Basket | Login


Freedom to live in least restrictive environment

2012: The ethical issues linked to restrictions of freedom of people with dementia

In section 2.1, we considered whether it is ethical to deprive people of their freedom on a permanent or temporary basis in a particular place. In this section, we look at different measures used to keep people in a particular place (irrespective of whether they are voluntarily or involuntarily in that place) as well as measures to prevent movement or choice of action, or to make people comply with another person’s wishes.

Such measures will be referred to collectively as measures of restraint. The use of coercion is slightly different to other forms of restraint as the emphasis is on forcing a person to do something rather than preventing them from doing something. However, as explained in section 1.1.4 (definitions), it amounts to a restriction of freedom. Other forms of restraint although often conceptualised in terms of a restriction also result in forcing a person to remain in a certain place, to be immobile or to do something other than that which they wanted to do. Whilst recognising that there is a subtle difference between the two terms, coercive measures will be included in the term “measures of restraint”.

In the following sections, different measures of restraint are briefly described and the ethical implications of their use considered. This will be followed by a more global consideration of the ethical issues related to the use of restraint in general and in particular in residential care settings.

1.2.1       Physical restraint

Description and examples of physical restraint:

The American Food and Drugs Administration (FDA) defines "protective restraint" (in 21 CFR Section 880.6760) as:

"a device, including but not limited to a wristlet, anklet, vest, mitt, straight jacket, body/limb holder, or other type of strap that is intended for medical purposes and that limits the patient’s movements to the extent necessary for treatment, examination, or protection of the patient or others"

This is clearly set in the healthcare context, is heavily influenced by notions of intent and purpose, and focuses on measures which actually touch the person’s body. The American Health Care Financing Administration (HCFA) (in 42 CFR Section 483.13 (a)) defines physical restraint under the Interpretive Guidance in the State Operations Manual as:

"any manual method or physical or mechanical device, material, or equipment attached or adjacent to the individual’s body that the individual cannot remove easily which restricts freedom of movement or normal access to one’s body."

The HCFA definition is functional and focuses on the effect on the individual. It is not limited to devices which touch the body but includes those which are adjacent to the person. This would include anything from a strait-jacket, to a tray table or to a side rail. It would even include a sheet if that had the effect of restricting the person’s freedom of movement (Westmoreland and Feigal, 2000). Variations of the HCFA can be found in several European studies and whilst it was devised to be applicable in nursing homes, it could also be applied in a range of settings, including the hospital and in people’s homes. It does, however, combine measures which are applied directly to the body and devices which are adjacent to a person, which are qualitatively quite different and it does not include any reference to the degree of restriction imposed by the measure. As mentioned in section 1.3.4, with regard to the difference between restriction of freedom and deprivation of freedom, the degree or extent of the restriction is important.

Examples of direct physical restraint (i.e. touching the body) would include various types of straps and belts, strait-jackets, limb holders or mittens and in certain circumstances, various medical devices such as tubes and drips. Physical restraint, involving less direct contact with the body in the sense of not binding it to something or to itself, would include bed rails, bed nets or cages, trays fixed to chairs, wheelchair brakes and bars (Gallinagh et al., 2001; MDAC, 2003; Retsas, 1998). Furniture may also serve as means of restraint if it restricts or prevents mobility such as chair which tilts backwards (Retsas, 1998). This emphasises that the way that objects are used and the effects of this are important in determining whether it is a form of restraint. For example, a sheet is not a form of restraintper sebut it is when it is tucked in so tightly that a person cannot move, irrespective of whether this was the intention. The same applies to chairs which tilt backwards. There may be no intention to restrain and people may use them willingly, especially if they are the only chairs available to sit on, but they may result in a restriction of freedom of movement.

A cage bed is a cage fitted on top of a bed which serves to confine a person to the bed (MDAC, 2003). The cage may be made out of metal bars or a net but the end result is the same. Examples of the use of this form of restraint in psychiatric hospitals and residential social care homes have been documented (MDAC, 2003). It is slightly different to other forms of restraint in that it is neither attached to the body nor adjacent to it. The person is inside the bed and cannot get out. It is nevertheless a device which restricts people’s freedom.

A person’s freedom to move about or even to move parts of their body may be restricted by a medical device such as a drip or tubes attached to a machine. If the person believes that this is essential or fears the consequences of disconnecting it, it may serve to restrict their freedom of movement. Depriving a person of necessary medical aids such as a walking stick, a Zimmer (walking) frame, a wheelchair or a mobile drip to which they normally have access may also prevent a person from moving about freely and could therefore be considered as a means of restraint as illustrated in the following quote.

“A resident in a wheelchair had been moving around the corridor during the night and staff had assisted him back to bed. The staff had removed the wheelchair from the room.”(Commission for Social Care Inspection, 2007, p.38).

The above example suggests a situation involving the intent to restrict freedom of movement. However, it would still consist of restraint, were such intent lacking.

The issue with regard to medical devices is somewhat different and less clear cut in that people’s freedom of movement may be restricted solely for the purposes of treatment, which is in many cases standard practice (e.g. the intravenous administration of hydration, antibiotics or blood, and the drainage of bodily fluids). Yet such measures may restrict a person’s freedom of movement. Presumably consenting to the procedure after having been fully informed of what is involved might determine whether the procedure is treatment or restraint. This raises further issues as to whether restraint consists solely of restrictive measures to which a person has not consented and also to whether the intent is to restrain. So far, we have argued that the main criteria are that something restricts freedom of movement and this irrespective of questions related to consent and intent. However, it would be unrealistic to consider the provision of hydration to a person who needs it as a form of restraint. Perhaps with regard to medical devices, issues such as intent, purpose and necessity are in fact relevant to the debate about whether they represent a form of restraint.

The prevalence of the use of physical restraint

Reviews of the literature on the use of physical restraint in residential and semi-residential care facilities suggest that it is widespread (Goethals, Dierckx de Casterlé and Gastmans, 2012; Hantikainen, 1998). However, it is difficult to obtain a reliable indication of its use and the prevalence of the use of physical restraint on people with dementia tends to be lost amongst general statistics on its use on nursing home residents.  Figures for the prevalence of the use of physical restraint on people with dementia in their own homes are lacking.

Reports in the literature of its frequent use on older residents in nursing homes vary between 25% and 85% (Hantikainen, 1998). In Norway, the use of restraint in care homes is reportedly quite common with an average of 77% of residents in care homes having dementia and 69% having experienced the use of restraint (Engedal et al., 2000 in Slettebø and Haugen Bunch, 2004). Hammers and Huizing (2005), on the other hand, estimate the prevalence of physical restraint in care homes as being between 41% and 66%. In Spain, the Spanish Untie the Elderly Programme maintains a database on the use of restraint in a representative sample (check) of 667 facilities covering 27,332 residents of care homes. The data, which was obtained by means of a survey completed by the various facilities, reveals that in 2011 23.67% used physical restraint (based on the HCFA definition of physical restraint mentioned above. However, the prevalence of the daily use of physical restraint solely for people with dementia amounts to 63%.

Gastmans and Milisen (2006) suggest that different estimates of prevalence may be due to differences in the categories of people studied, as well as cultural, legal and educational differences and the research methods used (e.g. observation, self-report or medical files). In countries where there are strict regulations on the use of physical restraint, depending on the methods used, it may be difficult to detect all cases of its use. Survey methods and interviewing may result in the desire to provide a more socially acceptable image of the institution or staff. On the other hand, direct observation runs the risk of organisations and staff being careful not to reveal the use of restraint whilst the observer is present. There is no ideal solution and the studies provide valuable information about the use of restraint against older people and/or people with dementia, bearing in mind that the actual use of restraint is likely to be somewhat higher.

The existence of guidelines or regulations on the use of restraint, provided that they are known about and sufficiently understood, may result in a less use of physical restraint in professional care settings (Karlsson et al., 2000; Weiner et al., 2003).

Reasons for and attitudes towards the use of physical restraint

A number of reasons for the use of physical restraint have been identified in studies. Some are linked to the person to whom the measure is applied, such as to prevent them from harming themselves, to prevent falls, due to suicidal tendencies and restlessness. However, the prevention of harm is clearly not the only reason and the use of restraint has been proven to cause harm. Some reasons are based on a disturbance to other people such as aggressive behaviour, sexual harassment of staff, disturbing other residents or going into other people’s rooms. The occurrence of BPSD (see section 1.3.1) or any behaviour which carers find challenging may sometimes result in the use of restraint if the carers are not sufficiently trained, unaware that the use of restraint may result in harm or do not have access to the support they need. The following table provides examples of challenging behaviours which are classified as either aggressive or non-aggressive.

Aggressive and non-aggressive challenging behaviour (CB) (Ian James, 2012)

Aggressive forms of CB

Non-aggressive forms of CB










Hair pulling


Throwing objects

Stick Prodding





Physical sexual assault

Verbal sexual advances

Acts of self-harm

Repetitive noise

Repetitive questions

Making strange noises

Constant requests for help

Eating/drinking excessively



General agitation

Following others/trailing

Inappropriate exposure of parts of body

Masturbating in public areas

Urinating in inappropriate places


Handling things inappropriately

Dismantling objects

Hoarding things

Hiding items

Falling intentionally

Eating inappropriate substances


Sometimes restraint is used because it has been requested by the person with dementia, by relatives or by a doctor. If prescribed by a doctor, nursing or care staff may feel obliged to carry it out and fear reprisals for acting against a doctor’s wishes (Karlsson et al., 2000). The decision to use restraint is not always made by one person. In some cases, it is made by the care or medical team.

Some reasons reflect a perceived benefit to other people (e.g. when it is due to understaffing, lack of competent staff, the desire for peace and quiet, lack of knowledge about alternatives to restraint and about the consequences of using restraint) (Gastmans and Milisen, 2005; Hantikainen, 1998; Kirkevold and Engedal, 2004).The routine use of physical restraint (especially bed rails) or for reasons which are unclear has also been reported (Cohen-Mansfield et al., 1992 in Hantikainen, 1998; Kirkevold and Engedal, 2004; Meyer et al., 2008). As physical restraint is often used arbitrarily, it is not always documented. Kirkevold and Engedal (2004) suggest that lack of legislation of the use of physical restraint may explain why it is not always reported or reasons given for its use.  

The perceived need to protect people from harming themselves, particularly due to falls, is a common reason given for the use of restraint (Hamers and Huizing, 2005; Karlsson et al., 2000) and is often linked to fears of litigation. However, it has been argued that such fears are often unfounded and result from a lack of knowledge about liability for negligence and physical injury (Robbins, 1996). Moreover, physical restraint may actually increase the likelihood of falls and cause severe injury. In addition, it could be argued that too great an emphasis is placed on the prevention of harm at the risk of overlooking the need to promote autonomy and general wellbeing.

A lack of legislation banning the use of physical restraint may allow it to persist. In Japan, for example, it is forbidden to restrain older people receiving long-term care but people with dementia in hospital are not accorded the same protection. Consequently, it is fairly common for hospital nurses to physically restrain patients with dementia in order to prevent them from removing intravenous infusion tubes (Yamamoto and Aso, 2009). A difference between the use of restraint in psychogeriatric wards of nursing homes and hospitals in Israel was also reported in a study by Weiner, Tabak and Bergman (2003) but this was explained in terms of the nurses in nursing homes having greater knowledge about guidelines on the use of restraint.

Just as carers may have reasons for using physical restraint, they also have reasons for removing it once applied. In a small Swedish mixed methods study, 30 registered nurses were given a semi-structured interview based on a vignette about a fall-prone person with dementia but with limited information about the situation (Karlsson et al., 2000). Part of the study involved asking the nurses whether they would respect the person’s wishes and remove the physical restraint which had been applied. Only 9 stated that they would and of the remaining 21, 8 initially said that they would but then changed their minds. Of those who would remove the restraint, reasons given were to prevent harm (through agitation or the person feeling poorly), to respect the person’s autonomy and on a conscious decision to take the risk themselves that the person may fall and injure himself. The researchers point out that the reasons not to continue the use of restraint were related to ethical values whereas those to continue restraint were mainly related to external factors such as lack of time. Nevertheless, all of the nurses opposed to removing the restraint would do so for at least one of the reasons provided in addition to the additional vignette (e.g. based on restraint being forbidden, the staff-ratio being high or removal having been requested by a doctor or the family or knowledge that the person previously said he could not imagine being restrained. Whilst the initial nine nurses were prepared to take risk, the other nurses’ decisions to remove restraint were largely dependent on the decisions of other people.

These results cannot be generalised to all nurses but they provide valuable information about the reasoning of a group of nurses with regard to the use of restraint which may be applicable in similar situations with regard to similar groups. Reflecting on the findings, the researchers suggest the importance of different attitudes towards the person who is restrained and personal responsibility. In particular, they interpret the reasons of the 9 nurses who were willing to remove the restraint as reflecting an understanding of the person’s behaviour being meaningful as opposed to a symptom of the disease, as well as to them being willing to take independent decisions.  

The impact of physical restraint

Numerous examples of the negative impact of physical restraint can be found in the literature. These include medical/physical, mental/psychological, social/behavioural consequences and mobility/agility. Examples are provided below. This is an artificial distinction as many consequences are linked to more than one area.

Medical/physical consequences include bruising, acute functional decline, decreased peripheral circulation, cardiovascular stress, bladder and bowel incontinence, constipation, muscle atrophy, pressure ulcers, infections, asphyxiation, strangulation, cardiac arrest and even death (Cotter, 2005; Cotter and Evans, undated; Evans et al., 2002; Gastmans and Milisen, 2005; Parker, 1997). Such effects are not linked solely to the use of physical restraint as older people with cognitive and physical impairments may be more prone to such harmful effects compared to those with no impairments (Gastmans and Milisen, 2005). Evans et al. (2003) suggest that as estimations of the number of actual deaths linked to the use of restraint are mainly based on retrospective studies (e.g. consulting death certificates), it is likely actual cases may be much higher. 

Mental/psychological consequences include serious psychological disadvantages, psychological trauma due to restraint serving as a symbolic reminder of past victimization, increased cognitive decline, apathy, depression, disillusionment (Evans et al., 2002; Flannery, 2003;Gastmans and Milisen, 2006).

Social and behavioural consequences include impaired social function and behavioural symptoms such as aggressiveness and regressive behaviour (Cotter, 2005; Evans, Wood and Lambert, 2003; Hamers and Huizing, 2005)

Consequences linked to mobility and agility include decreased muscle strength and balance, decreased mobility, increased risk of falls and injury, even resulting in death (Capuzeti et al., 2000 and 2002; Evans et al., 2002; Tinetti, Liu and Ginter, 1992).

With regard to mobility, Luo, Lin and Castle (2011) found that the use of trunk restraint was associated with a higher rate of falls and fractures amongst people with dementia compared to people without dementia. The use of full bed rails was associated with a lower risk of falls for people with and without dementia. This study was based on a survey covering a random sample of 13,507 residents in nursing homes. Staff members provided the responses based on medical records and other documentation. Residents were not interviewed.  It was found that people with dementia were more likely to be restrained and slightly less likely to have bed rails. Cotter (2005) describes the risk of people who have been physically restrained becoming trapped between the headrest, mattress and bedrail which can lead to asphyxiation and strangulation. People may fall as a result of struggling over barriers (such as bedrails) or as a result of fatigue or unsteadiness following an effort to free themselves from a restrictive measure (Cotter, 2005).

The impact of physical restraint is not limited to those who are subjected to it. Gastmans and Milisen (2006) point out that there has been very little research into relatives’ experience of the use of restraint on their loved ones. They suggest that relatives may be affected in various ways. Some, for example, consider the use of physical restraint as a sign that all hope is gone, some prefer to ignore that it is being used and yet others experience anger or disillusionment. The use of restraint may also reflect and even contribute towards unequal power relations.

Ethical issues linked to the use of physical restraint
Respect for dignity

Treating people with dementia with dignity should perhaps be considered as one of the most important ethical principles to be observed when providing care and specifically with regard to the issue of the use of physical restraint. Human dignity (in the sense of “Menschenwürde”) is often considered as inherent in being a human being and something that cannot be taken away from a person (Nordenfelt, 2002). However, this is just one interpretation of what dignity means.

There is also the dignity of identity which is the kind of dignity that people attach to themselves as integrated and autonomous people who have their own history, future and relationships with other people. A person’s perceived value and worthiness is reflected back in the context of interaction. Dignity can therefore also be considered as an attribute that can be subjectively felt but also something which can be denied, ignored, withheld and violated. It is often linked to a person’s self-image, which has evolved and been fostered over many year but which can be damaged by the acts of other people. This type of dignity can be shattered by other people, even unintentionally due to a failure to really understand each person i.e. who or what kind of a person they really are. The dignity of identity provides a possible explanation for the humiliation suffered as a result of being visibly or publicly restrained. Such humiliation may also result from cultural expectations linked to adulthood and associations with measures sometimes used to restrain children (e.g. putting rails around cots, putting them in play pens, tucking them tightly in bed, making them wear mittens and using harnesses, reins and wrist links to stop them wandering off.

Nordenfelt (2002) explains that the dignity of identity can also be considered objective because it can be violated even in cases where a person is not aware of the violation. Consequently, it would be a violation of a person’s dignity if they were physically restrained in a way which another person would find undignified even if the person restrained is unconcerned or even unaware of the measure. This would also be applicable in the case of chemical restraint which might have an impact on a person’s appearance and visible behaviour.

Personhood - The degree of restraint and objectification

As mentioned earlier, the distinction between the restriction or deprivation of freedom is based on a consideration of degree or intensity of a particular measure. Restraint of various degrees or levels of intensity may be used as the actual means to deprive a person of their freedom or to restrict it. However, rigorous legislation, recommendations or guidelines are sometimes lacking on the use of physical restraint. As most forms of restraint do not constitute a deprivation of freedom (as loosely defined, for example, in the ECHR), there is a risk that they may be tolerated. However, the degree or level of intensity is not entirely an objective measure.

What one person may consider as a slight restriction of their freedom or as being of minimum importance, may be devastating for another person based on the meanings attributed to the measure and the significance of being restrained in that way for that person. This may be based on the person’s history, experience, values and expectations, as well as on the importance placed on the impact of the restriction. For example, one person might not object in the slightest to being tucked in bed so tightly that they cannot move. That person may interpret that as a caring act, not be someone who moves much during the night or not even pay much attention to it. Another person may feel that it is an extreme form of restraint which even intrudes into their periods of sleep, liken it to the swaddling of an infant and consider it humiliating, malicious or an abuse of power. Failing to take into account each individual’s reaction to the measure amounts to treating the person as an object to which restraint is applied and not as a unique individual. Such practices fail to respect personhood.

As stated earlier, the intent to restrict liberty does not affect whether a particular measure constitutes restraint and also, the fact that a person does not oppose the use of a particular measure does not necessarily make it acceptable. On the other hand, understanding the way that a measure is experienced by the person on whom it is applied is part of providing good care and judging levels of restraint. Attention to such details may contribute to uncovering practices and methods which physically restrain people but which are not recognised as such.

Justice and equity – Discrimination, devaluation and power

Various forms of restraint have been practiced throughout the ages. Examples include putting people in small cages, fastening them to walls, the use of shackles and chains and putting people in public stocks. Such practices were inflicted on people who were considered as having no rights, no value or having committed a crime. The intention was to cause physical and mental pain or anguish, and/or to humiliate people. Other practices such as swaddling infants may have been linked to beliefs about care and the formation of the body and have not completely died out. Although restraint was often deliberately designed to cause pain in the past and this is presumably not the intention with current means of restraint, people may consciously or unconsciously associate measures to restrict their movement with such well known means of punishment.

According to the principle of justice and equity, people should be treated on the same basis and not be disproportionately subjected to risk, harm or even preferential treatment. However, restraint seems to be used against certain groups of people and not others. It is often used against people who do not have the right or ability to oppose it such as people who have been compulsorily detained, and potentially vulnerable groups such as infants, older people and people with physical or mental impairments. Whether this is because such groups are easy targets and cannot fight back (which would consist of an abuse of power) or whether it is because they are not valued, is not known. However, the groups which are subjected to such practices are groups in society which are often less valued than other groups and such practices are therefore tolerated. In a paper by Minkowitz (2006), then Co-Chair of the World Network of Users and Survivors of Psychiatry, she describes discrimination as occurring “first of all, by making an exception of actions against people with psychosocial disabilities, actions which would otherwise be considered torture” (p.15). 

The use of restraint, if not challenged, could be considered as representing a form of structural discrimination, based on the devaluation of people with dementia and on the unfair exercise of power which would also be against the principles of justice and equity. Structural discrimination is the kind of discrimination which is inherent in societal practices and traditions and often unrelated to a personal wish to treat another person or group unfairly. Sometimes such discrimination is so deeply engrained in society and daily life that people may be surprised at the suggestion that there is anything wrong with what they are doing. They may argue that there is no malicious intent and that it is just the way or even the only way that things are done.

Link and Phelan describe power as one of the components of stigma, the others being labelling, stereotyping, separating “us” from “them” devaluation, discrimination and emotional reactions. Power can be detected behind every component of stigma and without the social, political and/or economic power of the people who stigmatise, there would be no stigmatisation (Link and Phelan, 2006; Mahajan et al., 2008; Parker and Aggleton, 2003). People with dementia tend to be older, be considered as having a mental disorder and the majority of older people with dementia are women. This adds to the stigma of dementia that linked to mental illness, as well as ageism and sexism. Some people with dementia may be further stigmatised on the basis of other attributes linked to ethnic groups, religious beliefs, sexual orientation or lifestyle factors.  In addition, some people with dementia will have lost certain rights and no longer be considered competent to make certain decisions about their lives.   These attributes, which become socially salient due to the significance that they have for other people, often reflect power imbalances within society which are reflected in attitudes, presumed social status and socio-economic differences. Admittedly, many care homes have a high turnover of staff, who are predominantly female and from all groups in society, some of which are also traditionally devalued. However, it is the combination of various potentially discrediting attributes which render people with dementia vulnerable to abuse in the form of physical restraint.

Beneficence – For whose benefit?

The concept of structural discrimination does not exonerate individuals from blame because each person can take a stance against injustice. Often there are several issues at stake. Professional carers have a professional duty to care and this involves accomplishing a range of tasks within a given time. Failure to accomplish tasks on time may be problematic for them and be interpreted as a sign of laziness or incompetence. Certain practices, such as the use of restraint, may be overtly or covertly condoned or even promoted by a care home, thereby making it difficult for staff not to use them as it would mean going against the hierarchy. Some homes may simply turn a blind eye to various practices in the interests of efficiency and cost saving.

Informal carers may feel that they have a moral duty to care or simply care based on a loving relationship with the person with dementia. Some point out that they had no option, did not choose to be a carer or just drifted into care (Alzheimer Europe, 2001). Non-related informal carers, such as neighbours and volunteers, which are less common, may care for a variety of reasons (e.g. based on religion, community spirit, reciprocity or friendship). Family carers may also experience conflicting demands which lead to the use of physical restraint. This might include professional obligations, competing family demands, the need to leave the home (e.g. for shopping or for administrative matters) and personal reasons (e.g. based on leisure activities, socialising or having a break).

The reasons provided above are examples of real-life situations which sometimes contribute towards the use of physical restraint and not justifications for its use. They highlight the dilemma sometimes faced by professional and informal carers which consists of balancing competing demands, wishes and obligations. Caring for a person with dementia should involve doing what is in their best interests, which rarely if ever involves the use of restraint. Yet, as mentioned earlier, people with dementia often lack the power to protest, may be easily manipulated and, being dependent on the carer, are in a vulnerable position. The onus is therefore on carers to respect the person with dementia and to act in their best interests. This is an ethical perspective which does not resolve the conflict mentioned above.

A perceived problematic situation cannot be considered as a justificationper sefor the use of physical restraint. Equally, the problematic situation is not caused uniquely by the person with dementia. It may also be a reflection of the way that the care of people with dementia is organised by society which reflects economic, political and social factors, which in turn may be a reflection of the value accorded to people with dementia. Defining a person’s behaviour as problematic or challenging begs the question “for whom?” All too often the answer is not “for the person with dementia” but rather “for other people”. Physical restraint would rarely be in the best interests of the person with dementia and unless it is, it should be considered unethical.


Nonmaleficence, which means doing no harm, is particularly relevant to the issue of physical restraint. As seen in the section on the effects of physical restraint, physical restraint can have serious adverse physical, psychological, emotional and social effects, and even lead to death. Based on the principle of nonmaleficence, its use cannot in general be considered as ethical although it is always possible that there may be exceptional circumstances. On the other hand, there is a risk that the initial use of restraint may exceptionally be justified but that its use is continued thereafter, when there are no longer grounds to justify its use. This may also apply to the use of certain medication.

Respect for autonomy

The issue of autonomy with regard to the use of restraint applies first and foremost to the people with dementia whose freedom is restricted in that way. They are being deprived of the freedom to decide what they want to do and when, or what they would prefer not to do. However, this is not simply about having the freedom to choose one option over another. The consequences of the lack of freedom of choice are inextricably linked to other losses of freedom such as the freedom to physically move, to respond to one’s physical needs, to present a certain image of oneself, to protect oneself against humiliation and possible abuse, to socialise with other people and to accept risks. Respect for autonomy is not more important than other ethical principles but in the case of restraint, it lies at the heart of several ethical issues and the eradication of the use of restraint would help avoid certain unethical situations in the context of care.

The autonomy of care staff is not being respected if they are expected to use certain forms of restraint in order to fulfil their required tasks and if they are uncomfortable with this either professionally or personally. One way to express their autonomy is to voice their concerns or cease working for that establishment. However, if such use is widespread and accepted by apparently respectable establishments and more highly educated healthcare professionals, it is possible that they may question the legitimacy of their concerns. For this reason, it would be beneficial to include training in ethical reflection in the formal training of care staff and as in-house or external continuing education. Training in ethical issues related to the care of people with dementia would also be beneficial for managers of care homes as it might lead to a more positive attitude towards care staff who express ethical concerns. 

Security versus well-being

The use of restraint is often justified on the grounds that it is necessary to prevent a person from coming to harm. We have already argued that far from preventing harm, the use of restraint may actually lead to accidents and even death. Arguments based solely on the principles of beneficence and nonmaleficence are therefore unconvincing. Another point to consider is whether and if so to what extent people with dementia should be protected from harm. As a potentially vulnerable group (and in the light of fears about litigation for failing to protect them), people with dementia may be over-protected. This reduces their autonomy, may fail to respect their dignity and overall, is not compatible with enhancing their quality of life. Often greater importance is attributed to safety than to wellbeing especially when it concerns other people who are considered as vulnerable and for whom someone is responsible. It may be considered acceptable for people to engage in acrobatics and mountaineering, even to surf in shark-infested water, as a hobby or to cross busy roads, handle hot liquids and manipulate machinery but not for an elderly lady with dementia to get up out of her chair when she chooses in order to walk around a little and chat with other residents. Gastmans and Milisen (2006) suggest that although physical integrity may be considered as a fundamental value, this value should not always take priority over others and that consequently, overall well-being should be promoted. This would include social, psychological and moral aspects of wellbeing.

Goethals et al. (2012) reviewed qualitative literature on nurses’ decision-making in the case of physical restraint. Concerning the issue of security and wellbeing, they point out that nurses’ decisions to use physical restraint are nevertheless often dominated by concerns about safety even though they take into consideration other values such as beneficence, freedom and respect for autonomy when deciding on the use of restraint. They also consider wellbeing in terms of the wellbeing of staff and other patients as well as that of the person in question. This is an issue which probably also applies in the context of people’s own homes where there are other family members to consider. In a care environment which prioritises safety over wellbeing, nurses (and care staff) may be torn between their own values and their perceived or imposed professional responsibilities. Weiner et al. (2003) describe them as “acting as “double agents”, trapped between professional and ethical obligations towards patients’ rights and the obligation to carry out employers’ policies, which may involve cost-saving and efficiency measures” (p. 513).

Care may become a series of physical acts devoid of any genuine care for the person concerned. If nurses try to cope by rationalizing the use of physical restraint rather than prioritising total wellbeing, there is a risk of distancing which may make it easier for them to apply such restraint (Sullivan-Marx, 1996 in Goethals et al., 2012). This corresponds to the conceptualization of stigma of Link and Phelan (2001) in which making a separation in one’s mind between “us” and “them” is associated with a perceived loss of status of “them” which makes it easier for the people considered as “us” to discriminate.

Addressing issues often linked to the use of restraint

Most of the literature is on the use of restraint in professional care settings, but most of the literature on addressing issues which commonly lead to the use of restraint also seem applicable to the home setting. Some of the issues discussed in the literature are reflected in the recommendations of this working group.

Modification of the environment and of procedures

In addition to the removal of bed rails and other forms of physical restraint, Cotter and Evans (undated) suggest the need to implement necessary changes to the environment and to procedures. They suggest, for example, the use of surveillance, implementing measures to ensure bed safety and to protect against falls and injuries, personalizing the environment with objects and photos, encouraging families to take part in care and striving for consistency of both staff and care routines. Part of changing routines might involve providing care


The issue of communication has been addressed by several researchers. Some describe the need to speak clearly, slowly and calmly, using non-verbal communication if needed, whereas others emphasize conversation, negotiation, involving people in decisions about their care, giving them responsibility (Cotter and Evans, undated; Kontio et al., 2010; McKillop and Petrini, 2011; Slettebø and Haugen Bunch, 2004). These approaches all involve personal contact and involvement with the person. Such presence and interaction should be meaningful. Chiovitti (2011) mentions the importance of authentic relating which is reminiscent of the work of Buber (1970) as well as Kitwood (1997). Buber emphasized the importance of relating to people authentically (i.e. as people rather than as objects), whereas Kitwood described the downward spiral towards objectification, isolation, increased impairment and death.

Carers may experience difficulties communicating with people with dementia due to a lack of training or understanding of what dementia involves. It cannot therefore simply be suggested that they communicate more with the person with dementia in their care. Organisations have a responsibility to ensure that professional carers are adequately trained to maximise the possibility of effective communication and also to help them to establish a meaningful care relationship with people with dementia in their care. Numerous publications and guides are available on effective and meaningful communication with people with dementia, and many Alzheimer Associations run training courses on this issue.

Part of communication involves the interpretation of the message or behaviour of the person with dementia, Chiovotti (2011) also emphasises the importance of not taking behaviour personally. Some messages and behaviour may be correctly interpreted and were intended to be obstructive or against the carer. Everyone has the right to express themselves, get frustrated and react to others as they see fit. This includes people with dementia. However, when people have difficulty communicating with and understanding people with dementia and when people with dementia are unable to communicate what they wish to communicate, misunderstandings will occur.

Educational programmes

The findings of a few studies suggest that educational programmes may be beneficial in helping change attitudes and encouraging restraint free practices. Pelfolk et al. (2010), for example, evaluated the effects of a restraint minimisation education programme on staff knowledge, attitudes and the use of physical restraint by means of a randomized controlled trial. The participants in this Swedish study included 184 staff in the group receiving the educational programme and 162 staff in the control group (which received six 30-minute videotaped lectures). The education programme ran for 6 months and covered six themes including dementia, delirium in older people, falls and fall prevention, the use of physical restraint, caring for people with dementia and complications in dementia.  The emphasis was on understanding, preventive measures, the adverse effects of restraint and communication. The importance of fostering an attitude of understanding rather than of control was also echoed by Kolanowski et al. (2010). The researchers conclude that staff education can increase knowledge, change attitudes and reduce the use of physical restraint without any change in the incidence of falls or the use of psychoactive drugs.

Another study compared the effects of an educational programme on four Norwegian care homes for people with dementia (Testad, Aasland and Aarsland, 2005).  This study consisted of staff in two care homes attending a full day seminar followed by one day sessions of guidance per month for six months. The researcher rating the use of restraint did not know which homes were the ones where staff were receiving the programme. The programme focused on decision making and ways to avoid the use of restraint. The success of the programme was evaluated on the number of restraints used on residents and the level of agitation of residents in the care homes where staff were receiving the programme compared to those where staff were not. The researchers conclude that although the level of agitated behaviour remained unchanged or even increased slightly, the educational programme led to a significant reduction in the use of restraint in institutionalized elderly people with dementia thus improving the quality of care.

These two studies lend support to the claim that it is possible to care for people with dementia without resorting to the use of restraint. The outcome measures adopted in these studies (i.e. the criteria used to determine whether or not the interventions were successful) and the way that the results of these are interpreted are a positive reflection of the attitudes of the researchers towards the use of restraint. They suggest that the main objective of good care is not to rule out totally the possibility of falls or to reduce the occurrence of agitated behaviour but rather to provide humane, appropriate care which contributes towards the quality of life of residents. The educational programmes did not result in zero use of restraint although it is possible that this may result from continued exposure to this approach and on the ethos of the various homes. In the second study, a visit one year later revealed that staff were still carrying out the non-restraint programme.

Issues linked to the cost and organisation of restraint free practices

There is some difference of opinion concerning the cost of implementing restraint free alternatives (Wang and Moyle, 2004). Werner et al. (1994) stated that it is a complex and costly process, whereas Cohen et al. (1996) and Koch and Lyon (2001) argue that most alternatives are inexpensive, that additional staff are not required and that success is grounded in staff education and the commitment of staff, residents and families. In a qualitative study involving a series of focus group discussions with 35 staff members from six nursing homes in America, time emerged as a key barrier to the use on non-pharmacological approaches to BPSD (Kolanowski et al., 2010). The researchers concluded that it was necessary to implement staffing patterns which give staff the time to make a difference in the care of people with BPSD.

1.2.2       Chemical restraint

People with certain forms of dementia such as Alzheimer’s disease and dementia with Lewy bodies may be prescribed “anti-dementia” drugs such as donepezil, galantamine, rivastigmine and memantine. For many people, such drugs slow down the progression of symptoms and improve their cognitive abilities for a certain period, thus contributing towards quality of life. They are therefore often described in positive terms.

In addition, many people with dementia are prescribed other types of drugs such as sleeping tablets, tranquillisers and anti-depressants to treat the symptoms or behaviour which may arise during the course of the disease (Jacques, 2001). Concern is increasingly being expressed about the use of these drugs for people with dementia. Some of these drugs are used to treat the behavioural and psychological symptoms of dementia (BPSD) and are commonly called “anti-psychotics”. They date back to treatments originally produced in the 1950s as a treatment for psychosis which involves the inability to distinguish between reality and imagination. They are known as typical antipsychotics and include drugs such as chlorpromazine, haloperidol and trifluoperazine. Progress was made in the following decades resulting in a new type of antipsychotic drug becoming available in the 1990s known as an atypical antipsychotic (examples including risperidone, olanzapine and quetiapine), which had less disturbing side-effects. It is this second type of antipsychotic which is frequently given to people with dementia (Banerjee, 2009) and has come to be associated with chemical restraint.

Some experts believe that such drugs are not appropriate for people with dementia at all, whereas others believe that there may be situations which justify their use provided that other non-pharmacological approaches have been tried first. It is increasingly being recognised that antipsychotic medication should never be used as a routine measure. These debates have implications for the way that chemical restraint is defined in that most definitions of chemical restraint focus on the inappropriate use of medication and hence on the purpose of the administration of a pharmacological substance:

“Chemical restraint means a drug used to control behavior and used in a manner not required to treat the patient’s medical symptoms” (Title 22 of the California Code of Regulations).

 “A psychopharmacological drug (a drug prescribed to control mood, mental status, or behaviour) that is used for discipline or convenience and not required to treat medical symptoms or symptoms from mental illness or mental retardation that prohibit an individual from reaching his highest level of functioning” (Virginia Department of Health, 2004).

“The use of drugs and prescriptions to change people’s behaviour” (Commission for Social Care Inspection, 2007).

“Sedative or tranquillizing drugs for purely symptomatic treatment of restlessness or other disturbed behaviour” (Mental Welfare Commission for Scotland, 2006, p.20) [adding that drug treatments for medical or psychiatric conditions which underlie the disturbance are not included].

Clearly, intention/purpose is important as a drug which is used in an appropriate manner and in the appropriate dose if and when required for the benefit of the person concerned would not be considered as chemical restraint. On the other hand, there are differences of opinion as to what constitutes appropriate use, an appropriate dose and the need for the drug to be taken. Some would argue that the use of antipsychotic drugs is not the appropriate treatment for BPSD and does not effectively treat medical symptoms due to the risks and side effects linked to its use. This will be discussed further in subsequent sections of this report.

Prevalence and use of chemical restraint

The prevalence of psychoactive medication in the context of long-term residential care of the elderly is quite high (Meyer et al., 2008;). A comparison of the use of chemical restraint in 31 specialised dementia units and 32 traditional units housing a total of 625 people with dementia was carried out in America. It was found that 45.3% and 43.4% respectively of people with dementia in the two units were being given such medication on a routine basis (Sloane et al, 1991). In Germany, over fifteen years later, an observational study of 30 nursing homes with 2,367 predominantly female residents in Hamburg found that 52.4% were receiving at least one psycho-active drug (Meyer et al., 2008). However, there is some evidence that the use of anti-psychotic drugs may be declining. A national audit of nearly half of all general practitioner  (GP) practices in the UK revealed that the prescription of antipsychotic drugs to people with dementia In the UK, fell from 17% to 7% between 2006 and 2011 (Kmietowicz, 2012). On the other hand, it was also found that there were huge regional differences ranging from 2% in the London area to 13 % in the North West of England. Moreover, another study reported a stable use of anti-psychotics between 1990 and 1997 amounting to approximately 30% of cognitively impaired residents whereas 48% of 348 people with dementia were reportedly taking antipsychotics in three large cities in England in 2006 (Lindesay, Matthews and Jagger, 2003; Fossey et al., 2006). It is possible that there is some difference in the rate of prescription of such drugs across Europe and within individual countries. However, it is difficult to obtain generalisable and comparable data.  A large proportion of people with dementia do not have a formal diagnosis of dementia but may be being prescribed antipsychotics (Banerjee, 2009). Similarly, some studies are not limited to people with dementia but cover all residents in nursing homes or cognitively impaired residents. Also, different methods of data collection are used such as self-report, observation and statistics from suppliers of the drugs or patients’ records.

Common medical grounds for the prescription of various drugs to people with dementia

Chemical restraint is used as means of control and even punishment, sometimes to address staff shortages and sometimes due to lack of knowledge and skill to manage certain behaviours which are common in people with dementia, especially in residential care settings. Behaviour and mood, which are considered by others as being problematic or abnormal, are often considered as legitimate grounds for the administration of psychotropic medication but according to Banerjee (2009), antipsychotic drugs have only shown minimal efficacy for the treatment of BPSD. Benzodiazepines are also psychoactive drugs with a sedative effect. Their use is very common but poorly supported by evidence. On the contrary, there is evidence connecting them with cognitive decline, and increased frequency of falls and fractures.

As certain pharmaceutical drugs are commonly used to sedate people, there is growing concern over the “PRN” prescription of certain drugs such as psychotropics.  PRN is an abbreviation of the Latin “pro re nata”(as needed) and means that a medical practitioner prescribes a drug which nursing staff may then administer to patients or residents at their discretion. Consequently, when the drug is actually administered (i.e. when it is later given to the person with dementia as opposed to when it is prescribed), it is nursing staff who decide. They are more directly involved in the care of people with dementia and therefore may be affected in their decision making by their own needs based on the behaviour of the person to whom they are administering the drug.

Medical impact and possible side effects of such drugs on people with dementia

Possible side-effects may differ slightly from one drug to the next but, according to Jacques (2001), might typically include:

  • constipation, confusion, unsteadiness, low blood pressure and heart problems in the case of anti-depressants;
  • loss of effect and dependency on the drug in the case of hypnotics or sleeping tablets and anti-anxiety drugs;
  • tremor, muscle spasms (dystonia), restlessness (akathisia) and abnormal movement of the mouth and tongue (dyskinesia) and low blood pressure in the case of conventional anti-psychotics, as well as weight increase, sedation and unsteadiness in some of the atypical anti-psychotics, which nevertheless have relatively fewer side effects.

There is increasing evidence that the use of risperidone, olanzapine, aripiprazole and quetiapine is associated with a serious risk of adverse side effects, such as cerebrovascular adverse events (CVAEs), including death (Banerjee, 2009; Passmore et al., 2008). This has resulted in a series of warnings about the use of atypical anti-psychotics (e.g. from the US Food and Drug Administration (FDA) in 2003, the European Medicines Agency (EMA) in 2004 and the UK Medicines and Healthcare products Regulatory Agency in 2004) (Banerjee, 2009). More recent data (Trifiro et al., 2007) suggest that all antipsychotics, conventional as well as atypical, are associated with similar increased risk for all-cause mortality and cerebrovascular events. Subsequently the warnings were extended by the FDA (2008) and the CHMP (EMA, 2008) to include both types of antipsychotic drugs.

Risperidone is licensed drug for BPSD but with a strict indication, only for a short-term treatment (up to 6 weeks) of persistent aggression in patients with moderate to severe Alzheimer’s disease unresponsive to non-pharmacological approaches and when there is a risk to harm to self or others. Olanzapine has also evidence that support its use but in the same concept and with caution. See also the Maudsley Prescribing Guidelines in Psychiatry (11th edition) for evidence-based information about the treatment of BPSD.

In a full consideration of the negative effects of antipsychotic medication, Banerjee (2009) points out that the findings of some studies may be misleading in the sense of underestimating the risk. This is because in an attempt to focus uniquely on the effect of the drug, the researchers may have chosen subjects who are not representative of the people who will eventually take the drugs being tested (i.e. the latter are likely to have more severe dementia, more general medical co-morbidity and hence be at greater risk in the first place of cardiovascular adverse events or death). As an example of the risk posed by the use of antipsychotics to people with dementia, Banerjee (2009) makes the following calculation:

“If, at any one time, we are treating approximately 180,000 people with dementia with antipsychotic medication in any year, and we make the conservative assumptions that the average treatment episode is the 6-12 weeks used in trials, this equates to the following: an additional 1,800 deaths per year; and an additional 1,620 CVAEs, around half of which may be severe.” (p.29)

Ethical issues linked to the use of chemical restraint

The use of psychotropic medication without people with dementia having given their informed consent (i.e. either by concealing the medication, forcing them to take it or administering it without asking) fails to respect their autonomy. This deprives them of the right to decide whether to be treated.

In some cases, the behaviour of the person with dementia may be perceived as deviant in the sense of not fitting into the socially defined norm. This may result in treatment which suppresses their right to behave in a way which is congruent with their personality and feelings. Agreement that the person is acting out of character is not sufficient grounds to try to stop such behaviour. It cannot be presumed that everything that a person with dementia does is caused by changes in the brain. People react to the experience of dementia and to the way they are treated in different ways and having dementia may result in people being in different situations to those they may have experienced in the past. Also, their behaviour may reflect their emotional and psychological reaction to their symptoms and they have a right to express this. The issue becomes complicated when the person’s behaviour greatly disturbs or offends other people. This creates a dilemma whereby it is necessary to balance the interests of all involved which is linked to the principles of beneficence, nonmaleficence and wellbeing. Even if the interests of other people are eventually given priority, this does not mean that the use of medication is justified.

Basson, Brugeron and Herson (2012) caution against the misuse of psychotropic substances as a kind of chemical straitjacket which deprives people of the possibility to express their wishes, thus making it easier for other people to decide on their behalf. However, they also reflect on the appropriate use of medication, including psychotropic drugs, to facilitate decision making. They give the example of a person who is anxious, depressed or delirious and therefore cannot apprehend a situation in an objective manner due to their emotional state and distorted perception of the environment. In such a case, they argue, the use of certain psychotropic drugs might allow for a smoother decision-making process involving rather than excluding that person. However, it must be considered whether the risks linked to the use of psychotropic drugs outweigh the possible benefit of participating in decision making, whether such a response is not disproportionate and whether other approaches might be more appropriate.


Chemical restraint can interfere with the ability of people with dementia to communicate with others. This may have negative consequences on maintaining ties with loved ones, eventually leading to fewer visits. This, in turn, may contribute towards social withdrawal and isolation, and jeopardise relationships, which is not conducive to promoting the wellbeing of people with dementia. Moreover, apathy is generally considered as a cause for concern and not a state of wellbeing yet chemical restraint may lead to a state of being which is similar to apathy (e.g. a passive state, indifference and a blunting of emotions).

Respecting individuality

It is important to consider whether medication is the right/best response for the person with dementia: Often, behaviour is described as a behavioural or psychiatric symptom of dementia and attributed to neuro-degeneration. However, such explanations are increasingly being challenged along with the recognition that some of the behaviour which other people find difficult to manage is the result of frustration, boredom, fear and self-expression of people with dementia. A more constructive approach to managing various behaviours, which are considered as problematic (often to other people), is to try to understand what the person is communicating through that behaviour as it may be a problem or unmet need which can be addressed (Kitwood, 1997 – in Hughes, 2008). Some behaviours, such as repetitive or continual walking (often called “wandering”) would be better understood rather than stopped (Marshall and Allan, 2006). There is growing recognition of the need to find other ways to manage such “symptoms” or behaviour. For this reason, what might have been considered as routine treatment a few decades ago is considered as a form of restraint. Alternative terms currently used (such as “chemical cosh”) reflect this gradual change in perspective. 

In the context of the “users and survivors of psychiatry” movement, Minkowitz (2006) advocates the right to be different and for differences not to be the object of attempts to change a person from one state of being to another against his or her will. This could also apply to people with dementia who may be subjected to chemical restraint in order to make them calmer, passive and easier for others to “manage” rather than simply permitted to be as they are. Even in the presence of psychotic symptoms (feelings of persecution, delirium and hallucinations), the use of drugs is not necessarily the best option. Other options are possible. Other symptoms such as “wandering”, agitation and anger do not always even need to be stopped (Basson, Brugeron and Herson, 2012?).

Beneficence and necessity

A key issue with regard to the use of medication is whether it is necessary and in the best interests of the person with dementia. If it is, it would be ethical to prescribe or administer it but if not, it could be perceived as a means of restraint.

It would be unethical to prescribe any kind of medication to people with dementia if the need for such medication had not been accurately assessed and established. However, many GPs receive very little training in dementia and are not confident in their ability to manage and treat people with dementia (Commissioning Support for London, 2009; Koch and Iliffe, 2010). This problem is often exacerbated by the short duration of consultations and difficulties communicating with people with dementia (Banerjee, 2009: Boise et al., 1999; van Hout et al., 2000).

Based on a review of the literature on agitation, Bidewell and Chang (2010) point out that there is no agreed, coherent definition and measurement of agitation. They argue that despite the existence of reliable tools to measure agitation, they are often time consuming and consequently, there is a tendency to ask for caregiver reports which may be prone to bias. They consider models of agitation based on biological and unmet needs, as well as on behavioural and environmental factors, which suggest the need for non-pharmacological approaches to dealing with agitation.


Psychotropic medication may have a considerable impact on the body and mood of the recipient. The issue of proportionality is therefore particularly relevant. It is necessary to consider whether such a measure can be justified as a means to address an issue such as agitation for which a range of alternative options are possible with fewer and/or less serious side effects or consequences. In other words, is the use of psychotropic medication the appropriate level of intervention for the identified need?


The prescription of anti-psychotic drugs with serious side effects to people with dementia has been long accepted and only begun to be challenged in recent years. In addition to the medical arguments against this practice, based on the serious side effects and the ethical issues of autonomy, beneficence and nonmaleficence that this practice raises, there is also the issue of justice/equity. It is unjust to expose certain groups in society to more risk than would be acceptable for other groups. One might consider whether it would be acceptable to manage certain behaviour or symptoms of other groups of people by administering psychotropic substances on a regular basis. It is clear that this would not be acceptable in the case of infants who are particularly prone to crying or screaming or of hyperactive toddlers. The fact that this is still the response to managing the behaviour or challenging symptoms of people with dementia is perhaps a reflection of ageism or of the stigma of dementia, both of which involve devaluation and discrimination.

Covert administration of drugs

Often, people who are in hospital are given one or more tablets and not told what they are. Some people, including those with dementia, may take the tablets based on trust or obedience (i.e. doing as they are told) and implicit consent might be assumed, whereas others may ask for clarification and thus take them on the basis of informed consent, presuming that they are capable of giving informed consent (which may not be the case). Whilst such medication is given openly, there may be an element of psychological coercion based on unequal power relations between medical professionals and patients (McKillop, 2013). Nevertheless, in both cases, the person is aware that they are receiving some kind of medical treatment. This is not the case when medication is administered covertly. This can occur in care homes, hospitals and even in people’s own homes.

Crushing tablets or opening capsules (e.g. to disseminate in food or drink) may be harmful and therefore fails to respect the principle of nonmaleficence. Nurses who do this may be held accountable and prosecuted for negligence. Informal carers may be unaware of the dangers involved. Attempts to hide medication or alter its current form occur for a variety of reasons (e.g. due to a person’s refusal of the medication, their tendency to spit out tablets, difficulties swallowing, dislike of the taste or not understanding what to do with the tablet when administered). Irrespective of the reason, there are risks involved which are linked to the formulation of the medicine. For example, some drugs are specifically designed to prolong the time taken to be absorbed in the body or to deliver the drugs to a particular part of the gastrointestinal tract (Greenwall, 2003). This is achieved by the coating of the drug. Enteric coating, for example, does not dissolve in acid and is therefore released in the intestine rather than the stomach. Osmotic coating, on the other hand, allows water to pass through it which results in a slow release. Another option is coated beads for which a thicker coating results in a longer release time. Such tablets must therefore be swallowed whole in order to be absorbed as needed. Crushing tablets or opening capsules may therefore result in an initial overdose of the medication followed by a period with no medication. It is not always possible to know whether a drug has a modified release system. 

With the covert administration of drugs, the respect and dignity of the person with dementia is at stake in that administering drugs in this way involves deceit and the betrayal of the trust that the person with dementia places in those responsible for their care and wellbeing. Such practices may jeopardize the trust that they place in professional or informal carers in other respects. It may also result in distress and anxiety if the person experiences side effects and cannot discuss them openly with medical staff as the latter are not admitting that a drug has been administered. Alternatively, the person may fail to mention disturbing side effects and simply attribute them to their condition.

Covert drug administration also fails to respect the principle of autonomy because, as mentioned earlier, it does not give a person the opportunity to give informed consent or refuse the medication administered. The person is disempowered as they are not aware of what is happening and are thus denied the opportunity to react. In concealing the administration of a drug, medical staff are subjecting the patient to possible side effects of various degrees of severity without their agreement.

The Royal College of Psychiatrists Ethics Sub-committee states that the practice of disguising medication in the food of patients who are unable to consent is unethical and should not take place (UKPPG, 2006). The Sub-committee makes a distinction between a deceitful process and a co-operative process for patients who find taking tablets difficult. Administering drugs in food is not considered unethical in the case of a person who is unable to consent provided that the decision has been discussed within a multi-disciplinary team and with relatives and carers and the person’s advocate if there is one, that the decision is recorded and a doctor has prescribed the administration in this way. The person must nevertheless be informed each time that medication is administered in that way that it is in the food, which medication it is and why they are receiving it.

1.2.3       Zero tolerance of the use of physical and chemical restraint

Based on a reflection about the ethical issues linked to the use of physical and chemical restraint, there seems to be little if any justification for its use with people with dementia. Yet, it is persistently used throughout Europe. Alzheimer Europe is opposed to the use of all forms of restraint on people with dementia. Whilst accepting that there may be exceptional circumstances or isolated incidents which justify its use, this is considered unlikely and the rare exception to the rule. The working group was therefore very pleased to hear from Dr Antonio Burgueño Torijano about the “Untie the Elderly and People with Dementia Programme” in Spain (of which he is the Director). He argues in favour of zero tolerance of the use of physical and chemical restraint on people with dementia and looks at its prevalence in Spain. He explains how reasons frequently given for the use of restraint are not supported by empirical research and constitute popular myths.

Drawing on an extensive review of the current literature on restraint, a consideration of the ethical issues involved and over three years’ practical experience of zero tolerance of restraint in Spain since the start of the programme in 2003 (based on fieldwork and data from 687 facilities and 29,332 residents), Dr Burgueño Torijano presents a convincing argument comprised of ten key points which describe why restraint is unethical and poor clinical practice, how zero tolerance of restraint can be practically achieved and the benefits of this approach for all concerned. Part of the philosophy of zero tolerance is the acceptance of a certain degree of risk and restraint is not considered an option except in rare situations where there is a serious and imminent threat to the integrity of the person with dementia.

The term “zero restraint” should be considered as an attitude rather than as a goal or the description of the outcome i.e. total absence of the use of restraint. When talking about restrictions on people with dementia, discussion arises about the use of psychotropic drugs. Some clarification is therefore needed regarding the term "zero tolerance". The proposal of a zero tolerance attitude to the use of restraint is not a zero tolerance of the use of psychotropic drugs but a zero tolerance of the use of these drugs for purely restrictive purposes. There are authorised indications of psychoactive drugs. As described earlier, in certain situations, this may be justifiable and in keeping with good care. Psychotropic drugs are part of the essential medical tools used by doctors to manage certain problems and in some cases are justifiable (i.e. when their use is appropriate, effective, clinically justifiable and proportionate to the perceived need, and not prescribed or administered on a routine or prolonged basis).  According to Dr Burgueño Torijano, establishments which remain restraint-free are obliged to keep a tight control of the use of psychotropic drugs and polypharmacy, and to enhance the environmental and organisational measures to prevent falls and serious behavioural problems (as can be seen in Appendix 1, the inappropriate use of psychotropic drugs may result in the use of physical restraint) . He maintains that this can only be achieved and maintained by means of an attitude of zero tolerance.  In order to implement zero tolerance, professionals need to be well trained and able to act within the framework of a special management policy within the facility and in a supportive context. The Spanish programme was therefore created to serve as a reference. It provides a technical framework, performing outreach activities, training and recommendations etc. The accreditation system of restraint-free facilities in Spain, developed by the Untie Programme, has resulted in greater public awareness of restraint-free facilities. The full text of this argument can be found in Appendix 1.

Measures to rule out or limit the use of restraint can also be found in other countries. In the Netherlands, for example, the focus is on using fewer restrictions of freedom and the Health Care Inspectorate monitors progress with this. According to Frederiks (2013), zero tolerance policy involves much more than informing carers that the use of restraint is no longer permitted. A zero tolerance policy involves supporting management, raising awareness of carers and providing training to improve their expertise etc. This is essential to avoid the creation of new problems which might lead to an increased use of medication. This point is echoed by Burgueño Torijano who states that an attitude of zero tolerance is required in order to prevent new problems from occurring. However, there are cultural and also legal differences in the definition of restraint. In the Netherlands, for example, all psychotropic drugs, irrespective of their indication, must, due to recent changes in legislation, be considered as a form of restraint (Frederiks, 2013). This may also be the case in some other countries.

1.2.4       The use of force/physical intervention

People with dementia are sometimes physically restrained not through devices, equipment or chemicals but by means of hands-on contact. This presumably occurs when other people cannot manage a difficult situation and lack the skills to communicate with the person with dementia. The use of force might, for example, include pushing or pulling a person (e.g. into a van, building, chair or bed), grabbing hold of their arms to force them into a particular place or prevent them from leaving or pinning them down or immobilising them (e.g. in order to administer treatment or “care”). Physical violence might also be used such as slapping, hitting or shaking the person. This would also be classed as abuse and violence, as might any illegitimate use of restraint. As stated earlier, the use of restraint can rarely be considered as being legitimate. In some countries, legislation addressing abuse or violence contains special reference to potentially vulnerable groups such as older, dependent people and people with mental or physical disabilities. Sentences are often more severe in the case of abuse against a person for whom one is legally responsible for their care. The moral justification for the increased sentence perhaps reflects the moral obligation of solidarity and care towards people who are less able to protect themselves.

1.2.5       Coercion

Coercion involves the use of threat, intimidation or another form of pressure to force another person to do something that they would not do by choice. Again, this does not necessarily involve the use of devices or chemicals. Coercion may involve the infliction of physical or psychological pain or pressure but sometimes a credible threat may achieve the desired result. Often the initial use of force is what makes future threats credible. In other cases, the threat is based on deceit as the thing that is threatened could not occur but the person does not know this.

The use of coercion is relevant to the principle of autonomy as it involves denying a person the right to determine their own actions. It is also reflects failure to respect a person’s dignity and the abuse of power. The principle of nonmaleficence is not respected as coercion may result in significant psychological and even physical harm. Allowing coercion to occur represents a failure to protect people with dementia from harm.

In nursing homes, people have limited opportunities to deal with the situation. They have nowhere to go and in many cases cannot avoid contact with the person exercising coercion. They may also feel powerless in that there is an unequal power relationship between residents in a nursing home and care staff. In care homes, certain power relations may in some respects differ from those common in society at large and in other respects be similar. For example, people from ethnic minority groups may find themselves in a position of power over residents from the dominant ethnic group in that society (although not necessarily within the professional hierarchy). Other forms of pre-existing unequal power relations may nevertheless affect the experience of coercion (e.g. the use of coercion by male carers against female residents/patients). This also applies to the use of physical restraint in general. The life history of each resident/patient is not always known. People with unresolved psychological trauma, a history of physical abuse or having experienced abusive relationships may be more deeply affected by coercive measures than others.

The problem may be further exacerbated by the fact a person with dementia may have difficulties communicating and their credibility may be questioned. In nursing homes with poor standards where staff lack appropriate knowledge, certain coercive practices may even be tolerated or promoted by the management.

1.2.6       Environmental restraint

Environmental restraint would cover measures to restrict the freedom of movement of people with dementia by means of adaptations made to the living environment. Examples might include:

  • locking devices (either mechanical locks, code-operated locks or complicated opening mechanisms which people with dementia might find difficult to operate);
  • staircases with no handrail;
  • poor lighting;
  • confusing colour contrast on the floor;
  • mirrors on doors to make people think there is someone else there;
  • lack of transport.

Debatable/depends on circumstances:

  • camouflaged exits (e.g. using a curtain, a continuing barrier along a wall or signs);
  • walking areas which all lead back to the point of departure;

The above examples raise a few different issues. The first example could be construed as consisting of the unlawful restriction of liberty if used, for example, in a person’s home or day care centre which people voluntarily attend. Voluntary residents with dementia can sometimes be seen at the door of a day care centre, frantically beating on it with the hope of being let out. Sometimes, people with dementia beg visitors to let them out so that they can go home. This can be heart-breaking. They are clearly not enjoying the freedom and security to move above in a restricted area. Staircases with no handrail and poor lighting are dangerous especially for older people with dementia who may be unstable on their feet. Also, people with dementia may have problems with visual perception and tend to need more light than usual. Confusing colour contrasts on the floor may make people with dementia uneasy or even stumble (e.g. if the contrast appears like a hole or a step). Lack of transport forces people to stay where they are, especially if the amount of money they are permitted to carry about has been limited so that that they cannot get a taxi or bus. This is especially problematic in places which are situated at a considerable distance from towns. It means that people are not free, for example, to go for a drink in a pub or to a betting shop, the hairdresser’s or a cinema.

Nowadays, camouflaged exits and circular walking areas are generally considered as good dementia design but if not used appropriately may serve as a means of restraint. For example, whilst some people may enjoy ambling around the building, perhaps talking to other residents, or getting some fresh air in a garden in the security that they will find their way back, others may feel trapped. Circular walking areas enable people with dementia the freedom to move about freely and in security but only within the area to which they have been granted access. Some public buildings, such as such town halls, churches, museums and educational institutions, also prevent people from entering private areas. Nevertheless, people in such places have the freedom to leave the building, whereas this is often not the case for people with dementia. People with dementia are not only being denied access to certain areas but also of the right to leave the building. Used in an appropriate care context and for the individual needs of people with dementia, certain measures (such as camouflaged exits and circular walkways) could be considered ethical in the sense that they provide a safe environment, give people the opportunity to exercise and may encourage social interaction. Staff therefore need to be attentive to how various design features are experienced by individuals and to be ready to intervene when necessary.  

Environmental restraint therefore touches on a wide range of ethical issues, covering respect for people’s autonomy, what benefits, disturbs or harms them and whether it is justified to use such measures. The concepts of singularity and historicity are also important as environmental restraint is usually a global approach for which differences between people cannot easily be taken into account.

Nevertheless, according to Blackman et al. (2003), the lack of published guidelines on the design of outdoor dementia-friendly environments seems to reflect a presumption that the lives of people with dementia are “home-based,” although they acknowledge that there are greater technical difficulties in trying to control outdoor environments. The failure to respond to the needs of people with dementia outside of the home and to take into account their impairments results in a loss of mobility and hence of freedom to choose where to go and what to do. Loss of such freedom may result in social isolation, depression and premature loss of remaining skills. The environment in this case consists not only of the physical environment but also of attitudes and practices within the public domain which serve to restrain people with dementia.  

1.2.7       Psychological means of restraint

Deceit has already been mentioned in relation to the use of coercion and the covert administration of medication. It is also an unethical psychological tactic used to prevent people from moving about or leaving a building or area which involves taking unfair advantage of a person’s cognitive impairment. A typical example is the situation whereby a person is led to believe that someone is coming soon to pick them up (e.g. from a day care centre). They then spent all day by the door waiting for that person or van to arrive and take them home. Not only do they miss out on the opportunity to carry out enjoyable activities (if there are any available) but they may worry about missing the person and being left there. Meanwhile, staff have peace of mind in that they know where the person is and what they are doing (i.e. waiting).  There have even been cases, where a mock bus stop has been created for that purpose. The person can sit down and occasionally go back inside for a cup of tea whilst waiting. It could be argued that this reduces distress but at the same time it could be considered as disrespectful, amounting to a mockery of the person. Whether such practices are disrespectful depends to some extent on how they are implemented. If the bus stop, for example, becomes a place for social interaction where people sit together on a comfortable bench, protected from the wind and rain, and chat or look at pleasant surroundings, the effect may be beneficial despite the fact that it is misleading. If people are left to stand alone at a solitary post in a state of distress and expectation, it cannot be said to be beneficial in any way.  

The use of deceit may also have more far-reaching consequences in that it may deprive a person of their freedom and result in them losing their home, their belongings and their social contacts. This is the case, when a person is tricked into believing that they are going into respite care for a short time when in reality they are being moved into residential care. Their cognitive impairment renders them particularly vulnerable to such trickery.  

Other psychological means of restraint include the use of emotional blackmail (which would also be coercion), false promises and measures which challenge a person’s self-esteem and dignity such as ridicule, infantilisation, objectification, ignorance and humiliation. For example, a care worker may complain loudly when asked by a resident to be accompanied to the toilet thereby drawing attention to a private issue or using childish terms which are belittling.  This might result in the person restricting what they drink to avoid the embarrassing situation or suffering in silence, remaining seated and risking incontinence.

The power of the white coat may also serve as a psychological means of restraint. Some people, including those with dementia, tend to defer to doctors and may therefore comply with restrictive measures which are not necessarily in their best interests simply because they have been ordered by a doctor. Such deference may be based on courteous respect or difficulty challenging the perceived authority of doctors.

Cultural norms and procedures are sometimes also used against people to prevent them from doing what they want.  This might be based on pride in one’s appearance, common practice or respect for others. For example, ensuring that a person spends the whole day in nightwear, does not have access to their traditional headdress or only has slippers may be sufficient to prevent some people from leaving a building despite being free to do so. For example, some people would not go out in public without covering their head or body in accordance with their religious beliefs and traditions although many would wish to wear such attire indoors as well if other people were present. Some people might willingly do what others want to avoid getting another person into trouble. Others may adhere to restrictive rules and regulations as they have come to believe that they are not worthy of the freedom that other people enjoy or feel a burden to society. 

Freedom of movement and choice can also be limited by attitudes and practices which provoke psychological or cognitive responses such as shame, confusion, anxiety or fear. This is particularly reprehensible bearing in mind the memory loss and confusion often experienced by people with dementia. In some cases, this could also be perceived as intimidation tactics/coercion, as illustrated in the following example.

“One staff member prevented some people moving around by placing their hands on their Zimmer frames and asking several times where people were going, which resulted in some people looking anxious and submissively sitting back down.”(CSCI, 2008 cited in Hughes, 2008)

1.2.8       Assistive technology

Although assistive technology (AT) can be used to facilitate independence within the home or in residential care settings, and for communication, wellbeing and recreation, its use for surveillance and monitoring often receives most attention in relation to ethical issues as it tends to be associated with the restriction or loss of freedom. However, whilst surveillance and tracking devices can be used to restrict freedom, they can also be used in an ethical manner to promote autonomy and increase freedom (Niemeijer et al., 2010; Zwijsen et al., 2012). Examples of surveillance and tracking devices or systems include:

  • the direct tracking of a person by means of global positioning systems (known as GPS) which locate a person by satellite (due to an ankle or wrist device, a pendent or a mobile phone) and then transmit details of their location via a mobile phone network to a mobile phone, computer or call centre.
  • tagging which involves the person with dementia wearing a special device (e.g. on the ankle or wrist) which emits a signal if the person leaves a designated area such as their home or garden. This can be combined with a numeric paging device which alerts the carer. 
  • access control systems (equipment which recognises a chip card and acts on the environment accordingly by locking or unlocking doors, blocking elevators and recording access to certain areas).
  • the identification of a trail of locations which indicates where a person has been (known as “bread crumbing”)
  • the use of (infrared) technology to detect immobility within the home or that a person has not returned home by a certain time or has left the house at an unusual time such as in the middle of the night.

It has been suggested that it would be wrong to consider AT as “morally neutral” because some devices and systems include characteristics which affect the rights of those using them and cannot be removed as they are “substantially rooted in the conception of the application” (Casas et al., 2006). In the context of residential care, debates over the ethical issues linked to the use of AT tend to focus on the moral acceptability of the effects of its use, with conflict often arising when the interests of institutions do not correspond with those of the residents.

A major issue related to the use of the tracking and surveillance devices is that of balancing autonomy and freedom of movement with concerns about safety and the possible need to protect people with dementia. Tracking and surveillance devices can be seen as contributing towards autonomy by enabling people to go out alone or move about freely without having to ask permission. It gives people the freedom to decide when and where to go (sometimes within certain established or agreed limits). This may help maintain morale, self-esteem and social contacts, provide exercise and generally improve quality of life. Those which monitor movement may enable people to get along with their daily lives in relative privacy but with the assurance that assistance is possible if and when needed. 

Zwijsen et al. (2012) describe how surveillance technology can be used as a means to provide general safety, additional safety or more freedom. They highlight its limitations such as it not being able to prevent falls, not guaranteeing quick help, not always working properly and violating privacy. However, it is clear that it can be used constructively as one measure within an overall approach to care. The issue of privacy can also be addressed, assessed in relation to alternative approaches and attempts made to minimise the impact of the use of surveillance on privacy. Some loss of privacy might be preferred by some residents when compared to potential gains in freedom of movement (e.g. a preference for movement monitor or even a video camera in a bathing area to enable a person to bathe alone).

On the other hand, the same devices may be perceived as a means of restriction with an over-emphasis on safety (perhaps influenced by fears of litigation amongst professional carers), or a form of coercion or social control (preventing people from carrying out certain activities) and may lead to frustration and increased dependency on others. The principles of beneficence, nonmaleficence are important in this respect. There are also issues of privacy and dignity to be considered, and the risk of paternalism and of stereotyping people with dementia as vulnerable (i.e. applying a blanket label of “vulnerable” rather than assessing the possible vulnerability of each person in relation to a specific situation).

The way that AT is experienced depends on many factors (in addition to the needs and wishes of each person with dementia) such as the framework in which it is used, people’s individual judgements of its benefits, risks and drawbacks, the meanings they attach to it, its visibility and aesthetic qualities, and the extent to which users have been consulted and consented to its use. The ethical issues linked to this topic were discussed in-depth in a report published by Alzheimer Europe in 2010 and will therefore not be repeated here. They can be consulted at:

1.2.9       Recommendations on the freedom to live in the least restrictive environment

Government, policymakers and service providers

Policymakers and service providers should strive for the least-restrictive environment for people with dementia.

Allocation of resources at all levels should reflect the commitment to achieving least restrictive environments.

A legal framework and guidelines should be developed to protect people with dementia against the use of restraint.

An awareness raising campaign should be developed in order to inform people about the relevant legal and ethical issues linked to the use of restraint.

A definition of restraint should be included in the framework and guidelines which covers physical, chemical, psychological, electronic and environmental measures but is sufficiently broad to include any other measures which serve as a means of restraint through the way they are applied.

The use of restraint on people with dementia without informed consent should be considered as unethical and a form of abuse unless justification can be provided to prove the contrary.

The use of restraint on people with dementia, if personalised and respectful of each individual and of his/her unique situation and needs, may be considered as part of good care provided that the person has given informed consent.

An independent organisation should be established with responsibility for investigating the use of restraint to which people with dementia (and other residents in homes), carers and healthcare professional can report their concerns anonymously.

This organisation should be granted the power to make unannounced inspections, to issue recommendations to organisations to change their practices and take appropriate legal steps if necessary.

Governments should set targets to reduce the prescription of anti-psychotic medication for people with dementia. 

Clear guidelines should be developed for doctors on the prescription of anti-psychotic medication for people with dementia, covering the dangers, possible alternatives to dealing with BPSD and challenging behaviour and the need to reduce such prescriptions.

In cases where the police (or other relevant authorities depending on the country) might be called in to restrain a person with dementia or ensure his/her transferal to a care facility, they should have at least a basic understanding of dementia and take measures to minimise distress and discomfort linked to the use of any measures of restraint.


People with dementia should not be submitted to the use of restraint.

Restraint should only be tolerated in extreme situations where the physical and mental integrity of the person with dementia is in serious and imminent danger.

Such situations are considered extremely rare and should be avoided by careful planning and reflection.

The restraint of a person with dementia who is unable to consent should be permitted only after discussion within a multidisciplinary care team and with the relatives, carers and advocates of the person with dementia.

The use of restraint should not be considered an option but rather a failure on the part of carers to provide good care.

A person with dementia who has been physically restrained should not be left unattended.

Restraint should never be used on a routine basis even if initially prescribed by a doctor.

Before considering how to deal with BPSD and challenging behaviour, it should be determined for whom such symptoms and behaviour are disturbing.

The use of any form of restraint should be authorised by a doctor and documented.

Failure to obtain medical authorisation for the use of restraint should be considered as a negative factor in inquiries into suspected abuse linked to the use of restraint.

Medication should not be concealed in the food or drink of a person with dementia who is able to consent to drug treatment unless they have given such consent.

Whenever medication is administered through food and drink, the person receiving it should be informed and provided with the reason for this approach as well as details of the medication administered in that way.

Care establishments

Care establishments should develop a clear policy about not using restraint.

Care establishments should develop and implement a policy of zero tolerance of restraint.

Care establishments should support healthcare professionals in providing restraint-free care.

Care establishments should ensure that measures of restraint are not readily available to personnel who provide care or treatment to people with dementia.

Care establishments should ensure that the demands of the organisation with regard to the provision of care do not encourage care staff to restrain residents.

Care establishments should discuss with the carers and relatives of residents with dementia the need to promote autonomy and of the necessity to allow some degree of risk.

Care establishments should have a clear policy for dealing with possible injuries which residents may suffer in the course of their stay, including responsibilities in possible cases of litigation.

Healthcare professionals

Healthcare professionals should receive training in the use of non-pharmacological interventions to deal with BPSD and challenging behaviour.

Doctors should be obliged to justify that any prescription of psychotropic medication is appropriate, effective, clinically justifiable and proportionate to the perceived need.

Doctors should ensure that any prescription of psychotropic medication is time limited.

Doctors should consider whether BPSD and challenging behaviour are caused by other factors which might require specific treatment or non-pharmacological approaches.

Psychotropic medication should never be prescribed or administered on a routine basis.

The appropriateness of the prescription for any person with dementia already taking antipsychotic medication should be reassessed.

When psychotropic medication is administered on a PRN basis, the reason for administering it at a given time should be documented, as should the initial prescription. 

Healthcare staff and doctors prescribing or administering psychotropic medication to people with dementia should ensure that the potential benefits outweigh the risks for each person receiving them and be able to justify this.

Healthcare professionals should be informed about the risks involved in using restraint on older people and people with dementia.

Healthcare professionals should receive training on how to reflect on the ethical issues linked to the use of restraint.

Healthcare professionals should be informed about the ethical and legal implications linked to the provision of care as well as to the use of restraint.

Informal carers

More research should be carried out into the use of restraint at home.

Informal carers should be informed about the legal and ethical issues linked to the use of restraint.

Informal carers should be provided with training to deal with BPSD and behaviour which they may find challenging.

Informal carers should have access to affordable, appropriate and timely support as well as respite in order to address issues which might lead to the use of restraint.



Last Updated: Tuesday 14 May 2013


  • Acknowledgements

    The above information was published in the 2012 Report "The ethical issues linked to restrictions of freedowm of people with dementia" as part of Alzheimer Europe's 2011 Work Plan which received funding from the European Union in the framework of the Health Programme. Alzheimer Europe gratefully acknowledges the support it has received frm Fondation Médéric Alzheimer Europe and the Alzheimer Europe Foundation for the development and publication of this report.
  • European Union
  • Fondation Médéric Alzheimer