Can Health-care Assistant Training improve the relational care of older people? (CHAT) A development and feasibility study of a complex intervention

Phase 1

We conducted a telephone survey of all NHS trusts in England to understand what training as usual (TAU) looked like for HCAs caring for older people in hospitals in England. We wished to establish the structure, content and variability of HCA training and, in particular, training in providing relational care of older patients in hospital. Respondents to the survey were those responsible for HCA training within their trust. The methods and findings from the telephone survey are reported in detail in Chapter 3.

The qualitative component of phase 1 comprised focus groups of older people (or their carers) with recent experience of hospital care, together with interviews of HCAs and other hospital staff undertaken in each of the three study centres. These methods and findings are described and reported in Chapter 4. The purpose of these focus groups was to understand the care experiences of older people and their expectations of the training HCAs should receive. The purpose of the interviews with HCAs and members of staff who worked alongside them was to gain insights into staff perceptions of the challenges that HCAs face in caring for older people in hospital and to explore interviewees’ perceptions of training needs in this area of care.

A new training intervention for HCAs to improve the relational care of older people was developed as part of the study. The process of creating this training intervention, Older People’s Shoes™, is described in Chapter 5. The training intervention drew on several sources: the interviews conducted in phase 1, existing evidence from the literature, an expert panel and learning about customer care practices of four retail organisations.

Phase 2

The second phase of the CHAT study was a feasibility cluster randomised controlled trial and process evaluation. This compared TAU for HCAs with the new HCA training in relational care of older people, that is, the Older People’s Shoes intervention. The unit of randomisation was the hospital ward. Outcomes were assessed at the level of ward, HCA and patient. The purpose of the feasibility trial and the process evaluation was to determine the feasibility and viability of a definitive trial. Methods are described and findings reported in Chapter 6.

The process evaluation was conducted in parallel with the feasibility trial. This consisted of observations of the delivery of the intervention, follow-up interviews with trainers and HCA learners, and learners’ evaluation following training. The process evaluation is reported in Chapter 7.

Methodological frameworks

Owing to the nature of the study design and the range of methods used to address the aims, we drew on a number of methodological frameworks to inform our study. The HCA training intervention developed as part of this study and the feasibility testing of it as part of a trial was informed by the most recent guidance on the development and evaluation of complex interventions. 49 Of the four stages (or elements) of the process from development through to implementation of a complex intervention, the focus within our study was on development (phase 1 of the CHAT study) and feasibility/piloting (phase 2). Our aim was to follow this guidance when possible, up to but not including a definitive evaluation:

Best practice is to develop interventions systematically, using the best available evidence and appropriate theory, then to test them using a carefully phased approach, starting with a series of pilot studies targeted at each of the key uncertainties in the design, and moving on to an exploratory and then a definitive evaluation.

Medical Research Council,49 p. 8

In designing the feasibility study for a randomised controlled trial we used Kirkpatrick’s four-level evaluation model,50 and measured outcomes at each level: reaction (measured by course evaluation), learning (change in empathy and in stereotypical attitudes towards older people), transfer (observations of relational care delivery) and results (patient experience of the relational care they receive). The measurement of distal outcomes of health-care training is challenging. In the Older People’s Services and Workforce Interventions: a Synthesis of Evidence (OPSWISE) study for clinical support workforce developments,51 of the 76 papers identified, only two were reports of randomised controlled trials (Kuske et al. 52 and Clare et al. 53) and only one observed level 4 (care home resident) outcomes. 53

The Kirkpatrick training evaluation model has been criticised for a lack of attention to the environment in which trainee skills are practised. 54 This was in part addressed by the phase 2 process evaluation for which we drew on recently published guidance. 55 A range of methods was used to inform our understanding of the different contexts in which the training intervention was delivered, the process of intervention delivery and the mechanisms of impact.

Presubmission of the grant

Prior to the activation of the grant we worked with the Public and Patient Involvement in Research (PPIRes) group, an organisation hosted by the South Norfolk Clinical Commissioning Group. 59 The PPIRes group brings together volunteer members of the public to collaborate with researchers in local trusts and universities in Norfolk and Suffolk and to develop proposals from the initial idea through to dissemination. At the time of writing it has a panel of approximately 70 lay members. Prior to submission of the grant application we worked with the PPIRes co-ordinator to plan the PPI in the study and to invite panel members to be involved in the development of the application. Twenty-six volunteers responded and a summary of the study document was circulated via the PPIRes co-ordinator for review. The purpose of this was twofold: first, to get informal feedback from panel member views on the questions the study sought to address and on its proposed methods; and, second, to identify potential panel members who might wish to play a more active role should the study be funded. Views on the study were positive. A question was raised on whether or not the staff group should be extended beyond HCAs to OS. This highlighted the potential breadth of application for the intervention, but the focus of the commissioning brief prevented us from incorporating this suggestion. Some panel members expressed uncertainty as to the role of a HCA, and this was an early reminder of the need to check our assumptions about the ability of patients and relatives to distinguish members of the HCA workforce from other care staff. A discussion group was also organised in which all available documents were circulated in advance and six volunteers attended a 3-hour meeting to discuss the application in detail.

Recruitment and study documents

Prior to our application for ethics clearance to conduct phase 1 staff interviews and focus groups with older people, the PPIRes co-ordinator arranged a meeting (7 November 2013) of four panel members and the principal investigator. The purpose of the meeting was to review participant-facing study documents. Consent and participant information sheets based on NHS template documents were adapted in the light of detailed discussion at the meeting. Changes were made to simplify expression of interest forms and participant information sheets. The focus group prompt guide was also adapted, with suggestions made as to how to explain what we meant by relational care to focus group participants. At this point, two of the group became the PPI representatives for the CHAT study and remained so for the duration of the study period. Margaret McWilliams has been a PPIRes member for over 10 years. Her interest in this project stemmed from a carer perspective and a short hospital stay, which emphasised the importance of HCAs and how essential it was to be kept informed of what was going to happen as part of your daily routine. Margaret runs a hearing aid clinic for Norfolk Hearing Support Services where she speaks to many older people about their experiences. Janet Gray has been a PPIRes member for 2 years and is the carer of her parents and relatives who have experienced many hospital stays.

Focus groups

A later section of the report details our work with older people’s organisations to assist with raising awareness of, and recruitment to, the focus groups (see Chapter 4, Recruitment). In addition, we were keen for PPI representatives to play a key role in the conduct of the focus groups themselves. As our PPI representatives were based at one of the three study centres, local PPI representatives were recruited for this purpose at the other two centres. The contribution of the PPI representative was determined by their own preference and, therefore, varied at each centre. At one focus group, for example, a PPI representative chaired the discussion. At all three focus groups the PPI representative worked with the facilitator to welcome participants as they arrived, clarified facilitator topics and participant discussion as needed and alerted the facilitator to participants who indicated that they had a view to express but were reticent about joining in the discussion.

Intervention development

The process of intervention development is described fully in Chapter 5. The core intervention development team included our two PPIRes representatives together with a HCA from one of our partner trusts working on a ward caring for older people. Collectively, the PPI members worked to keep the focus on the needs of older users of hospital services and to ensure that the training intervention was designed with HCA learners firmly in mind. The group met on four occasions and formed a close knit team to produce what became the Older People’s Shoes training intervention. Roles inevitably became less demarcated and all team members became involved in all aspects of intervention development including structure, content, delivery, and proofreading training materials. We consider that the final product was substantially strengthened by this invaluable contribution. In addition, our HCA representative worked with researchers shortly prior to intervention delivery to ensure that activities were credible to reflect the work experience of HCAs in busy hospitals.

Study oversight

Details of the trial steering group are provided in Study Management. Membership of that group included our two PPIRes representatives as well as a HCA representative recruited via the Royal College of Nursing Health Practitioner Committee. The steering committee provided oversight to all aspects of the study. Our PPI and HCA representatives were vocal and enthusiastic members of this committee, providing sound and thoughtful advice at each stage of the research. They were also very supportive of the research team at points in the process when we hit challenges.

Feedback and reflection on the process of public and patient involvement in the CHAT study

For a relatively short project (2 years) we felt that both the process and outcome of PPI within the CHAT project was successful. We forged strong relationships over a short space of time. Soon after the study end point, the PPIRes co-ordinator conducted an informal meeting with our two PPI representatives to hear their views on the PPI process. Both PPI representatives commented on how much they had enjoyed being part of the team and that the experience had been rewarding. They felt their contribution was valued and they appreciated being included in communications beyond formal meetings. They felt that their views had been sought and respected by the steering group, with the chairperson of that group ensuring that they were actively involved in discussions. They were appreciative of travel arrangements for meetings being organised well in advance. Working alongside our HCA representatives had assisted them in understanding the nature of a HCA’s work, and, by extension, the focus of the study from both a user’s and a caregiver’s perspective.

Purpose

The purpose of the telephone survey was to understand the current provision of HCA training, particularly with regard to relational care for older people. This would provide insight into how a new training intervention in relational care for HCAs could be effectively delivered within the context of current training provision in acute NHS hospitals. The objectives of the telephone survey were to understand (1) current training and support processes, (2) the extent of training content with respect to relational care and care specific to older people and (3) perceived challenges in delivering HCA training.

Sampling frame and eligibility

All NHS acute hospital trusts in England were eligible to take part. Trusts were identified from the Health and Social Care Information Centre,60 which places each trust into one of six categories (large, medium, small, multiservice, specialist and teaching). The one key contact at each trust eligible to take part in the telephone survey was a person with responsibility for designing, managing, delivering or overseeing the training of HCAs.

Recruitment

Recruitment to the telephone survey was carried out by four researchers employed on the study grant. The recruitment process is illustrated in Figure 3.

FIGURE 3.

Telephone survey recruitment process.

Data collection

The survey was carried out over two periods: between February 2014 and September 2014, and then between July 2015 and October 2015. This was because of the early departure of a researcher at one centre and the period of time that elapsed before a replacement could be made.

Data management, coding and analysis

Data were collected in a paper-based case report form and in audio files. Audio files were recorded using a portable digital voice recorder connected to a standard telephone. A unique identifier code was assigned to each trust. Audio data were uploaded and stored locally on secure servers. Structured telephone interview data were extracted from audio files and case report forms to a spreadsheet. Extracted data were anonymised and either coded for analysis or described accordingly. To categorise HCA training content, two researchers (CA and FN) coded data retrospectively using a shared template.

Counts and percentages of non-missing data were used to describe categorical data, and means with their standard deviations (SDs) were used to describe continuous data. A key development in the interim between the two periods of time during which the survey was conducted was the introduction in March 2015 of the Care Certificate. To check for any bias that this may have caused, we compared trusts interviewed before and after the national launch of the Care Certificate, using unpaired t-tests for continuous variables and chi-squared tests for categorical variables. A similar process was used to examine non-response bias comparing trusts that took part with those who did not. For categorical variables, where one or more cells had expected cell counts of five or less, Fisher’s exact test was used. All data analysis was conducted in Stata version 14 (StataCorp LP, College Station, TX, USA).

Ethical considerations and ethics approvals

We were mindful that the care of older people in hospitals has been subject to recent criticism. 1 The telephone survey was undertaken at a time at which HCA training has been the focus of national attention. 2 This required our approach to both recruitment and the conduct of the telephone interview to be sensitive. Potential and actual participants were assured that the focus of the survey was to get a national picture of HCA training in acute NHS hospital trusts rather than identify particular failings. The researchers made it clear to the key contacts interviewed that individual trusts would not be identifiable in any reporting of survey findings.

Permission to undertake the telephone survey was provided by the Faculty of Medicine and Health Sciences Research Ethics Committee at the University of East Anglia on 19 December 2013 (reference number 2013/2014-19).

Sample

Of the 161 acute NHS trusts approached to take part in the survey, a total of 113 (70.2%) structured telephone interviews were completed (Table 1). Of those trusts that took part, the mean number of whole-time equivalent staff was 4646, and there was no evidence that size of staffing establishment differed between participating and non-participating trusts (p = 0.43). Across Health and Social Care Information Centre trust type (small, medium, large, multiservice, specialist or teaching) the proportion of trusts that responded did not vary (p = 0.94). Trusts were surveyed in one of two time periods over the study duration and the proportion participating was lower during the second period (56.9% vs. 80.9%; p < 0.001). Trusts approached in the second period included those that had not refused in the first period but were more difficult to establish contact with. The second period took place after many trusts had been involved in preparing for the introduction of the new Care Certificate that was officially launched in March 2015. Two-thirds (66.1%) of the key contacts at participating trusts were involved in the direct delivery of HCA training, whereas in the remainder the key contact had a more strategic role in HCA training.

Structure of health-care assistant induction training

Key contacts at just under half of participating trusts (50/110, 45.4%) reported induction programmes of 1 week or less, with the remainder having longer induction programmes and 1 in 10 having HCA programmes of between 2 and 3 weeks (Table 2). It was the norm for new HCAs to have a mentor or buddy (98/113, 86.8%), with only eight trust key contacts (7.1%) saying this was not the case. When respondents reported the type of mentor for those trusts with a system of mentoring or buddying, these were senior HCAs (n = 50, 46.3%), registered nurses (n = 16, 14.8%) or either (n = 17, 15.7%). New HCAs were accorded supernumerary status at most of the participating trusts (n = 81, 71.7%), the remainder reporting that new HCAs were not supernumerary or that supernumerary status was dependent on other factors. Many trusts indicated that the duration and type of support on wards was at the discretion of the ward manager. There was no evidence of differences between trusts surveyed at each time period with respect to how HCA induction was structured (analysis not shown).

Content of health-care assistant induction training

One-third (n = 37, 32.7%) of trust key contacts reported content within their HCA training induction programme that we considered to be relational care (Table 3). When asked specifically about induction training that was related to the care of older people, 43 (38.1%) trust respondents referred to subject areas such as privacy, dignity and respect (n = 30, 27.3%) and communication skills (n = 24, 21.8%), all considered by the researchers to involve relational care. Only two respondents (1.8%) said that their trust covered the subject of ‘customer care’. Dementia care was reported as being included in HCA induction programmes by the majority of respondents (n = 94, 85.5%). Other training induction content relevant to older people and reported by survey respondents was nutrition and hydration (n = 31, 28.2%), falls (n = 25, 22.7%) and sensory/physical impairment (n = 23, 20.9%). Nearly one-third of respondents (n = 35, 31.5%) said that they made no distinction during induction training between the needs of older people and those of any age group. Nearly all trust respondents interviewed prior to the national launch of the Care Certificate reported plans to develop HCA training (56/57, 98.3%) compared with 73.7% (28/38) of trusts surveyed after the national launch (p < 0.001), suggesting changes were just starting to be introduced in the intervening period.

Challenges of training health-care assistants

Reported challenges related to training HCAs were categorised under four headings: the wider context, resource limitations, ward engagement and HCA-related challenges (Table 4). The most frequently cited challenge for delivering training to the HCA workforce was getting staff released from wards to attend (n = 53, 46.9%). Whether this was because of a lack of ward manager engagement with HCA training delivered at trust level or simply because of a lack of staffing resource is not possible to determine from our data. However, many respondents unsurprisingly cited resource limitations as a challenge. Trust key contacts reported challenges of being limited not just in terms of funding but also in relation to the availability of assessors, mentors and training venues. The wider context of opportunities (or lack of opportunities) for HCAs to develop their role was highlighted by some, together with the difficulties associated with HCA training rarely recognised beyond an individual trust. The highly diverse nature of the HCAs in terms of their care experience and academic ability was the most cited challenge relating to members of the HCA workforce.

Training beyond induction

The variability of the extent and nature of training after induction meant that insights into this area of HCA training were gleaned through open-ended questioning. Once the initial training/induction period was over, many trusts reported that HCAs had access to ongoing training, although the emphasis was on training newly employed personnel. Only one trust reported that they had received funding allowing them to put both new and existing HCA staff through the Care Certificate. One trainer suggested that restricted access to training affected the ability to retain good HCAs, thereby increasing HCA turnover and creating a ‘vicious circle’. Although many trusts reported having postinduction training available, this varied greatly in terms of structure, focus and content. The target group for this training also varied greatly between trusts. Some reported holding regular HCA study days covering an array of specialist skills; however, these sessions were rarely mandatory and tended to be at the discretion of ward managers. Owing to time and resource constraints some trusts had opted for an e-learning approach and offered packages in areas including dementia and end-of-life care.

One trust offered monthly open-access support worker sessions, which could be tailored to the needs of the individual, and another trust ran a weekly skills refresher day open to both registered nurses and HCAs. However, an ad hoc approach to training was the norm for most trusts. Many telephone survey respondents were unaware of the content of specialist training available to staff, as this was carried out on the ward by clinical trainers. Again, this training was governed by managerial requirements and limited by time constraints.

Purpose

To inform the content of the HCA training intervention we ran three focus groups (one in each centre) of older people and carers with experience of acute care. We wished to identify these groups’ experience of relational care as provided by HCAs. We wanted to understand the values-based training needs of HCAs in maintaining the dignity of, and affording respectful care to, older patients in acute NHS settings from the perspective of those they care for. Each focus group aimed to gather a broad range of perspectives from older people who had been an inpatient, or a carer of an inpatient, at an acute NHS trust. The purpose of focus groups is to explore people’s experiences, attitudes and feelings on a topic in a way that capitalises on group interaction. 61,62 Interaction enables participants to build on other people’s input and to ask questions of each other, as well as to re-evaluate and reconsider their own understanding of their specific experiences. 63

Setting and eligibility

Focus groups were carried out in non-clinical settings in venues that had disabled access and transport links. At each centre transport was arranged for participants who required it, and costs were reimbursed for others. Eligible participants were former hospital inpatients at any acute NHS trust aged ≥ 65 years or the carer of a former inpatient aged ≥ 65 years. Although not an eligibility criterion, we prioritised those whose experience of an inpatient stay was at least 3 months and no longer than 6 months prior to the focus group meeting on the basis that this would avoid any very raw emotions in a group setting, while maximising the chances of recall.

Recruitment

Data collection

Data management, coding and analysis

An analysis of focus group transcript data was carried out in NVivo version 10 (QSR International, Warrington, UK). Data were initially coded in NVivo, using a framework of codes aligned to the broad themes suggested by topic guide, for example what people want when they arrive on a ward, examples of good relational care (in hospitals and in other settings) and what staff should know about individual older patients. This was followed by more detailed analysis using an inductive approach in which data within themes were examined and interpreted to draw out more refined themes and conceptual nuances.

Given that the purpose of the focus group was to inform the development of a training intervention for HCAs to improve their delivery of relational care, analysis focused on thematic content, and not behaviour or non-verbal data. 64 As focus groups are valuable for the interactions between participants, instances of consensus, contradiction and controversy were sought and used in presenting the findings. 61 Data were not analysed for differences between groups or along lines of sex or ethnicity. However, the relationship between patient and carer needs was examined.

Ethical considerations and ethics approvals

At the start of the focus group meeting, participants were reminded that data would be anonymised and kept confidential and were asked to maintain the anonymity and confidentiality of other participants. Thinking about and discussing experiences of and around hospital stay can be upsetting. At each focus group one member of the research team was given responsibility for looking after any participants should they be upset and wish to withdraw from the discussion. At the end of the discussion participants were provided with details of the local trust’s patient advice and liaison service should they wish to discuss their experiences further. Six months after the focus group, an update on the study was sent to all focus group participants, letting them know how their views were being used.

Permission to undertake the focus groups was provided by the Faculty of Medicine and Health Sciences Research Ethics Committee at the University of East Anglia on 19 December 2013 (reference number 2013/2014-19).

Purpose

Semistructured one-to-one interviews with HCAs and OS (principally nursing staff) in the three centres were conducted to elicit their perspectives on what good relational care of older people looks like, what a training intervention for HCAs should contain and what style of training delivery was likely to be most effective. These interviews allowed us to understand the context of providing relational care to older patients, any barriers to training access or implementation of training and to investigate the perceived training needs of HCAs with respect to relational care.

Setting and eligibility

At each centre we worked with a partner acute NHS hospital trust. The three trusts were all teaching hospitals: one in London, one in a rural county and one in the Midlands. Wards caring for older people in the trusts were purposively sampled to reflect a wide range of HCA experience on different types of ward (health care of older people, general medical and orthopaedic). Our intention was to ensure that the training intervention developed would be relevant to HCAs with different levels and types of workplace experience. Eligible ‘OS’ were those who directly manage HCAs on recruited wards (ward managers and staff nurses), who work alongside HCAs on recruited wards (e.g. allied health professionals) or managers with responsibility at the division or trust level.

Recruitment

Data collection

Data management, coding and analysis

The interview data from each trust were coded in NVivo by the local researcher using a coding framework developed from initial readings of the transcripts and agreed by the study team. This collaborative work to identify themes ensured validity and reliability of the analysis. The coding framework included broad themes specifically directed towards the aim of the study (to develop a training intervention for HCAs to improve relational care of older people). The analysis used both deductive and inductive approaches.

Examples of deductive themes were organisation- and patient-related challenges in HCAs’ work, the role of HCAs in relational care (categorised using the study team’s understanding of what relational care consisted of), experiences of good customer care and perceived gaps in training. Other themes were imposed to inform how we framed the intervention and managed practical arrangements, as well as giving important contextual data to help us interpret our findings on the feasibility of the intervention.

Following this process of deductive data extraction, a more detailed thematic analysis of the whole data set was then carried out in NVivo by one researcher, using an inductive approach and the constant comparative method in order to enhance analytical rigour65 and the credibility and ‘trustworthiness’ of the findings. 66 At this stage, subthemes such as ‘tensions inherent in HCAs’ work’ emerged from interviewees’ accounts.

The interview and analysis process was iterative. This meant that we were able to use findings from earlier interviews to inform subsequent interviews. For instance, we used early findings on ‘challenges in HCAs’ work’ to frame a question used in later interviews about whether or not interviewees thought such challenges could usefully be addressed in training.

Ethical considerations and ethics approvals

Ward managers agreed to participate in the study prior to fieldwork commencing. The research team was keen to ensure that ward staff were not under pressure from ward managers to take part in interviews. We therefore approached HCAs directly and made it clear that participation was voluntary. Verbal consent was taken at the initial approach and written consent immediately prior to the interview. All interviewees were free to refuse consent to being audio-recorded and to withdraw from the study at any time. Confidentiality and anonymity were assured.

Permission to undertake HCA and OS interviews was provided by the Faculty of Medicine and Health Sciences Research Ethics Committee at the University of East Anglia on 19 December 2013 (reference number 2013/2014-19) and from the research and development departments of each of the three participating NHS trusts.

Sample

Thirty people were invited to participate in one of the three focus groups, of whom 29 attended (Table 7).

Length of hospital stay ranged from 1 day to 4 weeks, and time since discharge ranged from 2 months to 3 years, although one carer had a friend in hospital at the time of the focus group. Participants drew on experience of care from 14 different hospitals. Focus group discussions were carried out in each centre between June and July 2014 and lasted an average of 1 hour and 45 minutes.

What is relational care?

Founded as it is on relationships and interaction, relational aspects of care can be discerned in terms of both how it makes patients and families feel ‘cared about’ and ‘feel in safe hands’ and what is required from staff in order to elicit that feeling. Attitudes, behaviours and communication styles were all implied. We asked focus group participants to tell us about experiences during a hospital stay when they had felt that they (as patients) or their family member or friend had been cared about. The findings presented here, which use pseudonyms, include comments from both patient and carer participants on what gave them a positive feeling of being cared for.

Five major themes emerged through our analysis of staff activities associated with effective relational care: building relationships, showing kindness and concern, noticing and being proactive, offering choice and individualising care and making patients and families feel welcome and secure. Our findings echo those from other studies of the importance placed by older patients and their families on the relational aspects of care. 11,13

In participants’ accounts, positive experiences of care hinged on relationships. Chatting to patients and getting to know them were important in building those relationships. A degree of mutual disclosure was also involved:

My father is very frail and doesn’t talk easily to people and will just keep himself to himself. But one day we went in to see him and he was chatting really happily with this health-care assistant, and they were having a laugh and he knew all about this health-care assistant’s life, he was another nice Jewish boy and whatever and blah, blah, blah. And it was just lovely, it was heart-warming.

Hannah, carer, centre 3

Staff showing kindness and concern was noted as important. An example was given of a HCA really going the extra mile in putting together a photo album for a patient who had great difficulty in communicating. Often these qualities were manifested in doing ‘little things’, such as charging a mobile phone, tending to a patient’s appearance or offering a cup of tea. These displays of good relational care played an important part in making patients feel cared about and secure in an alien environment:

[T]he most important thing to me was the kindness of the nurses. I couldn’t sleep and a nurse came round and she said ‘Are you OK?’ and I said ‘I’d love a cup of tea’ and she went and got me one. That was very important to me. It gave me a feeling of security, peace of mind and that sort of thing. It’s something I think most elderly people would value most, to have peace of mind because they’re away from their secure surroundings.

Evelyn, patient, centre 1

Participants in all three focus groups thought that ward staff should be proactive in noticing care needs and offering care. This was thought to be relevant for older people with communication difficulties or cognitive impairment but also to be applicable beyond that as older people often felt reluctant to ask for help:

[M]y mum would never ask anything anyway . . . So somebody to come and just talk and check on you regularly I think is really important.

Vera, carer, centre 1

It was also noted in all groups that one element of good relational care was offering choice and individualising care:

[T]he health-care assistant that was looking after my mum came to her in the morning, [. . .] and asked my mum ‘What do you want to happen as far as your care is concerned, like washing, dressing and stuff like that?’. Asked her what she actually wanted, and would she rather have her wash earlier or later? Did she want a shower, does she want a bath? She actually asked her what she wanted and what she preferred. [. . .] my mum really appreciated that.

Julia, carer, centre 3

Giving this kind of choice helped to incorporate familiar routines into an alien environment and gave patients a degree of self-determination. Talking over patients as if they were not there was given as a prime example of poor relational care. Such treatment was experienced as depersonalising. One person said it made her feel as if ‘you might as well not be there’. Another told of a friend’s distress and sense of being ‘demeaned’:

[O]n one occasion [. . .] the bed was being changed and there was a carer on both sides helping [my friend] because they couldn’t do it by themselves, and they talked over her, and just as they were leaving they said ‘Did you ever work?’ And in fact she was a very intelligent person who’d held down an important job in the National Health Service. And when I went in she said ‘They just didn’t relate to me in any way’. And she was almost in tears because she said ‘I know I’m old and it’s a long time ago, but at one point I was somebody’ [. . .] She was really upset by that because whether she’d worked or not was important, but she felt demeaned and I thought that was very sad.

Wendy, patient, centre 1

Carers valued being made to feel welcome and at home. They also wanted to be kept informed. Carers’ and patients’ feelings of comfort and support were intimately entwined. The treatment of one impacted on the feelings of the other:

I think it’s also very important when you’re a patient in hospital, lying there all day, waiting for the visiting time to start, that you’re confident that your family and your visitors feel confident enough to talk to the staff and find out if you’re OK, if anything else is going to crop up, so that everybody can work together and look after you as a team.

Sophia, patient, centre 1

One interviewee summed up a general view of relational care:

[B]y and large, it’s a matter of meeting that patient’s specific individual needs and engendering a relationship where the patient feels that he or she is being well looked after and has the confidence and the ability to rely on those who are around him or her.

David, carer, centre 3

There was a broad consensus among focus group participants of wanting patients to feel welcome, known, secure and ‘at home’, and there was an iterative relationship between the feelings of older patients and carers.

Experiences of relational care beyond health-care settings

As an alternative way to unpick what relational care looked like, we asked participants to talk about occasions outside health-care settings (e.g. shops, banks or restaurants) in which they had felt very well treated and what it was that had engendered such a feeling. There was a great deal of overlap in responses to this question and the question of relational care in hospital. Ingredients of relational care noted outside acute settings were a timely response and staff being kind, helpful and informative, knowing who you were, making you feel special, listening attentively and noticing your needs:

[W]hen you go out for lunch somewhere [. . .] when it really works and you feel really special and well treated is because they’re actually kind of noticing what’s going on all the time, noticing whether it’s time to come over and shift the plates rather than leave you there for ages. Noticing whether it’s the right moment and moving at the right time. So that observation but having the time to observe obviously and then to react accordingly and they’re quite busy of course in restaurants.

Vera, carer, centre 1

[In] Canada [. . .] you go into a shop, it’s ‘Good morning, how are you?’ and everything [. . .] which I think is most important because if you’re feeling down and you go into a shop and someone smiles at you, you feel so much better, you know.

Maureen, patient, centre 2

Although some participants voiced a wish that people should try to understand what it is like being old, this was not straightforward and a danger of being stereotyped was also acknowledged in one focus group:

I’m in a care home and we have people who are not very well trained. They do get training, particularly in all the things like infection control, and health and safety and so on and so forth, but it’s those other little things – them knowing how you think when you get older. And they think you’ve aged a lot more than you really have.

Mm.

And I sometimes sort of feel like turning round and saying ‘Look, I may be the age you can see on the care plan but I’m sorry, I don’t want to talk about that, I want to talk about something interesting, something out of the newspaper’. [. . .] they really do think we were – well we were born in the last World War. But we don’t always want to go around singing ‘Pack up your troubles in your old kit bag’!

Eileen, patient, centre 2

Not only did some participants complain of being stereotyped, but they felt they were dismissed, or even invisible, because of their age:

[. . .] I feel now they just see you as old.

They do, if they see you at all.

If they see you at all, they see you as old and dismissed.

[. . .]

There are some nice younger people.

Yes.

Yes, there is, there’s a lot of them they treat you with respect. But they’re just – you come across some that don’t, and it hurts sometimes because you think ‘I’m a human being still, even with my wrinkles and my ruddy arthritis and everything’, you know.

Avril and Joan, patients, centre 2

This feeling of being dismissed resonates with the depersonalisation that patients felt when staff talked over them as they carried out their tasks.

Patients’ and carers’ expectations of staff when first arriving on a ward

We asked participants what they felt they should be able to expect when they first arrived on a ward. This line of questioning was based on the assumption that ‘first impressions count’ and drew on research on older people’s experience in urgent care settings,67 which found that older people frequently experience a diminished sense of significance, a feeling that they do not matter, which the authors attributed in part to a lack of attention to older patients’ wider psychological and informational support needs. Our findings that patients and carers place great importance on being made to feel welcome and ‘at home’ suggest that these feelings need to be established early on:

A very good welcome. Make them feel comfortable. Make them feel valued. Make them feel like they’re in good hands. They are free to ask for whatever they feel will make them feel better.

Shola, carer, centre 3

A ‘good welcome’ included being greeted with a smile and being offered a warm drink, and staff introducing themselves. Being made to feel at home also included introductions to OS and their roles, being introduced to other patients in the bay, being orientated to the ward by being told or shown where things were and being informed about routines of the day, such as meal times and visiting times.

Patient and carer confidence and sense of ease was tied to feeling that staff made the effort to get to know a patient’s particular needs, such as whether or not they need glasses or a hearing aid, whether or not they had dementia, their dietary requirements and other preferences. One participant said:

Time at the beginning is really important because mum doesn’t understand, she’ll forget within 1 minute what you’ve said to her. And the carer needs to know specifically about her and what her needs are, otherwise she’s going to feel at sea and completely lost. So some time with somebody to get all the background.

Vera, carer, centre 1

What patients and carers want staff to know about them

Good relational care involves building relationships between staff and patients, and staff getting to know about patients’ individual care needs and preferences on arrival in a ward. With an average length of hospital stay of 9 days for an older person,3 we asked participants what they would wish or expect staff to know about a patient over this length of time. A positive experience on admission to a ward is crucial, but in the longer term participants want staff to know things about them that help to build a relationship and make them feel known as individuals. Such knowledge includes past occupation and something of their history, where they live, hobbies and interests and a bit about their family. Some participants pointed out that the relationship with staff members is stronger when information sharing is reciprocal:

I think it’s all part of the settling process, you know, if they could see one of these forms that you fill in, I think you get them from the Alzheimer’s Society, you look at it and you think ‘Oh, you know, he was in the RAF [Royal Air Force]’. ‘Oh you were in the RAF were you, [name]?’ and it perhaps just strikes a chord and he thinks ‘Oh, you know, they know something about me, I’m not just something that’s going to lay on a bed’.

Trish, carer, centre 1

Well I suppose generically whatever one needs to know in order to build a relationship. I’m not quite sure what the ingredients are. [. . .] It’s like when I go to the barber, you know, we talk about football or his children, my children, whatever it may be. Just those little details that sort of make the difference between a closer relationship and a more distant one.

David, carer, centre 3

I think it’s appropriate to have a little bit of background on the patient that would build a relationship [. . .] The children, the husband, just a little bit about the family. So that’s sort of starting the relationship, or the conversation. Then the nurse might also tell you a little bit about themselves so you have something in common to discuss about – as you said [David], about football and the rest of it – or something for discussion.

Gloria, carer, centre 3

Recommendations for health-care assistant training in relational care

Given the importance of relational care to focus group participants, the fact that not all of their experiences of this were positive and in acknowledgement of the fact that there is no standardised training for HCAs, there was strong support for HCA training in those aspects of relational care noted above: building relationships, kindness and compassion, being friendly and approachable, getting to know patients as people, finding out, noticing and expecting care needs, listening attentively [‘It’s listening and hearing isn’t it?’ (May, patient, centre 3)], individualising care, involving patients in decision-making and not depersonalising patients.

When asked about other areas of content they would wish to see in the training of HCAs, participants talked about communicating with people with dementia, doing stimulating activities with patients, ways of encouraging patients to eat and drink, not passing on work-related stress to patients and non-verbal communication. On this last point one participant said:

On the non-verbal, actually to be aware of body language and also bustle. If you’re bustling people are going to think OK, she’s in a hurry. If you can be kind of relaxed when you come to talk to the patient, it’s not just ‘How are you? Good morning’, it’s how you say it, not just what you say.

Vera, carer, centre 1

The focus groups threw around ideas about what the core messages of the training might be, and how to get those across. The first idea was that relational care was not an add-on task. Rather, it was something that could and should be woven through everyday care activities. As one participant put it:

I mean, when they’re actually doing observations with patients, taking temperatures and things like that, that’s the time they should be talking to the patient for 5 minutes. To just talk with the patient and, you know, get to know the patient a bit better.

Avril, patient, centre 2

The second idea was that it was important to understand what it is like to be an older person:

[R]emember that we are older (and I’m talking for myself now and probably for some other people) and we’re slower. And whereas somebody else can just sort of jump out of bed and that’s it, it takes me quite a time.

Anthea, patient, centre 1

The third idea regarded understanding what it is like to be an older person in hospital:

[U]nderstand the position that that person was in. [. . .] [T]hat person is in the hospital for a reason. They’re not in there because they’ve asked to go [. . .] they’re in there for a specific reason and I think it’s understanding that that person is in a totally strange environment, perhaps somewhat frightened (because most of us are frightened of the knife or whatever we’re in there for), and understanding that that person is totally out of the environment that they are used to being in, and it’s trying to get those person to make them feel at home, welcome and that they’re not on their own.

Clive, patient, centre 1

One noted aspect of being an older patient was how being dependent on staff for urgent needs created a feeling of powerlessness:

[W]hat I felt particularly was if only the person there could see how powerless these people feel when they’re so vulnerable and they’re in their beds. And have some understanding of that. Then they could be a lot more generous towards them I think.

May, patient, centre 3

This participant appears to be asking for staff to respond with empathy, a deep understanding or shared feeling, of what it must be like to need help and to have to rely on others to provide it. Other discussions raised the importance of HCAs understanding how difficult it is to lose one’s independence, a situation that arises both as a cause and as a result of an inpatient stay.

To help HCA trainees to ‘get into older patients’ shoes’, participants recommended and supported the idea of role-play, using methods to ‘mimic’ the experience of conditions associated with ageing and having older patients contribute to the training.

Sample

At each trust semistructured interviews were carried out with 10 HCAs (n = 30 in total) and eight other staff members (n = 24 total). The total number of semistructured interviews was 54. The majority of interviews were carried out between May and July 2014, with five HCA and two ‘OS’ interviews at two of the three trusts being held back until November 2014 to allow feedback on an early outline of the training intervention. The duration of the interviews ranged from 21 to 62 minutes, with a median length of 33 minutes.

Across the three study trusts the HCAs we spoke to had been working as HCAs for between 5 months and 15 years. Many of them had experience of paid and unpaid care work elsewhere prior to or concurrent with their work at the trust. ‘OS’ interviewed were matrons (n = 2), ward managers (n = 7), other nursing staff at bands 4–6 (n = 12), those with a role in HCA training (n = 2) and an allied health professional. Further details of the face-to-face interview samples are given below (Tables 8 and 9). In what follows we use ‘OS’ to denote all non-HCA interviews and interviewees and ‘staff’ to cover both groups of hospital staff.

Experiences of health-care assistant training with respect to relational care at the three trusts

During the period in which the interviews were undertaken there was a 2- or 3-week mandatory induction programme for all HCAs joining each of the three trusts, although in one trust the induction training was being rolled out to long-standing staff. This meant that there was great diversity in the training HCAs at the trusts had received. In our interview sample 11 of the 30 HCAs had joined prior to the current induction programme being in place and said that they had received little or no training prior to starting work as a HCA. On the other hand, besides mandatory induction and update training, some of the HCA interviewees had undertaken training prior to or concurrently with their employment at the trust: six had nationally recognised qualifications in health care (National Vocational Qualification level 1–3), six had taken short courses in end-of-life care for patients and families run by a local hospice and seven had attended trust-based additional training on caring for people with dementia beyond the mandatory minimum.

Examples of topics covered in the current induction programmes across the three trusts that may be regarded as ‘relational care training’ were respect, privacy and dignity, compassion, communication, culturally sensitive care, palliative and end-of-life care and care of the confused older person. This training, perhaps through necessity, is far from extensive. In the trusts with the 2-week induction programme each of these topics were sessions lasting 60–90 minutes. The trust with the longer induction programme (trust 2) had some longer, more interactive sessions, including input from patients and a dementia charity.

It appeared, from a number of sources, that the level and style of the induction training in two of the trusts was largely theoretical. One interviewee felt that the induction training on topics related to relational care at their trust was focused on trust policies and expectations, rather than the practice of delivering relational care in a real-life setting.

Each trust had policies on ward-based supervision and ongoing assessment of competencies during the HCAs’ first year. However, difficulties in finding appropriate staff willing to mentor, supervise or sign off competencies for HCAs were mentioned in interviews at two of the trusts. There was a discrepancy between the competencies HCAs should have achieved within a year of starting work and the training gaps identified by staff interviewees. This meant that ward-based support often fell short of that required to facilitate the trust’s intended programme of training.

What is relational care for older people in acute care settings?

As in focus group discussions, interviews with hospital staff showed that relational care could be discerned through how it made patients and carers feel and what staff did to engender those feelings. In staff accounts, four key themes emerged as central to good relational care for older people in acute care: making patients and visitors feel welcome, seeing patients as people, getting to know patients and verbal and non-verbal communication (including noticing and reading non-verbal clues).

Recommendations for health-care assistant training in relational care

There was widespread, although not universal, support for additional HCA training in relational care. A few registered nurse interviewees believed that (any) HCA-specific training beyond mandatory requirements would make HCAs feel more valued. The aspects of relational care that hospital staff saw as most important in caring for older people in acute settings were making patients and visitors feel welcome, seeing patients as people, getting to know patients and verbal and non-verbal communication (including the art of noticing and reading non-verbal clues). HCAs and OS also identified a need for training in dealing with their own emotions and those of patients and relatives, in dealing with bad news and in caring for patients with dementia.

In terms of delivery, almost without exception, interviewees stressed that HCA training should be ‘practical’. There were many dimensions of practicality, for example not being ‘too theoretical’, practising skills, being relevant to the realities of life on a ward, role-modelling good practice, role-play and bringing situations ‘to life’. There was a strong belief that training should be interactive.

Consistent with focus group participants, staff interviewees thought it important to help HCAs to ‘get into older patients’ shoes’, to gain insight into the vulnerability and fears involved in being a patient. The value of ‘getting into older people’s shoes’ was about raising awareness of the challenges an older person in hospital might face and understanding how older patients might feel. There was strong support for using actual patients’ experiences in the training and for using age simulation suits and other types of simulation:

[W]hen I was at school we had like someone come with like different goggles on to check like with like tunnel vision and like blind in one eye and then we had to use a wheelchair and we had to go in the wheelchair and get from out the wheelchair to the toilet. Then we had like our legs strapped together and we did all things like that so we knew how it would be. I think that’d be good in that sort of thing ‘cause sometimes when a patient takes like 20 minutes to walk to the toilet you’re like ‘urgh!’ But then actually doing it yourself you sort of understand why it’s taking them so long.

Ailsa, HCA, trust 2

One interviewee talked about using imaginative journeys to understand what it is like being an older patient:

[T]ry to make us see the patient’s point of view. Like, try to make us, you know, understand that – you know? Like exchange roles with the patient. Yeah, like, say, ‘Imagine yourself. You’re in there’. Kind of take them on a journey to imagine that it’s you [. . .] How would you want to be cared for? So I think just taking them on that journey, to make them [. . .] imagine and travel in it and then they’re able to get inspired and deliver better than – or the best that they can deliver.

Aliya, HCA, trust 3

One HCA stressed that the training should explicitly recognise the important role that HCAs play:

Me, if I will train [a HCA], I think I need to explain [to] her how important the role of the health-care assistants for the patients. How can you make them comfortable, ‘cause they are vulnerable to come here because of what happen to them. So how can you look after them. How important your role is to that patient.

Jade, HCA, trust 1

We were advised by interviewees not to rely too heavily on e-learning because of difficulties in accessing computers, possible technical difficulties with trust computers, lack of information technology (IT) skills and because staff are rarely released from duty in order to do e-learning. Releasing staff for any training is always contingent on staff numbers and the demands of the ward at the time of training. It was clear from answers to our questions about access to training that, in order to maximise the chances of staff being released to attend training or attending training on off-duty days, training should be in blocks of a whole day. Prebooking bank staff to cover the work of HCAs who are attending a day’s training was also recommended.

Findings from focus groups of older people and staff interviews

We placed great weight on the views of older people, HCAs and OS as to the content, structure and style of our training intervention. An overview of our findings from our focus groups and staff interviews is reported in Chapter 4, Summary. These findings are summarised in Table 10 to illustrate the similarities (and occasional differences) between those receiving and those providing relational care in hospital settings.

Reviews of research evidence

Two reviews of current evidence were key in shaping the HCA training intervention, Older People’s Shoes. The first was a review and synthesis of qualitative evidence of older people’s and relatives’ experience of acute hospital care. 13 The second was a realist synthesis and review of the evidence for workforce development interventions to improve the skills and care standards of support workers in older people’s health and social care services. 51 As with the design of phase 1 of our own study, by using these two complementary reviews we were able to use evidence from two perspectives: (1) older patients and their relatives and (2) members of the clinical support workforce.

Bridges et al. 13 synthesised findings from 42 original studies and a systematic review that examined older people’s experience of acute care. The authors argued that, while technical aspects of care are often taken for granted, there are three aspects of relational care that, when adequately addressed, are associated with positive experiences of acute hospital care. The first, ‘connect with me’, relates to the relationship between the person providing care and the older patient and his or her relatives. A lack of a sense of reciprocity in the relationship makes people feel anxious and burdensome. The second, ‘see who I am’, illustrates the importance of care staff recognising and helping patients maintain their identity while an inpatient. The third, ‘include me’, refers to the importance of shared decision-making and the involvement of older people and their relatives in treatment and care. The desire for participation in decisions may vary, but the need to have an understanding of what is happening or planned is widely held but often not met, particularly for older patients. The authors conclude that the actions of individual care staff and the relationships they build with older people and their relatives have the potential to make a powerful difference.

The OPSWISE study51 was commissioned as part of the same funding call as the present study. Given the complementarity of the two studies, the study teams maintained close contact, including attending study steering meetings. Using realist principles, Rycroft-Malone et al. 51 identified a number of context–mechanism–outcome configurations (what works, in what context and in what way) from published evidence and stakeholder interviews. These eight configurations or explanations as to what elements of workforce development interventions work for the older person’s support workforce are (1) keep interventions close or ‘real’ to the work of the support worker, (2) pay attention to individual support worker’s starting point and role expectations, (3) tap into learners’ motivations, (4) develop interventions in the context of the organisation’s wider goals, (5) co-design training interventions using the right mix of people to reflect the complexity of the workforce, (6) get the right mix of people to deliver interventions to promote shared learning and improve cohesion, (7) take a planned approach that draws on theory and (8) ensure that interventions are comprehensive and multilayered to embed and spread impacts across organisations. The implications of these configurations are summarised in Box 1.

Initiatives and tools to improve relational care

Although rarely evaluated, there have been a number of initiatives that have sought, explicitly or implicitly, to enhance the quality of relational care. Few, if any, are targeted at the care of older people exclusively, and those that were tended to specifically focus on the care of older people with dementia. We investigated the content of current initiatives that we were aware of in order to learn from existing tools, avoid overlap and be aware of where our own HCA training intervention was situated in the broader context of related interventions and initiatives.

The Setting, Ask, Gather, Empathy and Talk, Help, You, Me, End (SAGE & THYME) model was developed by clinical staff and a patient at University Hospital of South Manchester NHS Foundation Trust in 2006 to meet the 2004 National Institute for Health and Care Excellence’s guidance on ‘improving supportive and palliative care for adults with cancer’. 68 It was designed to be relevant to all grades of staff and to improve skills in how to listen and respond to patients and carers who are distressed or concerned. The title is a mnemonic guiding a health worker into and out of a conversation with someone in distress. Its approach is to encourage health workers to offer psychological support by holding back advice and prompt patients to consider their own solutions. It is taught using a foundation-level workshop by three facilitators using a mix of small-group work, lectures and rehearsals. It includes ‘live pause’ technique to facilitate direct feedback and discussion. There is evidence that these workshops have a positive effect on self-confidence, self-perception of competence and willingness to explore the emotional concerns of patients. 69

Barbara’s Story is part of a dementia awareness training initiative for hospital and community health staff developed by Guy’s and St Thomas’ NHS Foundation Trust. The initial Barbara’s Story film shows the experience of an older woman (Barbara) through her own eyes as she attends a hospital appointment and is admitted for investigations. The film was shown regularly from September 2012 to April 2013 and attendance was mandatory for all trust staff, with a total of 11,054 clinical and non-clinical staff attending during that time. Barbara’s Story was also embedded into the corporate induction programme for new trust staff. Subsequently, a second series of films was developed, which show Barbara’s health deteriorating and her receiving care in hospital and community settings. The films were shown from September 2013 to March 2014, with one new episode available each month. Staff are expected to gain an understanding of issues faced by patients with dementia in order to recognise the problems they face. Staff have reported that Barbara’s Story engaged them emotionally and prompted empathetic responses. 70 Reported changes included giving more time to patients, improved communication, giving more information and assisting patients who are looking lost or confused. 70

Active Caring for Everyone (ACE) is a programme developed at Worcestershire Acute NHS hospitals Trust designed to improve day-to-day interactions between staff and patients. As staff attitude accounts for a high proportion of patient or relative complaints in any NHS trust, the initiative includes a card carried by all staff and ‘shown’ if it is believed the care being delivered is falling below a certain standard and accompanied by the phrase ‘you didn’t play your ACE card’. It aims to show that each point of contact, no matter how small, can result in a positive patient experience, increase staff understanding of their own role and responsibilities in delivering high levels of customer service and recognise good customer service and actively seek ways to solve problems and handle concerns. There has been no external evaluation of this programme as far as we are aware.

A fairly common approach to increase empathy among health staff is to simulate the experience of disability and/or ageing. For this, some trusts have used equipment designed for this task or used more impromptu methods. Age simulation suits aim to get the wearer to experience the disabling effects of ageing on simple tasks, movement, orientation and energy levels. Training helps to highlight the importance of communication and limitation of the patient in the hospital environment (and others). An evaluation of a 2-day simulation training programme that included the use of an ageing suit found a significant increase in confidence across all staff grades, and staff reported spending more time getting to know patients. 71 Similar results of a rise in confidence and positive behaviour change are reported in an evaluation of a simulation-based educational programme to equip HCAs with non-technical skills to undertake their role safely and effectively. 72

Life-story work

Life-story work is considered to be an approach that can be used to transform how care staff think and feel about older people. 73 It is thought to be beneficial to older people, and the use of life-story instruments that capture important personal information about patients is increasingly encouraged by care organisations and patient groups. Older stakeholders and members of the steering committee also considered evaluation of routinely used life-story instruments important. Life-story instruments are known to have both advantages and disadvantages22 and these are summarised in Table 11. One of the disadvantages from our perspective is that they are often assumed to be relevant only for patients with a diagnosis of dementia. To help HCAs ‘see the person inside the patient’, we planned to develop and include a modified form of life-story work into our training intervention.

A number of life-story instruments in common use were identified from internet searches and local knowledge. These were the Alzheimer’s Society’s This is Me,74 the Getting to Know Me book, based on original work by NHS Lanarkshire,75 the Getting to Know Me form from University Hospitals Coventry and Warwickshire NHS Trust76 and Worcestershire Acute Hospitals NHS Trust’s About Me – Lifestyle and Capabilities Booklet. 77 The areas covered in varying levels of detail include care preferences, physical ability to perform daily activities, communication impairment, mobility and relatives’ involvement in care. Something we were particularly interested in, because of our desire for our training to assist HCAs to engage with older people as people first and patients second, was the inclusion of items about personal history and particular preferences beyond those related to physical care. Items such as ‘my life so far’, past employment, live events, interests and hobbies were present in some of the instruments but, perhaps understandably, tended to be given lower priority than information required to perform physical care tasks.

Learning from customer care

Four large commercial/retail organisations agreed to speak with us prior to the development of the HCA training intervention. Our interest was in their approach to improving customer care provided by their staff, including training materials and modes of training delivery. Trainers from each of these organisations met us to provide insights that we expected would inform the content and delivery of our HCA training intervention.

The retail organisations involved were Boots Opticians, a company with 600 outlets in the UK, serving an older customer base; Aldi, a multinational company with approximately 400 supermarkets in the UK; Domestic & General, the leading UK domestic appliance care company providing protection plans for household appliances through telesales; and B&Q, a do-it-yourself store which has been recognised for its policy of actively promoting the employment of older staff.

Discussions with staff who had responsibility for customer service training in these organisations covered a range of topics, including any learning theories and principles used to underpin their training. Although commercial sensitivities prevent us from providing extensive detail and ascribing this to individual organisations, we learned about the use of neurolinguistic principles in training and the ways in which acronyms and mnemonics are used to reinforce key messages and encourage consistency in good customer service. Particularly important customer care learning points that were transferable to the health-care setting included how to make each contact count, the importance of active listening and the conscious actions of smiling and greeting customers. A list of key learning points that were drawn on for the HCA training intervention is given in Box 2.

Panel of expert witnesses

Core members of the study team identified a number of national and international experts in relevant fields that would help us build the HCA training intervention. We invited these experts to provide ‘evidence’ by being informally interviewed by members of the study team that met on 8 September 2014. The members of the study team plus an experienced HCA from one of our partner NHS trusts formed a panel and each of the expert ‘witnesses’ (Table 12) were interviewed by the panel either by teleconference or skype™ (Microsoft Corporation, Redmond, WA, USA). All our experts were briefed prior to the panel about the study, the composition of the panel and our areas of interest. The specific areas of focus were policy, context, transmission, experience, evidence and content. The panel members used an informal checklist of areas to explore with each expert witness (Box 3).

The key messages distilled by panel members from evidence provided by these expert witnesses were:

1. Values-based training is important but there are difficulties in providing training for established/existing HCAs as many trusts restrict training to new starters.

2. Be aware of barriers to training in the workplace, such as poor access to technology, lack of time, limited IT skills, attitudes of managers (‘HCAs do not need training’), the fact that HCAs themselves do not feel that they need training and the negative impact of the label ‘untrained workforce’.

3. HCAs need to be enabled to be self-aware, emotionally resilient and clear about their own self-compassion, but bear in mind that this may need to be handled with sensitivity and require adequate training of the trainer.

4. It is desirable to integrate ‘life-story work’ into everyday tasks and to encourage HCAs to ‘be a detective’ to find out things that highlight the individuality of each patient.

5. Recognise the lives and invaluable contribution of the HCAs themselves; adopt an assets-based approach (what strengths do HCAs bring?) rather than a deficit model (what skills are lacking?).

6. Small improvements can make a big difference.

7. Situate relational care within practical clinical care and, if using customer care practices, it is important to ensure that these are not superficial.

8. Be explicit and realistic about expectations and recognise the power that HCAs have ‘in the moment’, if not organisationally. HCAs are influential in patients’ and relatives’ experiences of care, yet this is seldom recognised and their contribution may remain invisible.

9. In terms of training design, learning should be participatory and interactive. Staff appreciate having a tailored training package but supervisor buy-in and post-training support are crucial.

10. Links to practice and use of real-life examples are important for authenticity.

Intervention development workshops

The development of the HCA training intervention was guided by members of the interdisciplinary study team comprising nurses, methodologists, education and social science researchers, together with a HCA and PPI representatives. Members of this group were based across the three centres and worked remotely but came together at eight intensive workshops, some residential, spread across the intervention development period of the study (Table 13).

The purpose (and result) of these workshops was to refine the HCA training intervention and to set milestones in order to maintain progress as we worked through each stage of intervention development. The process drew on the group members’ familiarity with the data sources and other inputs, as well as with the experience that they collectively brought. HCAs and PPI members provided invaluable input ensuring the materials being developed were relevant to stakeholders. An example of how decision aids were used to process inputs from experts into the developing structure of the training intervention is provided below (Figure 4).

FIGURE 4.

An example of one of the decision tools used to inform structure and content with inputs.

Workshops were used to decide which elements of the inputs should be included in the HCA training intervention. All inputs were considered but not all were included. Some important points were not included in the training or were included minimally because they were considered to require specialist trainers (such as doing stimulating activities with patients) or dealt with a particular specialist need. For example, HCAs were very keen to acquire training that would help them to manage better dementia care but our remit and focus was on the relational care of older people. A small section on relational care and dementia was included, but attention remained on the needs of older people more generally.

Theoretical teaching and learning frameworks

The training intervention design process used a step-by-step approach to curriculum design (identifying what is to be learned), guided by instructional design frameworks (how it is to be learned). 78 The importance of using theoretical principles of instructional (pedagogical) design to develop educational interventions in health-care contexts is often overlooked, but it is essential to ensure that training builds on existing knowledge and values, harnesses intrinsic motivation and actively engages learners.

The overarching theoretical basis for the design of the training package is derived from Carver’s framework79 for applying the principles of experiential education. This framework provided elements appropriate for the practical nature of a HCA’s role that could be applied throughout the development of the training intervention. Carver proposes four key elements to experiential education: (1) authenticity (activities being relevant to the participant’s role), (2) active learning (engaged and active learning activities), (3) draws on experiences (what happened to them, how it felt, how they reacted, what resulted, what they observed) and (4) provides mechanisms for connecting experience to future opportunity (learners reflect on their participation in activities or on their potential roles as community members to make experiences relevant to their future endeavours). Carver’s four key elements to experiential education and examples of how these were applied in developing the activities within the HCA training intervention are shown in Table 14.

When developing training interventions it is important to provide a pedagogical framework to ensure the materials and activities are structured for learning to take place optimally. Gagné’s model considers three important domains that impact on learning (affective, cognitive and psychomotor),80 making it particularly suited to a values-based training approach. Gagné’s original model has nine steps: (1) gain attention, (2) identify objective, (3) recall prior learning, (4) present stimulus, (5) guide learning, (6) elicit performance, (7) provide feedback, (8) assess performance and (9) enhance retention/transfer. The HCA training intervention was designed using an adapted version of this model to structure individual learning activities within each unit. A simpler model with a smaller number of steps was more appropriate for a short training programme. The following five steps were included in our pedagogical framework:

1. create learning goal

2. provide a trigger

3. present content or learning material

4. design some guided practice, simulation or reflection

5. reinforce key messages.

The intervention development group used design templates scaffolded by these five steps to ensure that each session contained the pedagogical elements as the training content and learning activities were developed. An example of how this framework was applied is illustrated in the following learning episode. One of the three units in our HCA training intervention was ‘Learning from customer care’ (details of all units are given in Intervention development: product). In this unit the (first) learning goal was to understand what is meant by ‘customer care’. The trigger activity was to ask the HCA learners ‘Think about the customer care you have experienced recently? What made this experience a good or bad one?’. The learning content presented by the trainer then covered the concepts of active listening, every interaction matters and the art of noticing. These were some of the learning points drawn from our retail partners. This was followed by asking HCAs to reflect on how these concepts might apply to their own practice. This learning activity is concluded by reinforcing the key message ‘good customer care can be provided by noticing customer cues, by listening to needs and by providing a consistent standard of care’.

Content development, review, production and editing

Having identified our theoretical and pedagogical frameworks, these were applied to each unit of the training and the units were developed using a learning design template. Learning objectives for each of the three units were generated. Each unit underwent a critical, in-depth, iterative quality review by the intervention development group and the steering committee, which included HCAs, PPI representatives and trust representatives from the three centres. During this phase, media assets such as still images and film clips were sourced, reviewed for relevance and appropriateness and formatted. Permissions were obtained to use any clips from other sources.

The course book for participants and a guide for trainers that elaborated on the intended delivery process for each stage were written and reviewed in detail by the intervention development group. Professional designers were used to produce the two manuals. The equipment required to support the training was purchased. This included age simulation suits (GERontologic Test suits, Niederstotzingen, Germany), pyjamas and utensils for role-play exercises, together with copyright licences. Slide presentations and a bespoke online learning site were also developed to support the trainers and HCAs who participated in the training.

Structure and mode of delivery of Older People’s Shoes

Older People’s Shoes training comprised three units: (1) getting into older people’s shoes, (2) getting to know older people and (3) learning from customer care. Each unit was divided into two sessions, one per day, so learning on the first day was consolidated and built upon on the second day, approximately 1 week later. At the end of each session on day 1 HCA learners were asked to undertake brief individual work-based exercises prior to day 2. On day 2, each unit began with a brief discussion of these exercises so that learners could share from others’ experiences and learning. The structure and content of the 2 days are presented in Tables 15 and 16.

Each training day began at 09.00 with tea or coffee and introductions. Training was scheduled to finish at 16.20, with a 45-minute lunch break. The trainers were given guidance about approximately how long to spend on each section, and each session was allocated between 90 and 160 minutes of training time.

Training took place close to HCAs’ work places or in familiar training rooms. Rooms were laid out informally to foster a relaxed atmosphere. Rooms required computer and projection facilities and internet access; however, in case of technical problems, online materials were also available on a USB memory stick. It was important to look after the HCAs themselves during the training in case any aspect of the training caused any distress. This was done in a number of ways. The sessions were relaxed and interactive, HCAs were given time for reflection and could raise any concerns they had either within the group or individually with the trainer. The training was asset driven, affirming the importance of the HCA role in patient care, and the HCAs were shown how to access local support networks for any issues that may arise as a result of the training intervention.

Content of Older People’s Shoes

Training materials

Successful delivery of the course to HCAs requires a complete training package comprising the trainee course book, an online support tool, the trainer guide and a train-the-trainer course. These four components of the training package are described below and are available on request from the principal investigator.

Training the trainer

Trainers were practice development nurses or nurses with responsibility for training HCAs and all were experienced educators involved in HCA training in their own trusts. Members of the research team ran ‘Train-the-trainer’ days at each centre. These training days consisted of a 1-day face-to-face session, with additional time on another day for the trainer to raise further questions about any aspect of the training. It was essential to spend time with the trainers explaining the context of the training, the findings from phase 1 of the CHAT study as well as going through each of the units of the training. A guide for researchers ‘training the trainer’ was provided (see Appendix 11).

Purpose

The main purpose of the feasibility cluster randomised trial was to assess the feasibility of a definitive trial to compare the newly developed HCA training package Older People’s Shoes in relational care with current training in improving the care of older patients in acute NHS settings. To inform the feasibility of a definitive trial83 (and then, if feasible, the design of such a trial), the following issues were explored:

1. the acceptability of the intervention to trainers and HCA trainees

2. the willingness of ward managers, HCAs and older patients to participate in a cluster randomised controlled trial

3. the willingness of ward managers for wards to be randomly allocated

4. the level of non-response and item non-response to outcomes at the level of ward, HCA and patient

5. the acceptability of outcome measures to participants

6. the ability to monitor levels of resource use and quality-of-life data

7. the variability within and between ward, HCA and patient

8. the appropriateness of ward as the unit of randomisation.

Design

A pilot cluster randomised controlled trial was conducted to compare Older People’s Shoes with ‘HCA TAU’. Clusters were wards within three acute NHS hospital trusts in England with outcomes observed at the level of the ward, HCA and patient (Figure 5). A brief description of these trusts was provided earlier (see Chapter 4, Setting and eligibility).

FIGURE 5.

The CHAT study pilot cluster randomised trial design and target recruitment. AGED, Age Group Evaluation and Description; AWES, Assessment of Work Environment Scale; EQ-5D-5L, EuroQol-5 Dimensions, five-level version; PEECH, Patient Evaluation of Emotional Care during Hospitalisation; QUIS, Quality of Interaction Scale; TEQ, Toronto Empathy Questionnaire.

Eligibility

Recruitment

Baseline measures

Allocation of interventions

Stratified by NHS hospital trust, wards were randomly allocated by the Norwich Clinical Trials Unit. Each ward had an equal chance of receiving either Older People’s Shoes training for HCAs or TAU. Random allocation was generated via computer-written code using block sizes of four. To conceal allocation from those responsible for recruitment, randomisation took place immediately after baseline measures were completed and 4 weeks ahead of the start of the intervention (set-up period) to allow appropriate arrangements, including HCA staffing cover to be arranged.

Interventions

Outcomes and other measures

Sample size

As the aim was to test feasibility and the study was a pilot cluster randomised controlled trial, it was not powered to determine superiority of HCA Older People’s Shoes training compared with HCA TAU.

Data management

A data management plan was developed by the Norwich Clinical Trials Unit. Data were entered into a central database, password protected and accessible only to the principal investigator, members of the research staff and the database manager. Data entry was done via web pages created using Microsoft.NET technology (Microsoft Corporation, Redmond, WA, USA). All internet traffic was encrypted using standard secure sockets layer methodology. The data entry system validated data on entry to ensure that they were of the expected type (e.g. integers, dates) and within the expected range of values. The database was linked to an audit tool in which all data additions, modifications and deletions were recorded with date/time and user identifier.

Data analysis

Ethical considerations and ethics approvals

For observational sessions, we attempted to be as unobtrusive as possible. No observation was made of interactions that occurred behind curtains. We alerted staff to our presence on the ward and explained the nature of the study. We spoke to individual patients in the ward bays that were being observed. If any patient or member of staff asked not to be observed we made no record of any interaction that included them. For HCAs and patients we went to great lengths to ensure that they were fully aware of what study participation would involve. We did not have access to named patient data. These were collected (and anonymised) by trust-employed research nurses.

A favourable ethical opinion for this study [CHAT (feasibility randomised controlled trial) Protocol v2 9.2.2015] was granted by Cambridge South Research Ethics Committee on 13 February 2015 (application number 15/EE/0025, Co-ordinated System for gaining Permission reference number 162616). The study was adopted onto the UK Clinical Research Network portfolio (study identifier UKCRN18280).

Trial participation

Wards and health-care assistants

A total of 12 wards were enrolled into the study following agreement by ward managers. During the baseline period, 91 of the 96 planned observation sessions were successfully undertaken. There was a shortfall of five observation sessions as two wards were closed to visitors for a short time during the baseline period because of the presence of norovirus. The flow of HCA participants is reported in Figure 6. Of the 150 HCAs screened for eligibility, 134 fulfilled eligibility criteria and 112 agreed to take part in the study. Of the HCAs who consented to take part, 59 were based on wards randomly allocated to receive Older People’s Shoes training and 53 were based on wards allocated to receive TAU. Of the 59 who were allocated to receive Older People’s Shoes training, 45 successfully received the intervention. Baseline, first and second follow-up questionnaires were returned by, respectively, 40, 28 and 22 of those in the Older People’s Shoes arm and 32, 24 and 18 in the TAU arm. HCA recruitment took place between March and June 2015, with follow-up questionnaires sent out between June 2015 and October 2015.

FIGURE 6.

Flow of HCA participants through the feasibility trial.

Patients

Patients were recruited to the trial during a 4-week period pre randomisation and during weeks 9 to 12 post randomisation. The flow of both groups of patients is reported in Figure 7. During the baseline period, 97 of 129 eligible patients agreed to take part in the study. Patients were recruited to the baseline period between March and July 2015. As some patients who were eligible at the point of consent subsequently became ineligible (e.g. because they were transferred to another ward, were not discharged or withdrew for health reasons), data from 40 Older People’s Shoes patients and 52 TAU patients were available at baseline. Of these patients, 23 in the Older People’s Shoes arm and and 27 in the TAU study arm returned a questionnaire.

FIGURE 7.

Flow of patients through the feasibility trial. a, Randomisation occurred after patients consented but it is displayed in this way to compare differences at baseline.

During the follow-up period, 74 of 114 eligible patients agreed to take part in the study, of whom 67 were eligible for inclusion in analysis (26 patients from Older People’s Shoes wards and 41 patients from TAU wards). Of these, 16 patients in the Older People’s Shoes arm and 22 patients in the TAU arm returned questionnaires. Patients were recruited for the follow-up period between June 2015 and October 2015.

Groups at baseline

Inter-rater reliability and Quality of Interaction Scale

Eight paired observation sessions were undertaken, within which a total of 40 interactions were observed independently by two observers. These took place across all centres, with observer pairs the same within, but not between, each centre. Weighted kappa statistics are reported in Table 20. Although the overall consistency of 0.61 [95% confidence interval (CI) 0.32 to 0.89] suggests moderate to substantial agreement overall,101 there was some evidence of heterogeneity between pairs (p = 0.057). In 34 of the interactions observed by both observers, there was agreement (Table 21). Where there was far less agreement was in the number of interactions recorded by each observer within a session. Of the eight sessions in which at least one interaction was observed, there was no difference in the number of interactions recorded by each observer in four sessions, but in one session 12 interactions were recorded by one observer compared with 29 by the paired observer (Figure 8).

FIGURE 8.

Histogram of the difference in the number of interactions observed per session between raters.

Outcomes and sensitivity analysis

Health-care assistant–patient contact time

Interactions took place within 179 of the 187 observation sessions conducted during baseline and follow-up periods. The end time of two of these sessions was not recorded. For one of these the end time for the last recorded interaction was after 50 minutes (the guideline time for each observation session), so the end time of this interaction was assumed to also be the end time for the session. For the other session the last observation recorded ended after 18 minutes and it was thus difficult to estimate the end time; therefore, we assumed that this session and another eight sessions (four in each arm at follow-up) in which no interactions took place (and, therefore, no start of end times were automatically recorded) lasted 50 minutes. Based on these data, the mean session length for the 187 observation sessions that took place was 50.92 minutes (Table 30). A total of 1441 timed interactions took place in these sessions. The mean interaction time was 2.85 minutes, which when costed at £36.20 (the assumed cost per hour of HCA patient contact time; see Table 28) gave a mean cost of £1.72 per interaction. HCA staff undertook the vast majority of interactions on their own (see Table 30). Nurses were the most common type of other staff involved (in 103 of the 1441 interactions), with other health professionals being the next most common (19 interactions). OS time in the interactions was not costed on the basis that it was relatively low. It should also be acknowledged that, although we do not seek to undertake formal comparisons between groups, there seems to be no suggestion that interaction resource use will be reduced after Older People’s Shoes training. Both the mean length of interactions and the mean number of OS involved were higher in the Older People’s Shoes arm post training.

Consultation interaction times were also assessed by asking HCAs whether or not the average contact time with an older patient had changed since the start of the study as part of the follow-up questionnaire administered at 8 and 12 weeks post randomisation. This information was provided by fewer than half of the HCAs who were involved at the start of the study. Among those who did respond (Tables 31 and 32), there was certainly a trend towards an increased contact time rather than a decrease (of the 22 responses at the second follow-up only nine had not previously responded at the first follow-up). At 8 (and 12) weeks, the estimated mean change in contact time was 2.46 (2.89) minutes in the Older People’s Shoes arm, compared with 0.80 (1.53) minutes for TAU, a difference of 1.66 (1.36) minutes. After assigning the previously estimated unit cost of HCA patient contact time (£36.20 per hours) to these times, the implied extra cost per contact would be £1.00 per contact at 8 weeks and £0.82 at 12 weeks. Assuming that 7.7 contacts take place per hour (based on the mean number of interactions observed per session) this would equate to a cost equivalent to £6.32 per hour. We consider this to be a conservative estimate as individual patient bays were observed and there were likely to have been other unrecorded interactions occurring outside the observed bay. Assuming that HCAs work 1575 hours per year (as reported for a day ward nurse102), this would equate to an annual cost of £9949.98 per HCA. This is an estimate of the cost associated with the time that would need to be sacrificed from other activities if the HCA were to increase all patient contacts (in 1 year) by 1.36 minutes. On the assumption that each HCA might see 2622 patients per annum, this would equate to a per-patient cost of £3.79.

Quality of life using the EuroQol-5 Dimensions

The EQ-5D was sent to 159 of the 171 consented participants. Those who withdrew for health reasons (n = 3) or who became ineligible as a result of a change of ward (n = 8) or because they were not discharged (n = 1) were not sent the questionnaire. The details of the number of questionnaires returned and completed are given in Table 33. Overall, the response rate was just over 50% for both the EQ-VAS and EQ-5D-5L index value. For all participants, the resulting mean scores were 64.1 for the EQ-VAS and 0.606 for the EQ-5D-5L index value.

For those aged > 75 years, the population norms for the EQ-VAS are 72.90 for men and 74.07 for women, compared with 0.75 and 0.71 for the index value (based on the previous three-level version). As such, it can be seen that the participants in this study tended to have worse levels of health, according to the EQ-5D, than the population average for those with similar age or sex.

Hospital stay

Length of hospital and study ward stay were extracted for 97.5% of participants in the study (Table 34). Over all participants, the mean length of hospital stay was 7.3 days, compared with 5.9 days for the study ward. When a cost per bed-day of £275.05103 was assigned to each day in hospital/study ward, then the mean hospital stay cost was estimated to be £2017.62, compared with a study ward cost of £1635.82.

Purpose

The aims of the process evaluation were to (1) understand the processes involved in implementing the HCA training intervention and the trial, in order to better assess the feasibility of a definitive randomised controlled trial, (2) learn what adjustments could be made (in either the intervention or the trial) to improve them and (3) gain some understanding of the potential impact of the intervention.

A number of different methods were used in the process evaluation. This enabled us to capture different types of data on different aspects of the trial and intervention and (when appropriate) to allow methodological triangulation. 104 The four research methods used were:

1. observations of all training sessions (‘course observations’) to capture ‘in the round’ process data for each training session and variation in delivery between centres

2. course evaluation forms from HCA learners attending Older People’s Shoes training to capture immediate impressions of the intervention and its predicted impact on practice

3. interviews with trainers to investigate trainers’ perceptions of Older People’s Shoes training, and their experience of delivering it

4. interviews with a sample of HCA learners to investigate perceptions of the training intervention in greater detail than was possible from evaluation forms, including reports of its impact on practice; and to explore experiences of participation in the trial.

Observations of training

All training intervention sessions were observed by at least one member of the local research team using a common template (excerpt in Appendix 19) that recorded deviations from the trainer guide, evidence of learning, the strength of the relational care focus, timing and practical issues arising. In vivo quotes were also captured. One member of the research team additionally observed the training intervention being delivered at each of the three centres to record differences in delivery between trainers and centres.

Course observation notes were analysed by researchers at each centre to draw out key messages for refining the intervention and to identify deviations from the course as prescribed. These findings were then collated and analysed thematically. Findings from observations were compared with other data in order to identify areas of congruence or divergence.

Evaluation forms from health-care assistants participating in the training intervention

At the end of each training day time was built in for all HCA learners to complete an anonymous evaluation form (see Appendix 20), which used closed and open-ended questions to ask for their views on the training as a whole and the different activities within it, the resources, perceived impacts and any expected changes in their practice. Descriptive statistics were used to analyse responses to closed questions. The responses to open-ended questions were analysed thematically.

Interviews with trainers who delivered Older People’s Shoes training

All trainers who delivered Older People’s Shoes training were provided with a participant information sheet (see Appendix 21) and invited to take part in an interview about their experience. Verbal consent to take part in interviews was obtained after potential interviewees had had the opportunity to read the participant information sheet and a time and date for the interview was then arranged as soon as practicable after all training had been delivered. Interviews took place during interviewees’ work time, in a private room on the ward or elsewhere on-site. Written consent was taken immediately prior to the interview. A topic guide was used (see Appendix 22), with interviews designed to take 30–45 minutes. These semistructured interviews explored their views of the training and support they received to deliver the intervention, the content of the training, any suggestions for improvement and the relevance and perceived impact of the intervention for HCAs. The interviews were audio-recorded, with the interviewee’s permission, and the audio files were transcribed verbatim. The transcripts were then anonymised.

Interviews with health-care assistants receiving Older People’s Shoes training

At the end of the follow-up period (to avoid differential treatment of HCAs in the two arms of the trial) a subsample of 12 HCAs who had participated in the training were provided with a participant information sheet (see Appendix 23) and invited to interview. This number was expected to represent around one-third of trainees. Of those HCA learners who gave initial consent to interview, purposive sampling was used to maximise variation of interviewees in terms of sex, length of experience as a HCA at the trust and (drawing on the course observations) observed levels of engagement in the training.

Procedures for consent, recording, transcription and anonymisation were the same as for trainers. A topic guide was used (see Appendix 24), with interviews designed to take 30–45 minutes. These semistructured interviews explored HCAs’ expectations of the training intervention, their experience of the intervention (including any suggestions for improvement), any impacts on their practice and their experience of participating in the trial.

Framework analysis was used in NVivo to manage all interview transcripts and analyse them thematically. Framework analysis is particularly useful for applied research designed to meet specific information needs while remaining true to the accounts of the interviewees. 105

Sample

In total, there were 25 sets of structured observations of the 2-day training intervention, which ran three times in centres 1 and 2 and twice in centre 3. All 16 days of training were observed by at least one researcher, with a second researcher observing one delivery of day 1 and one of day 2 at each centre. In addition, the researcher from centre 3 observed the delivery of day 1 training at centre 1 and of both training days at centre 2.

Across the three centres, 40 HCAs attended day 1 of the intervention and 41 attended day 2. There was a 100% response for evaluation forms for each of these days. All six trainers agreed to be interviewed. The interviews lasted an average of 35 minutes.

Twelve interviews were carried out with HCA learners. We had planned to recruit equal numbers across centres. However, because of delays in training delivery at one centre, HCA learners were drawn from two centres only. The interviews lasted an average of 35 minutes. Two HCA learner interviewees were men and 10 were women. Their length of experience as HCAs working at the trusts ranged from 5 months to 12 years. The median time in post was 22 months.

As different research tools (course observations, evaluation forms, interviews with trainers and interviews with HCA trainees) were used to capture and triangulate data in the process evaluation, there is necessarily some overlap in data from different sources. Therefore, we present the findings from the process evaluation data thematically, drawing on all relevant sources within each theme, rather than presenting findings from each data source in turn. This reflects the dual nature of the overall study, which comprised intervention development as well as a feasibility trial. The findings will be discussed under (1) findings related to the HCA training intervention and (2) findings related to the trial.

Findings related to the intervention

Findings related to trial participation

Training needs of health-care assistants for delivering relational care to older people

We aimed to understand the values-based training needs of HCAs in maintaining the dignity of, and affording respectful care to, older patients in acute NHS settings (study aim 1). This aim was addressed in the first phase of the study through (1) a telephone survey of NHS hospital trusts in England, (2) focus groups with older people and (3) semistructured interviews with HCAs and staff who work with HCAs.

To understand training needs, we needed to understand what training was currently given to HCAs working in acute hospitals. Key contacts were identified at each of the 113 trusts who took part in the national telephone survey and provided details of HCA training within their trust. Our findings suggest that induction training varies widely in terms of length, structure and content. Variability across and between trusts in how HCA training needs are met (or not) is a concern for policy-makers2,38,39,106 and one that the new Care Certificate has been designed to address. 57 Although the introduction of the Care Certificate took place midway through the time in which the survey was conducted, at this early stage there was no evidence from our survey that variability was diminishing.

We did not restrict the survey to questions about training that HCAs received at induction but aimed to glean a picture of HCA training for those in post as well as those newly appointed. We found that the greatest emphasis is placed on induction training, and the ongoing training of HCAs is typically devolved to ward managers through mentorship. The design of our survey did not enable us to pursue ward-level training beyond that known to the key contact at the participating trust. It became apparent that survey respondents who were employed at trust level were often unclear of the detail of such training. We cannot say whether or not postinduction training is adequate, but without trust-level oversight it is likely to be highly dependent on the motivation of individual ward managers. For survey respondents, the most frequently reported challenge was getting HCAs released from ward duties to undertake training. Others have noted that HCAs have been adversely affected in the past by a workplace culture that does not afford a particularly high priority to HCA training. 40

The prevalence of dementia among hospital patients aged ≥ 70 years is estimated to be 42%. 107 The nature of HCA work means that they will play a key part in managing challenging behaviour of older patients with dementia. Therefore, it was unsurprising, and welcome, that almost all trust respondents reported that HCA training included the care of patients with dementia. Aspects of training that could be considered to fall within our broad definition of ‘relational care’ were reported only by one-third of trust respondents. One-third of the trust contacts who participated in the survey stated that their HCA training did not distinguish between the needs of older people and patients of all ages. Although it is important that training should not stereotype older patients, there is a risk that this approach ignores the real challenges faced by older people when they are admitted to hospital. Calasanti108 refers to this as ‘age-blindness’, whereby age-associated bodily change and the marginal status of older people are overlooked.

Although older people’s needs are addressed in HCA training, we found no evidence from the telephone survey that relational care is considered a priority in such training. This was in contrast to our findings from the focus groups with older people with recent experience of being a patient in hospital. For older people and their relatives, their experience of hospital care often hinged on the quality of the relationships they had with staff who cared for them. Making connections at a personal level with staff transforms care for older people, allowing them to feel less like ‘patients’. Although this finding is not new,6,13,15,19 findings from our telephone survey would suggest that it does not currently explicitly inform the way HCA staff are trained to perform their role.

In interviews with HCAs and OS, participants were aware of the effect that poor communication can have on relationships with patients, although they were not always aware of the reticence some older people felt in asking for help. Staff identified a sense of conflict between the need for efficiency and the importance of providing good relational care. This is a finding supported in other literature,19,109 but evidence from our focus groups would suggest that older people acknowledge, and are sympathetic to, the pressure that care staff are under. HCAs were keen to learn how to have difficult conversations with patients and relatives and how to avoid projecting work-related stress. In our study older people and their relatives questioned whether or not staff were always aware of the sense of powerlessness older patients feel when they are in hospital. Pressures of time mean that HCA work is responsive to expressed need, but this is at the expense of attempts to meet needs, which older people are hesitant to voice.

Development of a health-care assistant training intervention

We aimed to develop a values-based training intervention for HCAs designed to address the needs of older patients for high-quality relational care (study aim 2). On the strength of our analysis of (1) HCA and OS interviews, (2) focus groups with older people and their relatives and (3) the telephone survey of hospital trusts in England, we established certain principles to guide the development of a relational care training intervention for HCAs:

1. Training should be connected to everyday practice to ensure credibility and allow learners to draw on experience and implement new skills.

2. Training should be interactive to retain learner interest, reflect the way HCAs work in teams and acknowledge the highly practical nature of their role.

3. Training should be assets based, whereby the starting point for learners is building on existing strengths rather than addressing weaknesses or deficits.

4. At the heart of training should be patient experiences conveyed using the words of patients wherever possible.

5. Web-based learning should be used as an adjunct to the training and not its primary mode of delivery.

6. Training should use novel experiential learning techniques to facilitate learners’ ability to consider the perspectives of older people in their care.

These principles were used to develop Older People’s Shoes, an intervention designed to meet the learning needs of HCAs working with older people. These principles were consistent with, and enhanced by, findings from the OPSWISE project,51 a realist synthesis of evidence to inform clinical support workforce developments. Through a series of intensive workshops and drawing on expertise within and outside health and education sectors we produced a 2-day training intervention for HCAs working with older people.

Feasibility of a definitive cluster randomised controlled trial

We aimed to assess the feasibility of a cluster randomised controlled trial to compare the performance of the developed training intervention for HCAs with current training in improving the care of older patients in acute NHS settings (study aim 3). To address this aim we conducted a feasibility pilot cluster randomised controlled trial and a parallel process evaluation. In total, 12 wards (four from each partner NHS trust) were randomised to either HCA training in Older People’s Shoes or TAU. Although we did not seek to establish superiority of HCA training in Older People’s Shoes over TAU because of the feasibility study design, there was evidence that the direction of effect, after adjustment for baseline differences, was towards better HCA outcomes in the Older People’s Shoes arm of the trial. Feasibility was to be determined by the answers to a series of predefined questions (see Chapter 6, Purpose). These are dealt with in turn below.

Implications for practice

Although the study conducted was predominantly methodological, it raises a number of implications for the training of HCAs:

1. ‘Downward substitution’ within the health-care workforce is perhaps inevitable with increasing pressure to contain costs. Even if the size of the workforce does not change, the proportion of HCAs within it will grow. The work of HCAs is critical in improving the experience of older patients and their carers by delivering good relational care, but there is little relevant training available to them. Older People’s Shoes training which is grounded in evidence about HCAs’ everyday work can help them to deliver relational care in the challenging context in which they work.

2. Current training for HCAs in relational care is limited and variable particularly for existing (rather than new) staff. The Care Certificate includes dignity, communication and person-centred care. Relational care brings these aspects together into a coherent approach. Support for training in this area is imperative and greater oversight at the trust level is required beyond the induction of HCAs to ensure that efforts to equip the HCA workforce to provide relational care is not lost when responsibility is devolved to wards.

3. A number of factors (a degree of public distrust, the relative lack of investment and support by trusts for HCA training, staff hierarchies and restricted opportunities to come together as a section of the workforce) can make HCAs feel unsupported and unvalued. Training designed specifically around the needs of the HCA workforce but embraced by the wider system, can bring HCAs together to reflect and share good practice, boost morale and give HCAs a sense of value and purpose in their work.

4. The ‘front-line’ role of HCAs often exposes them to strong emotions (of anger, frustration, grief and despair) in patients and carers. Furthermore, their role often requires them to have ‘difficult conversations’. A lack of training in these areas leaves HCAs vulnerable and may result in a lack of appropriate support for patients and carers. Older People’s Shoes training allowed some exploration of these issues but further work is required in this area so that HCAs are given the skills to deal with these challenging situations and signposted to appropriate sources of support.

5. The demands of care on older people’s wards, in which patients have complex needs and comorbidities and need high levels of support with activities of daily living, can make it extremely difficult to create pockets of time through which to demonstrate relational care. Training needs to recognise this and help identify ways in which trainers can deliver good relational care while carrying out everyday care tasks.

6. Empathy is an important component of relational care. Training that deals with these concepts only in abstract form is unhelpful. Training that provides experiential learning and the opportunity to listen to older people’s accounts will help HCAs strengthen their relationships with patients to better understand their needs and feelings.

Things to look out for re: relational care

Is the term ‘relational care’ problematic? What (other) words do HCAs use? (At what point) does the term relational care enter the vocabulary of the HCAs?

At what points in the course is relational care kept in clear focus? Where does it get lost? What makes relational care slip in/out of focus?

Things to look out for re: learning

Give examples of HCAs talking about examples of relational care/building on discussion points/missing the point/losing the thread.

Do participants seem clear about what relational care is? At what point(s) is a grasp of relational care achieved?

Which activities seem to work best in terms of learning? (Give examples of participant engagement – contributions to discussions; flip chart brain storming; ideas and anecdote sharing or obvious times/some participants where this is not happening.)

What conditions appear to help/hinder learning?

Things to look out for re: course delivery

Does the trainer appear to ‘get’ the point of the training? (e.g. do the key messages emerge for each activity? Does the trainer manage to bring discussions back to relational care? Would you expect the learning goals for each session to have been met?)

In order to facilitate learning we are aiming to make the tenor of the group participative and not didactic. Can you give evidence for this?

Does the trainer appear to uphold the assets-based values underpinning the training? (Give examples)

Fidelity of implementation – How closely does the trainer stick to the manual? In what respects? Please mark timings per activity on the grid below, where indicated.

(PDF download)

Training and support to trainer

• Did you feel confident that you understood the purpose of the Older People’s Shoes training programme?

• [Recap on any feedback on the trainer manual already given by the respondent] Do you have any further suggestions on how could we can improve it to make it easier to use?

• Did you feel confident in delivering the training programme? (If so what helped, if not what could we have done to support you more?)

• Did you have previous experience of delivering something similar in terms of the style of learning used in Older People’s Shoes – 2-day, day-long, group size, interactive, peer-to-peer learning, shared discussion, reflection).

• Did you have previous experience of delivering something similar in terms of the subject matter?

• [Recap on pre-course training and support. This will include initial meeting to recruit trainers, Train-the-trainer sessions, any feedback after each day, all other e-mail/phone support]. Do you have any suggestions for improving the training or support you received in order to deliver the programme?

• There was some practical input on the training days from the researcher(s) [and the other Practice Development Nurse] (setting up the room, putting up signs, welcoming participants, sorting out IT and catering, etc.). Do you think this course could be delivered without any additional help? (If not) what help do you think would be needed?

Course content

• Were there sections or elements of the training programme that you felt didn’t work very well? Can you tell me more about that/those?

• What sections or elements of the training programme did work well? Which bits do you feel the HCA trainees engaged with most?

• Do you have any comments on the e-learning resource?

View of impacts on practice

• What are your thoughts on the relevance of the training to HCAs working with older patients?

• Do you know whether the training has had any impact on HCAs’ practice?

Suggested improvements

• Are there any other steps you think we could take to improve it? (Prompts: title?/timetabling/number of days, gaps, rostering/support from research team/participants manual?, etc.)

• Is there anything else you’d like to say about the training?

Thank you for your time.

Expectations

• What were your thoughts about the Older People’s Shoes training before you came on the course?

• What (if anything) did you hope to learn or practice during the training?

Course content and delivery

• What did you enjoy the most?

• Was there any of it you think we should cut?

• Was there any topics you’d have liked more on?

• Or anything else we should have covered?

• The course included periods of watching and listening; reflection; discussion; sharing knowledge and experience; and doing. Do you feel the balance was about right? If not, what would you have wanted more/less of?

• What about the timing of the course? Pace? Length of day? Number of days? Gap between days?

• Have you accessed the e-learning at all? (If not) Why was that? (If yes) Between training days and/or since? Which elements did you look at? Was it easy to use? Was it useful?

View of impacts on practice

• Overall what do you think you’ve gained from the course?

• Has being on the course made any difference to the way you feel about older patients?

• Has being on the course made any difference to the way you feel about your work?

• Have you managed to put anything you learned into practice? (If so) Can you tell me more about that?

• Has it been difficult to put any of it into practice? (If so) Can you tell me more about that?

Suggested improvements

• We used the term ‘relational care’ in the training. Do you think that’s a useful term to describe what we were talking about? Is there a better term you can think of?

• We’re looking at any ways we could improve the course if we run it again. Do you have any (other) suggestions? (Prompt if necessary re: what do you think of the title? Course book? Timetabling? Number of days? Gaps? Rostering? Did it cause you any difficulties attending the training?)

• Is there anything else you’d like to say about the training?

Thank you for your time.