Vulnerable People and Groups

The imperative for informed consent and the practicalities involved are set out in the Core Issues section (here). However, there are a variety of specific issues which can apply when working with vulnerable people and groups. It is likely that the inherent power imbalance between researcher and researched will be greater. Moreover, in some instances, there can be institutional pressures on participants, which may not be fully visible to the researcher. It has been argued that it is critical to encourage participants “to draw their own boundaries of privacy and emphasise participants’ prerogative to withdraw materials from the study at any time.” 1 Providing the opportunity to review materials and involvement is especially important in longitudinal studies. Phased consent may therefore be appropriate, providing the potential to renegotiate consent as required throughout an emergent research process.

However, whilst obtaining informed consent is especially critical when working with vulnerable people, it is also the case that with some groups – e.g. some people with learning disabilities – there may be potential problems of understanding what the research is about, what their role in the research will be and how the research will be used. Hence obtaining informed consent can be difficult and special care needs to be taken to develop appropriate strategies for communicating the implications of involvement in research.

When working with vulnerable people, there can be particular dimensions about which potential participants need to be informed. As set out under the ‘Anonymity and Confidentiality’ tab of this section, in some situations participation in research can make already vulnerable people more vulnerable. The researcher has an obligation to do everything possible to minimise any dangers of participating and an obligation to be completely frank about any potential danger that cannot be eliminated.

When working with some vulnerable groups it is likely that potentially distressing topics will be covered in the course of the research. Therefore extra care is needed to ensure that participants understand the implications of involvement in terms of the emotions that may be stirred.

In situations, where an adult participant may be unable to make a fully informed decision, or lacks the mental capacity to make a decision about whether to participate in ‘intrusive’ research, the Mental Capacity Act 2005 applies in England and Wales. Scotland has separate legislation. The Mental Capacity Act does not apply to clinical trials (CTIMPs). Chapter 11 of the Mental Capacity Act Code of Practice defines intrusive research as ‘research that would be unlawful if it involved a person who had capacity but had not consented to take part’. In accordance with s.32 of the Act researchers must consult with specified persons, usually those involved in the person’s care, to obtain advice as to whether the person should take part, and what their wishes might be if they had capacity to say. The consultee is NOT empowered to give a proxy consent. Research that falls under the Act MUST be approved by an ‘appropriate body’, essentially a ‘flagged’ NHS REC or approved social care REC.

It should also be recognised that some vulnerable people may find being asked to sign a consent form frightening. In this situation consideration should be given to alternative strategies for obtaining legitimate verbal consent.


Where research involves children (under the age of 16) consent should be obtained from parents or those in loco parentis. However, it is highly desirable that children themselves should be actively engaged in the consent process. It is arguable that children are capable of being partners in research and that they have rights to receive information, to be listened to, have their wishes and feelings taken into account and to give or withhold consent if judged competent to do so. From this perspective, assumptions about lack of competence to give informed consent can function to deny children valuable opportunities for involvement, thereby eroding their rights and excluding them from processes of developing shared understanding of social relationships. As with other vulnerable groups, there can be circumstances where denying children agency on grounds of their ‘vulnerability’ could result in the perpetuation of injustice (e.g. abuse). It may, therefore, be appropriate to seek consent from the parents of participating children and from the children themselves. However, if consent is gained from the relevant adult but the child clearly withholds consent or shows distress, the wishes of the child should prevail. Enabling children to engage in a meaningful way requires careful attention to practical considerations, including the use of information sheets and consent forms written in a child-friendly way. (See sample consent form with pictures.)

Ethics, vulnerability and diminished mental or cognitive ability

Informed consent requires sufficient competence, voluntariness and autonomy to understand what is involved. This can present particular problems when researching people with some levels of learning disability, those with dementia or diminished cognitive ability, or sense of self – e.g. some people experiencing mental ill-health or people under the influence of drugs. In these situations, judging competence can be very difficult. Factors that researchers can bear in mind when making an assessment include the individual’s abilities to:

  • communicate a preference
  • understand the purpose of the research
  • understand any potential risks and/or benefits to themselves and others in the future
  • understand the rights to refuse consent or withdraw consent
  • understand confidentiality and its limits
  • retain these understandings throughout the course of participation

It should be remembered that all these factors require specific assessment with regards to the particular study, rather than generalised assumptions of competence or otherwise. Researchers should also keep in mind that competence is not only a question of the individual’s ‘ability’; it is profoundly influenced by the person’s experiences and circumstances, linked to the issues of voluntariness and autonomy. Voluntariness and autonomy can be influenced by:

  • A lack of experience of meaningful, assertive decision-making
  • The powerless position of people within institutional and social contexts
  • Loneliness and a lack of meaningful social relationships
  • The increased likelihood of compliance and acquiescence when answering questions
  • Assumption that informed consent procedure is a ‘test’ requiring the ‘right’ answer

A basic, but important question to ask oneself is: can this person say “no”?

Context specific strategies should be adopted to maximise understanding of what is being proposed and the option to refuse involvement. In general, very clear, simple language should be used to describe the project. Michelle McCarthy’s work on sexual experiences and sexual abuse of people with learning difficulties illustrates the difficulty of gaining informed consent and the strategies she adopted.

Michelle McCarthy example

McCarthy elaborates on her attempt to gain informed consent in this study:

Because people with learning difficulties may find it difficult to understand what is being put to them and can find it difficult to make choices, careful attention needs to be paid to seeking informed consent to participate in this kind of process. With regard to the women’s consent to come to my sessions in the first place, I would explain what it might entail, then they were asked if they wanted to come. Almost all said yes. As a member of staff and therefore in a position of authority, it would clearly have been difficult for them to have said no to me. Because of the power imbalance that exists between service providers and service users with learning disabilities, there was little I could do at this stage to decrease this acceptance, other than make it explicit that they did not have to see me if they did not want to.

McCarthy, M. 1996. ‘Sexual Experiences and Sexual Abuse of Women With Learning Disabilities’.
pp.119-20. In Women, Violence and Male Power: Activism Research and Practice.
edited by M.Hester, L. Kelly and J. Radford. Buckingham: Open university Press

McCarthy also tells us that it was not easy to obtain informed consent from these women, as the words and concepts of ‘research or ‘methods’ or ‘policy’ did not have much meaning for them. She then used a very simple explanation: ‘I’m talking to women about sex, so I can learn more about it and to try to make sure women get the help they need. I might want to talk to other people and write about what I learnt, but I won’t tell them anything personal about you or use your name’. With this attempt, many women with learning difficulties agreed to take part in McCarthy’s research and all allowed her to tape-record their interviews.


Above text (with permission) from: Liamputtong, P. 2007.
Researching the Vulnerable. London: Sage. p.34

Consent by proxy

As noted above, research with minors requires parental consent, although it is highly desirable to obtain informed consent from the children themselves. Other categories of individual who may also be unable to provide informed consent include those listed above and also terminally ill patients with limited lucidity. The Mental Capacity Act 2005 will apply if the research falls into the definition of 'intrusive' research and it has been ascertained that:

  • the person lacks the necessary capacity to make a free and informed choice; and
  • it is deemed that the research is of sufficient importance to be undertaken without the individual’s direct consent;

In these circumstances there is no mechanism for obtaining a proxy consent by law. Instead the Mental Capacity Act requires that the researcher consults with the patient's care giver(s). The following questions should be addressed:

  • Who can provide proxy consent on the person’s behalf? i.e. make judgement of
    • the person’s likely preference;
    • whether the research is in the person’s best interests.
  • Who thinks they should have a right of veto?
  • Does the person still have to provide assent, and how would this be demonstrated?

Most typically the next of kin or another family member would be consulted. Alongside this, although it is not required by law, ethical best practice suggests that the participant should also provide assent. If assent is to be sought attention should be paid to determining how would this be demonstrated and recorded.

Arguably, there can be a strong ethical imperative to do research in some situation where informed consent cannot be obtained directly from the individual concerned:

… what is immoral is neglecting or avoiding to do research with this population simply because it is impossible to obtain consent in the usual way. Such an attitude denies this group the possibility of any improvement in care that might result from our work.

Morse, J. M. 2000. ‘Researching Illness and Injury:
Methodological Considerations.’ Qualitative Health Research 10(4): 538-46

Institutional Issues

Institutional contexts require special care with regard to ensuring the voluntariness of agreement to participate in research, due to deep-rooted, habitually reinforced compliance. In this regard the most obviously relevant situations are prisons, care homes and other residential settings; but the institutional dynamics of compliance can also be in play in situations such as outpatients’ clinics, or any other situation where a person’s welfare depends, or is perceived to depend, on the continued goodwill of a caregiver.

Prisons literally, and other institutional settings effectively, provide a ‘captive audience’ for researchers. This can be attractive to researchers because of the easy to ‘motivate’ participants and closed research environment. However, the institutional context should never be allowed to blur the issue of voluntariness. Involvement should always be entirely voluntary. Moreover, people in institutions should never be chosen as a consequence of the convenience of their situation: research in such situations should have a demonstrable intention to benefit those researched. Furthermore, it should be kept in mind that people in institutions are likely to be vulnerable for ‘personal’ as well as ‘institutional’ reasons. Prisons have a high prevalence of issues likely to have a bearing on people’s capacity to make an informed choice about involvement, such as mental health problems, substance abuse, learning disabilities, literacy problems, language problems and ethnic and cultural minorities. Whilst those in hospital or care home settings will have a high prevalence of mental, physical or age related health problems and frailties.

As far as possible, researchers should ensure that the institutional context does not compromise the individual’s right to make a free choice about participation. Refusal should not result in any form of sanction, adverse criticism or the loss of privileges. However, it must be recognised that the institutional effect is pervasive and that subtle and not-so-subtle institutional pressures may be at work. As illustrates in the diagram below, the researcher may not necessarily be aware of the pressures to which potential participants are subject:

the researcher has limited control over pressure exerted by institutions on those being researched

In institutional settings, where conformity and compliance are rewarded, people may not feel that they have a real choice. Older people in care homes may feel disempowered and give consent as a matter of habit. Prisoners are aware that behaviour is continuously monitored and assessed, and that this can have very real consequences. Less subtly, access to residents may be filtered by staff who select those they think will give a positive account of the institution, or will be able to tell the researcher something interesting. Institutions may also control access to individuals on ostensible grounds of the individual’s wellbeing, e.g. withholding access to people where research is deemed by them to be controversial or having the potential to ‘cause distress’. While these decisions may, in some cases, be well founded, researchers should not necessarily acquiesce immediately. Consideration should be given to whether, for those people in institutions the greater risk can be in not taking part in the research. For example, without the research, is there a genuine risk that people are at risk of poor services, neglect or abuse? Further, it is arguable that excluding these people from the research process only serves the status quo of institutions; assists institutions to protecting themselves from scrutiny; and serves to reinforce constructed “vulnerability stereotypes”. A focus on individuals’ capacity could be more helpful.

Researchers must be alert to the well-being of residents in institutions. Mistreatment in institutions can take many forms, including psychological and verbal abuse, physical assault, neglect, stealing and sexual abuse. In this regard, researchers should never agree to total confidentiality as there is an overriding obligation to alert authorities if there is evidence of mistreatment.


1 Daly, K. 1992. ‘The fit between qualitative research and the characteristics of families’, p.10 in Qualitative Methods in Family Research, edited be J.F. Gilgun, K.Daly and G.Handel. Newbury Park, CA: Sage publications.



↑ Top of Page