in the informed consent process. It also highlights the possible danger in placing too much emphasis on obtaining assent at the risk of failing to find more meaningful ways of ensuring that people with dementia really do agree to participating in research.

Cooperation and capacity of proxies

Ethical problems may arise if there are doubts about the mental capacity of the proxy decision maker and where there is more than one proxy decision maker (Bramstedt, 2003). In some cases, relatives may be opposed to decisions made by an appointed proxy decision maker, yet due to their relationship to the person with dementia, their cooperation or assent to the study may be essential for the effective participation of the person with dementia.

[1] This phrase is taken from the book “Catch 22” by Joseph Heller. The term is now generally used to describe a certain type of vicious circle whereby doing one thing may be mutually dependent on doing another thing but for some reason this is illogical, impossible or creates a dilemma.

Assessing the capacity to consent to research

A person should be deemed capable of consent unless proven otherwise.

The knowledge or belief that a person has dementia should be considered as reasonable grounds for doubt concerning his/her capacity to consent to research and to justify the assessment of his/her capacity.

A diagnosis of dementia (or the belief or presumption that a person has dementia) should never be considered as sufficient proof that a person lacks the capacity to consent to research.

Attempts should be made to ensure that researchers have access to relevant and appropriate tools for the assessment of the capacity to consent to research, as well as to training in how to use them.

Additional research should be carried out to develop and validate appropriate, short and reliable assessment tools.

Such tools should be validated on different groups of people with dementia.

The provision of information for the purpose of consent

Researchers should have the necessary skills to provide relevant information, communicate it in a manner adapted to the potential participant, respond to questions and facilitate the communication of the decision by the potential participant. Alternatively, they should appoint someone who has such skills.

Researchers should ensure that potential participants and their representatives understand the difference between receiving medical treatment and taking part in medical research, emphasising the fact that the main aim of research is not to benefit the individual participant.

Potential participants must be informed about other options (i.e. what treatment options are available, how alternatives might be better or worse than participating in the study)

When providing information, researchers should be aware of differences between participants, taking into account their cultural backgrounds and other factors such as level of education, linguistic matters, differences in perceived power, impairments and psychological state etc.

The use of visual or other aids should be considered if necessary.

Researchers should check whether the person with dementia has understood the information that s/he has been given about the study and his/her participation in it.

Potential participants should be provided with a participant information sheet and consent form. These two documents should be combined and each page contain a reference number for the study and a date.

Once signed, participants should receive a copy of the combined participant information sheet and consent form.

Willingness to and factors affecting consent to research

Researchers must ensure that people who take part in research agreed to do so freely after having been given all relevant information, having understood this information and having received satisfactory responses to any questions they may have had.

Researchers should be aware of the possibility of undue pressure from relatives to prevent or persuade a person with dementia to take part in a study.

Staff or students in a hierarchical relationship to the researcher should only be asked to volunteer to participate in research carried out by that researcher if adequate measures have been taken to address the issue of coercion with regard to their recruitment.

Potential participants should not be rushed into signing the consent form and in any case, be given at least one week to think about the implications of participating and to ask any questions.

Exceptions to the time limit for reflection should be considered in the case of urgent or end-stage dementia research.

Cultural traditions should be respected. In case of doubt regarding the way to approach members of a particular cultural community, researchers should seek advice from people who are representative of the target population of the study.

When seeking consent, particular attention should be paid to environmental factors (such as noise level, time of day and lighting), wording and sentence structure, the avoidance of medical jargon and the amount of information provided at a given time.

The need to obtain consent for emergency research (e.g. into life-saving treatment or procedures) involving people with dementia who are unable to consent should be waived but assent from relatives sought.

If relatives are not available to consent to emergency research, researchers should proceed but if relatives are available and object, they should not.

Ongoing consent and withdrawal from the study

Informed consent should be obtained at various intervals throughout the study. This provides regular opportunities for participants to withdraw from the study.

It should be stated on the consent form, close to where participants sign that they consent to participate in the study, that they are free to withdraw at any time.

It may be necessary to support the person with dementia in the decision-making process. In cases where s/he lacks the capacity to consent but is nevertheless involved in research as a result of proxy decision making, his/her assent should be sought at regular intervals

Participants must be aware that they can withdraw from the study at any time, freely and without having to provide justification for their decision. In addition, researchers should be attentive to signs of distress linked to participation and if necessary check with the participant whether s/he wishes to withdraw from the study.

Researchers should follow-up participants who withdraw from a study in case they are experiencing any negative effects from the study or resulting from their withdrawal.

Researchers need to be alert to signs of distress, be ready to stop the consent process with that person either temporarily or permanently, and to have envisaged a way to deal with distress following the consent procedure (e.g. by having a counsellor available).

Researchers should include a clause in the consent form in which the participant states whether s/he would like to continue participating in the study, should he/she lose capacity before the end of the study.

Issues surrunding loss of capacity to consent

A person who loses capacity during a study, and did not indicate prior to the study his/her wish to continue in the event of this occurring, should be withdrawn from the study.

Additional attention should be paid to a person’s willingness to continue participating in the study if s/he loses capacity after having given valid consent (but previously consented to remain in the study in such case).

Researchers should be encouraged to include a clause on consent forms where participants can state whether or not their data can be used (for the purposes of the study to which they are currently consenting) in the event of their future/further incapacity or death.

Involvement of third parties in the consent procedure

Attempts should be made to involve the spouse or partner in the consent procedure (e.g. through joint discussions).

If the study involves researchers entering a home, which is also the home of other people, the researchers should try to ensure that the other residents agree to this intrusion into their private sphere.

Objections and concerns from carers should be taken into consideration by the researchers out of respect for the carers and also as they may reflect legitimate concerns about the participation of the person with dementia in the study.

Researchers should not discuss with carers the participation of a person with dementia in a study if they know or could reasonably believe that the person with dementia would object to their being consulted.

If the person with dementia has the capacity to consent but the carer is strongly opposed to his/her participation, this should not prevent the person with dementia from participating in the study.

If the carer’s consent is to be sought as the person with dementia lacks the capacity to consent, care must be taken to avoid making the person with dementia feel infantilized, humiliated or devalued.

Advance directives for research

People should be encouraged to consider writing an advance directive for research (in which they state whether they would or would not like to take part in research and state any particular preferences).

People should be encouraged to consider writing a values history, which could be combined with an advance directive for research.

When consenting to research in an advance directive, an indication should be provided of the type and level of risk/burden that would be acceptable as the actual nature of the future research is unlikely to be known when consent is given in the advance directive.

The current wishes of the person with dementia should be considered alongside those expressed in an advance directive covering research.

Researchers’ own personal beliefs about personhood and the criteria for status as a person should not result in people with dementia having made an advance directive for research being denied the right to participate in such research.

A person’s treating doctor should not propose the involvement of his/her patient in medical research, in which s/he has a personal interest, based on the latter having made an advance directive for research.

In exceptional circumstances and where such participation might be in the patient’s best interests, the treating doctor should seek a second opinion from an independent doctor.

Proxy decision makers

The assent of a person with dementia to participate in research should be sought if s/he lacks the capacity to given informed consent.

If possible, the opinion of the person with dementia about letting someone decide on their behalf should be sought before approaching a proxy decision maker for this purpose.

When appointing a proxy decision maker, people should strongly consider discussing their wishes and views about research with the chosen person.

Proxy decision makers should try to make decisions based on their understanding of what the person with dementia would have wanted rather than what they feel would be in his/her best interests.

Proxy decision makers should strive for authenticity (i.e. making decisions which conform to the known values of the person with dementia).

A proxy decision maker should not have a significant connection to the project (i.e. s/he should not be under the personal or professional influence of the research team).

A proxy decision maker should not directly benefit from the participation of the person with dementia in a particular study.

The proxy decision maker should be informed that s/he has been appointed and given the chance to refuse such appointment. 

Proxy decision makers should be asked to declare any potential conflict of interests linked to providing consent on behalf of the person with dementia.

The person appointing a proxy decision maker should be aware of whether the proxy would have the legal authority to consent to research on his/her behalf and whether there would be any restrictions regarding the type of research to which the latter could consent.

Proxy decision makers should try to ensure the accurate interpretation of wishes contained in an advance directive or values statement if the person with dementia has made one.

Proxy decision makers should be involved in determining whether the research that is eventually proposed is in line with the wishes expressed in the advance directive.

If a proxy decision maker loses capacity, another person should be asked to take over the role as questions may arise during the study which necessitate further decisions being made.

If there are doubts about the capacity of a proxy decision maker when the initial decision is being made, an assessment of his/her capacity should be made.

Issues surrounding the further use of data

Researchers should obtain additional informed consent for any research involving the use of existing data or human tissue for purposes other than those for which informed consent was originally obtained.

Researchers should be encouraged to include a clause on the consent form where participants can state whether or not they agree to their data being used for possible future studies. A space should be provided for the participant to state any conditions attached to such additional consent. 

Restrictions on the right to participate in research

No individual should be denied the right to participate in research on the basis of their group membership.

The absence of perceived direct benefit should not serve as a justification to prevent people with dementia from participating in any kind of research.

A broader definition/understanding of the concept of direct benefit should be developed covering, for example, social benefit and personal satisfaction (e.g. linked to feelings and self-esteem). 



Last Updated: Thursday 29 March 2012


  • Acknowledgements

    The above information was published in the 2011 Report "The ethics of dementia research" as part of Alzheimer Europe's 2011 Work Plan which received funding from the European Union in the framework of the Health Programme.
  • European Union