A manual-based intervention for carers of people with dementia and sleep disturbances: an acceptability and feasibility RCT

People with dementia

Adults with dementia (any type and severity) and clinically significant sleep difficulties [Sleep Disorders Inventory (SDI)39 item score of ≥ 4 on screening] that were judged a problem by the person with dementia or their family were included. If the person with dementia did not have capacity to give consent, a consultee’s declaration was sought. Patients living in a care home were excluded. Those with a primary sleep disorder diagnosis (e.g. sleep apnoea), rather than dementia-related sleep problems, were also excluded.

Carers

Primary family carers of the person with dementia were included for quantitative assessment interviews and intervention delivery. When the carer who spent most time with the person with dementia was not a family carer, then the paid carer was able to participate in the intervention. The family carer, however, completed the carer assessment measures. All carers provided emotional or practical support to the person with dementia at least weekly. Carers who were unable to give informed consent, or with probable dementia, were excluded.

NHS trusts

Participants were primarily recruited from two NHS trusts in London: Camden and Islington NHS Foundation Trust (site 1) and Barnet, Enfield and Haringey Mental Health NHS Trust (site 2). Both trusts provided funding to cover the excess costs resulting from training, intervention delivery and clinical supervision. Clinicians approached people with dementia and their family carers who were attending one of the five memory services: (1) Camden, (2) Islington, (3) Barnet, (4) Enfield or (5) Haringey. Those interested in the study were asked to give permission for the research team to approach them, and were given an information sheet. Following contact by our team, an appointment was arranged to obtain informed consent in the potential participants’ homes (or in the research team’s office if preferred). A baseline assessment was carried out after this. Intervention delivery and follow-up assessments also took place in the participants’ homes.

Join Dementia Research

In addition, the research team approached potential ‘matches’ through Join Dementia Research (JDR). This is an online and telephone service developed and launched in 2014/15 by the NIHR. It aims to make it easier for people with dementia and other interested members of the public to participate in dementia studies. People register their interest, providing basic or more detailed demographic and clinical information and their contact details (including their preferred means of contact). The registrant or a named representative can then be contacted by researchers to discuss the potential suitability of studies. To recruit from this site [site 3, University College London (UCL)/JDR], potential matches with any dementia diagnosis listed on JDR were identified. We ticked ‘yes’ to ‘Has next of kin?’, chose ‘Include empty or NULL values’, ticked ‘yes’ to ‘Has a carer?’, ticked ‘no’ to ‘Volunteer is a carer?’, ticked ‘no’ to ‘Volunteer cares/supports a person with dementia?’, and selected ‘More than 3 times per week’ for ‘Contact with carer’ (‘Include empty or NULL values’), and selected ‘Private Residence’ and ‘Assisted Living Accommodation’ for ‘Accommodation Type’ (‘Include empty or NULL values’). Owing to limited resources, we defined our search area based on proximity to UCL.

Demographic and clinical characteristics

1. Sociodemographic details (i.e. sex, date of birth, type of dementia diagnosed, age when left education, last occupation, current marital status, ethnicity) were collected at baseline. We had information on all potential participants’ sex, including those who did not provide consent to be screened, those who were not eligible and those who were not randomised.

2. Type of dementia was recorded from clinical information.

3. Severity of dementia was measured using Clinical Dementia Rating™ (CDR)40,41 through informant information at baseline. This is a reliable and valid instrument for rating the severity of dementia. 42 It is used to rate performance in memory, orientation, judgement and problem-solving, community affairs, home and hobbies, and personal care. This information was used to classify dementia severity of clinically diagnosed patients or those on JDR into very mild (0.5), mild (1), moderate (2) or severe (3).

4. Rescue medication’s role: all prescribed psychotropic medication and melatonin was measured at baseline and 3 months by completing the Client Service Receipt Inventory (CSRI),43 which incorporates a list of all medications in the previous 3 months. Use of the following was recorded:

   • anxiolytics and hypnotics

   • antipsychotics

   • antidepressants

   • other psychotropics.

5. Reported side effects: side effects were measured using a study-specific Safety and Tolerability Assessment (see Appendix 2) to record the occurrence of falls, dizziness, headaches and gastrointestinal symptoms (appetite or bowel symptoms) and any other side effects at baseline and 3 months.

Potential outcomes for the main trial

1. Sleep disturbance in the person with dementia was measured using the SDI39 at screening and 3 months. The SDI is a standalone tool for assessing sleep disorder symptoms in people with dementia, developed for use to measure outcomes in original melatonin trials. It has been used in pharmacological and non-pharmacological studies and validated against actigraphy and clinical variables. The SDI consists of the seven sleep subquestions of the sleep and night-time domain of the Neuropsychiatric Inventory (NPI). 44 These are: (1) difficulty falling asleep, (2) getting up at night, (3) wandering, pacing or conducting inappropriate activities at night, (4) awakening the carer at night, (5) awakening at night, dressing, planning to go out, thinking that it is morning, (6) awakening too early in the morning and (7) excessive daytime sleepiness. SDI item scores are based on the carer’s ratings on each of the seven symptoms separately. Each item is rated according to frequency (scale 0–4) and severity (scale 0–3) of sleep-disturbed behaviours and, when frequency and severity are multiplied, possible item scores range from 0–12. Those who scored ≥ 4 on any individual item were judged to have clinically significant sleep disturbance and were eligible for the study. The SDI mean global score, which is often reported, is the total frequency multiplied by severity, divided by 7.

2. Neuropsychiatric symptoms were measured using the NPI44 at baseline and 3 months. This is a validated instrument with 12 domains of neuropsychiatric symptoms. A single frequency and severity rating is given for all behaviour subquestions within a domain, with each domain scoring between 0 and 12; higher scores mean increasing severity. A summed score out of 144 captures total neuropsychiatric symptoms.

3. Daytime sleepiness was measured from the Epworth Sleepiness Scale (ESS)45 at baseline and 3 months. This is an eight-item measure assessing the tendency to sleep/doze in specific daily situations (possible score range 0–24; a score of > 10 indicates excessive sleepiness).

4. Quality of life was measured using the Dementia Quality of Life – Proxy (DEMQoL-Proxy)46 at baseline and 3 months. It is a 31-item interviewer-administered questionnaire answered by a carer. It is a responsive, valid and reliable measure of quality of life in people with dementia with satisfactory psychometric properties. 47

5. Services use was measured using the CSRI at baseline and 3 months. It is widely used for dementia trials and will delineate treatment as usual (TAU) as well as treatment for those in the active arm of the study.

Wrist-worn Actiwatches, worn 24 hours a day for 2 weeks at baseline pre randomisation and again 3 months after randomisation, provided estimates of rest–activity rhythms, light exposure and sleep. Actigraphy has theoretical advantages for patients who may not be able to remember well enough to accurately fill in questionnaires or sleep diaries. It has been validated against polysomnography (PSG) in populations including older adults without dementia (one study48 of 77 people, with a mean age of 35 years) for its accuracy at correctly detecting sleep, wakefulness and wakefulness after sleep onset. The Actiwatch detected sleep in the whole population with 96.5% accuracy, but was not as good at detecting wakefulness (32.9% accuracy), and this became worse with increasing age. Another study of older women without dementia (with a mean age of 69 years) found that wrist actigraphy detected sleep less accurately in this age group and was unacceptable for those with low measures of sleep efficiency. 49 However, it has been used in 10 nursing home residents with severe dementia to estimate sleep during night-time, and the measures significantly correlated with total sleep time from electroencephalographic recordings. 50 We used the MotionWatch 8 (herein referred to as ‘the Actiwatch’), a Conformité Européenne (CE)-marked class 1 (low-risk) medical device (CamNtech Ltd, Cambridge, UK). Such non-intrusive, reusable Actiwatches have been used successfully in previous trials of people with dementia and sleep disturbance. 51 The aim was to use a minimum of 1 week’s worth of data. The participants were given written instructions (see Appendix 3). It was ensured that that the person with dementia and their carer between them understood the instructions and had the opportunity to ask questions. Whenever possible, the watch was placed on the wrist of the participant’s non-dominant hand, and variability was controlled among the watches by ensuring that the participants wore the same watch at baseline and follow-up. The data were recorded in MotionWatch Mode 1, which uses a single-axis algorithm and peak detection, sampled at 50 Hz, and processed into 60-second epochs. Each recording was started in the evening on the day of the baseline or follow-up assessment (preferably at 17:00) and continued for a minimum of 14 days, after which the watch was collected. Carers were asked to use a sleep diary (see Appendix 3) to keep a record of the bedtimes and rise times of the person with dementia, or, as an alternative, to indicate these times by pressing the Actiwatch’s ‘event marker’ button. These were used to define each participant’s sleep analysis window. We also asked the carers to write down (e.g. in the sleep diary) any occurrences when the watch was not worn for > 1 hour. Participants were encouraged to contact the research team if they had any further questions about the watch.

• Sleep measures:

  • Sleep efficiency – sleep time expressed as a percentage of time in bed. This captures both initiation and maintenance of sleep, reflecting the proportion of time in bed spent asleep, and has been found to be reliably impaired in actigraphy studies of people with dementia and sleep problems.

  • Sleep time (minutes) – the total time spent in sleep according to the epoch-by-epoch wake/sleep categorisation. As the Actiwatch infers, time asleep is not ‘actual sleep time’.

  • Wake time (minutes) – the total time spent awake according to the epoch-by-epoch wake/sleep categorisation. As the Actiwatch infers, time awake is not ‘actual wake time’ but it is the label used.

  • Time of ‘lights out’ or going to bed.

  • Time of falling asleep.

  • Time of waking up.

  • Time of getting up.

  • Time in bed (hours) – the total time between ‘lights out’ and getting up.

  • Fragmentation index – the degree of fragmentation of the sleep period, calculated by summing mobile time (%) and immobile bouts of ≤ 1 minute (%).

• Circadian rest–activity rhythm metrics/non-parametric circadian rhythm analysis (NPCRA) measures to assess the timing, amplitude and stability of rest–activity rhythms:

  • Relative amplitude – the amplitude of circadian rhythm (range 0–1), calculated by dividing the difference between average activity in the most active (M10) and most restful (L5) periods by the sum of M10 and L5.

  • Interdaily stability – the degree of regularity in the activity–rest pattern, ranging from a total lack of rhythm (0) to a perfectly stable rhythm (1).

  • Intradaily variability – the degree of fragmentation of activity–rest periods (range 0–2), from prolonged periods of activity and rest over 24 hours to multiple short periods.

  • L5 – activity count for the 5 most restful hours.

  • L5 – start hour (of the 5 most restful hours).

  • M10 – activity count for the 10 most active hours.

  • M10 – start hour (of the 10 most active hours).

• Core night-time analysis sleep measures:

  • Core night-time (00.00 to 06.00) sleep efficiency.

  • Core night-time (00.00 to 06.00) sleep time (minutes).

  • Core night-time (00.00 to 06.00) wake time (minutes).

Carer measures

DREAMS START intervention development

Conceptualisation of manual elements

The DREAMS START intervention was developed using an iterative and collaborative coproduction process with the team that developed the START and Managing Agitation and Raising Quality of Life (MARQUE) manualised interventions for carers of people with dementia (GL, PR, CC), experts in manualised cognitive–behavioural interventions for sleep disorders (CAE, SDK), our collaborators from the Alzheimer’s Society (JAP, RH) and carers of people with dementia who have experienced sleep disturbances, and through practice sessions as part of therapist training. Figure 1 summarises the manual development process.

FIGURE 1.

Development process of the DREAMS START manual.

Development of the intervention began in February 2016 when the project management team met and agreed that DREAMS START would be a six-session, multicomponent intervention, based on the best existing evidence at the time,26,35,37,38,58–61 comprising a cognitive–behavioural component (including psychoeducation, coping skills for families, and activities for people living with dementia) and light therapy. Figure 1 provides an overview of this process.

The first stage of intervention development was to share the existing StrAtegies for RelaTives (START)35 and Oxford’s manuals on cognitive–behavioural therapy for insomnia62 within the team and to consider which elements to build into the DREAMS manual. One month later, Gill Livingston, Simon D Kyle and Penny Rapaport met face to face and, joined by Colin A Espie on the telephone, agreed the main initial structure and content of the six sessions. At this stage, it was agreed that the intervention would be both interactive and individualised, using the actigraphy data collected at baseline to inform a shared understanding of the person with dementia’s night-time settling and waking problems, and their daytime sleepiness/fatigue and to help generate an optimal ‘sleep window’. As in the team’s previous work, a collaborative, non-prescriptive model was adopted, explicitly encouraging family carers to build on their own experience of what they have found works and to develop and use new techniques and behavioural strategies, recording what works in their manual, and continuing to use successful strategies. A specific focus was incorporated within sessions on identifying and overcoming barriers to changing behaviours and routines of people with dementia and considering how to minimise the risk of harm when people with dementia are awake at night. In line with the form of the START intervention, each session included a stress reduction exercise, a between-session practice task, a recapitulation on the previous session and troubleshooting around putting strategies into practice. Five of the same stress reduction techniques from the START intervention were used in the DREAMS START manuals (sessions 1–5). All manuals from the START intervention are available freely online. 36,60

Patient and public involvement in the intervention development

After making initial revisions, family carers were invited, via the Alzheimer’s Society Research Network (ASRN), to attend a focus group (in May 2016) to provide feedback on the drafts and to share their experiences of what strategies they did and did not find useful in caring for a relative with sleep difficulties. The group was facilitated by Penny Rapaport and Kirsi M Kinnunen, based on a semistructured topic guide (see Appendix 4). All carers consented to the discussion being audio-recorded, and discussions were then externally transcribed and the data were entered into NVivo 11 software (QSR International, Warrington, UK) and analysed using thematic analysis. 63,64 Two researchers (BH and LW) independently coded the transcript for the main themes that occurred and each labelled the themes and generated a thematic framework. They then met with Penny Rapaport to discuss and agree a consensus on their coding and generate the final framework, according to which the data were organised. Two current and two former carers of people with dementia attended the focus group and, having further revised the manual based on this initial feedback, a virtual reference group of family carers and people living with dementia was consulted by e-mail to obtain further feedback on the draft manuals. The virtual reference group included the four focus group participants and three additional members of the ASRN, including one person living with dementia, one current carer and one former carer.

The key themes elicited from the focus group related to the sleep disturbances that people with dementia experienced and the explanations carers had for these difficulties, the effects that sleep disturbances had on the carers, what they found helpful, and suggestions that they had for the DREAMS START intervention.

The main difficulty that the carers described was their relative waking repeatedly in the night and getting out of bed. This often increased the risk of coming to harm, for example by falling down the stairs or by using electrical appliances unsafely. Carers attributed this night-time waking to a number of causes, mainly confusion, worrying about their symptoms, and dementia type and severity. Confusion meant their relative with dementia mixing up night and day, which one carer felt was exacerbated by being indoors and inactive during the day.

All four of the carers in the focus group spoke of the emotional, physical and practical effects that their relative’s sleep disturbance had on themselves, their relatives and other family members. The main effect on carers was that they themselves became sleep deprived. This happened because they were woken up directly by their relative, and also because they were kept awake by the worry that their relative might awaken, get up and come to harm. Carers described how the lack of sleep over time had negative physical effects on their health and affected their ability to work and function during the day. Carers highlighted various successful and unsuccessful strategies that they had used. Gentle touch and physical comfort had a relaxing effect on their relatives at bedtime and during the night, and keeping the evenings calm and winding down before bed was helpful. They also highlighted the importance of carers finding ways to manage their own stress and sleep problems. There was a common consensus between the carers that using medication to manage sleep problems in their relatives was not helpful. Some also found that there sometimes did not seem to be any pattern to the sleep problems and no helpful strategies.

These findings were used to further refine the manual, ensuring that the emergent themes were included and, in particular, that there was a greater focus on carers finding ways to minimise the impact of sleep disturbances on their own well-being and ensure that the person with dementia was physically and emotionally comfortable. At this point, direct quotations from the carers about their experiences were added to the manual to situate the content and privilege the experiences of family carers. In addition, both the focus group participants and the virtual reference group participants gave feedback and made suggestions about different aspects of the manual sessions. These are summarised in Table 1.

Intervention content

The DREAMS START manual used in the trial was designed to optimise both the individual’s sleep at night and his/her wakefulness during the day. To optimise night-time sleep pressure and the circadian regulation of sleep, the intervention focused on supporting carers to use practical zeitgebers (cues that influence a person’s biological rhythms, e.g. regular timing of bed and rising, morning wake-up light, standardised mealtimes) and to establish adaptive stimulus control (e.g. pre-bed settling routine, management of wakeful episodes). The intervention introduces strategies to promote de-arousal at night (e.g. relaxation, bedroom comfort, no caffeine or alcohol intake before bed, no activities in bedroom) and behavioural activation during the day to maintain alertness, reduce daytime naps and increase engagement to strengthen both central and peripheral clock timing. The intervention also focuses on helping carers to develop coping strategies to manage concerns about their own sleep health.

Sessions 1–5 all included one or two key topics for discussion, a specific plan or goal to try out between sessions, a stress reduction exercise with an accompanying compact disc (CD)/MPEG-1 Audio Layer III (MP3) file and a sleep diary for the carer to fill in for monitoring progress between sessions. Although manualised, during each session the participants would develop specific goals, which were combined to make an individualised plan.

The six intervention sessions covered the following:

1. Understanding sleep and dementia – psychoeducation on the importance of sleep, sleep process, the impact of dementia on sleep and what causes sleep problems in people with dementia. This session also discussed lifestyle and bedroom environment factors that can affect sleep, and carers were encouraged to identify potential changes that they could try out before the next session. In addition, the impact that sleep problems can have on the person with dementia and their relative was discussed.

2. Making a plan – psychoeducation on the importance of light for sleep and the relationship between light, sleep and dementia. This session was built around the light and activity data collected by actigraphy at baseline, with the individual’s data inserted into the manual in advance of the session and an explanation given. The carers were given a ‘light box’ (Lumie Arabica SAD Light, Lumie, Cambridge, UK) to keep and it was switched on and left on during the sessions to encourage people to understand that they habituated to the bright light (10,000 lux at 25 cm). A time switch was provided if carers wanted it, so that it could be switched on when they were otherwise engaged. It was recommended that the light box be used for 30 minutes at the same time every morning. During the session, the participants made an individual plan for increasing natural and artificial light for the person with dementia. Based on the individual’s sleep data, a ‘sleep efficiency’ score (time asleep/time in bed) was calculated, and any suggested changes to an individual’s time to bed and rise and for reducing daytime naps was integrated into their individual action plan.

3. Daytime activity and routine – this session highlighted the importance of daytime activity and routine and focused on building pleasant activities and exercise/physical activity into the day. This session included a seated exercise video for less physically able individuals. It also included the importance of establishing a good day and night routine and ways to strengthen the link between bed and sleep. In addition, an exercise and activity plan were built into the individual plan.

4. Difficult night-time behaviours – this session began with troubleshooting around putting the individual plan into action and identifying potential solutions to any barriers. The rest of the session focused on describing and investigating difficult night-time behaviours that were specific to the individual with dementia.

5. Taking care of your own sleep – this session included using the information collected on difficult night-time behaviours to create strategies for managing these difficulties. The rest of the session focused on the carer managing their own sleep, including ways to challenge unhelpful thoughts and feelings and make time for themselves.

6. What works? Using strategies in the future – this session recapitulated on earlier sessions and focused on what carers found useful and what worked. From this, an individualised sleep action plan was finalised, which included strategies for both the person with dementia and the carer.

Details of how the interventions were standardised

The intervention was standardised primarily through being manualised. This allowed the research team to maintain tight control over how the intervention was delivered, and provided a clear and detailed structure for the graduate psychologist therapists to adhere to. It also meant that the carers could keep the manual and refer back to notes, plans and information between and after the sessions. Variation came with the introduction of personalised goals and strategies, and the use of individual actigraphy data. Delivery of the interventions as intended and in a standardised way was assured by training the therapists and ensuring that they were ‘signed off’ as individually competent in each session, and by offering regular and structured clinical supervision, as detailed in the following section.

Training the therapists

Four research assistants who were psychology graduates with no clinical training were employed to deliver the intervention and received in-depth training prior to delivering the intervention. The research assistants participated in five knowledge and skills-based sessions on the following topics:

1. Dementia (CC)

2. Sleep (SK)

3. Introducing DREAMS START (GL)

4. Clinical skills for delivering the intervention part 1 (PR)

5. Clinical skills for delivering the intervention part 2 (PR).

Training was delivered through a combination of seminars, discussion, reflective learning and guided reading. Skills-based competencies were learnt through role-play, small-group exercises and clinical simulation in pairs. Training drew on the curriculum for psychological therapists devised by the Department of Health and Social Care for its IAPT programme, and the successful training programme developed for the START intervention. There was a strong practical focus in the training programme on how to deliver the therapy, potential clinical dilemmas, collaborative goal-setting, managing sessions with more than one participant, working with interpreters, empathic listening skills, effective use of supervision, safe working practice, and when to ask for help. Throughout the training, an emphasis was placed on the researchers facilitating carers to develop and practise using their own strategies and finding their own solutions rather than feeling the need to instruct carers and provide solutions within the session. In addition to the knowledge and skills-based training, therapists were trained to adhere to the manual by practising repeatedly in pairs, and required to demonstrate, by role playing the entire intervention for one of the clinical members of the research team, competence in delivering each session of the intervention to an agreed standard.

Supervising the therapists

The process of formal clinical supervision of the therapists began at the start of the intervention delivery period and continued until the final sessions had been delivered. The clinical psychologist, Penny Rapaport, met with each of the two teams (two therapists in each team) for 1.5 hours of group supervision per fortnight. In addition to this group supervision, she was available for individual supervision, which was either requested by the psychology graduates on an ad hoc basis, or (on occasion) was initiated by the investigators. If they had any urgent clinical or procedural concerns or questions relating to their clients, the therapists could approach Penny Rapaport, Claudia Cooper or Gill Livingston at any time, for example if risk issues arose during a session.

The group supervision format was seen as the most effective use of available resources, with psychology graduates benefiting from both the professional expertise of their supervisor and the clinical experiences of their peers; it was successfully applied in the START randomised controlled trial (RCT). It was expected that the supervision format would also maximise peer support within and outside the supervision sessions, and facilitate effective teamworking. The format was tailored to reflect the specific needs of the present research study. During the course of the project, supervision performed a number of functions including case management, clinical skills development, ensuring safe practice with clients, and staff support. Each of these functions is explored in turn in the following sections.

Details of how adherence of care providers to the protocol was assessed or enhanced

Primary outcomes

1. Feasibility of recruitment: this was assessed by (1) the proportion of participants consented out of those meeting the eligibility criteria at screening and (2) the proportion randomised after baseline assessment.

2. Feasibility of the intervention: this was assessed by recording the proportion of participants randomised into the intervention group who by the end of the trial had attended four or more of the six sessions.

3. Acceptability of the intervention: this was assessed through qualitative interviews with up to 20 intervention group participants after follow-up, post unblinding.

Secondary outcomes

1. Referral rates from the recruitment period were measured from records about all and eligible referrals at the end of the recruitment period.

2. Follow-up rates were measured after the last follow-up visit, from records indicating which participants completed follow-up assessments at 3 months.

3. Reported side effects (patient falls and comorbid physical illnesses) were recorded using a study-specific DREAMS side-effects questionnaire at baseline and 3 months.

Acceptability of outcome measures for a future trial of clinical effectiveness and cost-effectiveness was assessed through recording the completion rates of instruments (see below) at baseline and 3 months, the acceptability of tools from the qualitative interviews post unblinding, and calculating the sample size that would be required to detect a clinically important difference in outcome.

Sample recruitment and procedure

A purposive sample of carers who had received the intervention for qualitative interviews was recruited in order to assess acceptability of the intervention and whether or not there were groups that the intervention was unsuitable for, and to detail any required refinements in the intervention or the assessment procedure. They were invited to participate in the interview after follow-up assessment and unblinding.

Purposive sampling was used to ensure that people from a range of sociodemographic backgrounds were interviewed. 68 Included were people of differing ages and relationships to the person with dementia, family and paid carers, those living with the person with dementia and those living separately, people from varying ethnic backgrounds, those who had finished the intervention and one carer who had not finished the intervention. Carers were recruited until theoretical saturation was reached.

After this, Penny Rapaport and Lucy Webster met with the ASRN members who had been part of the development of the manual for a focus group to consider any changes post trial.

Qualitative interview content

A semistructured interview guide was developed for the carers who had been participants in the trial (see Appendix 9) of open-ended questions based on our study objectives. These explored the acceptability and practicality of both the intervention and the assessment measures, to gain suggestions for refining the trial. The interview guide was revised iteratively during the interview phase, adding themes as they were brought up by interviewees. All interviews were audio-recorded and recordings were destroyed after analysis.

After completing the trial and gaining qualitative feedback from the participants, the family carers that had contributed to the intervention development were invited to attend a focus group at the Alzheimer’s Society premises in September 2017. The aim of this focus group was to hear group members’ thoughts on the qualitative findings, the final version of the manual used in the trial and the materials being presented, and any suggestions for further refinement of the intervention.

Sleep and non-parametric circadian rhythm analysis measures

The sleep analysis function of MotionWare was used to produce sleep measures from overnight actigraphy data. The NPCRA analysis in MotionWare is based on an approach that does not assume that the data fit any predefined distribution. 69 First, the whole recording period was highlighted, and the editing tool was used to remove recordings exceeding 14 days and periods of missing data (when the watch was off the wrist based on notes in the sleep diary or the carer’s verbal report on Actiwatch data collection, or the recording indicated no data). The maximum recording period used for the analysis was 14 days (in the majority of cases from 17.00 on the first day to 17.00 on day 14, but in some from 18.00/19.00 on the first day to 18.00/19.00 on day 14). Second, using the edited data, each sleep period was defined, choosing as the start and end points the bedtimes and rise times recorded in the sleep diary or by using event markers (showing as blue lines in the data). If the carer did not complete the sleep diary or use the event maker button, the bedtimes and rise times were defined based on the carer’s verbal report on Actiwatch data collection. In addition, the activity and light data were used as guidance for editing the sleep periods. When unsure, two researchers edited a sleep period and reached a consensus. Each sleep period was saved. The edited recording period was also saved to derive the NPCRA measures, and the summary option used to record average light (lux) over the same period. Finally, a report was produced, including all the sleep analyses and NPCRA measures evaluated in the feasibility trial as possible outcomes for the main trial. In the statistical analysis, sleep data based on fewer than seven nights were excluded. To produce NPCRA measures used in the statistical analysis, copies were made of each previously saved edited recording. The length of each missing data period was then checked, and any 24-hour periods with ≥ 3 hours of missing data were excluded. 70 Having saved the NPCRA period, a new report was produced containing the non-parametric measures only.

Core night-time measures

To produce sleep measures for what the research team defined as core night-time (0.00 to 06.00), a copy was created of the edited recordings previously saved. The existing sleep periods were removed and then the sleep summary table option was used to limit each new sleep period to start at 0.00 and end at 06.00. Having saved these periods, a report was produced containing the new sleep measures.

Sensitivity analysis

In a sensitivity analysis, only participants who lived with a family carer at baseline were included on the sleep and NPCRA measures. The sleep data that were based on seven or more nights and the NPCRA data without ≥ 3 hours of missing data for each 24-hour period were used.

Summary of baseline data

Baseline data (sociodemographic characteristics, actigraphy measures and other scores) for participants and carers were summarised by treatment group using means (with SDs), medians [with interquartile ranges (IQRs)], counts and proportions, as appropriate, to gauge the balance in characteristics between the randomised groups.

Three-month follow-up

Follow-up scores, actigraphy measures and other scores at 3 months were summarised using means (with SDs), medians (IQRs) and counts (%) as appropriate. For continuous measures, correlations between baseline and follow-up measurements were calculated. The number of participants with completed values was summarised for each outcome.

Measurements were compared between randomised groups using appropriate regression models to provide estimates of the effect of the intervention with 95% CIs (e.g. difference in means), adjusted for baseline score and site. For actigraphy data analyses, the following were focused on: sleep efficiency, relative amplitude, sleep fragmentation index, start hour of most restful hours, activity count for most restful hours, start hour of most active hours and activity count for most active hours for those with ≥ 7 days of data. Core sleep data between 0.00 and 06.00 were measured and a sensitivity analysis excluding participants who did not live with a carer at baseline was carried out.

Use of psychotropic medication

The frequency (%) of participants in each randomised group who had taken each type of medication (anxiolytics and hypnotics, antipsychotics, antidepressants, other psychotropics and melatonin) during the 3-month period prior to both the baseline and follow-up assessment was calculated.

Side effects

The frequencies (%) of comorbid physical illnesses and participant falls were summarised by randomised group at baseline and 3 months.

Primary outcomes

Secondary outcomes

Potential outcomes for the main trial

Referral rates

An average of four potential participants were referred by the memory clinics weekly.

Follow-up rates

Of the 62 randomised participants, 57 (92%) were followed up at 3 months. The number lost to follow-up was 4 out of 42 participants (9.5%; exact 95% CI 3% to 23%) in the intervention group and 1 out of 20 (5%; exact 95% CI 0.1% to 25%) in the TAU group. Two people with dementia withdrew consent, two carers were uncontactable and one person with dementia died (see Figure 3).

All psychotropic medication prescription

Table 4 shows the number and frequency of psychotropic medication at baseline and 3 months, with 45% of each group receiving at least one psychotropic medication.

Table 5 shows the number of patients in each group who at the 3-month follow-up had a prescription for psychotropic medication or melatonin that had not been prescribed at baseline. The odds ratio of being prescribed at least one of these medications in the intervention group compared with the TAU group was 0.51 (95% CI 0.06 to 4.01; n = 55).

Reported comorbid physical illnesses and side effects

Table 6 shows comorbid illnesses at baseline and comorbid illness and possible side effects at 3 months. At baseline, both groups reported similar rates of all symptoms. There is no clear pattern of difference in change over time between the two groups.

Fidelity

One randomly selected intervention session was recorded for 34 out of 40 (85%) participants who started the intervention. Three participants refused the audio-recording, two did not continue the intervention up to the session randomised to be recorded and one recording was partial and could not be scored. Across all sessions (Figure 4), managing the carer’s concerns and keeping the carer engaged in the session was rated 5/5 (IQRs 4.00–5.00 and 4.25–5.00, respectively), while keeping the carer focused on the manual and keeping the session to time was rated 4.00 out of 5.00 (IQRs 4.00–5.00 and 3.25–5.00, respectively).

FIGURE 4.

Median scores from fidelity monitoring of all sessions of intervention delivery.

Table 7 shows the strategies that the 37 carers who attended session 6 wrote down as their plan to use in future. Most of them intended to use several strategies.

Completion rate and results of validated interviews at baseline and 3 months’ follow-up

Table 8 summarises completion rates and scores of patient- and carer-validated questionnaire scores at baseline. The NPI, PSQI, SCI and HSQ-12 were completed for 61 out of 62 (98.4%) participants. All other measures were completed for everyone. Summary scores are similar between the two groups, although carers in the intervention group had consistently better scores than carers in the TAU group.

Table 9 summarises the patient- and carer-validated questionnaire scores at 3 months. More than 90% of those who were randomised at baseline completed the SDI, HADS, ZBI, SCI and HSQ-12. Over 85% completed the other questionnaires. Summary data indicate generally better scores for the intervention group for both patient and carer measures. After adjusting for site and baseline score, significant improvements due to the intervention are evident for ESS total score, DEMQoL-Proxy and ZBI score.

Analysis of actigraphy data

Table 10 shows the baseline sleep data for those who wore the Actiwatches for seven or more nights and had complete NPCRA data, that is not including 24-hour periods with ≥ 3 hours of missing data. Only one of the 62 randomised participants with dementia did not wear the watch for seven or more nights at baseline. The sleep diary or event markers were used by 50 out of 62 (81%) randomised participants to record the person with dementia’s bedtimes and rise times for actigraphy. Of the remaining 12 participants, eight (13%) gave a verbal report of sleep pattern, including any atypical bed or rise times (e.g. due to medical appointments), whereas four (6%) did not.

Table 11 shows the sleep data at follow-up for those who wore the watch for seven or more nights and NPCRA data with all 24-hour periods with ≥ 3 hours of missing data excluded. Of the 57 people with dementia who were randomised and still in the trial at follow-up, six refused to wear the watch again; 49 out of 51 (96%) of those who did wear it had actigraphy data for seven or more nights. Carers of 42 (82%) of these participants provided their relative’s bedtimes and rise times via the sleep diary or event markers. Of the remaining 11, eight (16%) provided a verbal report of their relative’s sleep pattern; three (6%) did not.

Sensitivity analysis of actigraphy estimates including only those co-resident with a carer

Table 12 shows the follow-up actigraphy data for the people living with a family carer.

Table 13 shows the correlation between scores on the baseline and follow-up questionnaire and actigraphy measures to contribute to power analysis for a full trial.

Assumptions and sample size calculation

It is envisaged that the main DREAMS trial will retain the same basic characteristics as the feasibility study. It will be a parallel, multicentre trial with two arms and 2 : 1 randomisation to the DREAMS intervention versus TAU and randomisation at the individual level. The primary outcome will be SDI score at 6 months (to allow more flexibility in intervention delivery time). As the intervention is delivered by therapists (with each therapist likely to see approximately 15 people), it is expected that there could be therapist clustering in the intervention group. Analysis will compare groups allowing for this clustering and making adjustment for baseline SDI score. Table 14 shows calculated estimates of the required sample size based on variance and correlation of SDI scores observed from the feasibility work, and plausible ranges for other unknown parameters.

The analysis would compare groups allowing for clustering with adjustment for baseline SDI. We wanted to detect a difference in mean SDI (on the SDI 0–12 scale) scores of 0.6 (effect size 0.3) or 0.8 (effect size 0.4) because a mean difference of 0.8 was observed in the feasibility study, although there was a much larger difference in medians as data are skewed. SD estimated from baseline feasibility data was 2.24. Correlation between baseline and follow-up measurements was assumed to be 0.57 from feasibility study estimates.

Intracluster correlation (ICC) could be between 0 and 0.08 (0.08 indicated by upper confidence limit from START trial. This was used as the feasibility study was too small to provide a useful estimate.

We will aim for 2 : 1 randomisation (achieved by calculating unequal sample sizes before adjusting for clustering and inflated by the design effect to achieve a 2 : 1 ratio), with 90% power and significance of 0.05. Inflation for a non-parametric analysis is included for the case of data being non-normal (feasibility study data indicate that this may be the case for SDI). This calculation should provide a conservative estimate of the sample size needed in the case when analyses are based on transformed data (e.g. log or square-root transformation, which may be appropriate). 79

Demographics

Sixteen carers who received the DREAMS START intervention were interviewed, 15 of whom had completed all six sessions and one who had completed only two. The demographic characteristics of the carers are displayed in Table 15. The 16 carers included two paid carers and 14 family carers. The majority of carers were women and they were from a diverse range of ethnicities. Most spoke English as a first language (n = 11). Their ages ranged from 22 to 87 years (median 53.5 years, IQR 48.5–64.0 years). Thirteen carers lived with the person with dementia, one of whom was a paid live-in carer Monday–Friday; the other three carers comprised two family carers who were non-resident and a paid carer who worked shifts as part of a team of carers. The qualitative findings are presented in three main headings: (1) the acceptability of the assessment tools and intervention, (2) aspects of the intervention that the carers found helpful and (3) the acceptability and practicality of the intervention.

Acceptability of the assessment tools and intervention

Actigraphy

Carers discussed both the person with dementia’s experience of wearing the watch and the feedback they received about the person with dementia’s sleep during the intervention sessions and after completion of the trial.

Intervention overall

Almost all the carers interviewed spoke positively about the intervention overall:

I found it really quite empowering. And with someone who suffers with dementia, or a family member of someone [who suffers with dementia], everything is just so bleak and, sort of, grey. And you’re dealing with so much that actually having some idea of clarity and putting things into perspective and knowing that you can, [that] there are some things that you can do rather than all being completely unmanageable.

C7; live-out daughter

Only one participant, who dropped out of the intervention, reported that it was not helpful for her relative:

I have been on a course [about] how to deal with dementia in the past . . . So in a way I didn’t find that very beneficial to be honest with you.

C5; co-resident daughter

The positive feedback mentioned how the intervention was enlightening, interesting, well-designed and provided a new perspective. Six of the interviewees planned to use the manual in the future:

It was really enlightening; it was, on a personal level, it was very, very . . . I found it to be really useful because there were things that I would overlook that I didn’t realise were that important.

C7; live-out daughter

No, I think, really, it was just the total thing . . . was just interesting for us because we’d never have even thought about any of these things before.

C8; co-resident husband

Look, I got nearly 25 or 30 certificates. And I think this was the most complete. The most useful, the easiest. And compared to the others, I think there’s nothing to add to this one . . . very well designed.

C9; paid carer

The relevance of the intervention was discussed, and participants highlighted how the relevance varied according to the individual presentation of their relative. Even when participants did not find some parts of the intervention relevant, they found other parts useful and judged the overall intervention to be interesting and useful:

But anyway, we were slightly difficult in our case because we’re both quite active, you know, and we were already doing a lot of the things . . . so some of it didn’t mean an awful lot to us. But, I mean, overall, somebody’s done an awful lot of work, putting all this together and, you know, this package is quite useful, I think.

C8; co-resident husband

I think the DREAMS programme was quite interesting. Even though some of it didn’t apply to my mum. A lot of the things were for the ones that move about, you know, are more mobile and everything. But some of the points were quite useful, you know, for me to use on my mum.

C10; co-resident daughter

Manual design

Participants found the manual layout user-friendly and easy to read, considered the balance of pictures, vignettes and text appropriate and felt the direct quotations from carers to be important. Two people wanted more examples and quotations:

It was well explained, it was in plain English, it was all understandable.

C13; live-out daughter

Pictures are always good . . . but I think more importantly is the case studies, you know, and anecdotes, yes, that’s more important.

C1; co-resident daughter

One carer would have liked an electronic version of the manual and two would have liked to be able to use it on their mobile telephone:

It would’ve been better if the paperwork was more electronic, that way it doesn’t get lost.

C12; live-in granddaughter

It might’ve been easier for me to just talk into my phone at the end of the day or, you know, to keep an audio diary.

C2; co-resident wife

Some application could be developed that would send you a message.

C5; co-resident daughter

Another carer suggested organising the manuals with tabs for reference to make it easier to find things.

Content of the manuals

Participants found the content of the manual relevant and useful:

In my mind, there isn’t anything that I think you’ve missed or anything different or that could be improved on. I think you covered an extensive range [of] matters in terms of relaxation, breathing, sleep, and understanding, all of those things.

C11; co-resident daughter

I found it very good, I didn’t find anything boring about it; it was well calculated, and well put down on paper. I found every part of it very imaginative.

C3; co-resident husband

Participants also liked receiving information about caring for someone with dementia, the biological processes of sleep and how dementia can affect an individual’s sleep. The diagrams used in the manual to describe circadian rhythms and sleep pressure were described as easy to understand for carers, but not for people with dementia. Having this specific information clearly explained helped the carers to make sense of sleep difficulties in dementia and justified making behavioural changes, such as increasing natural light:

The sleep cycle, REM [rapid eye movement] and all of this, and that it’s in cycles, it isn’t just one cycle that lasts the entire night . . . I think it’s useful to know that, actually, if I, you know, if I do this it affects this and that’s why then it has a knock-on effect.

C1; co-resident daughter

I really liked the fact that how sleep and dementia are correlating with each other, I found it very helpful. Because with people that have dementia, as their mind is slowly becoming less integrated, they’re very more sensitive as to how it can impact their sleeping. Because if you have dementia . . . you don’t have a tough shell against having a routine which can change dramatically, like myself.

C12; co-resident granddaughter

Now I’m really aware that being out in the conservatory is really good for her because of the light. Whereas before I thought, ‘oh, it’s better because it’s a bit oppressive in the front of the house, a bit dark’ . . . But now I know that there is a physical reason for it, it’s not just, you know, a hunch.

C7; live-out daughter

Sleep diary and record forms

One aspect of the manual that received mixed feedback was the sleep diary, designed for participants to use between sessions to record sleep patterns and strategies tried. Completing a sleep diary was described as useful both for understanding the sleep patterns of the person with dementia and in enabling carers to provide further details to others, such as health professionals:

The diary, the timesheet, everything you know because it helps me to learn, it helped me to be able to see exactly what’s going on. And I can give exact times to people now as to when she’s awake whereas before I might’ve missed it.

C11; co-resident daughter

Although some carers found the sleep diary easy to complete, a number of carers spoke about the difficulties of completing it. They attributed this to the design of the diary and the added stress of record-keeping, especially when the person with dementia awoke in the night:

But it was a nightmare because of the ways, yes, because it goes from noon . . . sometimes if you forget also to do it you can think, ‘oh, you know what we didn’t . . . let’s leave yesterday, let’s just do today’ . . . but by the time the next day came it just got really, yes, really confusing so we just left it.

C7; live-out daughter

But it was crazy because when you try to sleep and she’s woken up you have to memorise the time, or otherwise you have to leave the bed and go and write it. Next day you think, what time was she awake? It was 3? It was 2?

C9; paid carer

One suggestion was to change the timings of the sleep diary to start recording at an earlier time of day, instead of noon, and to use day and night symbols:

I don’t know what time would be good, maybe 6 a.m., you know, so you definitely knew that that was Tuesday, all that line.

C2; live-in wife

Other record forms within the manuals were described as easy to use and useful.

Process of intervention delivery

Carers valued a number of aspects of the process of intervention delivery. These included, first, having the space to talk; second, developing individualised strategies; and, third, being flexibly guided through the intervention.

Using the light box

Of the 16 participants interviewed, 14 had tried to use the light box to increase their relative’s light during the intervention. Only three of them were using the light box every day at the time of the interview:

Every morning he asks me to switch on the light, you know . . . he totally seems to count on the light.

C14; co-resident wife

However, five of those not using it every day spoke about using the light box again, particularly in the autumn or/winter when there is less light:

Well, we didn’t use it after, sort of end of March really, mid-March because she was out so much then and the nights were getting lighter. But certainly I think it was February time for a period of about 3 weeks or 4 weeks that we did we just put it on while she was in the room . . . It probably did help; it did because she wasn’t going to sleep so much during the day.

C7; live out daughter

The others spoke of their relative disliking the light because of its brightness or because it did not seem to be working:

She didn’t want it on a lot of the time; she said it was too bright.

C13; live-out daughter

My son said, you know, it’s making their nan sleep more.

C10; co-resident daughter

Increasing natural light

Five people spoke about trying to increase natural light, often indoors, by sitting nearer to windows or opening curtains, or spending more time outdoors:

We changed the furniture, so she was near the window. So she has more light.

C9; paid carer

What we still try and do is take my mother out daily. So still trying to give her some exercise. Still to get her out into the sunlight for at least half an hour.

C15; co-resident son

Carers had mixed opinions on the effects of increasing natural light; some felt that it aided sleep, whereas others felt that it had no effect:

Yes, she’s more settled when she goes out and she got more light. And then she becomes more tired and she sleeps better.

C9; paid carer

Getting my mom up in the morning, lots of daylight, it didn’t really seem to have that much effect.

C6; co-resident son

Increasing exercise

Practising the relaxation exercises

Although many of the interviewees enjoyed the relaxation exercises within the sessions, some felt that the exercises were not for them and struggled to practise between sessions. Carers were more likely to use elements of the exercises that did not require the CD, such as the breathing exercises, or used it as a reminder to take a moment to relax and notice when they were feeling stressed:

I try and incorporate the breathing when I’m feeling a bit, like today I thought, ‘ugh, God, I just haven’t got the patience today to deal with mum’s agitation’ . . . but it’s, like, you do it, for that moment of time you’re relaxed and then it’s, like, you have to go back into it.

C1; co-resident daughter

One carer also described using the imagery relaxation exercises with their relative, adapting it to make it more relevant and briefer for use over the telephone:

I did try and get my mum using that kind of technique. Imaging herself in a particular environment. I changed it sometimes to the Caribbean. To somewhere that she is familiar with . . . Sometimes I’d speak to her over the phone and try to describe . . . an imagery to think of that she’d like that made her feel happy.

C13; live-out daughter

Carer looking after themselves

Carers spoke about how the session around looking after themselves was important, but that they needed more in-depth opportunity to address more long-standing patterns and habits:

What it did do was it stopped you and it said, ‘hang on a minute, you know, this is a management programme it’s not just about the person, it’s about the family member and looking after . . .’ but actually the fundamentals of looking after yourself, you know, when it comes to practice it’s difficult.

C7; live-out daughter

I think it’s absolutely key, you know, it’s probably the most important aspect of the whole thing, but also very, very difficult because if one’s spent, you know, nearly 70 years of one’s life practising, you know, all the buzzing.

C2; co-resident wife

Changes to the bedroom environment

Changes to the bedroom environment covered a range of topics, from making the room more comfortable to adding more personal objects so that, when the person living with dementia woke up, they recognised that they were at home. Mostly carers felt the bedroom was too warm, which some confirmed using the thermometer provided:

I think that is part of this really just to try and keep things in order in her bedroom, so when she goes to bed she doesn’t have to go to bed in chaos.

C7; live-out daughter

And I did put the heating, central heating, on the timer. So it came off a bit earlier, so that it was slightly cooler. And yes, she was saying she was freezing . . . You know, and it was just mild.

C13; live-out daughter

Some participants had, before the intervention, thought about changes that they could make to improve the bedroom environment, such as blackout curtains, either themselves or with help from their memory service. For others, changes to the bedroom environment were not possible: for example, one person with dementia had never liked having their bedroom curtains closed at night, and their carer felt that this was impossible to change:

Just before we started this I actually got blackout curtains, which had already helped a bit to make her sleep a bit.

C6; co-resident son

Changes to bedtime routine

The changes made to the bedtime routine encompassed making the wind-down period start earlier, trying to decrease stimulating activities before bedtime and changing the time the person with dementia went to bed, which seemed to improve sleep:

I would phone my mum sometimes about 10 o’clock just to say goodnight and I’ve just put that a bit more forward now so I phone her at 9 o’clock. And then that 9 to 10 o’clock is a sort of period of winding down.

C7; live-out daughter

Trying to keep her up as long as possible so she’ll sleep better through the night. All those sort of things, they did help, because obviously when she goes to bed early she will get up early.

C15; co-resident son

Some discussed how the bedtime routine was fixed because their relative lived alone and depended on paid carers, or because they thought changing the routine may be too confusing for their relative.

Managing night-time behaviours

Strategies for managing night-time behaviours included thinking more about the causes of behaviours and carers reacting differently to being disturbed in the night:

I’ve now figured out, look, she’s not going back to bed because it’s not right, she doesn’t know what to say and what to do. And she’s actually started coming into my room and at first I couldn’t understand why. Now I know she’s trying to tell me something’s wrong . . . So I know now to just go in and, I have a spare, and I can have mom’s bed changed so that she’s back in bed and asleep in 15 minutes flat.

C6; co-resident son

One carer also adapted the home environment by putting signs on bathroom and bedroom doors, which helped their relative find their way to and from the bathroom in the night, and prevented them wandering around the house in the night.

Lifestyle changes

The main lifestyle change mentioned was reducing the consumption of caffeine from the late afternoon onwards, and especially before bed, which most people did. However, this was not always successful. Another carer discussed how her relative drank caffeine at night but did not think that had any impact on his sleep:

We talked about having a snack before bed and cutting out caffeine, which I tried, again, for a little while, but we seem to have slid back to . . . Well, P does have, does actually drink decaffeinated tea now, which he didn’t before . . . But he likes a particular form of coffee, which is very sweet and has caffeine, and I haven’t been able to wean him off that.

C2; co-resident wife

Reducing daytime naps

Carers used a number of the strategies, such as going out and increased activity, to try to reduce daytime napping in the person with dementia. Some of the carers had some success with these strategies, but other carers discussed finding it an ongoing struggle to reduce naps:

But a couple of, like, past photos, family photos and everything, you know, like we recorded and had on a CD . . . A few other videos, places that she was familiar with, like her village, where she comes from, that kept her awake a few times.

C10; co-resident daughter

If you wake her up she’ll get really angry, so my dad phones me and he says, ‘she’s been asleep for about 45 minutes, can you phone?’. So, he’ll put the phone right next to her to wake her. So I’ll phone up, she’ll pick it, I’ll say, ‘hi Mum, how are you? What you doing?’ you know, and I’ll try.

C7; live-out daughter

And it’s impossible if she wants to go to sleep, she will go to sleep, no matter what you do. Napping, you know.

C4; co-resident daughter-in-law

Overall benefits for carer and person with dementia

Overall, nine of the carers interviewed spoke about their relative sleeping better after taking part in the intervention, but five of them were unsure whether or not it was due to the intervention or other reasons, such as medication or progression of dementia over 3 months:

You’ve created a miracle, because it really did work.

C11; co-resident daughter

For years, she’s been very bad at sleeping and it seems to have improved quite a bit, but whether that’s anything to do with this course, I don’t know.

C8; co-resident husband

Mum was put on mirtazapine and we started the research. So, it is very difficult to say what . . . where the benefit came from . . . And again I don’t know whether it’s the mirtazapine. She’s actually gone down on her mirtazapine from 30 to 15 [mg] and her sleep is still OK.

C7; live out daughter

Some attributed the improvements to a changed routine:

She definitely stays upstairs in her room during the night for longer. She came less times downstairs to eat or just listen to the radio, as she used to . . . Yes, it’s a better routine.

C9; paid carer

One carer described how the intervention improved not only the sleep but also the mood of the person she cared for; this person with dementia had taken part in the sessions with the carer:

But I have to say he enjoyed it every week, he looked forward to it. And he seemed to think it was making him better. And his mood, everything changed and, touch wood, he’s still the same.

C14; co-resident wife

In terms of benefits for carers, two described how the intervention had also improved their own sleep and two spoke about feeling calmer from the skills that they had learnt from the intervention:

I really was and even people at work noticed a difference with me. It was like, you seem full of life. I’d say it’s because I’m sleeping at night now . . . The programme as I saw it for us, for my mum and for me, worked really well.

C11; co-resident daughter

I’ve learnt to be calmer though and more patient by sort of . . . To stop thinking about it and think of something else.

C10; co-resident daughter

One carer also mentioned the feeling of altruism she felt from taking part in the research:

And I think there’s something, it sounds a bit weird, but there is something comforting about having done this, having been involved in something that is hopefully going to lead to some helpful outcomes for a lot of people, so I think that in itself is good.

C2; live-in wife

One carer of a relative with very severe dementia felt that the intervention had a limited effect:

I’m not saying don’t offer it to them [people with severe dementia] . . . but if you find that, actually, it’s more helpful in people with mild to moderate dementia, you might feel that it’ll be of more benefit and you’ll be able to reach more people within those categories, and help more people as a result.

C1; co-resident daughter

Acceptability and practicality of the intervention

Recruitment and follow-up

The referral rates were four potential participants per week from two memory clinics, which informs the number of trusts required for a full trial. We used JDR and recruited two people over the recruitment period. There was clearly more potential, but 19 out of 25 (76%) of those registered on the list did not respond to contact and only two of those we contacted consented. Thus, it seems that JDR is most useful as a supplementary method to recruit people rather than as the sole or main method. Generally, ≥ 80% follow-up is regarded as satisfactory, and this trial achieved 92%.

Completion of outcome measures

Very high completion rates of the validated questionnaire measures were achieved, using carers as informants. At baseline, all but one person with dementia wore the watch and 50 (81%) carers completed the sleep diary or event markers to record the person with dementia’s bedtimes and rise times. However, at follow-up 49 (79%) of the 62 people with dementia in the randomised group had ≥ 7 days of actigraphy data and 42 (82%) of their carers provided the bedtimes and rise times on the sleep diary or by using event markers to aid interpretation.

Actigraphy as outcome and tool

It was originally envisaged that actigraphy data would be the primary outcome in a full trial. However, in the feasibility study, it was available for only 79% of randomised participants at follow-up (for ≥ 7 days; the usual ‘gold’ standard). In addition, the sleep diary or event markers, which give times of going to bed or getting up, are required to interpret the data. Only 68% of people randomised provided this, so data interpretation was difficult for some of those with records, as it relied solely on verbal reports or the researchers’ supposition. Therefore, our feasibility study indicates that we are unlikely to have the level of reliable data required to use actigraphy as a primary outcome. The research team suggest that it would be helpful to validate the use of Actiwatches in this population before considering actigraphy measures as a primary outcome in any efficacy trial.

There is, in addition, a paucity of validation data about actigraphy in this population. Although sleep efficiency is often taken as a good summary sleep measure from actigraphy in younger people, it appears not to be as well measured in older people possibly because sleep in actigraphy is inferred from time in bed and movement. Movement is, in contrast, directly measured. According to the actigraphy data, both groups were still spending a long time in bed after the intervention and it did not look as if the intervention changed this sleep behaviour in any substantial way. There was an indication from our actigraphy results that the intervention participants were more active during the day and less active during the night. These findings accord with the qualitative feedback from the carers in the study as well as the results of the validated instruments. The carers in the study and the PPI group judged the important outcomes to be that the person with dementia was less restless during the night, more awake during the day, disturbed them less during the night-time and seemed happier (which are measured in the SDI). In these circumstances, they were unsure that these measurements of sleep added additional outcome information, or were accurate.

In contrast to their disappointment with actigraphy as feedback, the carers and therapists found the information from actigraphy valuable to use at baseline to consider the rest–activity pattern and help make a plan and this would continue to be incorporated in the manual as part of the intervention in a full trial.

Overall, the place of actigraphy in a full study would, therefore, be to measure changes in daytime and night-time activity as a secondary outcome and as part of the intervention.

Validated interview measures

In the qualitative assessment participants commented on the combined length of the questionnaires, and felt that this was acceptable if participants need to be reassured that this information is being collected for the purposes of the research and will be useful. The completion rate for all validated instruments at baseline was very high, ranging from 98% to 100%.

Instruments for the person with dementia

The validated instruments to measure sleep disorder had a high rate of completion at follow-up, and the SDI was completed at follow-up by 90% of those initially recruited to the trial. This appeared to be the most practical way to measure sleep. The qualitative feedback from carers participating in the trial and PPI indicated that they felt that it reflected their experience of sleep as well and had relevance. Therefore, it is proposed as the outcome in a main trial.

Summary data for the carer-reported instruments indicated generally better scores for the intervention group and, after adjusting for site and baseline problems, there was significant improvement in daytime sleepiness and in quality of life of people with dementia, despite the small numbers in the study. There was also no increase and a possible reduction in the numbers of people who were prescribed at least one medication for sleep. These may be related to each other. Although power for these results had not been considered, the consistency in the direction of all the results suggests that they may be real. It is important that the intervention may reduce daytime sleepiness, whereas sedative medication sometimes increases it. Similarly, it is always important to consider quality of life for someone with dementia, as an intervention may improve a specific domain while reducing overall quality of life. We would therefore conclude that the DEMQoL-Proxy and ESS should be measured in a full trial. We also administered the CSRI, to assess the feasibility of its use, and suggest that it too should be included in a full trial, as this would enable the cost-effectiveness of an intervention to be calculated.

Harms

There was no clear increase or difference between groups in the use of psychotropic medication and melatonin, with the intervention group possibly being prescribed slightly fewer medications afterwards, taking into account baseline measures. This suggests that any effects were not due to psychotropics being used as rescue medication in the intervention group. There was no indication of important harms in terms of side effects in either group.

Carer outcomes

The carers in the intervention group reported significant improvements in ZBI-measured burden, and the direction of results suggested that there may be an improvement in depressive symptoms (HADS). It is important to consider the effect of sleeplessness on carers’ stress levels and mental health, and we would conclude that these measurements should be retained for a full trial. A few of the carers disliked the questionnaires used to measure carer sleep as they found it difficult to report averages. It is possible that they might be willing to record their own sleep using a sleep diary, attributing any sleep disturbance to disturbance of their relative’s sleep or any other cause. However, this was not tested and the carers were not very positive about sleep diaries for their relatives.

As carer sleep is often disturbed by the person with dementia, it seems worthwhile to measure it. In order to select one measure for future trials, two instruments were used: the PSQI and the SCI. The rate of carers who filled in these questionnaires was satisfactory, with 61 (98%) of the randomised carers completing both at baseline. Fifty-six (90%) completed the PSQI at follow-up and 57 (92%) completed the SCI. The SCI gives the information to consider whether or not criteria for insomnia have been fulfilled, whereas the PSQI does not. Overall, more than half of the carers fulfilled the SCI criteria for poor sleep at baseline. In the qualitative interviews, two carers were of the opinion that the PSQI did not measure relevant features of sleep and that it was too long. They preferred the SCI to the PSQI, and we will go with their preferences.

Additional data

One individual with dementia present during the interview also suggested that it may be important to record whether or not carers and the person they care for share a bed or bedroom as part of the assessments, as this can affect how the questionnaire is answered. This would be added to an assessment for a full trial.

Power for a full trial

To calculate the sample size required for the main study, feasibility estimates were used by calculating the SD of baseline SDI scores (2.24) and the correlation between baseline and 3-month measurements (0.57). As there is no estimate of clinically significant difference, a SD of 0.4 was used, different sizes were calculated considering the CI for ICC from the earlier START study. A full study with 2 : 1 randomisation (which is feasible from the results above) will require 230–296 participants in the intervention arm and 115–148 participants in the TAU arm (assuming an average of 15 participants per therapist, 2 : 1 randomisation and a drop-out rate of ≤ 15%, and including an inflation for the case of non-normality).

Manual design and delivery

Generally carers felt that the number of sessions was appropriate and liked the balance of pictures, vignettes and text. They appreciated the direct quotations from carers. Some thought that they would continue to use the manual itself as well as strategies within it. To make it easier for participants to use the manuals in between the sessions and after the intervention, one participant’s suggestion of tabs will be used.

Topics

Carers liked the information about caring for someone with dementia, the biological processes of sleep and how dementia can affect an individual’s sleep, and found the diagrams helped them to understand why they should make changes, such as increasing natural light and activity, including exercise. They tried different components depending on individual needs. Many of them asked for specific help from relatives, which would continue after the study. They often made adaptations to the environment, for example putting signs on the bathroom door. Some realised that their relative got up during the night for a reason that they were able to address, which helped improve their relative’s comfort and reduce their own sleep disturbance.

At session 6, all carers made a plan for the future and clearly appreciated a multimodal intervention as they wanted to continue using a range of strategies. All intended to persevere with the light box (although not necessarily in the summer) and continue to increase activity or physical exercise. Many intended to carry on with a later bedtime routine, usually going to bed around half-an-hour later than before, and continued to be aware that a comfortable bedroom is important. Most carers became more aware of looking after themselves by giving themselves time, challenging negative thoughts or using relaxation. They felt that this had improved their own quality of life and made them a better carer, and so they would continue with these self-care strategies.

The carers interviewed wanted to ensure that, in future, therapists did not go through all the content of the manual on topics that were not relevant to them, for example sections about increasing activity if the person with dementia was very active. For a full trial, it will be stated at the beginning of the sessions that not all of the topics will be relevant, and, if they are irrelevant, they will be omitted, but participants will still have the information within their manual should it later become relevant (e.g. if their relative’s sleep disturbance or physical health changes, or as their dementia progresses).

The carers felt that using telecare and other assistive technology (which was often part of plans) could be further emphasised in a future trial, as this would be useful for those people with dementia alone during the day and could help manage difficult or risky behaviours.

Some carers mentioned in the qualitative interviews that they would have preferred electronic versions of the sleep diaries that were used for the actigraphy and in the intervention, and found that it was difficult to complete the next day when they had to remember when someone had been awake in the night. They suggested that the timings of the sleep diary be changed to start recording from 06:00 instead of 12:00, and day and night symbols be used to make it a lot clearer; this will be done. The research team believe that it is worth investigating the possibility of, for example, sending an automated message to prompt a carer to report information, such as bedtimes or rise times. This could be done by simply replying to the text or by writing the information in the diary.

Delivery by therapists

The therapists were clinically supervised psychology graduates who were delivering the intervention to varying combinations of carers and people with dementia. Both the participants and the PPI group felt that the therapists were taking a facilitative and supportive approach. This was essential to the therapeutic process, as it was complex for participants to overcome the barriers to helping relatives who were often unable to formulate and remember plans themselves. They valued face-to-face contact and that the strategies were delivered to them and tailored to their individual needs.

Complexity

The intervention was complex and could be individually tailored, but took only six sessions. As sleep problems are complex and diverse, this allowed it to be appropriately individualised.

Fidelity

Fidelity ratings were high, suggesting that different therapists were able to deliver the intervention consistently.

Flexibility

The intervention was offered weekly; the median time taken to deliver the intervention was 7 weeks. Most participants thought it was ideal to have 1–2 weeks between sessions, rather than necessarily aiming for weekly sessions. Therefore, a degree of flexibility in timing would be offered for an intervention in future, allowing for the intervention to take up to 3 months.

Introductions

Thank you for agreeing to take part in this group. My name is . . . . As you know, I am a researcher from UCL and I will conduct and record this focus group. Everything you say is confidential but I would like you to introduce yourselves so that the typist can identify you.

Description of the research

We are in the process of developing a manual, which, over the next year, we are going to be testing out with people with dementia and their family carers as part of a trial. It will be delivered to individuals at home, and we hope that it will help with sleep and lead to improvement in quality of life for people with dementia and their families. The intervention will be based on what we know works from research and practice with people who do not have dementia and on our specialist research and clinical knowledge of people with dementia and their families.

Before we finalise the intervention, we want to hear your views and opinions about what we have developed, and then make further changes to the intervention manual. We are asking you because we know that you have responded to the Alzheimer’s Society invitation as this is something you are interested in and probably have experience of. We want to make use of your expertise.

Sleep difficulties/current understandings

Question (Q). Could you tell us about any sleep difficulties that your relative experiences?

• What do you notice happening at night and during the day?

• What effect do these problems have on you and your relative?

• How do you understand what may be causing these sleep difficulties for your relative?

• Which sleep difficulties do you find most difficult to manage?

• What has worked well in managing sleep difficulties?

• What have you tried?

• What has not worked well?

We are particularly interested in how to make the manual practical and to fit in with people’s lives.

Introduce the manual

Show them the draft manual/structure and present an overall summary.

• People with dementia and their families will receive their own copy of the manual to write in and keep.

• They will wear an actigraph (special watch) to see how their sleep rhythms are.

• There will be practical suggestions to try out between sessions.

• There will be one member of staff working with the person with dementia and their family.

• The manual contains information about sleep and dementia.

• Each session will include a combination of information giving, discussion and making an individualised plan with practical exercises, including increasing light at particular times and increasing activity.

• There will be a relaxation exercise and people will be given a relaxation CD or MP3 files.

• Each session will be around 1 hour.

• The staff member delivering the intervention will be supervised by a clinical psychologist and their general practitioner (GP) and memory service team will know that they are in the study.

The sessions of the DREAMS intervention comprise the following:

• Learning about sleep and dementia.

• Relaxation and bedroom comfort/pre-bed and mealtime routine.

• Using natural light, morning wake-up light.

• Pleasant activities.

• Management of daytime naps.

• Relaxation, especially at night.

• What works? Using skills and strategies in the future.

Give them a few minutes to look through and make any general comments. These prompts are only if not already covered. (Explain that this is only a draft/outline).

Q. What are your initial thoughts on the manual?

Q. What do you think about the design of the manual?

Prompts:

• How do you find the layout?

• Is it easy to read and follow, for example not too much text on each page/fonts/colours?

• What do you think of the pictures and images – more, less, different?

• How do you find the balance of information-giving and discussion?

Q. What do you think about the outline structure and content of the sessions?

Prompts:

• Do the topics and examples fit with what your experience is or has been?

• Is there anything important that you feel is missing?

• Does the order of the sessions make sense?

• Is it pitched at the right level for a range of people?

• Is it easy to understand?

• Do the key points stand out?

Q. Do you have any other feedback regarding layout, design, content, etc.?

Q. What do you think might make it harder for people to participate in the sessions and use the manual? (Prompt about wearing actigraph, using light box, increasing activities.)

Q. What do you think might make it easier for people to participate in the sessions and use the manual? (Prompt about wearing actigraph, using light box, increasing activities.)

Q. Before we finish, is there anything else you would like to mention that we have not already covered?

Thank you for taking part today.

Introductions

Thank you for agreeing to take part in this group. My name is . . . . As you know, I am a researcher from UCL and I will conduct and record this interview. Everything you say is confidential but I would like you to introduce yourselves so that the typist can identify you.

Description of the research

We have now completed the feasibility and pilot study of the DREAMS START intervention and wanted to meet with you all today to discuss our findings and any potential changes to the manual in the future. Before we begin talking, we will briefly tell you about the final version of the manual we used in the study and how we delivered it and then XXX will tell you about how the study has gone and what the people who received the intervention told us about how they found it.

Following our discussion today and based on the findings of the study so far, we will probably make further changes to the manual and how we deliver it, which we will hopefully test in a randomised controlled trial in the future. Therefore, we want to hear your views and opinions about what we have developed and tested and how it could be improved before further testing.

Revisiting the manual

Show them the manual used in the study and present an overall summary (present summary slides of content and process).

• People with dementia and their families received their own copy of the manual to write in and keep.

• They wore an actigraph (special watch) to see how their sleep rhythms were.

• There were practical suggestions to try out between sessions.

• There was one member of staff working with the person with dementia and their family.

• The manual contained some information about sleep in dementia.

• Each session included a combination of information-giving, discussion, looking at the results of the sleep actigraph from the week before and practical exercises, including increasing light at particular times and increasing activity.

• There were relaxation exercises and people were given relaxation CD or MP3 files.

• Each session was around 1 hour.

• The staff members delivering the intervention were supervised by a clinical psychologist, and the participant’s GP and memory service team will know that they are in the study.

The sessions of the DREAM intervention include the following:

• Learning about sleep and dementia.

• The importance of routine pre bed and at mealtimes.

• Relaxation and bedroom comfort.

• Using natural light, morning wake-up light.

• Pleasant activities.

• Management of daytime naps.

• Relaxation, especially at night.

• What works? Using skills and strategies in the future.

Give them a few minutes to look through and make any general comments. These prompts are only if not already covered.

Q. What are your initial thoughts on the manual we used?

What do you think about the design of the manual?

Prompts:

• How do you find the layout?

• Is it easy to read and follow, for example not too much text on each page/fonts/colours?

• How do you find the quotes? Prompts relevance, length.

• What do you think of the pictures and images – more, less, different?

• How do you find the balance of information-giving and discussion?

Q. What do you think about the content of the sessions?

Prompts:

• Do the topics and examples fit with what your experience is or has been?

• Is there anything important that you feel is missing?

• Is it pitched at the right level for a range of people?

• Is it easy to understand?

• Do the key points stand out?

Q. What do you think about the structure of the sessions?

Prompts:

• Does the order of sessions make sense?

• What do you think about the length of each session?

• Is there anything else we could do to make it easier to understand?

• What do you think about having relaxation and homework in each session?

• Do you think it is better to give out the manual session by session or all at once?

Brief presentation of findings to date:

• What are your first thoughts on the findings?

• Did anything surprise you that you heard?

• What do you think we could do to address some of the barriers raised? (Put up slide about what made it harder?)

• Do you have any suggestions about how to make it easier for people to make use of DREAMS START? (Put up slide about what made it easier?)

• Based on what you have heard today, do you have any other suggestions about changes we could make to the design/content/structure and process the sessions?

Q. Before we finish, is there anything else you would like to mention that we have not already covered?

Thank you for taking part today.

Before starting the recording

Give the carer a case report form to look at, a copy of the manual and the DREAMS START session card. Let the participant look at these for a few minutes.

Introduction of researchers

Thanks . . . . My name is . . . (As you know) I am a researcher from University College London and will be recording this interview. I am interested in your opinion about and experience of the DREAMS sessions. Everything you say is confidential, but I would like you to introduce yourselves for the recording so that the typist can identify you.

Description of research topic

We are asking you because you have tried out the sessions. We want to know about your experience and what was good and what was less good and how it has made a difference to you and your relative.

As a reminder, here’s a card listing the six sessions:

• Session 1: Understanding sleep and dementia.

• Session 2: Making a plan.

• Session 3: Daytime activity and routine.

• Session 4: Difficult night-time behaviours.

• Session 5: Taking care of your own sleep.

• Session 6: What works? Using strategies in the future.

You can use the manual too, if having your copy on hand is helpful.

Q. Can we start by hearing about how you found the intervention in general?

Prompts:

• What did you like best about the sessions?

• Was there anything that you did not like about the sessions?

• Is there anything important that you feel was missing/not covered?

• What were the key points that stood out for you?

• How did you find the length of the sessions? How about the number of the sessions? Were the sessions spaced apart enough?

• What aspects of the sessions/manual have you continued to use? Have you gone back to the manuals since the intervention stopped?

• What difference has the intervention made to you and/or your relative during the day? How about during the night?

Q. What did you think about the content of each sessions?

Prompts:

How did you find . . .

• Increasing natural light.

• Using the light box (and are you still using the light box now?).

• Increasing daytime activity/physical exercise.

• Making a new routine/strengthening the link between bed and sleep.

• Bedroom and lifestyle changes to improve sleep.

• Managing night-time behaviours.

• Looking after yourself/challenging unhelpful thoughts.

• Relaxation.

Is there anything else you would like to mention regarding content?

Q. How did you find the manual itself?

Prompts:

• How did you find the layout?

• Was it easy to read and follow, for example not too much text on each page/fonts/colours?

• What did you think of the diagrams and pictures – more, less, different?

• Were the sleep diary/record forms easy to use? How could we improve them?

Q. How did you find the plans and trying things out between the sessions?

Prompts:

• What made it easier for you to try things out between the sessions?

• What made it harder for you to do the tasks between the sessions?

• Did it make a difference trying things out together?

Q. How did the intervention fit with the sorts of sleep difficulties that your relative has been experiencing?

Q. How did your relative find wearing the watch?

Q. What did you think about the feedback from the letter comparing the difference in sleep from baseline to 3-month follow-up? What other information would you have liked to know? For example, about naps, if time to bed or rise has changed.

Q. What did you think of the initial and follow-up assessments, particularly the measures (questionnaires) we used?

Q. Before we finish, is there anything else you would like to mention that we have not already covered?

Thank you for taking part in the study and meeting me today.

Standard operating procedure: processing Actiwatch data