Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Practice review

Using dolls to enhance the wellbeing of people with dementia

  • Comment

Doll therapy in dementia care remains a controversial intervention but it may well provide people with sensory stimulation and purposeful activity


Patricia Higgins,RMN, MSc, QN is a memory service nurse specialist, Bromley memory service, Oxleas NHS Foundation Trust. 


Higgins P (2010) Using dolls to enhance the wellbeing of people with dementia in residential care. Nursing Times; 106: 39,early online publication.

An increasing body of evidence suggests the use of dolls can have a positive impact on people with dementia  in residential care. This literature review explores the benefits and potential problems of using dolls as a therapeutic intervention in dementia.

Key words Dementia, Wellbeing, Therapeutic intervention

  • This article has been double-blind peer reviewed


Practice points 

  • The use of dolls should be considered as part of a range of therapeutic activities available to people with moderate to advanced dementia. 
  • Not everyone will respond positively to a doll. Like any other intervention it needs to be provided as part of a person-centred care plan.
  • Introducing a doll should be in addition to the activities people with dementia already engage in, and not be a replacement for them.
  • Further studies are needed to help provide more robust evidence on which to base practice.
  • See Moore (2001) and Mackenzie et al (2007) for guidelines on the use of dolls in dementia care.



There are over 700,000 people living with dementia in England and Wales, a third of whom live in care homes (Department of Health, 2009). The well-being of someone with dementia is very much dependent on the environment they are in, and finding ways to enhance the individual’s well-being is the goal of person-centred care.

A growing number of anecdotal reports highlight the therapeutic benefits of dolls for people with dementia (Bailey et al, 1992; Moore, 2001; Gibson, 2005; James et al 2005; Verity, 2006).  Bailey et al (1992) found the dolls provided comfort and companionship for some of residents with advanced Alzheimer’s disease in their care home. They felt dolls provided sensory stimulation and seemed to promote purposeful activity.

Dolls can promote positive changes in behaviour, reducing aggression and agitation (Moore, 2001). They also increase interactions between staff and residents who can talk about the doll and carry out activities relating to it together such as folding its clothes.

Gibson (2005) describes how a doll helped provide comfort to her mother who had dementia and was living in a care home. The doll seemed to help her connect to a place in time where she felt in control. Seeing her mother treat the doll so lovingly gave her daughter comfort to think of her caring for her in that way when she was a baby.

Nursing Times Learning

Online training units, written and reviewed by experts. Earn two hours’ CPD and a personalised certificate for your portfolio.

Subscribers get five FREE learning units and non-subscribers can access each learning unit for £10 + VAT.

Click on the topics below to get started:

James et al (2005) report on how the introduction of a doll to a resident who was very withdrawn resulted in increased interactions with other residents and staff. And Verity (2006) describes how a resident who was always busy trying to “help” the other residents by, for example, moving them out of their chairs, was diverted from doing this once a doll was introduced. This seemed to give her the purpose she was looking for and focussed her attentions on caring for the doll.                    

Empirical evidence

Despite this growing anecdotal evidence, there is still little empirical proof to support the use of dolls as a therapeutic intervention for people with dementia. Mackenzie et al (2006), James et al (2006), Ellingford et al (2007) - all from the Newcastle Challenging Behaviour Service - and Minshull (2009) have endeavoured to provide this.

Mackenzie et al (2006) carried out a pilot project examining the use of dolls. All of the residents at the two care homes were invited to use a doll. The researchers investigated the impact of the dolls on the residents’ level of activity, interactions with staff and other residents and their happiness, agitation and amenability to receiving care. The impact on residents was generally positive across all six domains. Care staff felt the dolls had beneficial effects on the well-being of their residents, and 93% of key workers believed they helped them to communicate with residents.

James et al (2006) introduced dolls and teddy bears into a care home and the impact of the toys was assessed using the same domains as Mackenzie et al (2006), apart from ‘amenability to receiving care’. 93% of residents who chose a toy selected a doll. They compared the doll users to non-doll users across the five domains, and asked staff to describe the perceived benefit of that toy to the resident.

As with Mackenzie et al (2006) there were general improvements in well-being for those residents who chose a doll. There was no observed change in well-being for non doll users. Staff believed the dolls were beneficial and noted how each person formed their own unique bond with their doll. The authors describe how a patient who used to be a GP was seen examining his doll, while a  number of residents sat together with their dolls, prompting staff to call them the “mother’s group”.

Ellingford et al (2007) carried out a retrospective case note analysis investigating the frequency of positive and negative behaviours and incidences of aggression among 66 residents, 34 of whom were doll users, in four care homes.  They examined the notes for three months before and after the dolls were introduced. They found that positive behaviours increased and negative behaviours and aggression reduced in the doll users following their introduction. There was also a significant difference in scores of positive behaviour between doll and non doll users. All the doll users improved on all the behavioural measures compared to those who did not have a doll.

One of the limitations of these studies is that staff were aware who was using a doll and of the rationale underpinning the questionnaires. Another restriction was that the tools used to measure the domains of interest were not validated. Despite the methodological weakness of the studies which the authors acknowledge, they still add weight to supporting the use of dolls as a therapeutic intervention for people in the later stages of dementia.

Doll therapy sessions

Minshull (2009) took a different approach by carrying out doll therapy sessions. She brought a doll to a ward each week for an hour to see if doll therapy could improve the well-being of people with dementia. She used the Bradford Dementia Group Well-being Profiling Tool (2002) to assess the nine patients who showed interest in the doll before and during sessions. The majority of patients reacted positively to the doll, and over the four weeks most experienced an increase in well being during the session. Limitations of the study were the small number of participants and the fact that there was no comparison group.

There are some arguments for and against the sessional use of a doll. By using the doll purely for the sessions, the resident is less likely to form a strong bond with it. This means the problem that James et al (2006) found - where residents became too involved with the doll - are less likely to occur. However, using a doll on a sessional basis means it can only offer comfort and security when it is present and a resident may wonder where it is at other times.  However, it is important to investigate different approaches to help gain a fuller understanding of using dolls and to clarify the most effective approach.

Potential difficulties

Despite these promising results, James et al (2005) urge caution over the use of dolls because of the lack of detailed study or research on their long term use, and the effect they may have on people with dementia. The use of dolls does not come without problems. James et al (2005) report there had been arguments between residents over ownership of the doll. Some residents put the doll’s needs before their own, with one even giving up their bed so the doll could have a good night’s sleep. James et al (2006) report problems with a resident taking several of the dolls and of occasional difficulties when care home workers tried to separate residents from their dolls to carry out personal care.

Preserving dignity

While there is growing support for using dolls in dementia care they remain a controversial intervention. Those opposing their use suggest they are “infantilising” people by treating them as children, which is demeaning and fails to treat the person with respect and dignity (Cayton 2001).

The RCN (2008) states: “When dignity is present, people feel in control, valued, confident, comfortable and able to make decisions for themselves.” This statement can apply to someone with dementia who has chosen to care for a doll.

Andrew (2006) suggests that when a person with dementia has chosen to care for a doll and family members disapprove, it is the family who experiences the indignity rather than the patient.

A doll can give someone with dementia the opportunity to take on a familiar role, which they may identify strongly with and which may have been rewarding for them earlier in their life. It can provide the chance to care for someone and to fulfil a natural maternal instinct and need. Caring for the doll is a meaningful activity, which can provide a sense of purpose.  Moore (2001) suggests that a doll can also fulfil attachment needs, as described by Bowlby (1951) by providing a sense of security and comfort. The dolls can also fulfil the person’s need for play (Moore, 2001). This is usually suppressed in adulthood, but due to the reduction of social inhibitions that can occur in dementia this need may become more prominent.

Ethical issues

Mackenzie et al (2007) note there is an ethical issue if the person with dementia believes the doll to be a real baby and staff collude with this idea. They found most of the people with dementia they observed knew it was a dolls.

One of the four ethical principles is that of “non-malfeasance” - to do no harm (Beauchamp and Childress, 2001). Minshull (2009) asked patients in her study if they would like to hold her doll but if the person with dementia chose to consider it a baby she did not correct them. Mackenzie et al (2007) advise going along with whatever the person with dementia believes and to use the same name for the doll as they do. Andrew (2006) describes this approach as “an avoidance of an unnecessary truth”, and reports that Marzanski (2000) describes this as “justifiable benevolent deception”.

Accepting the individuals understanding of the doll is a form of validation. Entering into their reality and engaging with them in that place in time can lead to a shared experience and may increase well being. The alternative would be to correct them and explain the reality of the situation. The person is unlikely to accept this and may even become angry or decline to engage further either with their doll, which could then have a negative effect on their well-being.

James et al (2005) raises a further ethical dilemma: the person with dementia who chooses a doll but who might have been firmly against the idea before they developed the condition. Although people often think they know how they would feel or what their preferences would be in a given situation, until we are actually in that situation we cannot be sure. It is therefore important to consider the person’s wishes in their present circumstances.


The two case studies in Box 1 add to the anecdotal evidence supporting the use of dolls as a therapeutic intervention for people with dementia.  Not everyone will respond positively to a doll, and like any other intervention it needs to be provided as part of a person-centred care plan. It is important to remember that if a doll is introduced it should be in addition to the activities people are already engaged in, rather than being a replacement for them (Mackenzie et al, 2007). The use of dolls can be considered as part of a range of therapeutic activities available to people with dementia, both in care settling and in their own homes.  It is not easy to predict who will like one (James, 2006), and although it is mostly women who have responded there are some reports of men who have also benefitted from this intervention (Moore, 2001; Verity, 2006; James et al (2006).

A lack of a standardised approach to using dolls with people with dementia remains a concern. However, there are some useful guidelines available (Moore, 2001; Mackenzie et al, 2007). Further studies are needed to help provide more robust evidence on which to base this practice, but in the meantime the anecdotal evidence continues to grow.

Resistance to the use of dolls comes from those who are cognitively intact and lies in their preconceptions. However, people with dementia are able to decide themselves if they would like to engage with a doll. And if their eyes light up when they see it, then it is likely that it is having a positive effect on their well-being.  

Box 1. Case study: ‘The doll gave Mrs Brown some responsibility’

In my role as a memory service nurse specialist I have observed the positive effect of a doll on two patients. Both Jean Brown and Doris Edgar (not their real names) live at home with their husbands and have advanced dementia. 

Mrs Brown’s husband was concerned that as his wife’s dementia progressed it was becoming increasingly difficult for her to engage in activities at home. I had read several articles about the use of dolls and suggested he might try her with one. I gave him a copy of an article (Moore, 2001) and some guidelines for using them, to help him consider the idea.

Mrs Brown’s husband contacted me soon after to tell me the doll had been a great success. He had introduced the doll by placing it near to his wife so she could decide if she wanted to engage with it. She responded to the doll immediately.  Mr Brown felt his wife believed the doll was a real child and that she was looking after her. He described how she cradled the doll and stroked her hair and face. He felt the doll had given his wife some responsibility - something that had not been possible for a while - and he has since introduced a second doll.  As well as being a mother, Mrs Brown had been a teacher for many years looking after young children. She has many experiences to draw on in making sense of the situation. Six months later, the dolls are an important part of her care.

Following this positive response by Mrs Brown, I suggested Mr Edgar offered his wife a doll. Mrs Edgar has increasing difficulty engaging in conversation due to expressive dysphasia. She also has very poor concentration so that she is no longer able to engage in many of the activities that she used to, such as sewing. When I visited Mrs Edgar soon after she was given the doll, I found her sitting with it in front of her. She was quite animated throughout our meeting, commenting on the doll and observing where it was looking. She was captivated by the doll, and much more talkative than at previous meetings. Her husband reports how much she loves the doll and that her mood lifts when she has it with her.

Both Mr Brown and Mr Edgars felt the introduction of the dolls had improved communication with their spouse as they could provide a focus for conversations. This was also reported by Moore (2001), Gibson (2005), James et al (2006) and Minshull (2009).


  • Comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Links may be included in your comments but HTML is not permitted.