Our first challenge was to frame the problem in behavioural terms. The review of the literature, best practice guidelines, findings from interviews and discussions with researcher-participants, found the problem to be multifaceted: psychotropic medications are being used as a first line response to managing challenging behaviours among care home residents with dementia; care home staff do not understand the risks of psychotropics; although many care home staff were aware of person-centred care, they did not conceptualise it as a non-pharmacological treatment for managing behaviours that challenge and so did not see it as an alternative to medication; and although aware in principle of person-centred care, it was not always used in everyday practice. We turned to the literature and best practice guidelines to begin to explore ways in which these issues have been understood and what recommendations for practice have been made.
A useful overview of work on psychosocial (i.e. non-pharmacological) interventions for dementia care for the treatment of challenging behaviours was the pan-European project, INTERDEM [26,27]. The INTERDEM project identified and appraised the quality of existing expert guidance across Europe.
In the INTERDEM review, the UK guideline produced jointly by the National Institute for Health and Care Excellence (NICE) and the Social Care Institute for Excellence (SCIE), Dementia: supporting people with dementia and their carers in health and social care, first published in 2006 and updated in 2017 , was judged highest in quality across all domains and so was ideal to inform our intervention. Other UK guidance was also identified, which stems from the same evidence base and offers the same key messages: first published in 2013 and updated in 2017, The Dementia Guide , and Common Core Principles for Supporting People with Dementia .
The NICE and SCIE guidance  recommended staff training in person-centred care and outcome-focused care for people living with dementia. While such guidance enables us to identify priorities, it does not express the problem behaviourally. To seek a behavioural focus we explored literature that has attempted to operationalise the concepts underpinning person-centred care. A good example available at the time came from Passalacqua & Harwood’s  extension of Dawn Brooker’s VIPS . Their training programme offered concrete examples of communicative strategies demonstrating Valuing the person, Individualised care, focusing on the Perspective of the person with dementia, and the Social environment and increased care staff’s empathy toward and hope for those living with dementia. Following their training, Passalacqua & Harwood showed less depersonalisation of residents living with dementia and more time was spent to engage residents in leisure activities that were important to them. Something that was supported by our stakeholders:
“An activity can be something as simple as painting somebody’s nails, giving somebody a hand massage, looking through a reminiscence book on a one-to-one. We are trying to get [care staff] to do more one-to-one interaction that is more meaningful to that individual and doing lots of life history work.” (community psychiatric nurse)
Defining the Target Behaviour
We were now getting closer to a behavioural focus, i.e. what behaviours care staff need to employ in order to deliver person-centred care. To further explicate this, we examined the literature to specify the target behaviour as precisely as possible.
A number of systematic reviews were identified which provided a synthesis of the evidence base [6,12,21,22]. O’Connor et al.’s  summary of the paradigms which have historically informed the development of interventions to improve the management of challenging behaviours was helpful in judging different kinds of approaches. The ABC model (antecedents, behaviours, consequences) seeks to reinforce appropriate behaviours and discourage less helpful ones, drawing on the Social Learning Theory [23,24]. The unmet needs hypothesis states that challenging behaviours stem from unmet human physical, emotional or social needs . The unmet needs hypothesis was supported by our stakeholders:
“We very much look at every person as an individual and every behaviour, there is a reason behind that behaviour. So it’s looking for that reason and trying to connect and understand that that behaviour has happened.” (care home manager)
The stress threshold model suggests that dementia reduces a person’s ability to tolerate stress, which results in inappropriate behaviour . This work helped us conceptualise part of the problem – the need to re-frame care staff’s understanding of the antecedents of challenging behaviour and in turn their perceptions of people with dementia.
The unmet needs hypothesis resonated across best practice guidelines and findings from interviews and discussions with researcher-participants. Alongside this, there was an emphasis in interviews and discussions on communicating with compassion as a founding principle of person-centred care. We therefore considered both responding to challenging behaviour as an unmet need and communicating with compassion as potential target behaviours for our intervention (see Tables 1-2).
This exercise enabled us to translate the concepts into actions, e.g. problem-solving, using distraction techniques, and it helped us realise the potential impact of changing this behaviour; if care staff were able to change every interaction with residents and behave in a person-centred way at all times, this would have a hugely positive impact on the residents’ wellbeing and could potentially provide a successful alternative to psychotropics as a way of managing challenging behaviour .
The behaviours identified to achieve compassionate communication were clear across the systematic reviews and guidance documents cited above, but were best illustrated by Passalacqua & Harwood’s  adaptation of Brooker’s VIPS . We contacted these authors to obtain the video material used in their training which clearly demonstrated the significance of eye contact, direct communication, and not ignoring the resident. This helped us delineate some of the likely content in our intervention.
Behavioural Analysis and ‘Diagnosis’
In order to develop an intervention that would have a chance of being effective at changing behaviour, our next step was to understand the precursors of behaviour in order to identify ways that we might be able to intervene to change behaviour. This took the form of a behavioural analysis . We did this by coding data from stakeholder interviews and discussions using the COM-B model. We then compared this coding to recommendations in the literature and best practice guidelines. The results of this process are reported in Tables 3-4.
Table 3 unpacks the target behaviour – responding to challenging behaviour as an unmet need. It was clear from the literature that resource investment was required for person-centred care to be implemented  and so in terms of physical Capability, this target behaviour required staff to have stamina for long, emotionally intense working hours and from consultations with stakeholders it was clear that changes to shift patterns were unlikely to change. Support from the organisation in the form of regular breaks, supervision, good working conditions was required. Generating the psychological Capability of staff meant providing them with information and skills to boost their knowledge and understanding of dementia and the use of psychotropics as well as person-centred care. Physical and social Opportunities required buy-in from managers to ensure materials were available for initiatives such as about me books, and to ensure that responding to challenging behaviour as an unmet need became a normalised way of dealing with it across care staff. Reflective Motivation required staff to become self-aware and for that to translate to automatic Motivation staff would need to want to believe in the notion that challenging behaviour is an expression of unmet need thereby taking ownership of this idea.
Table 4 outlines the behavioural analysis of communicating with compassion. In a literal sense, staff already have the physical Capability to communicate with staff, but adding compassion to that communication may require a change for some individuals. Some staff may already have psychological Capability in principle, especially if they have received training on person-centred care, caring with dignity, and so on, but they may still require a refresher. Others may not have received training and so will need to boost their knowledge and understanding of how good communication skills can improve the wellbeing of residents with dementia. To create the physical Opportunity the staff need space and time to reflect on their communication with residents. This is contingent upon the social Opportunity which would endorse this reflective behaviour if it became normal practice to share things that worked well in positive interactions, for example. Reflective Motivation requires staff to be self-aware and to separate the person from the condition in order to behave compassionately toward them. For that Motivation to become automatic, care staff need to be personally invested in the benefits of communicating compassionately with residents.
To summarise, the behavioural ‘diagnosis’ was as follows:
- To achieve our goal that care staff would respond to challenging behaviour as an expression of an unmet need we needed to facilitate the use of problem-solving techniques to help care staff determine the underlying needs of people with dementia that may result in challenging behaviour.
- To do this, we would need care staff to understand this principle.
- We would also need to demonstrate that using tailored distraction techniques, which can reduce stress levels among people with dementia, may help to prevent behaviours that challenge.
- For our intervention to achieve the goal of care staff communicating with compassion, we would need them to have good working conditions with opportunities for reflection and discussion with managers and colleagues.
- Furthermore, we would need them to understand the benefits of compassionate communication to residents.
Identifying Intervention Functions
All work to date, reviewing the literature and best practice guidelines, interviews with stakeholder researcher-participants, and going through the stages of the Behaviour Change Wheel pointed to the intervention having a Training function; it was clear that we would need to impart skills. The training would also need to be Educational because it was clear from our interviews with stakeholders that care staff were not knowledgeable about the side-effects of psychotropics, the recommendation that challenging behaviours were better treated with non-pharmacological (or psychosocial) treatments, or that person-centred care constitutes one such treatment. Due to the significance of generating the social Opportunity in achieving change to our target behaviours we deduced that our intervention would also need the functions of Modelling and Enablement. Our analyses pointed toward collective change, or rather individual behaviour change supported by role-modelling from peers and management (modelling) with the recruitment of VIPS champions; and structured support from management in the form of supervision, handover meetings, training etc. (enablement).
Selecting Behaviour Change Techniques
In order to perform the functions outlined in the previous section we selected a number of Behaviour Change Techniques (BCTs) to integrate into our intervention . As a training intervention, it was largely based on Shaping knowledge, especially Instructions on how to perform the behaviour of communicating with compassion. Shaping knowledge also involved providing Information about antecedents and Re-attribution, in relation to challenging behaviour as an expression of unmet need, the side-effects of psychotropics, and the recommendation to offer non-pharmacological interventions to manage challenging behaviour.
Key elements of the education provided involved Natural consequences – Information about health consequences to residents and care staff themselves from engaging in compassionate communication was included in the training as well as Information about social and environment consequences of better communication within care teams and among residents. Monitoring of emotional consequences was encouraged.
Social support was embodied in the intervention in a Practical sense, i.e. that it was group training and involved group activities. Staff were also encouraged to provide each other with Emotional social support both in the training and in practice.
Once we had viewed the video material demonstrating Brooker’s VIPS  developed by Passalacqua & Harwood , we sought permission from the authors to use it in our own training. This formed the BCT, Comparison of behaviour through the Demonstration of the behaviour.
Working through the Behaviour Change Wheel brought us to realise that a fundamental characteristic of the training was that it was interactive and involved care staff role-playing their behaviours, thus engaging in Repetition and substitution – behavioural practice and rehearsal formed part of the intervention.
We also wanted to inspire Self-belief through Focus on past success as care workers, and Self-talk in small groups about their interactions with residents and how they made them feel.
Assessing the APEASE Criteria
Throughout the selection of these BCTs we had to keep in mind the APEASE criteria : affordability, practicability, effectiveness and cost-effectiveness, acceptability, side-effects/safety, and equity. The affordability of the intervention was determined by the funding body (National Institute for Health Research, Research for Patient Benefit reference number: PB-PG-0613-31071). In projects funded this way, interventions require treatment costs to be provided by the appropriate National Health Service Trust, and in the case of primary care, funding decisions are made by regional Clinical Commissioning Group (CCG). This meant the training would be restricted to a one-off session lasting three hours.
The training was delivered on site in participating care homes and so worked well in a practical sense. We use the Maslach Burnout Inventory  and the Attitudes toward Dementia Questionnaire  to measure the effectiveness of the intervention. As a feasibility study, we were not able to fully assess the cost-effectiveness. Acceptability of the intervention was something we had prioritised from the start, which is a key reason we recruited stakeholders to help inform it. A central tenet of the intervention was to provide equitable care to people with dementia, but a likely side-effect was that all residents would receive an equitable level of person-centred care. We did not identify any risks to safety as the intervention was discussed in detail with care home managers and was found to meet safety policies.
Description of the Intervention
The major component of the intervention was the training session itself. This involved a three hour session with care staff which used powerpoint for information delivery and workbook for interactive exercises and role-play. Care staff were able to take a copy of the slides and workbook away with them. Stakeholder researcher-participants highlighted that there was a need for GPs to also receive this training. We therefore made it available to GPs in person or online.
We contemplated for some time what our training should be called. We wanted the title of the training to sum up the key message: Unmet needs are hidden inside; they are difficult to reveal to the outside world when communication skills are lacking; we need to access what is on the inside to understand what is going on outside. We therefore named our training, Inside Out. Indicative content of the slides is shown in Figure 1. The prominence of problem-solving skills in caring for people with dementia across literature and best practice guidelines also resonated with this name: to understand what is on the inside, care staff need to play the part of a detective to problem-solve the situation, to identify those hidden causes of challenging behaviour, and to use the residents’ personal histories to develop ways of reducing or eradicating those triggers. This label also worked when considering self-care: care staff need to be aware of how they are feeling on the inside in order to prevent those feelings from showing on the outside in their interactions with residents.
Details of the training intervention are presented in Table 5.
Other materials included behavioural cues reiterating the central messages of the training to be posted on noticeboards and sent via email to staff. Meetings were also held with care home managers to encourage them to introduce the VIPS as an agenda item in team meetings and to raise the use of VIPS in supervision sessions.