Hearing and vision support in people with dementia living at home: Outcomes  from the intervention development programme for the European SENSE-Cog Trial

Step 1: Scoping the gaps

In this first step of the programme, to scope thoroughly the gaps in understanding, awareness, and service provision for the support care needs of PwD and hearing and/or vision loss, we sought stakeholder input and evidence from the literature.

Systematic literature review. We used a scoping review to identify (i) the existing evidence base, and (ii) the theoretical basis to support the new intervention. We have detailed the full result in (Dawes et al., 2019b). Briefly, we included studies that reported on interventions for hearing or vision loss in PwD in relation to: (i) cognition; (ii) rate of cognitive decline; (iii) behavioural/psychological symptoms of dementia; (iv) hearing/vision related QoL; (v) general QoL; (vi) activities of daily living; and/or (vii) caregiver burden for adults with dementing conditions. Our search revealed studies of low to moderate quality. Hearing interventions generally reported some reductions in behavioural and psychological symptoms of dementia and in hearing disability. However, there was inconsistent evidence for the impact of hearing or vision interventions on cognitive function, rate of cognitive decline, general QoL or caregiver burden, and consistent benefit across the small number of studies of vision interventions for any outcome. It was clear that there is an urgent need for a fully powered RCT of an intervention to improve hearing and vision loss to improve outcomes in PwD.

International Expert Reference Group – 1 (ERG - 1). The purpose of ERG-1 was to scope ‘gaps and solutions’ in the support care needs of people with concurrent problems from the perspective of professional experts from the relevant disciplines (hearing, vision, and dementia care). This knowledge elicitation involved 17 experts (clinicians and academics), who were purposively selected for their clinical or research expertise in that area. In the ERG, interactive activities were undertaken including sharing of expertise, small mixed groups guided discussions of prototype case studies. We have described this activity and its findings fully in Leroi et al., 2019 (Leroi et al., 2019). The consensus was that there are significant gaps in the knowledge, skills and practice regarding the detection, diagnosis, and support of comorbid sensory-cognitive conditions. Recommendations included the need for flexible, individualised, person-centred solutions to optimise well-being in PwD with sensory loss.

Survey of professionals’ knowledge, skills, and practice. Following the ERG-1 phase, we conducted a national UK survey involving 653 actively working clinicians with expertise in hearing, vision and dementia care (Leroi et al., 2019). The survey instrument was based on the output of ERG-1. Respondents answered each statement regarding knowledge, awareness and practice of crossover hearing, vision, and dementia care in their respective domains. In general, professionals across all domains reported awareness of the overlaps between hearing, vision, and cognitive impairment. A high proportion of dementia and hearing professionals reported that they are commonly asked about impairment in the other domains, but this was not the case for vision professionals. All three groups reported low awareness of brief screening assessments in the other domains for use in conjunction with their primary assessment. The majority reported insufficient skills and training or even awareness of referral pathways. Over 90% within each group agreed that guidelines would be useful.

Step 2: Modelling the intervention

In this step, our aim was to start to inform what the specific components of the intervention should be. The perspectives of PwD and their care partners were gathered, both quantitatively and qualitatively, regarding their support care needs (SCN).

Quantitative support care needs’ (SCN) analysis. To compile the necessary components of the intervention, we explored the perspectives of PwD and their care partners regarding: (a) the types of support care needs (SCN) in different EU sites; and (b) whether hearing, vision and cognitive status was associated with QoL and care partner burden, loneliness, and mental health outcomes.

We adapted the Supportive Care Needs Survey (SCNS-LF59), a fully validated instrument (Bonevski et al., 2000), for our population of PwD with sensory loss. Both the PwD and care partner completed our adapted SCNS, by consensus.

It was found that the majority of PwD reported needs in the ‘moderate to high’ range (Leroi et al., 2022). The highest number of needs overall were in the psychological domain, physical domain and need for health information regarding the affected system. Specifically needs in relation to anxiety, fear of loss of independence and memory getting worse were prevalent. Along with not being able to do things they used to do, the need to be given the relevant information to manage their problems at home, written information about the details of their care and being informed about support groups in the area.

Visual and hearing loss was shown to be significantly associated with depression in the PwD. The more severe the sensory loss, the greater the depression. Greater depressive outcomes were also observed for care partners, related to an increase in dependency of the PwD. Increases in care needs and levels of hearing loss was also adversely associated with impaired QoL, however visual loss did not exhibit the same effect.

Overall, there was a demonstration that greater SCNs are associated with more severe visual loss, but lower QoL is associated with more severe hearing loss. Increases in severity relate to a greater amount of need.

Qualitative exploration of support care needs. To explore these same needs in greater depth, we conducted a qualitative evaluation using focus groups (FG) and/or semi-structured interviews (SSI) with a purposive sub-sample of participant dyads. We delivered the first part of the FG/SSI in an exploratory style to determine what life is like for individuals experiencing concurrent dementia and sensory loss.

The interview guide (see Table 2) focused on: (1) the type of SCN for the PwD; (2) perceptions of living with dementia and sensory loss and related psychological needs; and (3) feedback on the components of the newly developed support intervention. The interview material was analysed according to Mayring´s (2000) summative content analysis (Mayring, 2000). We have outlined the methods and outcomes previously (Leroi et al., 2022).

Three key themes were identified: (1) the need for correction of sensory dysfunction and tailoring of support to individual needs; (2) care partner burden, social isolation, and loneliness; and (3) communication with professionals and education regarding the use of corrective devices for sensory loss. The three themes all indicated a need for a tailored individualised support intervention that targeted these key areas.

Step 3: Refining the prototype intervention

The aim of this step was to obtain feedback from relevant stakeholders to develop a draft intervention ready for the SENSE-Cog Field Test. Specifically, views and opinions regarding the implementation, utility, feasibility, acceptability, and tolerability of the proposed components of the prototype intervention were elicited.

International Expert Reference Group – 2 (ERG-2). The goal of the second part of the Expert Reference Group (ERG-2) held in Athens in 2016 was to elicit the viewpoints of expert professionals regarding the best implementation methods used by existing trials of complex psychosocial interventions in PwD. For the consensus options’ appraisal, expert opinions were gathered on the implementation of four recently published protocols for the GREAT iCST, ATTILA and LOTSE trials for complex psychosocial interventions in PwD with intact or corrected sensory function (Clare et al., 2013b; Himmelsbach et al., 2015; Leroi et al., 2013; Orrell et al., 2012). Each of these previous trials used different implementation methods, including a structured manualised approach with a set number of sessions, a semi-structured approach with individually established goals and a set number of sessions, as well as a highly pragmatic trial with interventions varying according to the local service provider. Outcomes in all these trials included QoL, functional, cognitive, behavioural, health economic and care partner outcomes.

For each of these studies, the ERG appraised the options with a view to adopting the most suitable method for delivering the SSI, using ‘TOWS’ method (Proctor, 2000). Briefly, this is a decision-making tool in which a matrix of external threats (‘T’) and opportunities (‘O’) based on the different methods used in the trial are mapped against the internal strengths (‘S’) and weaknesses (‘W’) of what the study team can deliver. The findings from this exercise demonstrated that the most appropriate implementation methods were those used by the GREAT study (Clare et al., 2013a). The main strength (‘S’) of this model is that it provides a highly flexible, tailored, and acceptable approach for delivering a complex psychosocial therapy for PwD in their own homes. The use of an occupational therapist in the GREAT study is also a good model on which to base the development of the SST for the SENSE-Cog intervention. The number, duration and frequency of visits also appears appropriate to the type of intervention in SENSE-Cog. The main opportunity (‘O’) provided by the GREAT protocol is the ability to offer a highly personalised and individualised approach. In the SENSE-Cog context, the study sites will be in different EU countries, which will necessarily vary in available resources and local practices. Furthermore, the complexity of each participant’s needs will demand a very individualised approach. The limitations (‘W’ and ‘T’) of adopting the GREAT protocol is that its highly flexible approach may compromise standardisation, and this will need to be considered in the RCT analysis.

Including the “Patient and Public Voice”. In addition to the experts’ opinions on the most appropriate method of implementation, input from the international SENSE-Cog ‘patient and public voice’ (PPV) Research User Group (RUG) was gathered to allow for insight into the experience of the research participant undergoing a complex trial. This approach followed the guidance of INVOLVE UK (INVOLVE, 2015), based on the rationale that a person with ‘lived experience’ of a health condition or treatment can offer a unique perspective informed by direct personal experience. During RUG meetings, aspects of the inclusion/exclusion criteria, participant tolerability and acceptability and other issues pertaining to the intervention design were discussed. Key recommendations subsequently actioned by the research teams, further refining the proposed intervention, included changes to the number of sensory support visits, the inclusion of those at more advanced stages of dementia and the re-phrasing of participant information sheets to enhance comprehension. Details of the PPV RUG activities and findings are outlined in Miah et al. (2018) and Miah et al. (2020).

Articulating the theoretical framework. We next sought out an appropriate theoretical framework to support the developing intervention. Since it involved behaviour change in both the PwD and their care partner, we scoped change theories and found the Behaviour Change Wheel (Michie et al., 2011) to be the most pragmatic and to have the strongest evidence-base. In our protocol of this development programme, we have outlined how each element of the Behaviour Change Wheel, summarised by the acronym ‘COM-B’, corresponds to a component or set of components of the newly developed SSI (Regan et al., 2017). In Table 3, we detail how the components of the proposed SSI mapped onto these COM-B elements.

Step 4: Feasibility testing the draft intervention

The SENSE-CogField Trial. With the drafting of a prototype intervention complete, the final step consisted of field testing the intervention programme (study design, components, and intervention implementation) with a representative participant population who would later take part in the full-scale efficacy trial.

The key objectives of this field trial were: (1) to undertake a process evaluation of the components and implementation of the intervention, including reach and fidelity issues, acceptability and tolerability of the intervention for participants as well as the contextual issues; (2) to examine the operational aspects of a future trial design, including recruitment and retention and training methods and materials; and (3) to explore the feasibility and face validity of potential outcome measures for the future trial. To do this, a single-arm open-label feasibility study in three of the international clinical sites (UK, France, and Cyprus) using the draft intervention was undertaken. Enrolment included 19 PwD-care partner dyads and procedures were closely aligned to the proposed final RCT protocol. The details of this study have been outlined in Regan et al. (2017).

To evaluate the feasibility, acceptability and tolerability of the proposed intervention, ratings were obtained using questionnaires, participant diaries, therapist logbooks and semi-structured interviews. All dyads (n=19) received a basic version of the intervention, and a subset (n=4) received a 12-week extended version. The basic version of the SI was composed of a clinical vision and/or hearing assessment with prescription and fitting of corrective lenses, and/or hearing aids and information about device maintenance. The extended version which had additional components is outlined in Table 3. All intervention components were successfully delivered, received, and enacted as intended in all participants who completed the intervention, with acceptability ratings of the intervention falling within pre-specified target ranges.

Importantly, improvements in QoL and sensory functional ability were found between baseline and follow-up, with QoL exceeding the threshold for minimum clinically important difference. These findings, in conjunction with qualitative data evidencing improvements in PwD functional ability, communication and social engagement, and a reduction in care partner dependency and feelings of isolation, further supported the need for a full scale RCT to evaluate the effectiveness of the intervention. Comprehensive reports on the field-trial are available elsewhere (Hooper et al., 2019; Leroi et al., 2020b).

Synthesis and integration of findings. As both qualitative and quantitative data were gathered the study team first had to agree on the weights of lines of evidence. This was informed by the existing evidence available and the strength of the method of data collection. Since the qualitative data were richer in content and provided more detailed context on which to base the outputs than the quantitative data, the weight of evidence rested on it. However, we used corresponding lines of evidence from both the quantitative and qualitative sub-studies to validate findings from the different sources. We did not come across contradictory data that might have generated further research questions. A summary of the conclusions of each sub-study from each step of the intervention development is outlined in Table 4.

Description of the final ‘SENSE-Cog Sensory Support Intervention’. The sensory support intervention is highly individualised and involves the input of a companion. The overall aim of the intervention is broad-based and targeted towards improving QoL, reducing functional disability, and increasing social connectedness.

Based on the integration of the outputs described, the intervention to improve QoL in people with dementia by supporting sensory function was as follows:

i. Specific components of the intervention:

Using the COM-B model as a theoretical framework, it was determined that the intervention should include the following components: (1) identifying and correcting any vision or hearing loss; (2) supporting adherence to the sensory devices (i.e., hearing aids, glasses, listening devices) with on-going advice and fittings; (3) identifying vision- and hearing-related functional deficits, facilitators and barriers; (4) addressing any identified skill deficit with vision and hearing training for the PwD and their significant other; (5) addressing any identified knowledge deficit with information provision and sign-posting; and (6) improving functional deficits through home-based environmental aids.

ii. Implementation of the intervention:

Hearing and vision assessments were conducted by an appropriate specialist (optometrist or ophthalmologist for vision assessment and audiologist for hearing assessment. A “Sensory Support Therapist” (SST) conducted up to 10 home-based visits with the PwD and their companion and carried out the components of the intervention tailored to the dyads needs.