A Delphi survey of practitioner’s understanding of mental capacity

George Clerk (Buckinghamshire New University, Wycombe, UK)
Jason Schaub (University of Birmingham, Birmingham, UK)
David Hancock (Buckinghamshire New University, Wycombe, UK)
Colin Martin (University of Hull, Hull, UK)

The Journal of Adult Protection

ISSN: 1466-8203

Publication date: 10 December 2018



The purpose of this paper is to present the findings of a study considering the application of the Mental Capacity Act and Deprivation of Liberty Safeguards (DoLS). Practitioners from a range of professions were recruited to provide their views of how to respond to a variety of scenarios. GPs, nurses, social workers, physio/occupational therapists and care assistants were recruited to participate.


This study used the Delphi method to elicit participant views and generate consensus of opinion. The Delphi method recommends a large sample for heterogeneous groups, and round one had 98 participants from six different professional groups.


Participants did not respond consistently to the scenarios, but disagreed most significantly when patient decisions conflicted with clinical advice, and when to conduct a capacity assessment. These responses suggest that clinical responses vary significantly between individuals (even within settings or professions), and that the application of Mental Capacity Act (MCA) is complicated and nuanced, requiring time for reflection to avoid paternalistic clinical interventions.


Previous studies have not used a Delphi method to consider the application of MCA/DoLS. Because of this methods focus on developing consensus, it is uniquely suited to considering this practice issue. As a result, these findings present more developed understanding of the complexity and challenges for practitioner responses to some relatively common clinical scenarios, suggesting the need for greater clarity for practitioners.



Clerk, G., Schaub, J., Hancock, D. and Martin, C. (2018), "A Delphi survey of practitioner’s understanding of mental capacity", The Journal of Adult Protection, Vol. 20 No. 5/6, pp. 174-186. https://doi.org/10.1108/JAP-05-2018-0009

Download as .RIS



Emerald Publishing Limited

Copyright © 2018, Emerald Publishing Limited


The Mental Capacity Act (MCA) 2005 sought to provide a framework for decision making by health and social care providers and professionals, relating to adults who may not have the capacity to make decisions for themselves. Prior to its implementation, there was insufficient legal guidance in situations requiring self-determination; for example, an adult lacking capacity as a result of dementia who is unable to make decisions regarding her care. Although previous decision making was guided by common law, practice in health and social care was largely influenced by a paternalistic culture that reduced the agency of patients (Samuel, 2014). In an attempt to protect individuals and provide a clear legal framework to guide practice, the Law Commission initiated a process that culminated in the MCA.

Outline of the MCA

The MCA applies to people over the age of 16 in England and Wales where there are concerns about a person’s decision making ability. It is underpinned by five enduring principles of care, seeking to place the individual at the heart of decision making. These five principles are: the presumption of capacity; supporting individuals to make decisions; the right to make an “unwise decision”; the “best interest principle”; and the “least restriction principle”. Section two of the MCA requires professionals to presume that a person has capacity unless there is reason to believe otherwise. The legal presumption in favour of capacity is based on functional terms, including the ability to understand, retain, and evaluate relevant information, as well as being able to communicate any decision (Willner et al., 2013). The best interest principle (Section four) requires the care provider to act in the best interest of the individual instead of expedience or budget. The best interest principle also obliges care providers to encourage the participation of the individual in any decision about them. When a lack of capacity is proven, the individual without capacity must be given the necessary support to make a decision. In such circumstances applying a best interest decision is crucial to providing appropriate care. However, this means that decisions can be contentious as the response can vary widely depending on the decision-maker (Martin et al., 2012).

The 2005 European Court of Human Rights ruling on HL v. United Kingdom marked a watershed in mental capacity law and policy in the UK. This case found that a man had been deprived of his liberty, because he did not have capacity to decide whether to be admitted to a hospital, and he was held against his parents’ wishes. The UK was deemed to have violated article 5 of the ECHR and concluded that English and Welsh law were insufficiently robust to meet the requirements of the law guarding the right to liberty and security. As a result, the Deprivation of Liberty Safeguards (DoLS) were inserted into the law, making the rights of patients an increasingly important consideration for health and social care.

A Select Committee post-legislative scrutiny report indicated that there was lack of awareness and knowledge about the MCA (House of Lords, 2014). They noted that a prevailing culture of paternalism and risk-aversion had prevented the Act from becoming embedded into practice (House of Lords, 2014). After consulting with a range of stakeholders (including adults with learning disabilities and advocacy groups for people with dementia), the overwhelming evidence showed a range of concerns: poor implementation; low awareness; a lack of understanding; and ongoing tensions between the MCA’s empowering influence for patients and the safeguarding responsibilities it places on professionals (House of Lords, 2014).

The evidence from the report suggested a range of practice dilemmas, which influenced the study outlined in this paper. This study has sought to understand the experience of health and social care practitioners using the MCA and DoLS. The Law Commission (2017) also suggested that the DoLS is not fit for purpose, identifying that professionals lack understanding of its application as well the complexities associated with applying the procedure. Parliament is currently considering how to alter these procedures, although changes have been slowed down because of the focus on Brexit (McNicoll, 2017). Despite the potential change to the DoLS, much of the current procedures proposed are anticipated to remain the same (Age UK, 2017; Spencer-Lane, 2017).

After a review of the MCA DoLS, the Law Commission (2017) has concluded that it is not fit for purpose and proposed “Liberty Protection Safeguards”. Included in the recommendations are provisions for enhanced rights to advocacy, greater prominence to the rights of persons to be deprived of liberty, widening the scope to cover 16 and 17 year olds and simplifying the best interest process. Such concerns highlight flaws in the current law, and the challenges that practitioners have to contend with. A gap seems to exist in the literature about studies on the impact the law has had on practice and this study is an effort to contribute to bridging the gap. The challenge service providers have to contend with is translating the notion of the individual, i.e. patient or other service user (PSU) as any other rights bearing individual whose rights and liberties are not lost simply because of the condition, from theory to practice.

Service providers have a duty to ensure that PSU’s exercise the right to make their own decisions. However, where they lacked capacity, the best interest principle must underpin any decisions made on their behalf. The notion of best interest is not new, but for many years it has underpinned a paternalistic. According to the Law Commission (2017). “[…] health and social care professionals and the Court of Protection often failed to consider the person’s wishes and feelings, and that the concept of best interests was often interpreted in a medical and paternalistic sense”. In health and social care, the concept of paternalism has played a key role in the relationship between the patient or service user and the doctor or other professional proving a service (Komrad, 1983). It implies the physician makes decisions for the patient based what he or she believes is in the patient’s best interest, so they health system has thrived on the notion that the physician or professional knows best (Murgic et al., 2015). So while the notion of autonomy of the individual is a legally protected value, it is important to understand what the challenges are in practice, where the old ideas of best interest must be abandoned in favour of the wishes and preferences of the patient.

Research strategy

This study sought to understand health and social care practitioners’ knowledge of MCA/DoLS; how these frameworks are applied in practice; to determine practitioners’ consensus about good practice within the relevant policy frameworks; and sought to map out the understanding of representatives from the different areas of practice (Boulkedid et al., 2011). To gather their perceptions, participants were presented with everyday scenarios in hospital and social care settings and asked how they respond. This enabled the researchers to identify key differences in staff understandings of safeguarding and how the differences may impact on practice. The views of a range of professionals were sought in the study, including nurses, social workers and healthcare assistants and doctors.


A Delphi survey was used to study how practitioners understand and use the MCA. This method was chosen to engage both experts and stakeholders, without relying on one individual’s expert opinion (Olaf, 1967). A Delphi method typically elicits the views of a group of experts, who are individuals with knowledge of the subject. It is an iterative approach to collecting information with the anonymous results of the first round being shared with participants prior to the second and subsequent rounds (Skulmoski et al., 2007). This process allows a Delphi to develop convergences of opinion and build consensus. Surveys are generally used to identify “what is,” but the Delphi technique goes further by attempting to address “what could/should be” (Miller, 2006 cited in Hsu and Sandford, 2007). Because of the iterative process, it helps to transform opinion into group consensus, and broadens knowledge on an issue of interest (Hasson et al., 2000). While the Delphi method is the preferred group technique for consensus building, Kendall et al. (1992) cited in Okoli and Pawlawski (2004) have also pointed out this approach can use differences of opinion to inform further scenarios. Using the Delphi can test whether there was consensus on the understanding and practice of the MCA and whether consensus could be built, if found to be lacking.

Delphi methods can be used for both qualitative and quantitative data collection. Data from the first round of a Delphi can be analysed either by qualitative coding or statistical summarizing (Skulmoski et al., 2007). However, by focussing on consensus building, the method sacrifices the best opinion for a watered down version of it. This is not unique to the Delphi technique as other forms of group work such as nominal groups (Powell, 2003, p. 377) use structured meetings to elicit the views of experts. Using scenarios for decision making has the potential for exposing variations in decisions among professionals and their colleagues given the same circumstances. This has been demonstrated in studies using the vignette method such as the Hughes’ (1998) study of drug injecting and HIV risk and safer behaviour, Reader et al. (2017) and impossible decision making and the risk of trade-offs in ICU.

Investigating the MCA in practice is a complex issue, and required the participants from a range of professions and different areas of health and social care. It reflects the range of areas in health and social care that the law on mental capacity affects and what happens in practice. The Delphi Survey allowed for this complexity and meant a broad understanding of the reality on the ground could be reached. Participants in this study could be in different locations and still participate in the research because of their knowledge and practice experience. This allowed the participants a degree of anonymity, allowing them to express their views free of influences. Hanafin (2004, p. 10) suggests that anonymity encourages experts to make statements on the basis of their personal knowledge and experience, rather than a more cautious institutional position. Anonymity is a feature of strength because it allows participants to express their views without the influence of dominant characters and the inability of the researcher to trace back responses to an individual leads to a lack of accountability (Sackman, 1975).

Ethical issues considered in the research strategy were approved by a Research Ethics Committee. The study involved eliciting from respondents potentially sensitive information regarding their understanding and practice in the context of the Mental Capacity Act 2005. The sensitive issues included the potential for revealing poor practice. Key issues considered included confidentiality and anonymity.

Composition and sampling

Delphi methods utilise a panel of experts with specialist knowledge and qualifications as well as participants who are experts by experience (Iqbal and Pipon-Young, 2009). However, the idea of experts is problematic, and has been described as misleading and overstated with a potential for bias (McKenna, 1994; Sackman, 1975; Strauss and Zeigler, 1975). Keeney et al. (2001) have argued, “Simply because individuals have knowledge of a particular topic does not necessarily mean that they are experts” (p. 196). In an attempt to ensure participants had a working knowledge of the issue, this study recruited health and social care practitioners that use the MCA and DoLS in their practice. This provided a varied group covering the range of experiences of people working with the legislation. Participants were experts because of their ongoing practice experience of using MCA/DoLS, and not because they held leadership roles in their organisation.

Participants were purposively recruited. The aim was not to find a representative sample, but to find frontline staff with knowledge and experience using the legislative framework of the MCA. The study recruited people with first-hand experience of using the MCA regularly whilst caring for individuals in either hospital or social care settings. Snowballing was also used to expand the sample, allowing participants to invite their colleagues with relevant experience (Habibi et al., 2014).

It was important that the sample was a heterogeneous group, including a diverse sample of professionals that covered the range of scenarios that health and social care workers experience; whose practice involved decision making using the MCA/DoLS. Where there is a homogenous sample, Skulmoski et al. (2007) have argued that a sample size of 10–15 may yield sufficient results. But a heterogeneous group requires a larger sample, as the heterogeneity delays the important step of building consensus, creating a longer process of analysing data and verifying results (Delbeq et al., 1975 cited in Skulmoski et al., 2007). The study recruited five groups of professionals and included 50 nurses and 30 care workers, ten social workers, five occupational therapists and five doctors to participate in the first round of the survey. They were from acute settings and residential care for the elderly. The number of respondents dropped significantly in the second round, with only 12 respondents.

Even with such a large sample, the response rate in the second round fell drastically. It is generally agreed that experts who participate in such studies are hard to engage with because of other commitments (Hsu and Sandford, 2007). Ludwig (1997) suggests a drawback to using Delphi methods is that the questionnaire method may significantly slow the process, with several days or weeks between rounds (Hsu and Sandford, 2007). In addition, the iterative process causes further delays, posing a challenge to Delphi investigators to speed up the process. After several e-mails to participants, the response was still very low, which suggested the interest and commitment had waned. Hsu and Sandford (2007) recommend telephone contacts to overcome the problem however, this could easily lead to coercion, which would diminish the voluntary nature of the process. However, in this case non-respondents were slowing the process down, so several weeks after the deadline, the round had to be closed. Hence the low response rate.

The study sought a diverse array of situational contexts, from comparatively minor to life-changing and even life-ending decisions, because of the wide scope of the MCA. Because the topic considered how the MCA applies across a wide range of circumstances, as a result the sample included professional and non-professional staff such as nurses, social workers, care workers, physiotherapists and GPs (Alghrani et al., 2016, p. 312). The sample included this range because they all use the MCA in their daily practice.

Recruiting a diverse sample required engagement with a range of organisations. Organisational gatekeepers were able to identify relevant managers (from both acute and care home settings). These contacts were able to recruit members of staff for participation. Participants were invited to attend a workshop, and asked to complete the study questionnaires. This approach was used because of the advantages of face to face interviews following the first round, which helped increase the response rate (McKenna, 1994 cited in Hasson et al., 2000). Face to face encounters are time consuming, but significantly improve the quality of qualitative data.

The questionnaire required 30 min to complete. Both an online and in-person version were used, and each iteration was open for 6 weeks, with weekly participation reminders.

Questionnaire design

The questionnaire presented several scenarios describing practice situations, with questions requiring participants to demonstrate knowledge and understanding of the MCA and to describe how they would respond to these scenarios. The survey included questions related to each scenario, exploring the application of the MCA/DoLS and participants’ understanding.

Because the findings are so indelibly linked to the scenarios, it is necessary to outline them here. There were five scenarios in the first iteration of the questionnaire, and are outlined in turn below:

Scenario one:

Described a patient with a learning disability in hospital being treated for chest injuries. She was in distress, causing her to interfere with treatment. The scenario suggested the staff thought there was a need to sedate her in order to allow treatment to be provided. This accompanying questions sought to explore if participants would consider sedating this patient irrespective of her capacity, to establish how staff deal with clinical situations where a patient’s behaviour, attitude, decisions, and decision-making capacity presents a risk for the patient’s own safety.

Scenario two:

Described a 15 year old child in hospital for arrhythmia refuses treatment. Her parents agreed with clinical advice that she needed treatment, but the girl refused. The key issue was to find out how participants managed the different wishes, and how they recommended to treat her, since, due to her age, she was not covered under the MCA.

Scenario three:

Presented a situation involving a patient with a history of anxiety and paranoia who refused a life-saving blood transfusion, stating that he “did not want anyone’s blood”. This sought to gather participants’ responses when capacity and mental health issues are both present in a clinical situation. Would participants assess the decision making capacity of such a patient, or defer, as is recommended, to mental health legislation? Furthermore, would it make a difference if the patient refused a blood transfusion on religious grounds?

Scenario four:

Presented a widespread care home situation where a new resident arrived but did not to settle, and immediately insisted on returning to her home; notwithstanding that she lived alone and had suffered frequent falls on the stairs. This scenario sought to understand whether respondents feel that such a patient should be deprived of her liberties.

Scenario five:

Asked whether the capacity-making ability of an older, frail person should be questioned when they were making decisions that could harm him/her financially. It sought to probe participants’ understanding of the “unwise decision”.



A total of 150 practitioners were invited to participate, and 98 agreed. The sample included: 47 nurses (47.95 per cent); 29 care assistants (29.6 per cent); seven social workers (7.14 per cent); two physio/occupational therapists (2.04 per cent); and four GPs (4.08 per cent) (Table I).

Findings – round one

Considering that the scenarios targeted specific elements of the MCA principles, responses were not as consistent as might be expected. The following provides more detail, but in general, participants often disagreed about the appropriate response to the situation, even within professional groupings. However, it is worth noting that in the first round, while there was a fair degree of consensus about the need for best interest assessment (BIA) among professionals. It was clear that there was still a small group of professionals made up of nurses and occupational/physiotherapists who were of the view that best interest decisions should be made for patients and other service users regardless of whether they have capacity. They were of the view that the decisions made by these patients were unwise and detrimental to their well-being.

In another scenario, a large number (63 per cent) from all the professional and non-professional groups were of the view that a patient who refused a blood transfusion should be assessed for capacity, while an equally large number (56 per cent) were of the view that the patient’s view must be respected, including all GP participants. It was also significant that more than 25 per cent of participant care workers indicated it would make a difference if the patient refused blood transfusion on grounds of religion.

There was no consensus on deprivation of liberty issues, with an almost even spread of opinions among both professionals and non-professionals. For example, when responding to the case of an older man making an unwise decision about spending his money, there was an even spread of views with less than 20 per cent of participants suggesting variously that no action is needed; there was a need to raise a safeguarding alert; or a need for a BIA and checking for powers of attorney. It is worth noting that a higher percentage of care workers (25 and 29 per cent) than any other professional group proposed BIAs and powers of attorney. This may be an indication of the willingness of care workers to pass on the responsibility of making such decision decisions to more senior staff; one care worker stated “They can get approval from higher authorities to her care and what action must be taken based on her psychiatric report and behaviour”.

Scenario one

Almost half of respondents (45 per cent) stated there is no need to sedate the patient with a learning disability, even if her actions were putting her health and wellbeing at risk, as long as she does not lack capacity. However, 22.4 per cent disagreed, suggesting they thought sedation was necessary, contending that intervention is required, even if it means overriding the decision of a patient who is deemed to have capacity. In total, 60.2 per cent of respondents indicated there was a need for a BIA in a situation where the patient lacked capacity. However, a small number of respondents (6 per cent), all nurses, did not suggest there was need for a BIA. They appeared to suggest that a patient without capacity had a right to refuse sedation. A smaller number of respondents (3 per cent) including two nurses and one social worker suggested they thought this was an emergency situation, which could not wait for the formal process of conducting a BIA and recommended immediate treatment against the patient’s wishes.

Scenario two

This scenario generated a more varied response, with the largest proportion (38 per cent) suggesting the patient could be treated in accordance with parental consent against her wishes, because she was a minor. A much smaller proportion (10 per cent) suggested the patient’s capacity would determine whether to prioritise her or her parents’ wishes. However, there was no consensus on whether a patient who was 15 years of age has capacity to make decisions. In all, 6.12 per cent advised a mental capacity assessment before treating the patient against her wishes, but in line with parental consent. Another small proportion, 8.2 per cent, including three nurses, three social workers and two doctors, suggested the need for Gillick competence testing. In all, 12.24 per cent of nurses advocated applying the MCA while only 6.12 per cent suggested the MCA did not apply here. The case of a minor falls outside the scope of the MCA, however, it was included to gain an understanding of respondents’ knowledge of what was within and without the scope of the MCA. Also, it provided insight into how complex situations are dealt with in practice and whether staff recognised a situation where, for, e.g. Section 21 of the MCA provided a transfer of proceedings of a matter related to persons below 18 years. Such knowledge was limited, however, admittedly respondents were mainly from adult health and social care services.

Scenario three

Responses to this scenario showed no consensus about how to deal with a patient who refused blood. In total, 63 per cent of respondents (mainly nurses and care assistants) suggested staff could assess the decision making capacity of the patient. But a number of respondents indicated this approach depended on the qualifications of available staff. However, 13.2 per cent of respondents suggested the patient was too ill to undergo an assessment, and 56 per cent thought the clinical team should respect the patient’s decision. This last group included all GP participants, a small number of nurses, and some care assistants and social workers who agreed that the patient’s decision to refuse a blood transfusion should be respected. In all, 28.6 per cent of respondents stated their actions would be no different if the patient refused blood on religious grounds, but over a quarter (27 per cent), mostly care assistants, stated that refusal on religious grounds would make a difference to their assessment. This indicated a willingness to respect the views of the patient if refusal was based on religious belief. Such views potentially demonstrates a willingness to accept religious beliefs as valid reasoning for what would otherwise be seen as an unwise decision. While a majority of respondents suggested a capacity assessment was necessary, it was surprising that a third of respondents did not suggest this was needed.

Scenario four

In total, 26.5 per cent of respondents to scenario four, advocated that it was in the best interest of woman to have a mental capacity assessment. A small proportion of respondents, 6.12 per cent, suggested that a DoLS application was needed, primarily because her wish to return home was deemed an unwise decision. When participants were asked whether she should have her liberty deprived, 13 per cent of nurses, 17.24 per cent of care assistants and 14.3 per cent of social workers suggested a BIA was needed. A significant proportion of respondents (36.2 per cent of nurses, 34.5 per cent of care assistants and 50 per cent of doctors) argued the patient should not be deprived of her liberties with a smaller number of respondents (15 per cent of nurses, 17.24 per cent of care assistants and 25 per cent of doctors) arguing that the patient’s liberty must be deprived for her own safety. Participants on both side of the argument suggested the best interest of the patient influenced their decision. For example, one participant stated “No, in best interest of patient if at all possible”. Responses to this scenario did not provide a clear consensus about whether a capacity assessment was needed regarding her requests to return home. Generally, when considering depriving the patient of her liberty, slightly more than half of respondents disagreed, which is not a strong consensus.

Scenario five

This scenario, about an older person’s use of money (an “unwise decision”), also created conflicting responses. A sizable minority, 14.28 per cent of respondents (23.4 per cent of nurses and 10.34 per cent of carers), suggested no action should be taken to stop the older person spending his money as he wished. A similar proportion of respondents, 14.28 per cent (17.02 per cent of nurse participants, 17.24 per cent of care assistants and 14.3 per cent of social workers) indicated they would raise a safeguarding alert, while a small number of practitioners (10.63 per cent of nurses and 14.3 per cent of social workers) suggested recommending Powers of Attorney and Lasting Powers of Attorney. A slightly higher proportion, 17.35 per cent of respondents (21.3 per cent of nurses 17.24 per cent of care assistants and 29 per cent of social workers), suggested a BIA was required. Some respondents (17.02 per cent of nurses, 7 per cent of care assistants and 25 per cent of doctors) suggested checking for lasting powers of attorney or applying to the Court of Protection. There were no strong views provided by respondents, even when considering their responses grouped by profession. Responses did not provide a consensus about how to deal with an unwise decision.

Second round Delphi survey

An important part of a Delphi study is to provide a summary of findings from the first round to the second round. This summary allowed participants to consider their responses in relation to the sample’s perspective of issues and to identify how their views related to the general consensus (Table II).

Participant attrition was a challenge in this study and there were significantly fewer respondents in round two. Round one included 98 participants, and nearly 50 per cent nurses. In the second round there were twelve respondents, with 66.6 per cent being nurses. Care workers were the second largest group in round one, nearly 30 per cent of respondents, but in round two this was reduced to 8.33 per cent. GP’s constituted 4.08 per cent of respondents round one, but none participated in the second round. Social workers and therapists increased their proportion of participation (Social workers; 7.14 per cent round one; 8.3 per cent round two: physio/occupational therapists increased from 2.04 to 16.66 per cent). Methodologists recommend that researchers individually engage with each participant, encouraging peer group affiliation, to reduce attrition (Stitt-Gohdes and Crews, 2004), which was attempted during round one. In this study, with a variety of professions involved, group heterogeneity was a challenge. The variety of professions and different hierarchical positions in health and social care may have hindered their group cohesion.

Findings – round two


Round two sought to refine the findings from round one, as a result the questionnaire was reduced to three scenarios, focussing on the responses with least consensus:

Scenario one was repeated, and focused on what participants would do if a patient in intensive care with a learning disability refused sedation. The questions were more focussed than in round one, such as “If he is assessed as lacking capacity, is there a need for a BIA?” This scenario sought to elicit responses about capacity to make decisions.

Scenario two presented a situation where an elderly woman arrives at a care home and immediately makes it clear she wants to leave and return to her own home. This scenario focussed on possible deprivation of liberty after the patient makes an unwise decision.

The third scenario sought to expose a key area of concern from round one, by considering specifically how responses differed depending on their perception of capacity. This scenario presented a middle aged man with a brain injury living in a nursing home. When approached by staff about his deteriorating self-care skills and body odour, he was irritated and refused bathing assistance.

Scenario one

There was a reduction in the proportion of participants that thought the patient should be sedated, 33.3 per cent, down from 45 per cent round one, even if he had the capacity to refuse treatment. In contrast, 53.3 per cent argued that there was no need for a capacity assessment, which is markedly higher than the 22.4 per cent in round one. In the second round, there was a significant increase in the proportion of respondents against sedation, but a subsequent drop in the proportion who believed sedation was necessary. The respondents that argued for sedating a patient with capacity who refused treatment is in sharp contrast with the majority view that there is no need for sedation.

Scenario two

Responses to scenario two provided a range of answers that suggested patience, encouragement, listening to and giving the patient her wish. Interestingly, 25 per cent of respondents suggested a BIA was needed. For example one respondent states, “Do a best interest assessment to determine what can be done? Appreciate Olivia wants to go home. Identify temporary measures to support her at home, such as a commode, until adjustments can be made for her to live at home”. Almost half of respondents, 41 per cent, suggested they would not deprive the patient of her liberty even with safety concerns. However, a significant minority of respondents (25 per cent, including some nurses) suggested they would deprive her of liberty in order to keep her safe in the care home until adjustments had been made to her home. Whilst respondents thought undertaking a BIA was in the patient’s best interest, there is a concern, raised by Griffiths (2014) that the clinical use of these has been paternalistic. There are a series of judgements about best interest that provide some suggestions about this issue. In the case of R(Burke) v. GMC [2005] the High Court stated that to determine the best interests of a patient who is deemed incompetent, a stringent standard must be applied to ascertain whether a patient’s life had become “intolerable” (Samantha and Samantha, 2005). The Court of Appeal subsequently ruled that the use of the phrase “best interests” is confined to an objective test used when considering the duty owed patients in need of care and treatment but lacking decision-making capacity. These judgements suggest a systematic approach to determining what is the best interest for a patient, which a quarter of respondents seemed to have suggested. However, responses provided by the majority of participants suggest a trend towards individualism to the extent that it fits with the Public Guardian’s individuated view on best interest; which, according to Martin et al. (2012, p. 22) is defined as “Anything done for a person who cannot make decisions for themselves must be in their best interests. This means thinking about what is best for the person, and not about what anyone else wants”. Martin et al. (2012) argue there is growing realisation that the needs of others may be a useful consideration when dealing with best interest issues, which contradicts the effects of hyper individualism.

Scenario three

Responses to scenario three were more consistent, with 58.3 per cent of respondents of the opinion that all support must be given to the patient, providing he has capacity to make decisions. Respondents stated, for example, “Keep encouraging him to have a wash but try different means – if prompts aren’t working, try giving him bath and shower gift sets and asking if he had any concerns about washing which was preventing him from doing so” “Try to encourage him to shower whilst remembering that no two people will necessarily have the same personal hygiene standards and this could easily be his normal”. Other respondents, 41.66 per cent, did not suggest anything contrary, however, they were somewhat unclear about what they would do making such comments as “stop telling”, “approach Josh privately” and “with dignity”. When considering a response if he did not have capacity, 91.6 per cent of participants argued for all necessary support to be provided to help him make the decision. Responses prioritised the rights of the individual to make decisions, contrasting with the paternalistic tendencies often associated with the care of patients, which appears to be contrary to the established practice (Griffiths, 2014, p. 1,220). This may be an indication that efforts to embed the principles of the MCA in practice are beginning to yield positive results.

Summary and discussion

Using scenarios in a Delphi study originates from the Rand study in the 1960s regarding military technology, and explored the most possible scenarios of the given topic for future predictions (Renzi and Freitas, 2015); scenarios were based on possible interactions reflecting the reality of practitioners. Vollmar et al. (2015) have pointed out that scenarios have limitations, which include the time-consuming nature of creating (therefore cost-intensive), in particular because they consume project resources. Because creating scenarios uses the imagination, information and competency of the experts taking part, there is potential risk of bias, preferences of the experts taking precedence or rejection of what seems too unorthodox, with opinion leaders dominating.

The scenarios used presented a moral dilemma, which is a technique for generating stories from a range of sources to examine beliefs, perceptions and attitudes. The scenarios were stories generated from research findings in collaboration with professionals in the field (Hughes, 1998). They provide a snapshot of a given situation and yet have the advantage of providing respondents the opportunity to discuss issues from a non-personal and less threatening perspective. However, the very non-personal nature of vignettes can be considered a weakness. Hughes (1998) argues that it does not provide the necessary interaction and feedback that is part of social life. While this study might have suffered such disadvantage, there is currently no research tool that can exactly mirror real life experience. However, it has the potential to prompt participants to reflect on similar experiences, meaning they are notably more realistic than traditional surveys (Steiner et al., 2016). In order to make the scenarios as realistic as possible case studies were researched and social workers, nurses, care workers and safeguarding leads were consulted.

These findings highlight a conundrum in practice about the lack of consensus when practitioners attempt to apply the MCA. The capacity of patients to refuse treatment is a key area of contention requiring further research to resolve uncertainty. This is a key area of concern emerging as an important field of legal, clinical, and behavioural research, which has its origins in studies on psychiatric patients in the 1970s and 1980s (Marson and Ingram, 1996; Appelbaum et al., 1981). While mental capacity assessment has become a familiar concept for practitioners, the point at which the assessment is triggered was identified inconsistently by participants. It can be argued that the scenarios used in this study demonstrated not a lack of capacity to make a decision, but unwise decisions that challenge practitioners’ duty of care for patients. Appelbaum (2007) argues that although standards for decision-making capacity for consent to treatment vary somewhat across jurisdictions, they generally include several considerations, including: the abilities to communicate a choice; to understand the relevant information; to appreciate the medical consequences of the situation; and to reason about treatment choices. In a clinical setting, such as a hospital, with a need for quick decisions under pressure, this setting conflicts with considerations that require time and reflection, as well as gathering a variety of views (from patients and caregivers). From the findings of this study, it appears that practitioners are struggling to identify when to assess capacity, and when an unwise decision may conflict with clinical advice.

The lack of consensus demonstrates both consequentialist and deontological positions of ethical judgements in professionals and non-professionals roles. While the consequentialist approach explores consequences of actions in order to identify which choice is likely to yield the most “good” and least “bad”. Deontology emphasises that certain acts are inherently right or wrong, independent of the outcome and asks what one “ought” to do in relation to duty or obligation. The MCA presents potential ethical conundrums which are a challenge for a Delphi study which aims at consensus building. In studies using the vignette method, a lack of consensus has been attributed to a number of reasons potentially reflected in this study include, a lack of awareness or familiarity with protocols, lack of agreement with guidelines because it limits the autonomy of professionals, professionals have bias, a prevailing paternalistic attitude (Nguyen et al., 2014).

The lack of consensus in MCA practice has been acknowledged by others (e.g. Clare et al., 2013; Bartlett, 2014), arguing there is no consensus on how to implement this in the future. However, where there have been indications of consensus, it has been suggested that conclusions or decisions are open to challenge and especially so in cases where the criteria for detention under the Mental Health Act have been met (Selmes et al., 2010, p. 224). Our analysis suggests the lack of consensus was due in part to a lack of understanding of key concepts underpinning the law, and possibly due to the complexity surrounding such issues as capacity decisions, best interest and deprivation of liberty. The complications that arise from these issues include the autonomy of the individual, which is in direct opposition to paternalism, which has been prevalent for many years. In the sedation and blood transfusion scenarios, for example, a lack of consensus arose from the tensions existing between notions of duty of care from professionals and the growing importance of the autonomy of the individual.

There is need for the law to be reviewed in order to incorporate recommendations made by the law commission, but further research is needed to understand the challenge of transferring the aspirations into practice, so that there will be more consensus on the issues to avoid decision paralysis in health and social care.

Delphi survey round 1

Profession Number(%) Percentage
General Practitioner Doctor 4 4.08
Nurses Nurse 37 47 37.75 47.95
Practice Nurse 8 8.16
Community/District Nurse 2 2.04
Care Assistants Healthcare Assistant 2 29 2.04 29.59
Care Assistant (care/nursing home) 27 27.55
Social worker 7 7.14
Allied Health Professionalsa (e.g. Occupational and Physio therapists) 2 2.04
No profession indicated 9 9.18
Total 98 100

Note: aGroups with few respondents collapsed to obscure individuals

Delphi survey round 2

Profession Number Percentage
Nursesa (including practice nurses) 88 66.6
Social work, care assistants and Allied Health Professionalsa 4 33.32
Total 12 99.92

Note: aGroups with few respondents collapsed to obscure individuals


Age UK (2017), “Deprivation of Liberty Safeguards”, available at: www.ageuk.org.uk/globalassets/age-uk/documents/factsheets/fs62_deprivation_of_liberty_safeguards_fcs.pdf (accessed 11 October 2017).

Alghrani, A., Case, P. and Fanning, J. (2016), “Editorial: the mental capacity act 2005 – ten years on”, Medical Law Review, Vol. 24 No. 3, pp. 311-7.

Appelbaum, P. (2007), “Assessment of patients’ competence to consent to treatment”, New England Journal of Medicine, Vol. 357 No. 18, pp. 1834-40.

Appelbaum, P.S., Mirkin, S. and Bateman, A. (1981), “Empirical assessment of competency to consent to psychiatric hospitalization”, American Journal of Psychiatry, Vol. 138 No. 9, pp. 1170-6.

Bartlett, P. (2014), “Reforming the deprivation of liberty safeguards (DOLS): what is it exactly that we want?”, Web JCLI, Vol. 20 No. 3.

Boulkedid, R., Abdoul, H., Loustau, M., Sibony, O. and Alberti, C. (2011), “Using and reporting the Delphi method for selecting healthcare quality indicators: a systematic review”, PLoS ONE, Vol. 6 No. 6, pp.1-9, e20476.

Clare, I.C.H., Redley, M., Keeling, A., Wagner, A.P., Wheeler, J.R., Gunn, M.J. and Holland, A.J. (2013), “Understanding the interface between the mental capacity act’s deprivation of liberty safeguards (MCA-DoLS) and the mental health act (MHA)”, a report for CIDDRG, Cambridge University, Cambridge, available at: www.psychiatry.cam.ac.uk/ciddrg/files/2013/10/MCA-DoLS-MHAINTERFACE-FINAL-REPORT.pdf (accessed 20 August 2018).

Griffiths (2014), “Best interests of adults who lack capacity part 1: key concepts”, British Journal of Nursing, Vol. 23 No. 22, pp. 1220-1.

Habibi, A, Sarafrazi, A. and Sedigheh, I. (2014), “Delphi technique theoretical framework in qualitative research”, Vol. 3 No. 4, pp. 8-13.

Hanafin, S. (2004), “Review of literature on the Delphi technique”, available at: https://pdfs.semanticscholar.org/38d8/baf4f555fe5ff230dd75eb8483eb9298cfaa.pdf?_ga=2.151635001.1223554173.1511355814-338125709.1511355814 (accessed 22 November 2017).

Hasson, F., Keeney, S. and McKenna, H. (2000), “Research guidelines for the Delphi technique”, Journal of Advanced Nursing, Vol. 32 No. 4, pp. 1008-15.

House of Lords (2014), Mental Capacity Act 2005: post-legislative scrutiny, Select Committee on the Mental Capacity Act 2005, London, available at: www.publications.parliament.uk/pa/ld201314/ldselect/ldmentalcap/139/13903.htm (accessed 9 October 2018).

Hsu, C.-C. and Sandford, B.A. (2007), “The Delphi technique: making sense of consensus”, Practical Assessment, Research & Evaluation, Vol. 12 No. 10, pp. 1-8.

Hughes, R. (1998), “Considering the vignette technique and its application to the study of drug injecting and HIV risk and safer behavior”, Sociology of Health and Illness, Vol. 20 No. 8.

Iqbal, S. and Pipon-Young, L. (2009), “The Delphi method”, The Psychologist, No. 7, pp. 598-600.

Keeney, S., Hasson, F. and McKenna, H. P. (2001), “A critical review of the Delphi technique as a research methodology for nursing”, International Journal of Nursing, Vol. 38 No. 2.

Kendall, J.E., Kendall, K.E., Smithson, O. and Angell, I.O. (1992), “SEER: a divergent methodology applied to forecasting the future roles of the systems analyst”, Human Systems Management, Vol. 11 No. 3, pp. 123-35.

Komrad, M.S. (1983), “A defence of medical paternalism: maximising patients’ autonomy”, Journal of Medical Ethics, Vol. 9 No. 1, pp. 38-44.

Law Commission (2017), “Mental capacity and deprivation of liberty”, available at: www.lawcom.gov.uk/app/uploads/2017/03/lc372_mental_capacity.pdf (accessed 11 October 2017).

Ludwig, B. (1997), “Predicting the future: have you considered using the Delphi methodology?”, Journal of Extension, Vol. 35 No. 5, pp. 1-4, available at: www.joe.org/joe/1997october/tt2.html (accessed 6 November 2005).

McKenna, H. (1994), “The Delphi technique: a worthwhile research approach to nursing?”, Journal of Advanced Nursing, Vol. 19, pp. 1221-5.

McNicoll, A. (2017), “Government eyes emergency measures to ease DoLS pressures. Community care”, 25 July, available at: www.communitycare.co.uk/2017/07/25/government-eyes-emergency-measures-ease-dols-pressures/

Marson, D.C. and Ingram, K. (1996), “Competency to consent to research: a growing field of research”, Journal of Ethics, Law, and Aging, Vol. 2 No. 2, pp. 59-63.

Martin, W., Freyenhagn, F., Hall, E., O’Shea, T., Szerletics, A. and Ashley, V. (2012), “An unblinkered view of best interest”, British Medical Journal, Vol. 345, No. 7887.

Murgic, L., Hébert, P.C., Sovic, S. and Pavlekovic, G. (2015), “Paternalism and autonomy: views of patients and providers in a transitional (post-communist) country”, BMC Medical Ethics, Vol. 16 No. 65, pp. 1-9, doi: 10.1186/s12910-015-0059-z.

Nguyen, Y., Perrodeau†, E., Guidet, Trinquart, L., Richard, J.M., Mercat, M., Jolliet, P., Ravaud, P. and Brochard, L. (2014), “Mechanical ventilation and clinical practice heterogeneity in intensive care units: a multicenter casevignette study”, Annals of Intensive Care, Vol. 4 No. 2.

Okoli, C. and Pawlawski, D.S. (2004), “The Delphi method as a research tool: an example, design considerations and applications”, Information & Management, Vol. 42 No. 1, pp. 15-29.

Olaf, H. (1967), “Analysis of the future: the Delphi method”, Document No. P-3558, The RAND Corporation, Santa Monica, CA, May.

Powell, C. (2003), “The Delphi technique: myths and realities”, Methodological Issues in Nursing Research, Vol. 41 No. 4, pp. 376-82.

Reader, T.W., Reddy, G. and Brett, S.J. (2017), “Impossible decision? An investigation of risk trade-offs in the intensive care unit”, Ergonomics, Vol. 61 No. 1, pp. 122-33.

Renzi, A.B. and Freitas, S. (2015), “The Delphi method for future scenarios construction”, Procedia Manufacturing, Vol. 3, pp. 5785-91.

Sackman, H. (1975), Delphi Critique, Lexington Books, Boston, MA.

Samantha, A. and Samantha, J. (2005), “End of life decisions; clinical decisions are Increasingly shaped by legal judgements”, British Medical Journal, Vol. 331 No. 7528, pp.1284-5.

Samuel, M. (2014), “‘Risk-averse’ social workers need fundamental attitude change to comply with mental capacity act, finds inquiry”, Community Care, 13 March, available at: www.communitycare.co.uk/2014/03/13/risk-averse-social-workersneed-fundamental-attitude-change-comply-mental-capacity-act-findsinquiry/ (accessed 13 June 2017).

Selmes, T., Robinson, J., Mills, E., Branton, T. and Barlow, J. (2010), “Prevalence of deprivation of liberty: a survey of in-patient services”, The Psychiatrist, Vol. 34 No. 6, pp. 221-5.

Steiner, P.M., Atzmüller, D.M. and Su, D. (2016), “Designing valid and reliable vignette experiments for survey research: a case study on the fair gender income gap”, Journal of Methods and Measurement in the Social Sciences, Vol. 7 No. 2, pp. 52-94.

Stitt-Gohdes, W.L. and Crews, T. (2004), “The Delphi technique: a research strategy for career and technical education”, Journal of Career and Technical Education, Vol. 20 No. 2, pp. 1-13.

Strauss, H.J. and Zeigler, L.H. (1975), “The Delphi technique and its uses in social science research”, The Journal of Creative Behavior, Vol. 9 No. 4, pp. 253-9.

Skulmoski, G.J., Hartman, F.T. and Krahn, J. (2007), “The Delphi method for graduate research”, Journal of Information Technology Education: Research, Vol. 6 No. 1, pp. 1-21.

Spencer-Lane, T. (2017), “Mental capacity and deprivation of liberty: the law commission’s review of the deprivation of liberty safeguards”, The Journal of Adult Protection, Vol. 19 No. 4, pp. 220-7, doi: org/10.1108/JAP-04-2017-0016.

Vollmar, H.S., Ostermann, T. and Redaèlli, M. (2015), “Using the scenario method in the context of health and health care – a scoping review”, BMC Medical Research Methodology, Vol. 15 No. 89, pp. 1-10.

Willner, P., Bridle, J., Price, V., Dymond, S. and Lewis, G. (2013), “What do NHS staff learn from training on the Mental Capacity Act (2005)?”, Legal and Criminological Psychology, Vol. 18, pp. 83-101.

Further reading

[2016] EWHC 3473 (Fam), A local authority v. D & Ors.

Hemer, O. (1967), Analysis of the Future: The Delphi Method, Rand Corporation, Santa Clara, CA.

Hill, T.D., Burdette, A.M., Angel, J.L. and Angel, R.J. (2006), “Religious attendance and cognitive functioning among older Mexican Americans”, Journal of Gerontology: Psychological Sciences, Vol. 61B No. 1, pp. P3-9.

Holman, A. and Moore, S. (2007), Making Decisions: An Easy Read Guide, 2nd ed., Office of the Public Guardian, London.

Corresponding author

George Clerk can be contacted at: george.clerk@bucks.ac.uk

About the authors

George Clerk is Lecturer in Criminology at the Buckinghamshire New University, Wycombe, UK.

Jason Schaub is based at the University of Birmingham, Birmingham, UK.

David Hancock is based at the Buckinghamshire New University, Wycombe, UK.

Colin Martin is based at the University of Hull, Hull, UK.