Feasibility of a two-part person-centred care initiative for people living with dementia in acute hospitals: A mixed methods study

Abstract

Background

The personal profile offers a potentially useful tool to support person-centred care of people living with dementia in hospital. To date, how profiles can be implemented into nurses’ work practices is not established.

Objectives

The aim is to establish the feasibility of a Person-Centred Care package, including a personal profile and staff education program to enhance implementation.

Design

Exploratory convergent mixed methods approach.

Settings

Four units of a tertiary health service, two intervention units and two comparison units set in southeast Queensland, Australia.

Participants

Person living with dementia and family carer dyads and staff.

Methods

Practicality was determined using participation logs, audit and review of meeting minutes. Acceptability was determined using interviews with family carers and nursing staff. Efficacy was evaluated using pre-post comparison survey design, assessing staff knowledge using the Dementia Knowledge Assessment Scale and person-centred care using the Person-centredness of Older People with cognitive impairment in Acute Care-revised scale.

Results

Practically, the personal profile was distributed to 95 and 73% of patients in the two intervention units. Of the 18 people living with dementia who consented to participate, only 6 (33%) had a This is Me form completed. The three-part education program was well attended (n = 190 participants). In terms of acceptability, carers’ (n = 5) experienced variable quality of engagement from nurses. In interviews, nurses (n = 18) experienced increased confidence to engage carers, in part attributed to local leadership, but attitudes towards care appeared to be influenced by perceived time constraints. For efficacy, completion of both surveys at all time points and in all units was 50% and higher. Dementia knowledge significantly improved in the intervention group (p < .01) however there was no difference in self-ratings of person-centred care.

Conclusions

The feasibility of a Person-Centred Care package, including a personal profile and a focused program of staff education was partially achieved, with the education component adopted into the organisation's continuing education program. Implementation research is required to enhance the element of coherence, how completing the personal profile is an investment in person-centred care rather than simply completing another form.

1. Introduction

Globally the number of people living with dementia has more than doubled from 1990 to 2016, from approximately 20 million to 44 million, respectively (GBD 2016 Collaborators, 2019). Dementia is now considered a global health priority, noting that many people living with dementia have increasing co-morbidities, and require hospitalization (Alzheimer's Disease International, 2016).

A recent meta-analysis reported the pooled relative risk of hospitalization for people living with dementia after adjusting for age, sex, and physical comorbidity was 1.42 (95% CI, 1.21–1.66), which was significantly higher than those without a dementia diagnosis (Shepard et al., 2019). Once hospitalized, people living with dementia experience more complications (Bail et al., 2013; ADI, 2016) and extended length of stay (Digby et al., 2017) than people without dementia. Hospital-acquired complications, such as delirium, have been attributed to the design of hospital services (Parke et al., 2017), specifically the intrinsic orientation towards fast and efficient services (Parke and Hunter, 2014), and lack of dementia knowledge and expertise amongst hospital staff (Digby et al., 2017).

Person-centred care is the preferred model of care in hospitals for attaining high-quality care (Australian Commission on Safety and Quality in Health Care, 2018), and, for people living with dementia, is a way of connecting and partnering with patients and families that is meaningful to the person (Edvardsson, 2015). Person-centred care is deemed best practice for the care of people living with dementia in multiple care settings (Clissett et al., 2013). Recently, Scerri and colleagues (2020) demonstrated that person-centred dementia care practice was enhanced by improved staff knowledge and attitude. Although knowledge of dementia is important, knowing the person living with dementia is essential to promoting person-centred care practices.

There is general agreement that families have a significant role to play in the care of a person living with dementia in hospital (Boltz et al., 2015; Stockwell-Smith et al., 2018). Arguably families’ knowledge and expertise about the person living with dementia could promote more person-centred connections and improve hospital care (Kelley et al., 2019). However, in a recent meta-synthesis, researchers found that families consistently feel excluded from hospital care and decisions (Burgstaller et al., 2018). Developing these person-centred connections between staff and family carers is challenging in the hospital setting.

Kelley and colleagues (2019) highlight that there are no consistent practices or guidelines for how to include families/carers in the acute care of people living with dementia. Research suggests that the use of a personal profile may facilitate this process (Bray et al., 2015; Luxford et al., 2015; O'Reilly, 2016). Personal profiles are simple communication tools that articulate a person's usual everyday needs and can be effective in helping healthcare services to deliver more person-centred care for people living with dementia (Brooker et al., 2014).

Three studies in the United Kingdom reported the use of a personal profile, This is Me. It was helpful for both families/carers and staff in developing, strengthening and promoting more person-centred collaborative partnerships to care (Baillie and Thompson, 2020; Bray et al., 2015; O'Reilly, 2016). In Australia, another personal profile, TOP 5, was successfully implemented in an Australian study (Luxford et al., 2015). Consistent with UK research, the findings of Australian research indicated that the Top 5 profile was simple and useful for clinicians and promoted personalized care for patients living with dementia and engagement with families/carers (Luxford et al., 2015). Although implementation of effective interventions to support the use of personal profiles is emerging, studies to date are limited in their reporting of implementation activities.

One of the more common implementation activities is staff education or training (Johnson and May, 2015). Earlier research on the use of This is Me did not provide details of any parallel staff education programs (Bray et al., 2015; O'Reilly, 2016). Staff education was explicitly included as a component of the TOP 5 research (Luxford et al., 2015), however there was no evaluation of education reported. A UK-based research team found that an education program provided in six wards of a general hospital, and which included a personal profile, Getting to Know Me, increased staff knowledge, attitude and confidence (Elvish et al., 2014). This program was successfully implemented in another three NHS services, with consistent results (Elvish et al., 2017). However, in these two studies the completion of the personal profile was not evaluated.

The aim of this study is to establish the feasibility of a Person-Centred Care (PCC) package, which includes the personal profile called This is Me plus a focused program of staff education and support, to improve person-centred care to people living with dementia in an acute care setting.

2. Method

2.1. Design

This study was guided by the research question: ‘What is the feasibility of an innovative two-part Person Centred Care package?” The Bowen Feasibility Framework (Bowen et al., 2009) was adopted for this study. The framework elements of practicality, acceptability, and efficacy of new interventions (Bowen et al., 2009) guided this study. An exploratory convergent mixed methods approach (Creswell and Plano Clark, 2017) was used to gather multiple sources of evidence for each of the feasibility foci. Combining quantitative and qualitative orientations is highly recommended when conducting feasibility studies (Orsmond and Cohn, 2015).

The Bowen Feasibility Framework supports a participatory approach to obtaining end user feedback throughout the course of the study, providing direction for further development and delivery of the intervention (Bowen et al., 2009). Implementation of the personal profile was guided by a Project Reference Group, consisting of end users including consumer, nurses, doctors, allied health staff, and researchers and included the planned education program.

2.2. Setting and participants

The study was conducted in four units of a tertiary hospital, which is part of a state‐funded health service in southeast Queensland, Australia. Units were selected based on a population of at least 60% of patients aged 65 years and older. Two units were assigned to the intervention, a Medical Assessment Unit (MAU) and an Acute Care of the Elderly (ACE) unit. Two comparison units that were analogous to the intervention units were also selected.

There were two groups of participants: (1) person living with dementia and family carer (dyad) and (2) staff in the two units. The person living with dementia participants were recruited based on being aged 65 years or older with a documented diagnosis of dementia and a primary family carer willing to participate in partnership with them. There was no requirement for the carer(s) to be co-residents (i.e., residing with the person). Nurses were purposively recruited based on being employed by the health service on the participating units.

2.3. Intervention

The person-centered intervention package included two parts. The first part was a personal profile, known as This is Me (Alzheimer's Society, 2017), which was adapted for the health service. The revised version was condensed to two pages and reordered to provide more focused information, which included preferred name, details of the person who knows Me best, important information about Me, what worries Me, what makes Me feel better, My pain responses, and My routines. This is Me was distributed for the nurse to complete in consultation with the patient and family/carer.

The second part was an education program. Considering the recommendations from a systematic review of dementia training (Surr and Gates, 2017), the program was designed to be relevant to nurses’ role and included:

• (1)Work time to complete an on-line course on dementia, called ‘The View from Here: Skills in dementia care for acute care settings’ (Dementia Training Australia, n.d.), which consists of nine modules, taking approximately 6 h to complete.
• (2)Group learning in 30 to 40 min case discussions with experts (e.g., occupational therapist, social worker, dementia nurse consultant) held during usual handover time on a fortnightly basis; and
• (3)A high-fidelity simulation, with patient and carer actors, to practice using the personal profile in an unfolding case scenario.

2.4. Data collection and analysis

Multiple data sources were used in this study to address the three feasibility foci of practicality, acceptability and efficacy. An outline of mixed methods data collection is provided in Table 1.

Table 1.

Mixed methods data collection outline.

Feasibility	Pre-intervention	Intervention	Post-intervention	Quantitative	Qualitative
Practicality		Screening log		✓
		Patient participant log		✓
		Staff education attendance log		✓
		Personal profile audit		✓
		Meeting summaries			✓
Acceptability			Carer interviews		✓
			Staff interviews		✓
Efficacy	DKAS		DKAS	✓
	POPAC-r		POPAC-r	✓

2.5. Practicality

Practicality focused on recruitment capability, intervention uptake, success or failure of execution, and resources required to implement the intervention (Bowen et al., 2009). For recruitment capability, a screening log was maintained, with the number of patients admitted to the ward, and those that met the inclusion criteria, recorded. The frequency of eligible patients identified was then calculated.

For the This is Me personal profile uptake, a participant log was maintained recording the number of patients and carers enrolled into the study. The number of patient carer dyads who participated was calculated as a percentage of those who were eligible to participate. For education intervention uptake, a participation log of staff participating was maintained for each of the three elements. The participant numbers were recorded.

Execution was determined by auditing participating patients’ This is Me personal profile with completion recorded as fully, partially, or not completed. The number for each was calculated as a percentage of all audited profiles.

Resources for implementation were identified in the Project Reference Group meeting summaries, including meeting frequency, attendance, record of implementation issues and solutions, and were treated as data. The meeting minutes were subject to a qualitative descriptive content analysis by two research team members (LG & MK). A descriptive content analysis was used to describe the processes and negotiations required to finalize the implementation plan. Qualitative descriptive content analysis is the preferred method for providing straight descriptions of phenomena (Sandelowski, 2000).

2.6. Acceptability

Acceptability is focused on the extent that the intervention is judged by participants to be suitable, satisfying or appropriate (Bowen et al., 2009). Patient and carer experiences of using the This is Me personal profile was solicited through telephone interviews conducted approximately one-week post-discharge. The interview focused on the extent to which This is Me was judged to be suitable, or appropriate and included questions about carers’ perceptions and opportunities to contribute to care in hospital.

Nurses’ perceptions of using the This is Me personal profile and their participation in the education sessions were recorded in individual interviews held following six months of implementation. The interviews were conducted in person (during work time). Data collection continued until data saturation was achieved.

Patient/carer and nurse interviews were digitally audio-recorded and transcribed by a professional service. The interview transcripts were subject to independent thematic content analysis (Braun a Clarke, 2006) by two research team members for the carers (GSS & AT) and three research team members for the nurses (GSS, TS & DS), respectively.

2.7. Efficacy

Many feasibility studies are designed to test an intervention in a limited way (Bowen et al., 2009). In this study, two self-administered questionnaires, the Dementia Knowledge Assessment Scale (DKAS, Annear et al., 2017) and the Person-centredness of Older People with cognitive impairment in Acute Care – revised (POPAC-r, Grealish et al., 2017) were used to determine efficacy of implementing the This is Me personal profile and the education program, respectively.

The 25-item DKAS was used to assess nurses’ knowledge about dementia. A total DKAS score, ranging from 0 to 50, is calculated by adding the scores for each statement. DKAS addresses knowledge across four domains: Causes and Characteristics (7 items); Communication and Behaviour (6-items); Care Considerations (6-items); and Risks and Health Promotion (6-items). The DKAS is a validated instrument with an overall Cronbach alpha of .85 (Annear et al., 2017). The reported Cronbach alpha for the subscales range from .65 to .76 (Annear et al., 2017).

Nurses’ perceptions regarding person-centred practice were measured using the 14-item POPAC-r scale derived from the original scale (Edvardsson et al., 2013). There are three sub-scales within the instrument. Individualising Care (5 items); Using Cognitive Assessments and Care Interventions (5 items); and Using Evidence and Expertise in Assessment (4-items). The POPAC-r scale is a validated measure (Cronbach's alpha Total = 0.84; subscales from .63 to .82; Grealish et al. 2017).

Quantitative data analysis was conducted using IBM SPSS version 26. Descriptive statistical methods were conducted for the (1) demographic data for interview participants, (2) DKAS and (3) POPAC-r. Inferential statistical methods were used to compare the pre- and post-data for both the POPAC-r and DKAS for the intervention and comparison units including, t-tests, and mixed factorial ANOVA on the matched data. The independent variable for each analysis was group (intervention, comparison) and the dependent variables were the total and subscale for the DKAS (dementia knowledge) and POPAC-r (person-centred care). Further analyses were conducted comparing the intervention and comparison units on each question of the DKAS pre (T1) and post intervention (T2). Mann-Whitney U analyses were performed as the data was ordinal (0 incorrect, 1 probably correct, 2 correct). Bonferroni corrections was applied as to the alpha level when necessary.

2.8. Integration of data produced from mixed methods

The research team met for a three-hour workshop to integrate the data and determine practicality, acceptability, and efficacy of the two-part person-centred care package intervention. The quantitative data for recruitment capability, intervention uptake, success or failure of execution and descriptive qualitative data for resources were combined to make a descriptive account about practicality. The themes from the interviews with carers and those with nurses were compared and contrasted to draw conclusions about acceptability. Conclusions about efficacy were drawn from the changes in group means for the comparison and intervention wards over the two time points for measures of knowledge and person-centred care.

2.9. Ethical implications

The study was approved by the hospital (QGC 52652) and university (GU HREC 2019/511) Human Research Ethics Committees. The consent mechanism in this study acknowledges the specific requirements for inclusion of people with cognitive impairment in research (National Health and Medical Research Council, 2015, Australian Research Council, & Australian Vice-Chancellors’ Committee). Researchers were aware of the vulnerability of patients and carers while in hospital and conducted interviews post discharge.

3. Results

The study was conducted over six months from September 2019 to February 2020. Eighteen dyads (n = 7 MAU; n = 11 ACE) participated in the study. Data is presented to address the three feasibility foci: practicality, acceptability, and efficacy.

3.1. Practicality

Screening and Audit. Approximately 73% of patients admitted to MAU and 99% admitted to ACE were aged 65 years and older (see Table 2). Of those patients aged 65 years and older, 18% and 34% on MAU and ACE respectively, had a diagnosis of dementia in their notes. The This is Me personal profile was distributed to 95% of patients living with dementia in the MAU, and 73% on ACE. Three patients received the profile on the MAU, and it accompanied them to ACE. Of the twelve This is Me profiles audited, eight had no questions completed (n = 1, MAU; n = 7, ACE) and four had all the questions completed (n = 4, ACE).

Table 2.

Frequency of screening: the number of eligible patients living with dementia identified in each of the intervention units.

Dates Nov-19 to March-20	Screened n	65+ years n	Dementia diagnosis n	This is Me provided n (%)	This is Me provided on MAU n
MAUa	656	476	85	81(95)
ACEb	218	217	74	54 (73)	3
TOTAL	874	693	159	135 (85)

a. Medical Assessment Unit b. Acute Care of the Elderly (ACE).

Education Program. Nurses participated in all three parts of the program. The on-line program had 87 enrolments, with 46 completions at the time of study close. Attendance at the unit-based sessions were 51 nurses across five sessions (MAU) and 31 nurses across four sessions (ACE). Six simulations were attended by 30 (MAU) and 32 (ACE) nurses.

Project Reference Group Meeting Minutes. There were five Reference Group meetings. Representatives from each ward, research team members, a medical practitioner and an allied health practitioner attended each meeting. A consumer and Assistant Director of Nursing were recruited to the Project Reference Group during the project. Their participation enhanced discussion of implementation and contributed directly to key decisions, such as the development of a video infographic for patients and families (consumer) and supporting staff to attend educational activities (Assistant Director).

Using the General Theory of Implementation (May, 2013), an implementation plan was developed during the initial two meetings. The implementation plan was fluid, informed by nurses’ input about possible barriers and enablers, with additional actions added over time. There were three elements to the plan: capacity, capability and potential (May, 2013).

Capacity addressed the social structures available to support implementation and included safety scrums, establishing a mechanism for nurses to complete and gain credit for a pre-existing national on-line course, and co-designing the simulation with the health service's experienced team. Capability was focused on intervention workability in the social system and included: nurses’ familiarity with other personal biography forms, such as the Sunflower (Agency for Clinical Innovation), printing the profile on coloured paper (blue), and a forget-me-not label on the patient board to identify that a profile was in their bedside file. Potential was about securing nurses’ commitment to making the intervention work. Strategies to enhance potential included delivery of the education program and introducing a slogan competition.

3.2. Acceptability

Carer Interviews. Five carers of patients living with dementia participated in a post-discharge telephone interview. The carers’ age ranged from 56 to over 80 years, and the patients were aged between 67 to 93 years. There was one male and four female carers (3 daughters; 2 partners) and two male and three female patients. Two themes were identified from these interviews.

Quality of care. Carers’ perceptions of the quality of care provided were mixed. Carers shared positive experiences, including the patient being treated with compassion and respect and members of the healthcare team taking time to explain treatment and future care options. For example:

‘[The doctor] was very good with explaining what was going on, long term care, options and decisions, all that sort of stuff … I didn't feel hurried that he had to get out’ (Carer 100).

However, most carers considered that nurses did not fully understand the complexities of dementia care and this influenced the quality of care received by the person with dementia:

‘Nurses need to understand the different phases, the different levels of dementia. They need to understand how to communicate with them. That would be the main problem that I had. They had no conception of how mum was different from the person in the next bed’ (Daughter 106).

Inclusion and provision of information. Most carers considered communication to be a key area for improvement and would have liked more opportunities to be involved in decision making during admission. They experienced difficulty in obtaining information concerning their relative and noted a lack of inclusion in care related discussions, or decisions with the healthcare team:

‘I would have liked a more direct conversation with the doctor in charge because the nurses weren't either fully across the prognosis or unwilling to divulge information or speculate. We felt that we would have liked more daily communication from the senior medical team’ (Daughter 002).

They also emphasized the need for nurses to listen and to learn from, the carer during admission regarding the person's dementia care needs and behavior management strategies:

‘… it's really around being able to work with him [person living with dementia] on a short-term basis and being able to understand his needs and communicate with him. That can only be done by actually understanding him, which means listening to the information that's provided by the family or knowing what questions to ask the family’ (Daughter 101).

Despite the intervention there was a common sense of exclusion reported by the carers and that improvements, beyond knowledge of dementia, are needed to promote person-centred care practices.

Nurse Interviews. Seventeen interviews were conducted with 18 nurses from the two intervention wards: seven from MAU and 11 from ACE. Their mean age was 41.6 years (SD = 8.6) and 17 identified as female. One interview was conducted in a pair at the participants’ request. The two key themes included increased confidence to engage with carers and attitudes towards care influenced by perceived time constraints.

Increased confidence to engage with carers. Nurses expressed increased confidence in engaging with carers to invite participation in care and in supporting/managing the care needs of people living with dementia. They felt that relationships between nurses and carers were more positive and that both parties benefited from the rapport that developed. For example:

‘I think we're trying to communicate with the carers a bit more. Asking carers have you got some little tips’ (P04).

‘Just involving the families … I think with the forms and stuff it was a lot easier getting them to write it all down because before it was probably more verbally spoken’ (P12).

The nurses appeared to view this engagement as augmenting their approach to caring for people living with dementia.

It was apparent that local leadership and the development of project specific and local resources had improved nurses’ confidence in delivering the intervention and in mentoring new and junior nurses to support person-centred care.

‘[Unit based dementia lead] been very instrumental in being our dementia champion’ and ‘… we've got the little activity boxes and diversional therapy, all of the general staff are aware that that's what [Unit based dementia lead] doing and we've got these activities’ (P04).

Nurses valued the education and simulation opportunities provided by the project and considered it had increased their awareness and confidence in providing person centred care.

‘I definitely noticed that people who did the course had-were learning more-getting more knowledge and being like, oh I never knew that’ (P11) and ‘… the simulations were helpful’ (P04).

Nurses identified the importance of supportive leaders to release them to attend education and provide resources to better provide person-centred care practices.

Attitudes towards care is influenced by perceived time constraints. Many of the nurses interviewed expressed positive attitudes and behaviours towards people living with dementia. For example:

‘I think everyone has learned because I've seen changes and-like have a bit more of an understanding…we don't just put them in a bed now…and just leave them’ (P11).

‘I feel that some [junior] staff are actually getting a little bit more proactive and looking at their [people living with dementia] interests and actually sitting and talking to patients and letting them walk (P10).

They also reported that their attitudes towards the use of sedating medication as a first resort in managing agitation or aggression had shifted, noting that they were now more likely to explore the reasons for the behavior and use diversional activities to reduce distress or agitation. For example:

‘I feel like people were quite quick to give medications … but now [since the project] I feel like it's much more of a what is causing these behaviors, what unmet needs are behind it and more problem solving and spending those few extra minutes’ (P09).

There was also some, albeit limited, use of stigmatized language and terminology by nurses:

‘…would actually make life a lot easier, and a lot less frustrating, because sometimes these people can get quite aggressive as well’ (P03) and ‘…. there's ones that are quite combative’ (P15) and ‘… working with geris [sic] you don't really get a lot of positive feedback’ (P15).

There was also a custodial attitude with focus on safety and the risks that people living with dementia pose in an acute care setting, for example, falls, wandering, aggression and the impact this has on other patients due to the time-consuming nature of their needs. This is illustrated by the following quote:

‘They'll either fall … and the other patients in the ward become neglected … all your time is taken up caring for this patient, but you've got three other patients care for, they get neglected because the dementia patient [sic] … is taking up your time because of safety reasons’ (P03).

Nurses remained concerned about the time to complete the This is Me personal profile, and perceived it as ‘one more form’ despite acknowledging that the information could add value to their care delivery:

‘I think trying to sell the form is hard. It's perceived as another form, no one is going to read it, why bother, it gets left in the chart and forgotten about’ (P06).

‘… it was very helpful. But sometimes we didn't get time to fill it out’ (P04).

In summary the nurses appeared to have a positive attitude towards caring for people living with dementia, however the use of stigmatizing language was evident in some interviews. Nurses perceived that their capacity to provide care for people living with dementia had improved and reported improved knowledge and understanding in relation to dementia care and related person-centred care practices. However, their challenge appeared to be how to provide this care within the perceived constraints of the workplace.

3.3. Efficacy

The survey findings indicate that the education program was able to produce improvements in knowledge but not significant change self-perceptions of person-centred care delivery. Participants’ demographic details are outlined in Table 3.

Table 3.

Demographic details for the intervention and comparison pre- and post- for DKAS and POPAC-r.

	Dementia Knowledge Assessment Scale (DKAS)							Person-centredness of Older People with cognitive impairment in Acute Care–revised (POPAC-r)
	Pre-Intervention			Post-Intervention				Pre-Intervention			Post-Intervention
		Intervention	Comparison		Intervention	Comparison	Intervention		Comparison	Intervention		Comparison
Total (N)		53	49		43	33	52		38	46		35
Age (years) Mean (SD) Range		37.9 (11.8) 22-66	35.9 (9.8) 21-59		38.3 (12.1) 22-67	33.7 (8.4) 22-56	38.8 (11.9) 22-66		35.8(8.9) 21-56	38.6(11.8) 22-67		33.7(8.5) 22-56
Gender (n) Female Male		45 8	41 8		38 5	31 2	44 8		34 6	41 5		33 2
Role (n) Assistant in Nursing Enrolled Nurse Clinical Nurse# Registered Nurse Clinical Facilitator Clinical Nurse Consultant/ Nursing Unit Manager		3 4 9 35 - 2	2 1 6 34 2 4		1 6 11 23 - 2	1 4 5 20 - 2	3 5 8 34 - 2		2 1 6 25 1 4	1 6 11 26 - 2		1 4 6 21 1 2
Experience (years) Mean (SD) Range		6.6 (5.7) 0-34	6.2 (7.4) 0-37		6.3 (4.9) 0-21	5.4 (5.3) 0-21	6.7(5.81) 0.7-34		5.6(5.9) 0.5-30	7.3(7.36) 1-44		5.9(5.9) 0.2-22
Dementia Training (n) Yes No		17 36	12 37		14 27	14 27	32 17		28 2	32 14		28 7

^#A Clinical Nurse is a Registered Nurse with clinical leadership responsibilities

Dementia Knowledge. The completion rate for the surveys was 56% and higher, with more returned surveys in the intervention wards (See Table 4). Pre-intervention there was no significant difference in knowledge between the intervention and comparison groups, on the total DKAS or the four subscales. Post-intervention, total DKAS accuracy was statistically higher in the intervention group, with improvements in the areas of causes and characteristics of dementia, communication and behavior, and risk factors and health promotion.

Table 4.

Mean and Standard Error (SE) for DKAS and POPAC-r Sub-scales and Total Scores for the Intervention and Comparison Group Pre- and Post-Intervention

Dementia Knowledge Assessment Scale (DKAS)
			Pre-Intervention		Post-Intervention
Sub-scales		Intervention (n=53)		Comparison (n=49)		Intervention (n=43)	Comparison (n=33)
Causes and Characteristics^		69.4 (2.7)		63.8 (3.0)		71.4(3.1)	62.1 (3.0)*
Communication & Behaviour^		77.9 (2.7)		76.4 (3.2)		82.9(2.7)	71.2 (4.7)*
Care Considerations		83.9 (2.4)		76.4 (2.9)*		82.9(2.8)	76.7 (2.8)
Risk Factors & Health Promotion^		55.0 (2.7)		56.5 (2.9)		62.0(3.5)	50.8 (3.5)*
Total DKAS		71.5 (1.9)		68.1 (2.0)		74.7(2.2)	65.1 (2.5)**
Person-centredness of Older People with cognitive impairment in Acute Care–revised (POPAC-r)
Sub-scales		Pre-Intervention				Post-Intervention
		Intervention (n=52)		Comparison (n=38)		Intervention (n=46)	Comparison (n=35)
Individualizing Care		3.82 (.11)		4.13 (.11)		4.31 (.09)	4.22 (.13)
Cognitive Assessments & Care Interventions		4.48 (.11)		4.74 (.10)		4.67 (.12)	4.74 (.12)
Evidence & Experts in Cognition		4.35 (.14)		4.38 (.16)		4.91 (.11)	4.54 (.15)
Total POPAC-r		4.21 (.11)		4.38 (.09)		4.59 (.10)	4.49 (.12)

^=percentage correct; *=p < .05; ** = p <.01

Person-centred Care. Fifty-two percent and higher POPAC-r surveys were completed with more returned on the intervention units. Nurses overall rated their person-centred care of older people with cognitive impairment quite highly at both time points. The analyses demonstrated no statistically significant difference in POPAC-r total, or the three subscales between the two groups, pre-or post-intervention.

4. Discussion

A two-part intervention package, inclusive of the This is Me profile and a comprehensive nurse education program in the care of people with dementia in hospital, produced improvements in knowledge and may improve individualized care. However, feasibility of the package, in particular the implementation of the This is Me personal profile, was more challenging. Contribution, the fourth phase of the general theory of implementation (May, 2013) is focused on enacting the new practice and provides the framework for discussing the findings of this study. Contribution consists of four concepts: coherence, cognitive participation, collective action, and reflexive monitoring (May and Finch, 2009).

Coherence, what actors (in this case nurses) perceive as meaningful in the practice was addressed through the educational program, which offered multiple approaches to learning. Dynamic, ongoing training is well recognized to enhance implementation (Powell et al., 2015). While knowledge about dementia improved, nurses had not yet conceptualized the use of the personal profile as a valued instrument to enhance person-centred care. They perceived it as ‘another form’ that required completion, rather than a tool that would be useful in their work. For example, the request to make the form shorter resonates with other studies, where the time to complete the personal profile is reported as a barrier to implementation (Luxford et al., 2015). The pursuit of a shorter form to reduce the work of completion reveals an underlying focus on efficiency, focusing on doing things in a faster and easier way in order to make time available for other work (Parke and Hunter, 2014; Jones et al., 2016).

To address this desire for efficiency, future implementation of the personal profile could focus on how knowing the person might reduce unforeseen and sometimes highly disruptive responsive behaviors (Jones et al., 2006). By couching the personal profile and knowing the person living with dementia as an approach to prevent disruptive behaviors, the intervention may appeal to the dominant risk management discourse and therefore align with nurses’ understanding of current practice challenges. This approach is worthy of further investigation

The concept of cognitive participation is focused on how the nurses were enrolled and engaged in the new practice (May and Finch, 2009), in this case, using the personal profile. Enrolment was primarily achieved through the Project Reference Group, where members were influencers in the two intervention wards, the ward-based group discussions, and the simulation. The use of influencers, also known as ‘champions’ or ‘local expert’ is associated with implementation of new work practices (Surr and Gates, 2017; Johnson and May, 2015). The slogan competition, a form of mass media (Powell et al., 2015), reached staff across both intervention wards. While not evaluated in this study, this type of ‘fun’ activity has the potential for staff engagement in new practices.

Collective action in the new practice is dependent upon interaction with already existing practices (May and Finch, 2009). The education program aligned with existing systems and was relatively easy to implement. While there was significant investment in organizing the education program, once established it worked well, with an acceptable number of nurses from both wards completing at least one element of the program.

Completion of the personal profile was more challenging. It was a paper-based form expected to be completed in a work environment that was dominated by electronic medical records. This may have contributed to low completion rate. Completion required engagement with a family carer, which was challenging in the Medical Assessment Unit, which had a short (24 to 36 h length of stay). Whereas, engagement with family carers is usual practice in the Acute Care of the Elderly (ACE) Unit, and this may have contributed to the higher number of completed personal profiles in the ACE Unit.

Reflexive monitoring refers to how the practice is assessed and understood (May and Finch, 2009). During the interviews, participants noted that nurses’ confidence to engage with family carers increased during the period of the study. As noted earlier, nurses perceived the time spent on completing the personal profile as a loss of time, rather than an investment of time. While nurses’ language during the interviews suggested that older patients living with dementia may be stigmatized, they self-rated their practice as highly person-centred, suggesting that further investigation into the alignment between nurses’ self-perceptions and language is required. The carers’ experience of mixed quality of care and limited access to information resonates with previous research (Burgstaller et al., 2018). Nurses’ limited self-awareness of the impact of their care on older patients living with dementia and their families requires urgent investigation.

The process of completing the personal profile with a family member was an opportunity for nurses to develop a relationship with the carer. Previous research suggests that nurses do not routinely include families in their conception of person-centred care (Clisset et al., 2013; Goldberg et al., 2014; Jamieson et al., 2016). In this study, how nurses understood families and how families can enhance person-centred care in hospital for people living with dementia was unclear and may have affected implementation. Further research into how nurses understand the contribution of families to person-centred care for people living with dementia is required in order to enhance coherence during implementation.

5. Limitations

There were limitations to this study. While the educational intervention appears to have positively impacted on knowledge, this result should be interpreted cautiously. We had planned to compare results from the same nurse participants at two timepoints and created a self-identification system to do this. However, the turnover of nurses in the wards was significant due to staff movements and delays in data collection associated with preparing for the COVID-19 pandemic. As such an adequate sample of matched pairs could not be achieved.

The ethical requirement for consent from the patient to participate in the study before auditing the personal profiles limited the number of profiles audited. The numbers of completed profiles may have been much higher than reported here. An additional complication was the short-stay nature of the Medical Assessment Unit, with some patients discharged before the audit could be conducted.

There were only five carers who were interviewed. The study, and the recruitment of patient-carer dyads was halted due to hospital closure related to COVID-19. Initially 14 families/carers consented to be interviewed. Of these, four indicated that they no longer wanted to be interviewed, and five did not respond after three contacts. While the themes of these interviews were consistent with previously published findings on carers’ experiences (Burgstaller et al., 2018), more information may have been revealed with more carer participants.

As the number of people living with dementia continues to rise, health services will increasingly be expected to incorporate caring practices known to enhance person-centred care for this population. To sustain a practice, organizational leadership and systems modifications are required (Aarons et al., 2016). Subsequent to the study, the educational program was found to improve nurses’ knowledge, and as such executive support for the program was secured. Sustainability of the education intervention is dependent upon: incorporating the free on-line course into the hospital's internal on-line education platform; available experts to lead discussions on dementia into the ward-based education program; and educator capacity to insert simulated characters living with dementia into new simulations to illustrate the implications of care practices on this cohort.

Finally, the use of a mixed methods approach permitted a rich description of the feasibility of implementation of the personal profile. Implementation was enhanced by the education program and the Project Reference Group; confounded by the transition of hospital administrative services from paper to electronic format; and illuminated the insight from staff about their practices. We found an underlying drive for efficiency that contributed to seeing the personal profile as a chore rather than an investment to support person-centred care. This drive for efficiency has been reported elsewhere (Parke and Chappell, 2010; Dewing and Dijk, 2016) and must be addressed in future plans to implement new nursing practices generally, and in the care of people living with dementia specifically.

6. Conclusions

Including families in the care of people living with dementia in hospital is emerging as an important element in the provision of person-centered care. The personal profile offers a tool for nurses to develop a relationship with families while learning more about the person living with dementia. We implemented the personal profile in combination with a three-part education program. The education program improved nurses’ knowledge about dementia but how this translated into practice change is less clear. Auditing actual practice via observations may provide a better understanding of person-centred care practices undertaken in acute care settings. Importantly, we found that while education is critical to improving nurses’ understanding about the care of people living with dementia, nurses shared focus on work efficiency, and associated resistance to changes that may enhance care, must be addressed. Future research into the care of older people living with dementia in hospital should focus on how nurses’ conceptualisation of efficient practice as ‘good’ practice develops and what practice looks like.

Contribution of the paper

"What is already known about the topic?"

• •The personal profile is emerging as a useful tool to support person-centred care of people living with dementia in hospital.
• •Completion of the personal profile is enhanced when family carers are invited to participate.
• •Effective implementation of personal profiles into nursing practice requires further investigation.

"What this paper adds"

• •A shared nursing value for work efficiency appeared to limit widespread completion of the personal profile.
• •Nurses report that education and using the personal profile increases their confidence to engage with families but carers’ experiences did not resonate with this.
• •While education improved nurses’ knowledge of dementia, this did not translate to improvements in self-perceptions of person-centred care or person centred practice.

Declaration of Competing Interest

The authors claim no conflicts of interest