Introduction

Fifty million people world-wide presently live with dementia [1] and approximately 65.7 million people are projected to be associated with the disorder by 2030 [2]. Encompassing a range of neurocognitive disorders, dementia is distinguished by progressive decline in cognition and impairment in function. Providing care for people with living with dementia is often exhausting and stressful for families and often at a late stage of the disorder, the person will require formal care support, such as long-term residential care [3]. There is growing awareness among healthcare service providers that person-centred care (PCC) is the preferred model of dementia care [1,3,4]. PCC focuses on supporting the person’s remaining abilities, rather than on the losses occurring [4,5], and it recognises the importance of knowing the individual’s history, personality and preferences, bringing the person into shared decisions on their care [6,7], and customizing care and lifestyle support accordingly [7].

The social-psychological theory of personhood in dementia [7] provides the basis for the PCC approach; it proposes that people exist in a social, relational context, and that positive and enriching interpersonal relationships can prevent the disabling effects of dementia and promote a sense of well-being [7,8,9]. Since the person’s life experiences are constructed by the social-psychological milieu, this can often have a more significant effect on the person than the illness itself, by influencing their relationships with caregivers [10]. Kitwood developed a set of guiding PCC principles [7] to help caregivers support the person’s well-being, which include: creating and strengthening a positive relationship with the person through warm and accepting human contact; communicating respectfully, valuing and honoring the person; treating the person as a sentient and unique human being by valuing their innate nature; assisting the person to retain their remaining strengths; viewing the person’s world from their own perspective; and enabling the person to feel socially confident and maintain emotional attachments [7,8,9].

Multiple sources, including international policy [1, 3], dementia advocacy groups [10] and national Dementia Strategies, for example in the UK [11] and Australia [12] advocate PCC for people living with dementia. This recommendation is based on the evidence that PCC can reduce the incidence of clinical issues, such as agitation [13] and delirium [14], and it can help with deprescribing of psychotherapeutic medicines [15]. Despite evidence of the effectiveness of PCC, a pervasive challenge in applying the PCC across healthcare services is the construction of dementia as a master status through use of labelling and social positioning, in which the person’s dependency and a lack of autonomy are expected by caregivers. When this attitude prevails in healthcare services, care becomes focused on compensating for the loss of functional and cognitive abilities rather than supporting remaining abilities [16]. This attitude extends to normalizing the presence of agitation and other neuropsychiatric symptoms in dementia, attributing dementia as the cause rather than the care context [11]. These prevailing caregiver attitudes serve to distance the caregiver from the person living with dementia, resulting in their diminished personhood [7, 11]. These non-PCC practices largely occur because caregivers are unclear on how to integrate PCC principles within existing healthcare services that are constrained by established ways of delivering health care (micro and macro levels) and because of task-focused workplace cultures [17, 18].

Redressing these limitations demands reconceptualizing the organization’s approach to dementia care, instituting enabling and proactive organizational support, attending to managerial and senior staff leadership, and providing targeted staff education, training, direct supervision and oversight of PCC [19,20,21]. Organizational-level provision of PCC requires a top down, bottom up approach; leadership is required from the top and staff caregivers must be equipped with PCC knowledge, skills and attitudes. While individual caregivers need to be skilled in the delivery of person-centred care, the entire organisation must also be supportive. Organisational leaders need to demonstrate commitment and support by articulating a clear vision of PCC, allocating sufficient resources that enable PCC implementation, and establishing and exhibiting values of organization-wide respect, empowerment and choice for the person, their family/advocates and their caregivers [4,5,6,20,21,22]. Leaders need to demonstrate attitudes and behaviours that support PCC by placing relationships before tasks in the care planning and delivery and enabling caregivers to balance the values and wishes of the person (and their family/advocates) with organizational values, in order to provide personalized care delivery in daily practice [4,5,6,17,18, 20–23]. In addition, a shared governance system needs to be established such that caregivers become part of the decision-making process by seeking their input and feedback on decisions regarding changes to policies and procedures, redesigning the care environment, and determining the effect of changes on daily workflow [5,6,21,22,23]. Finally, the physical environment needs to be adapted in order to support the person’s right to privacy, maximise the person’s independence, enable the person to make the best use of their capabilities, provide the person with opportunities to participate in community life and maintain emotional connections, and empower the person to feel psychologically secure, and physically comfortable and safe [4,5,6,23].

In establishing this level of organizational support of PCC, a coordinated approach is needed to communicate the common values of PCC across the healthcare organization. This requires a shift in the organization’s strategic direction, whereby the work culture, leadership and support to individuals within the care relationship must align with person centered values [7,8,9, 21,23]. The organization’s executive and managers must also create a climate of understanding and acceptance among staff that the change process will take some time and will require a great deal of commitment and effort by all members of the organization [20]. Healthcare providers, managers and direct caregivers will need to be part of the change process by embracing a reconceptualized future for the organization, the care recipients and themselves, and by investing in staff who are future oriented [21,23]. To achieve this, organizational systems must facilitate clear communication on PCC requirements, provide capacity for speedy problem solving among all team members [19, 21, 23], encourage new ideas and incorporate an effective way of discovering what is/not working for the person with dementia and respond accordingly [21,22].

A coordinated and sustained cultural and structural transformation supporting PCC requires the healthcare service to focus on respectful and positive relationships between caregivers, people living with dementia and their families, improved capacity of caregivers to provide PCC through the development of knowledge and skills, supporting the dignity and autonomy of the person living with dementia, and promoting collaboration and team work in delivering care that aims for well-being [18, 22,23]. These requirements are encapsulated in the VIPS framework [24] which provides guidance on organizational-level implementation of PCC, paying attention to the following four key elements:

1. Valuing: valuing service user and service staff.
   The organization’s mission statement identifies provision of a person-centred service; Human resources management ensure staff feel valued by their employer; Management practices are empowering to direct service staff; Management supports training and development for staff to be skilled in person-centred care; Management provides supportive and inclusive physical and social environments for people with cognitive disability; and Continuous quality improvement mechanisms are in place that are driven by knowing and acting upon the needs and concerns of service users.
2. Individualized Care: Treating people as individuals.
   Strengths and vulnerabilities of service users are recognised across a wide range of needs; Care plans are individualized that reflect a wide range of strengths and needs; Individual care plans are reviewed on a regular basis; Service users have their own personal clothing possessions for everyday use; Individual likes and dislikes, preferences and daily routines are known about by direct care staff and are acted upon; Care staff are aware of basic individual life histories and key stories of proud times, and are used regularly; and A variety of activities are available to meet needs and abilities of all service users.
3. Personal perspective: looking at world from perspective of person with dementia.
   Service users are asked for their preferences, consent and opinion on a day-to-day basis; Staff show the ability to put themselves in the position of the person they are caring for and think about decisions from their point of view; The physical environment (e.g. noise, temperature) is managed on a day-to-day basis to help people with dementia feel at ease; Physical health needs of people with dementia, including pain assessment, sight and hearing problems, are given due attention; ‘Challenging behaviour’ is analysed to discover the underlying reasons; and Rights of individuals are protected in situations where actions of an individual are at odds with the safety and well-being of others.
4. Social environment: the total human relationship environment, including staff /service user relationships.
   Staff help all service users to be included in conversations and help them to relate to others, despite cognitive and mental ability; Service users are treated with respect, with an absence of people being demeaned by ‘telling-off’ or labelling; There is an atmosphere of warmth, acceptance and comfort to service users; Service users’ fears are taken seriously; Service users are not left alone for long periods in emotional distress; Staff help service users, including those with cognitive disabilities, to be active in their own care and other activities of daily living, and not treat them as objects with no feelings; and Service users are encouraged to use local community facilities and to encourage people from local community to visit regularly.

Two validated assessment tools can be used to assesses aspects of the organizational requirements to support PCC; the Short Observation Framework for Inspection, version 2 (SOFI 2) [25] and the 76-item Person-Centred Environment and Care Assessment Tool (PCECAT) [26]. The SOFI 2 [25] is one component of the Care Quality Commission (CQC) inspection toolkit used to assess the quality of social care services according to the VIPS framework [24], chiefly in the United Kingdom. CQC inspectors use SOFI 2 [25] to observe the mood and engagement of people using services and the quality of staff interactions, and other aspects of care practice during their observations. This assessment tool, therefore, focuses more attention at the individual level of care services. By contrast, the PCECAT [26] assesses both the organisational elements that support PCC for people living in aged care homes, as well as service quality according to PCC principles. It aligns with care home accreditation standards in the Australasian region. The PCECAT [26] includes items on organisational characteristics such as staffing numbers, mix, skills and education in PCC, the organisational culture, care delivery systems, service quality systems, social and therapeutic activity programs, interpersonal relationships and interactions between individuals and caregivers, and the physical layout and design. The PCECAT [26] enables care service providers and assessors to evaluate whether and how PCC is occurring at the individual level and how PCC is being supported at the organizational-level.

Methods

Results

Literature search and study characteristics

The search yielded a total of 8127 study titles and abstracts from ALOIS [30], 1022 remained after the removal of 7105 duplicates and were subjected to title and abstract screening, and 991 of the studies which failed to meet the review inclusion criteria [27] (i.e. experimental studies, focusing on people with dementia, organization-level PCC interventions) were removed. Only 31 of the remaining studies met the review inclusion criteria on reading the study abstracts. Full-text versions of these 31 studies were reviewed against the predefined eligibility criteria [27]. Nineteen of the 31 studies were excluded from full review as four breached the RCT criteria through planned control group exposure to PCC, seven included PCC trial protocols, seven were sub-studies of PCC trials, and one study included some participants without dementia. The full review yielded 12 studies fulfilling all review criteria (N = 3402 study participants) (Fig 1). Of these 12 studies two were conducted in Australia, one in Germany, two in the Netherlands, one in Scandinavia, four in the UK and two in the USA.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA [43] search strategy.

https://doi.org/10.1371/journal.pone.0212686.g001

A breakdown of the studies eligible for inclusion in the review included 11 cluster-randomized controlled trials (CRCT) [15,33,34,36,38,44–49] and one quasi-experimental study [37]. All 12 studies compared PCC with usual care, i.e. non-PCC, using multi-modal PCC approaches at the organizational level (Table 1). While there were some differences in the way that PCC was implemented, all of the 12 studies adhered to the four VIPS [24] elements and systems requirements as follows:

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Characteristics of included studies.

https://doi.org/10.1371/journal.pone.0212686.t001

Across all 12 studies, there was an emphasis on valuing the service user and the service staff. Executive and management staff established systems to support the adoption of PCC among all levels of staff, and they empowered the trained PCC champions/coaches to assist direct service staff in reconceptualizing care and therapy practices. All 12 studies focused on developing direct care staff and care manager knowledge, attitudes and skills through staff training in PCC, complemented by PCC skills modelling and supervision by PCC champions/coaches, and managerial leadership for PCC at a systems-level. One hospital-based study [38] included all members of the healthcare team in learning how to apply PCC in recognizing, preventing and managing delirium and agitation more therapeutically. Additionally, all 12 studies educated, trained and supervised direct care staff to undertake person-centered care planning in consultation with the service user and/or their family members, which formed the basis for providing individualized care. The same processes occurred for planning and delivering social, therapy and activity programs targeted to individual strengths, needs and preferences, in which family members were encouraged to participate. This emphasis on provision of individualized care and therapy programs, aimed to acknowledge and support unique service user strengths and vulnerabilities across a wide range of needs. Person-centered, non-pharmacological approaches to reducing neuropsychiatric symptoms, such as agitation, was a motif of all 12 studies.

This was enabled through targeted staff education and supervision, champion/coach role modelling, casework discussions among the team, and managerial leadership, which aimed to change staff attitudes to ways of caring for people with dementia. In achieving this goal, direct care staff were encouraged to seek out information on the individual life stories, personalities and achievements, and to employ this knowledge in delivering therapeutic care. Staff were also encouraged to draw on this knowledge in developing closer relationships with the service user and their family members, and to facilitate a more engaged social environment. This also occurred through provision of a greater variety of activities available to meet needs and abilities of service users. Four studies [15,34,38,49] adapted the care environment using person-centered environmental principles, making it less restrictive, more comfortable, recognizable, interesting and interactive, and helping to orient the service user to the current situation while supporting their capabilities.

Comparison groups in all 12 studies provided usual (non-person-centered) care, social and recreation activity programs in other care units, wards or services without any planned exposure to person-centered models and systems. Follow-up assessment ranged from 3 weeks to 3 months [36,44,45], 4 to 8 months [33,34,38,48] and 10 to 18 months [15,37,46, 47,49].

The mean participant ages were 81–86.45 years and 65% were female, and all participants had a diagnosis of dementia and presence of neuropsychiatric symptoms that caregivers found troublesome. Participant attrition ranged from 20–30% in the 10 studies undertaken in long-term care homes. Most of the participants had more than one comorbidity and all had some functional impairment associated with older age and/or dementia.

Table 1 summarizes the participant, site and study characteristics, study intervention and measurement, and the risk of bias ratings of all 12 studies.

Primary participant outcomes

The results of analyses of pooled data for the primary participant outcomes from studies measuring these outcomes are presented in Fig 2, followed by results of the sensitivity analysis in studies with low statistical heterogeneity in Fig 3.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Forest plots of primary outcomes from all studies.

https://doi.org/10.1371/journal.pone.0212686.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Forest plots of all studies with low heterogeneity.

https://doi.org/10.1371/journal.pone.0212686.g003

Neuropsychiatric symptoms.

Six studies [33, 37,38,46,47,48] reported a non-significant reduction in the standardized mean score of neuropsychiatric symptoms with person-centered care compared with usual care (SMD 0.13, 95% CI -0.21 to 0.49; studies = 6; I² = 88%) (Fig 2). Sensitivity analysis was not conducted since only two of the studies had low risk of bias [33,47] (Table 1). In four of the studies with low heterogeneity [37,38,47,48] there was a non-significant reduction in neuropsychiatric symptoms (SMD 0.06, 95% CI -0.08 to 0.19; studies = 4; participants = 895; I² = 0%) as shown in Fig 3.

Agitation.

Eight studies [15,33,34,36,44,46,47,48] produced a non-significant reduction in agitation (SMD -0.54, 95% CI -1.23 to 0.15; studies = 8; I² = 97%). Sensitivity analysis of five studies with low risk of bias [15,33,34,36,47] showed an increased overall effect for the person-centered care group (p < 0.00001, SMD 0.38, 95% CI -0.50 to -0.25). Analysis of five studies with low heterogeneity [15,34,46,47,48], however, showed no significant difference in agitation between the person-centered care group and the control group; SMD -0.05 (95% CI -0.17 to 0.07; studies 5; participants = 1043; I² = 0%).

Depression.

Six studies [34,36,44,46,47,49] produced no difference in the standardized mean depression score (SMD 0.00, 95% CI -0.21 to 0.21; studies = 6; I² = 59%). Sensitivity analysis of four studies with low risk of bias [34, 36,44,47] similarly showed no significant differences in mean depression scores for person-centred care or usual care (p = 0.87, SMD 0.01 and 95% CI -0.13 to 0.16). Five of the studies with low heterogeneity [34,44,46,47,49] showed no change in depression with person-centered care or usual care (SMD -0.06 and 95% CI-0.27 to 0.15, studies = 5; participants = 861; I² = 53%).

Well-being.

Six studies [15,33,43,38,45,49] showed a non-significant increase in the standardized mean well-being score in person-centered care compared to usual care (SMD 0.64, 95% CI -0.37 to 1.65; studies = 6; I² = 98%). After excluding two studies with high risk of bias [38,45] the overall mean score in the person-centered care group remained significantly higher than in the usual care group (p = 0.002, SMD 0.25, 95% CI 0.09 to 0.40). Four of the studies with low heterogeneity [15,33,34,49] however, produced a non-significant improvement in well-being with person-centered care (SMD 0.15, 95% CI -0.15 to 0.45; studies = 4; participants = 798; I² = 77%).

Quality of life.

Seven studies [33,34,37,38,47,48,49] produced a non-significant improvement in standardized mean quality of life scores with person-centered care, compared with usual care (SMD 0.11 and 95% CI -0.04 to 0.26; studies = 7; I² = 47%). Analysis of three studies with low risk of bias (33,34,47] showed a significant effect in the quality of life score in the person-centered care group (p = 0.02, SMD 0.2, 95% CI 0.04 to 0.36). Six of the studies with low heterogeneity [33,34,38,47,48,49] showed a significant improvement in quality of life with person-centered care (SMD 0.16 and 95% CI 0.03 to 0.29; studies = 6; participants = 1298; I² = 29%).

Discussion

The review identified 12 studies that met the inclusion criteria for assessing the effectiveness of PCC delivered at the organizational level for people living with dementia as recommended in the VIPS guidelines [23] and the PCECAT instrument [26]. While the results showed a significant effect for increased quality of life, and a non-significant improvement in neuropsychiatric symptoms and well-being, there was no evidence that demonstrated a reduction in agitation and depression. Apart from demonstrating no effect for agitation, these findings concur with the results of three systematic reviews of PCC for people living with dementia [14,50,51]. Similar with previous findings [51,52], delivery of PCC at the organizational level did not have a positive impact on level or rates of depression in people living with dementia.

In all likelihood, the mixed trial results for reducing agitation specifically, one of the most common types of neuropsychiatric symptoms experienced by people with dementia, may reflect the intention of most studies to obtain these data in relation to the person’s behaviour in general, rather than in relation to specific care events, such as during personal care when an agitation response is more likely to occur. Another possible reason for the mixed results for agitation is the complexity of an organizational-wide implementation of the PCC approach. Implementation of PCC is dependent upon a number of factors that are not always measured in PCC trials, for example, organizational culture, staffing levels, staff skill-sets and mix, and the physical and psycho-social care environment.

There can be limited confidence in the findings of this review, given the moderate to high methodological quality of the 12 included studies in relation to the clear reporting of the study designs and methods. A majority of the studies had low risk of bias overall, with the exception of performance bias common in CRCTs and quasi-experimental studies. All 12 studies included participants of similar ages and clinical characteristics, including a valid diagnosis of dementia, and exhibiting neuropsychiatric symptoms which caregiving staff found troublesome. As expected in this vulnerable population, there was a 25–30% participant loss-to-follow-up in studies conducted in the long-term care setting, mainly due to the death of participants. The variation in time to follow-up in the different studies, ranging from 6–18 months in the long-term care setting and two to 3 weeks in the hospital setting, is explained by the focus of these different settings and thus, the longer predicted length of stay in the long-term care setting compared with the hospital setting. Being able to achieve improved quality of life in people with advanced dementia through the delivery of PCC in both of these settings has important clinical and policy implications.

Another cause for confidence in the review findings is that all 12 studies obtained baseline and follow-up outcome data using validated measures for the primary outcomes and for most of the secondary outcomes. Many of the studies employed the same validated outcome measures (refer Table 1), which enabled robust statistical analyses of the combined results.

While all studies showed improvements in at least one primary outcome using multi-modal implementation of PCC as outlined in the VIPS guidelines [24], the implementation procedures were not standardized across studies. All of the studies did adhere, however, to the principle of aiming to support personhood through the provision of individualized care, therapy and activity programs by all direct caregivers and therapy staff, as well as adhering to the principle of requiring all staff to interpersonally communicate and engage with the person in ways that encouraged their self-determination, self-respect, dignity and well-being.

Another common feature of all studies was the operation of PCC at the organizational-level, as recommended in the VIPS framework [24]. Organisational-level interventions included a combination of: staff education, training, guidance and exposure to role modelling in how to understand and interpret verbal and non-verbal communication from the perspective of the person with dementia; and providing the person with interesting, purposeful, and meaningful things to do in their daily life, in consultation with the person, their family and caregivers. Person-centered care and leisure/social activities all included approaches that were meaningful for the person, such as communicating with them about their memories of family, friends, places and events. When the whole care team employed this approach, it was found to increase the person’s participation in care and leisure/social activities, which in most studies also improved psychosocial and functional outcomes, albeit non-significantly.

These outcomes support Kitwood’s [7] theoretical assumption that positive interpersonal relationships and enriched care environments can prevent the disabling effects of dementia and promote a sense of well-being for the person [6,8,9]. A recent study evaluating an organisation-wide PCC model in Norwegian aged care homes found a positive relationship between a person-centered organisation, care staff work practices and organizational systems established to support the model [52].

Supporting information

Acknowledgments

The authors acknowledge the contribution by invited reviewer EB, who evaluated and consulted with SL and AYW about the two published CRCTs of which LC and JSP were chief investigators and thank her sincerely for assisting with this aspect of the review. Also acknowledged are members of the Cochrane Dementia and Cognitive Improvement Group who accessed potentially suitable article abstracts for the authors.