An Occupational Therapy intervention for residents with stroke-related disabilities in UK Care Homes (OTCH): cluster randomised controlled trial with economic evaluation

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 08/14/30. The contractual start date was in September 2009. The draft report began editorial review in April 2015 and was accepted for publication in September 2015. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Catherine Sackley is Deputy Chair of the Health Services and Delivery Research Researcher-led Board; Caroline Watkins is on the Health Technology Assessment Commissioning Board; and Keith Wheatley is on the Health Technology Assessment Clinical Evaluation and Trials Board.

Copyright statement

© Queen’s Printer and Controller of HMSO 2016. This work was produced by Sackley et al. under the terms of a commissioning contract issued by the Secretary of State for Health. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

Prevalence of stroke in the UK

The population of the UK and elsewhere is living longer. The average lifespan since 1960 in England and Wales has increased by 10 years for men and 8 years for women. 1 One in six members of the UK population were aged over 65 years at the time of the 2011 census,2 and the over-85-year-old age bracket is the fastest growing sector. 3 It is predicted that, by 2031, 22% of the population will be over 65 years old. 4 The incidence of stroke increases significantly with age. 5 According to British Heart Foundation statistics released in 2012, there are approximately 152,000 strokes in the UK5 and approximately 65,000 people experience their first transient ischaemic attack (TIA) each year. 6

Survival rates following stroke have improved significantly over the last 20 years because of medical advances in acute care and increased public awareness of stroke symptoms. However, owing to the decreasing levels of stroke mortality there is a significant rise in the number of people living with stroke-related disabilities.

In 2012, it was estimated that there are 1.2 million stroke survivors living in the UK. 5 Stroke represents the third most common cause of disability-adjusted life-years worldwide. 7–9 The disabilities experienced as a result of stroke are complex,10 potentially involving a multitude of physical and mental impairments, including difficulties as detailed in Box 1.

Approximately 10% of patients are discharged from hospital directly to a long-term care facility,11,12 and 25% of stroke survivors require long-term institutional care as a result of the brain injury. 13 Clearly, clinically effective and cost-effective health technologies designed to ameliorate disability and improve quality of life for a growing population of older stroke survivors are needed.

Long-term care descriptors

In England, at the end of 2012 there were 4675 care homes with nursing facilities (218,387 beds) registered with the Care Quality Commission, and 12,917 residential care homes without nursing facilities (245,942 beds). 14 The distinction between homes that provide nursing care and those that offer only residential care is dependent on the skills of the care home staff, and not necessarily associated with the level of disability of the residents. Homes that provide nursing care employ qualified nurses, whereas homes that provide residential care are not required to employ qualified health professionals. It is estimated that between 20% and 45% of all people newly admitted to residential care settings in the UK have stroke-related disabilities. 11,15–17 The prevalence of stroke and dementia in the older population suggests a huge demand for long-term care facilities, and the provision of effective health-care technologies within those facilities, both now and in the future. 18

Health-care services within care home settings

Older people with complex health conditions are the main users of health and social care services. 19 From 1990 to 2010 in the UK there was a significant shift in long-term care for older adults away from geriatric hospitals, and more towards care homes. 20 Long-stay hospital wards benefit from established auditing systems which, prior to recent developments in social care provision, care homes did not. As a consequence, patients living in care homes have been described as ‘living on the margins of care’. 21 During this period there have been a number of initiatives, devised by government, in association with the Royal College of Physicians (RCP), that have introduced care standards to regulate, and improve, health care for the older population, and develop a more integrated service in care homes. 3,19,20,22–24

The National Service Framework for old age presented care standards with three themes: dignity in care; joined-up care; and healthy ageing – promoting exercise and activity, independence, well-being and choice. 19 The joined-up care theme outlined reforms to ensure a comprehensive health assessment is conducted prior to admission to a long-term residential facility, to establish individual health-care needs. 3 The National Service Framework listed standards for four main components central to the development of an integrated stroke service in older age: prevention, immediate care, early and continuing rehabilitation, and long-term support. 3,19

In addition to the National Service Framework, the RCP produced a series of guidelines to enhance the health of older adults in long-term care. 20 A component within these guidelines focused on ‘overcoming disability’ from a therapeutic perspective. 20 The guidelines highlighted the importance of the care home environment, and the use of aids, equipment and adaptations to address disability and improve function in long-term care facilities. The philosophy behind these guidelines was that small increases in functional capacity of older people are deemed to impact positively on quality of life and cost of care. 20 The section on ‘overcoming disability’ concentrates on providing access to resources to improve or maintain functioning in primary activities of daily living (ADL).

For the purposes of this trial, primary ADL are referred to as personal or self-care ADL. Personal ADL are defined as:

• mobility

• transfers (e.g. from bed to chair and back)

• using the toilet

• grooming

• bathing

• getting dressed

• feeding.

These initiatives have instigated considerable progress in raising standards, increasing awareness of health-care issues in older age and lessening the long-standing stigma associated with this population. However, despite this progress, care services for older adults with high support needs, such as stroke survivors residing in care homes, are notoriously inconsistent, and most often dictated by financial constraints at a regional level. 25

More residents with a higher level of dependency and complex care needs are being admitted to care homes than ever before. 26 For residents with high levels of support needs, there is more of an emphasis on providing specialist care for a short period towards the end of life to ease suffering and promote dignity throughout. 16 It is critical to design health services with the needs, circumstances and preferences of the service users in mind. 19,27 Establishing an evidence base for clinically efficacious and cost-effective therapeutic health technologies, suitable for use by the NHS in care homes to promote dignity, joined-up care and increased independence is a research priority.

Occupational therapy

Occupational therapy is the therapeutic intervention that promotes health by enhancing the individual’s skills, competence and satisfaction in daily occupations . . . to act on the environment and successfully adapt to its challenges.

Yerxa et al. 28 p. 6

Activity is essential to health and well-being. 29 In the re-drafted report published by the World Health Organization entitled International Classification of Functioning, Disability and Health (ICF),30 the term ‘disability’ is described in reference to the interaction between an individual’s impairments, activity limitations, participation restrictions and their environment. This definition focuses on the individual’s capacity to engage in functional activity.

Occupational therapists aim to improve the quality of life of their patients by attempting to augment functional activity and increase their capacity to engage in personal ADL. 31 A philosophy of occupational therapy (OT) is that the intervention is most effective when it is integrated into the context of the individual. 31 OT typically applies a patient-centred goal-setting approach, so that the therapy package is individualised for each patient, and the goals of therapy are continually reviewed in relation to progress. 32 The treatment programme is planned around the patient’s goals. The patient is given as much autonomy as possible in maintaining or improving his or her own quality of life. Task-specific training, guidance and supervision is given to reinforce safe and effective practice of personal ADL. 33 Where necessary, training involves the use of adaptive equipment (e.g. adapted cutlery or walking aids) to facilitate an increase in capability, ameliorate activity limitations and provide therapeutic aid. Enabling modifications, tailored to individual needs, can be applied to the environment to promote safe and effective practice of ADL (e.g. the installation of bed levers, grab rails or a raised toilet seat). Particular attention is given to communication, to engender an informal atmosphere that will enable the exchange of ideas, and the offering of peer support. In summary:

Occupational therapy is a complex intervention. Practice includes skilled observation; the use of standardised and non-standardised assessments of the biological, psychiatric, social, and environmental determinants of health; clarification of the problem; formulation of individualised treatment goals; and the delivery of a set of individualised problem solving interventions. 34

Historically, OT services within the NHS have been situated in acute hospital services; however, nowadays therapists also operate as a part of local authority social care services throughout the UK. The ‘joined-up care’ initiative has helped instigate service reform to better suit the needs of users in the local community. 3,19 A review of the effectiveness of OT administered by local authority social services to older people at home has shown high satisfaction levels for the service. 34 It has also been suggested that the provision of adapted equipment to reduce dependency on additional services may be cost-effective. 35

Occupational therapy for stroke rehabilitation

The National Institute for Health and Care Excellence guidelines for stroke rehabilitation recommend that OT should be provided for people after stroke to help ameliorate difficulties with personal ADL. 36 The guidelines also stipulate that stroke survivors should be monitored regularly by occupational therapists with core competencies in this area. 36

Occupational therapy delivered to stroke survivors in their own homes has good evidence of benefit. 34,37 A systematic review and meta-analysis were conducted by members of the research team to determine whether or not OT, focused on promoting increased activity and independence in performing personal ADL, improves recovery for stroke survivors. 34 Analysis of nine trials (1258 participants) found that OT increased personal ADL scores, measured using the Barthel Index of Activities of Daily Living (BI). The standardised mean difference was 0.18 [95% confidence interval (CI) 0.04 to 0.32; p = 0.01] in favour of OT, compared with receiving no intervention or usual care. This equates to a single-point difference (5%) on the BI (20-point scale). Furthermore, for every 100 people who received OT after a stroke, 11 (95% CI 7 to 30) would be spared a poor outcome, defined as death or deterioration in abilities to perform ADL [odds ratio (OR) 0.67, 95% CI 0.51 to 0.87; p = 0.003]. 34 The review concluded that targeted OT should be available to everyone who has had a stroke, to reduce disability and increase independence in performing personal ADL.

Although the systematic review concluded that OT is effective when administered in patients’ homes,34 the clinical efficacy and cost-effectiveness of OT administered to stroke survivors living in a care home setting was not known. Differentiating stroke survivors who live in their own homes from stroke survivors residing in care homes is important. Typically, stroke survivors living in care homes have increased physical and mental limitations as a result of their brain injury, and their functional capacity to perform personal ADL is often restricted. For instance, 78% of residents in a care home have cognitive impairment, 76% need some form of assistance with ambulation and 71% are incontinent. 17 Reduced functional capacity may limit stroke survivors’ ability to engage in, and respond to therapy. As a result, generalisation of results from community studies to care home settings should be treated with caution.

Occupational therapy for care home residents with stroke-related disabilities

In the Netherlands, 93% of care home residents regularly receive some form of OT;38 however, in the UK it is available to as few as 3–6% of residents. 39,40 An audit of over 1000 residents in England found that none had been assessed for OT. 41 The most recent stroke guidelines recommend reducing nationwide variability in rehabilitative care after stroke,42 including the care home setting. 43 Owing to the number of patients transferring directly from hospital to a care home environment following a stroke, as opposed to returning home,13,44 it is necessary for rehabilitation and social care services to achieve equivalent standards, especially for those patients with increased dependence.

Following admission to a care home, stroke survivors’ health state typically follows a downward trajectory. Observational data suggest that care home residents spend 97% of their daytime hours sitting inactive with eyes open or eyes closed. 45 Inactivity in older care home residents can pose further health risks, such as pressure ulcers, joint contractures, pain, incontinence and low mood. 46 The provision of OT as a means of augmenting levels of functional activity may reduce the likelihood of these further health conditions and reduce unnecessary dependence.

Current evidence evaluating the efficacy of OT across the whole care home population, not restricted to residents who have experienced a stroke, has shown conflicting results. 47,48–51 The evidence relating specifically to stroke survivors living in care homes is extremely limited. A systematic review considering the efficacy of providing OT to stroke survivors living in care homes was conducted by our research team. 52 Literature searches were performed within: MEDLINE, EMBASE, the Cumulative Index to Nursing and Allied Health Literature, the Cochrane Central Register of Controlled Trials, six trials registers and 10 additional bibliographic databases (all searches ended in September 2012). 52 The review process revealed only one relevant randomised controlled trial (RCT) conducted to date; this was the OT intervention for residents with stroke living in UK Care Homes (OTCH) cluster randomised Phase II pilot trial discussed in full in Chapter 2. No firm conclusions of efficacy of OT provided to care home residents with stroke-related disabilities could be drawn from the systematic review. 52

Depression in stroke survivors residing in care homes

Symptoms of depression are common in older residents residing in care homes,53–57 and very common following stroke. 58 Experiences of depression following stroke may be directly attributable to the brain injury or an adverse psychological response to trauma. 58 Communication problems following stroke limit residents’ ability to express feelings of low mood and may be difficult to recognise by unqualified members of staff. 59 Presence of depression in care home residents is associated with poor outcomes and increased mortality. 57 Symptoms of depression include:

• losing interest in everyday activities

• finding it difficult to concentrate or make decisions

• feeling worthless, guilty, helpless, hopeless or in despair

• changes in appetite. 58

A previous study, assessing the feasibility of a trial evaluating the effectiveness of an OT programme at reducing levels of depression in a care home population, found no significant effects. 60 However, the trial was not powered to evaluate the efficacy of OT in alleviating in symptoms of depression. The OTCH trial sought to assess the influence of OT on levels of depression as a secondary outcome measure to the performance of personal ADL.

Assessing health-related quality of life in stroke survivors residing in care homes

A central philosophy underpinning OT is that the intervention involves engaging in activities that hold meaning for the individual. The personalised meaning behind the activities is thought to help promote increased quality of life for that individual. 31 Health-related quality of life (HRQoL), assessed according to a number of physical and emotional dimensions of interest,61 can be used to measure the perceived impact of a chronic disease. 62 The purpose of including a measure of HRQoL was to provide an additional multidimensional scale that considers both physical and emotional functioning to evaluate potential effects of the OT intervention not captured by the BI.

Training for care home staff

In the UK, the majority of care for older stroke survivors living in long-term care institutions is provided by the staff of those institutions. 63 During the OTCH Phase I stage, a number of care home staff in one area of the UK were interviewed. 41 From the staff responses, it was evident that none of the care homes was providing aids and appliances effective in reducing physical decline, that is there was a difference between policy and practice. It is therefore doubtful whether or not the RCP guidelines on ‘overcoming disability’ could be universally implementable across the UK. 20 The guidelines highlight the importance of the care home environment and the use of aids, equipment and adaptations to address disability and improve function in long-term care facilities. The results from the Phase I interviews were a strong indication that any development of health services in care homes needs to directly involve the staff who provide the majority of residents’ care.

A later report highlighted several aspects of staff involvement deemed fundamental in establishing a more positive culture in care homes. 64 It promoted the importance of staff training to move away from the prevalent model of task-based care system of doing things ‘for’ residents, and more towards a system with a shared commitment (‘doing with’) that includes emotional care. Consequently, the involvement of care home staff in the evaluation of OTCH was regarded as integral, in order to increase awareness of the broad spectrum of stroke-related disabilities and to provide continuity in care practices between staff and visiting therapists.

Aims and objectives of the Occupational Therapy intervention for residents with stroke living in UK Care Homes trial

Disabilities affecting ADL are commonplace for stroke survivors living in UK care homes, and yet access to rehabilitation services, particularly OT, is very restricted. The purpose of the study was to conduct a Phase III RCT to evaluate the effects of a targeted 3-month course of OT (with provision of adaptive equipment, minor environmental adaptations and staff education) for people with stroke sequelae living in care homes.

The primary outcome measure assessed was the capacity to perform personal ADL.

The secondary outcome measures assessed were mobility, depression and health-related quality of life.

An economic evaluation of the intervention was conducted in parallel with the evaluation of clinical efficacy as part of the health technology assessment. Providing an OT service to stroke survivors resident in care homes was compared against usual care. The trial aimed to evaluate whether or not there is sufficient evidence to advocate the routine implementation of OT for all stroke survivors living in care homes.

Trial design

The OTCH study was a Phase III, pragmatic, multisite, cluster RCT with economic evaluation, across several regions in England and Wales. The cluster design was justified because of the inclusion of an education component for care home staff and the potential need to apply minor modifications to the care home environment (e.g. install raised toilet seats). Furthermore, cluster randomisation at a care-home level reduced the potential for between-group data contamination during interaction between carers, therapists and residents. The flow diagram for this trial followed the Consolidated Standards of Reporting Trials (CONSORT) extension for cluster randomised trials. 65

Setting

Local care homes for older people in the UK were located in the following 12 trial administration centres (TACs):

1. University of Birmingham

2. Bangor University

3. University of Central Lancashire

4. University of Nottingham

5. Solent Healthcare Primary Care Trust (PCT)

6. Plymouth PCT

7. Wolverhampton PCT

8. Taunton PCT

9. Stoke-on-Trent PCT

10. Coventry & Warwickshire PCT

11. Bournemouth and Poole PCT

12. Dorset PCT.

Recruitment and consent

Randomisation, stratification and blinding

Care homes and participants were recruited and consented into the study before the randomisation process commenced to reduce bias. 69,70 Once the study co-ordinator received confirmation from assessors that all residents in a participating home had given their consent and completed a baseline assessment, the homes were grouped and randomised (1 : 1) to receive either the OT intervention or usual care (control). The allocation sequence was generated in software (nQuery Advisor version 7; Statistical Solutions, Saugus, MA, USA) by an independent statistician using blocked randomisation (block size 2) within strata [type of care home (with or without provision of nursing care)] and geographical location of the TAC at the Primary Care Clinical Research and Trials Unit, University of Birmingham, independent from the research team. Blocked randomisation was used to ensure groups were balanced with respect to the stratification variables. The details of the sequence were concealed from the research team, independent assessors and the study co-ordinator. Once the study co-ordinator received notification that all consenting participants in a care home had completed baseline measures and that the strata data had been logged, care homes were randomised. Care home allocation was revealed to the study co-ordinator, who then informed the care home manager and corresponding site therapist. If a care home had been allocated the OT intervention, the site occupational therapist then contacted the manager of the home to make arrangements for them to visit and commence the intervention. Treatment allocation was concealed from the independent assessors, but it was not possible to mask aspects of the intervention from staff or residents.

Procedure

The trial protocol is published elsewhere. 71

Quality assurance

Outcome assessments

Sample size

A change of 2 points on the BI is widely accepted as being clinically meaningful. 82 In order to detect a difference of this magnitude between the treatment arms, a sample of 72 participants in each arm was required, based on an estimate of standard deviation (SD) of 3.7 points,47,48,49 90% power and 5% significance level. However, residents in this trial were cluster randomised by care home and, therefore, the sample size was inflated by a factor of 4.6, resulting in an increase to 330 residents per randomisation arm. This design factor was based on an estimate of intracluster correlation coefficient (ICC) of 0.4 and an average of 10 residents per care home observed in previous research. 48,49 The ICC used in the sample size calculation was estimated from data related to several pilot studies from a single site with relatively small numbers of care homes. A larger estimate of ICC was used than the ICC observed during the OTCH pilot trial in the interest ensuring an adequately powered study. 48 Based on the attrition rate of 26% from the OTCH pilot study,48 it was estimated that 45 homes with 10 residents per home would be required in each arm of the study (900 residents in total) to detect a clinically meaningful difference using the BI. The sample size quoted in the original proposal was estimated at 840 residents from 84 care homes; however, this sample size was initially incorrectly inflated for expected attrition (26%). The correct figure should have been 900 residents from 90 homes. The revised estimate was identified at the start of the trial.

Occupational Therapy intervention for residents with stroke living in UK Care Homes pilot study

A pilot study of the OT intervention for residents in care homes with stroke-related disabilities was conducted in Oxfordshire, UK, and published prior to the trial. 48 The pilot study was undertaken to refine trial procedures and ensure the OT intervention was acceptable to residents and deliverable in the intended format. The design was a cluster RCT with the unit of randomisation trial at the care home level. Twelve homes (118 residents) were randomly allocated to either the OT intervention (six homes, 63 residents) or control (six homes, 55 residents). The control group received usual care. Usual care did not include OT. In the intervention group, a 3-month course of individualised one-to-one OT was provided to residents with stroke-related disabilities. The aim of the therapy was to maintain levels of functional activity in self-care tasks, adapt the physical environment (when necessary) and address specific impairments that limit performance in ADL or cause discomfort.

In addition, the OT intervention included training for care home staff directly involved in the care of the residents. Assessments were made at baseline and at 3-month (immediately following the intervention) and at 6-month follow-up. The measures used were the BI and RMI. The trial indicated a potential for the OT intervention to have detectable and lasting effects on morbidity. From baseline to the primary end point at 3 months the mean BI score increased by 0.6 points (SD 3.9 points) in the OT arm, but decreased by 0.9 points (SD 2.2 points) in the control arm. The mean difference between the groups was 1.5 points (95% CI –0.5 to 3.5 points, allowing for cluster design). The between-group difference in BI score was maintained at 6 months (difference of 1.9 points, 95% CI –0.7 to 4.4 points).

The analysis of the data suggested that OT may have a significant influence on maintaining functional independence in personal ADL for residents living with stroke-related disabilities within care homes. However, the analysis was exploratory because of its small sample size. The pilot study demonstrated feasibility of the research design. The method was practical and provided the relevant information to conduct a formal sample size calculation for a subsequent suitably powered definitive trial.

Data management

All pre- and post-randomisation data were initially captured by a blinded assessor on paper forms. The data were then entered manually onto the purposefully designed trial database system, which was located in the Primary Care Clinical Research and Trials Unit at the University of Birmingham. The data entry application server was accessible over the internet via secure remote access. All traffic through the data entry application to the database was automatically encrypted using 128-bit secure sockets layer. The database was only accessible from within the University of Birmingham’s information technology network. Only senior members of the research team had access to the database. The firewall for the database was configured to allow access from specific machines only. The data entry application and database used role-based security controls, which restricted access to parts of the data (i.e. uploading, editing and viewing). The data entry application forms were designed and set up by a data manager and computer programmers in collaboration with members of the research team.

Queries about data completeness were referred by the study statistician to the trial co-ordinator. Any missing data that needed to be clarified were obtained by telephone call with the participant or a care home member of staff. Completed questionnaires were entered onto the trial’s secure database by the assessor or a member of the research team. If changes were required on the paper questionnaires, they were formally documented and subsequently uploaded onto the database. The baseline and follow-up data were validated continuously throughout the trial for correctness and completeness. Furthermore, computerised validation checks were incorporated to minimise errors in the data sets (i.e. ranges, limitations and categorisation). Interim summary reports were generated for the Data Monitoring Committee.

Statistical analysis

Analyses were conducted according to a pre-specified statistical analysis plan. 71

Ethical approval

Trial registration

This trial was registered as ISRCTN00757750 on 21 October 2009.

Governance

A TSC was established to monitor the governance of the study. The Trial Steering Group comprised the main research team, an independent chairperson, a geriatrician, an occupational therapist, a physiotherapist with expertise in rehabilitation research, a patient representative and a representative from the NIHR. A Data Monitoring and Ethics Committee (DMEC) was also established. This committee was independent of the trial and monitored accrued data at regular intervals to assess ethical, safety and data integrity aspects of the trial. The DMEC consisted of an independent geriatrician as chair, a statistician and an occupational therapist.

Amendments to the study protocol during the trial

The first substantial amendment was submitted on 15 March 2010 to approve changes to (1) the protocol to clarify the recruitment and randomisation process, the statistical analysis and the TACs and (2) the demographic front sheet, MMSE, OTCH resource usage form and the care home invitation letter. Furthermore, the substantial amendment included the addition of the OT treatment log, adverse events reporting form, OT leaflet and consultee Geriatric Depression Scale-15 items (GDS-15) to the study. The first substantial amendment was approved 23 March 2010.

A second substantial amendment was approved on 21 October 2010. This amendment included (1) a clarification of the consultee’s responsibilities and duties in the declaration process; (2) an update of the consent form and participant information sheet with grammatical changes for clarity; (3) a redesign of the resource usage questionnaire to capture the OT data after the 12-month follow-up assessment to ensure assessors remained blinded; (4) new paperwork including a GP letter notifying them that their patient was randomised to the control arm of the study; and (5) care home letters notifying them of the randomisation outcome.

On 17 August 2011, a third substantial amendment was approved and included (1) changes to the participant and consultee information sheets to provide more information regarding the security of the trial’s database; (2) a new covering letter and form to the participant’s GP to ask for confirmation of their patient’s diagnosis; and (3) shortening the GDS from the 30-item version to 15-item version to reduce the burden on participants and assessors, and to bring the questionnaire in line with the consultee GDS. A further minor amendment was submitted to revise the consent and declaration forms for version control, which was approved on 22 August 2011.

A final substantial amendment was approved on 26 July 2013 that confirmed the transfer in sponsorship of the OTCH trial from the University of Birmingham to the University of East Anglia.

Patient and public involvement

Patient and public involvement was an integral part of the OTCH trial. Patients and carers were directly involved as research ‘partners’ and not just as ‘data providers’ (using the INVOLVE guidance; see www.invo.org.uk). A representative from the local stroke community was involved in the study design, listed as a co-applicant on the funding submission, recruited as a member of the TSC and is a co-author of this report. An additional stroke survivor was recruited onto the TSC which convened regularly throughout the course of the trial. Service users (and carers) were directly involved in developing the written material contained in the study protocol and participant information sheets. All support for patient and carer involvement was provided by the Stroke Research Network members. The Stroke Research Network team had expertise in training and supporting service users (and carers) for involvement in NHS, service evaluation and development. A wider patient/carer audience was consulted about the findings and recommendations drawn from the project. This occurred at meetings convened by West Midlands Stroke Research Network, The Stroke Association and Different Strokes.

Randomisation

Participating care homes were randomised between 4 May 2010 and 28 February 2012. Participants were followed up at 3, 6 and 12 months after randomisation.

Participant recruitment within clusters: sample size

The flow of participants throughout the duration of the trial is depicted in the CONSORT diagram (Figure 1). Within the 237 consenting care homes there were 1556 out of 9840 (16%) eligible residents. The 16% figure represents the prevalence of stroke in the consenting care homes. Of those identified as eligible, 1055 out of 1556 (68%) were registered into the trial and 501 out of 1556 (32%) did not offer consent. A total of 13 participants (nine care homes) did not progress to the randomisation stage, two were withdrawn by the care home manger to receive end-of-life treatment and the remaining 11 participants were excluded because of the withdrawal of the regional TAC described above. Table 6 summarises participant and care home recruitment levels according to regional TACs.

FIGURE 1.

Consolidated Standards of Reporting Trials diagram.

TABLE 6 - Trial administrative centre recruitment figures

TAC withdrew prior to randomisation.

TAC	Care home recruitment			Participant recruitment
	Consenting care homes	Beds in consenting homes	Randomised care homes	Eligible residents screened	Consenting residents	Randomised residents
Bournemouth and Poole PCT	12	434	12	55	48	48
University of Central Lancashire	16	954	16	142	86	86
Coventry PCT	14	537	14	107	56	56
Dorset PCTa	4	179	0	11	11	0
University of Nottingham	22	897	22	175	126	126
Plymouth PCT	16	556	14	68	40	40
Solent Healthcare PCT	28	1006	26	180	110	108
Stoke-on-Trent PCT	10	400	10	76	48	48
Taunton PCT	8	392	8	71	45	45
Bangor University	17	604	17	119	104	104
University of Birmingham	74	3098	73	427	323	323
Wolverhampton PCT	16	783	16	125	58	58
Total	237	9840	228	1556	1055	1042

Participant recruitment exceeded the original target of 900. More care homes were recruited because the average cluster size was lower than predicted, but comparable between the two arms (intervention mean 5.0 participants, control mean 4.2 participants). The slight over-recruitment of 1042 participants was also because of staggered closure of sites that had already consented residents to the trial. Cluster size information is presented in Table 7. The number of care home residents with a history of stroke was lower than expected. A total of 1042 participants were randomised. More eligible residents resided in clusters randomised to the intervention arm (n = 568) than in clusters randomised to the control arm (n = 474), see Table 8. The disparity in participant numbers between the two treatment arms was a chance occurrence. Consent was obtained prior to randomisation. There were more eligible residents in care homes randomised to the intervention arm than in care homes randomised to the control arm. A larger percentage of participating care homes (53%) provided nursing care (see Table 5), and from these a larger percentage of participants (64%) were recruited (see Table 8).

TABLE 7 - Cluster size frequency (n = 228)

Percentages may not total 100 because of rounding.

Cluster size (number of participants)	Frequency (%)	Randomisation arm
		Intervention, n (%)a	Control, n (%)a
1	29 (13)	11 (10)	18 (16)
2	38 (17)	17 (15)	21 (18)
3	28 (12)	13 (11)	15 (13)
4	43 (19)	20 (18)	23 (20)
5	30 (13)	18 (16)	12 (11)
6	15 (7)	9 (8)	6 (5)
7	10 (4)	5 (4)	5 (4)
8	11 (5)	6 (5)	5 (4)
9	9 (4)	8 (7)	1 (1)
10	4 (2)	3 (3)	1 (1)
11	3 (1)	0 (0)	3 (3)
12	2 (1)	1 (1)	1 (1)
13	1 (0)	0 (0)	1 (1)
14	1 (0)	0 (0)	1 (1)
15	1 (0)	0 (0)	1 (1)
19	1 (0)	1 (1)	0 (0)
21	1 (0)	1 (1)	0 (0)
23	1 (0)	1 (1)	0 (0)
Total number of care homes	228	114	114
Median participants per home (interquartile range)	4 (2–6)	4 (3–6)	4 (2–5)
Mean participants per home (SD)	4.6 (3.3)	5 (3.7)	4.2 (3.0)

TABLE 8 - Distribution of participants by randomisation arm and type of care home (n = 1042)

Type of care home	Randomisation arm
	Intervention (N = 568), n (%)	Control (N = 474), n (%)
Residential	207 (36)	166 (35)
Nursing	361 (64)	308 (65)

Participant characteristics

The demographic information listed in Table 10 indicates that age, sex and ethnicity were balanced across treatment arms. Of the 1042 participants, 962 (92%) were white and 665 (64%) were female. The mean age of all participants was 82.9 years (SD 9.2 years, range 43–102 years). The distribution of age groups was similar across care home type and randomisation arm (Table 11). Stroke or TIA was present in care home notes for all participants; however, confirmation from GP surgeries was received in 721 out of 1042 (69%) cases. The details regarding the distribution of stroke and TIA among participants are given in Table 12.

Retention

Retention throughout the trial was good. In the intervention arm, 21 participants moved home during the trial, three died, one was lost to follow-up and the remainder (17) completed outcome assessments at the 12-month end point. In the control arm, 16 participants moved homes, all of whom provided data at the 12-month end point (see Appendix 15).

Withdrawal and lost-to-follow-up data, as well as completion rates for the primary outcome measure at all time points, are presented in Table 15. A total of 313 out of 1042 (30%) participants died during the 12-month trial duration. The percentage of participants that died were balanced between treatment arms at all time points (see Table 14). Two participants (one in each treatment arm) were withdrawn by the care home manager to receive end-of-life care and died within 3 months. These two participants have been included in the number of deaths (see Figure 1 and Table 15).

Baseline primary and secondary measures

Secondary measures

Rivermead Mobility Index

Baseline RMI scores are presented in Table 18. The measure was completed by 1033 out of 1042 (99%) participants. The mean mobility score was low and comparable between treatment groups.

TABLE 18 - Baseline RMI scores by randomisation arm (n = 1033)

0–15, 15 indicates maximum ability.

RMIa	Randomisation arm
	Intervention (n = 557)	Control (n = 456)
Mean (SD)	3.12 (3.81)	2.85 (3.70)
Median (interquartile range)	1 (0–6)	1 (0–5)

Figure 2 displays the baseline RMI data plotted as a function of the baseline BI data. The plot indicates that a high proportion of participants had significant limitations on functional activity at the beginning of the trial. This plot includes data from 1012 out of 1042 (97%) participants who completed both baseline BI and RMI measures. A total of 493 out of 1012 (49%) participants scored below 4 on both the RMI and the BI, suggesting that half of the sampled population had severe limitations on engaging with personal ADL and severe limitations on mobility (see Table 44 in Appendix 16).

FIGURE 2.

The relationship between baseline RMI scores and baseline BI scores (n = 1012). BI 0–20, 20 signifying maximum ability; RMI 0–15, 15 signifying maximum ability.

Baseline GDS-15 scores were recorded from 913 out of 1042 (88%) participants (Table 19). The measures of central tendency for the baseline GDS-15 scores were in the range indicative of mild depression, and comparable between randomisation arms. Twenty-four per cent of participants scored in the 10–15 score range, suggesting severe depression.

TABLE 19 - Baseline GDS-15 scores by randomisation arm (n = 913)

GDS-15 score	Randomisation arm
	Intervention (N = 498), n (%)	Control (N = 415), n (%)
0–4 (normal)	157 (32)	131 (32)
5–9 (mild depression)	205 (41)	200 (48)
10–15 (severe depression)	136 (27)	84 (20)
Mean (SD)	6.8 (3.9)	6.4 (3.5)
Median (interquartile range)	6.0 (4–10)	6.0 (4–9)

Intervention

Overall, 2538 visits were made to 498 residents in the intervention arm (mean 5.1 visits, SD 3.0 visits). Total therapy time was 1724 hours. Median session duration was 30 minutes (interquartile range 15–60 minutes). Therapy was administered according to categories. Table 20 shows a summary of the time allocated to each category.

The 3-month follow-up

The 6- and 12-month follow-up

Primary and secondary outcome measures

At the two additional end points, the BI data showed no significant differences between groups (see Table 24). There was no evidence of any difference between the arms from the BI composite analyses at 6 and 12 months, presented in Table 25. In addition, the results from the secondary outcome measures assessing mobility (RMI, see Table 22), and mood (GDS-15; see Table 23), showed no significant differences between groups, at the 6- and 12-month follow-up time points (Table 26). The BI scores grouped according to severity rating for both treatment arms across all time points are presented in Figure 3.

FIGURE 3.

Barthel Index of Activities of Daily Living severity ratings at all end points across both treatment arms. (The data are given in Table 45, Appendix 16.)

TABLE 25 - Comparison of Barthel composite outcome at 6- and 12-month end points

Based on change in BI score at 3 months from baseline (below 0 or death, ‘poor’; 0 to 1, ‘moderate’; 2 and above, ‘good’).

Proportional odds of improvement in outcome after OT compared with control; adjusted for care home as a random effect and type of care home and centre as fixed effects.

BI compositea	Randomisation arm		OR (95% CI)b	p-value
	Intervention	Control
6-month assessment
n	526	449	0.95 (0.71 to 1.27)	0.74
Poor, n (%)	306 (58.2%)	269 (59.9%)
Moderate, n (%)	161 (30.6%)	122 (27.2%)
Good, n (%)	59 (11.2%)	58 (12.9%)
12-month assessment
n	513	432	0.84 (0.61 to 1.15)	0.27
Poor, n (%)	350 (68.2%)	314 (72.7%)
Moderate, n (%)	121 (23.6%)	77 (17.8%)
Good, n (%)	42 (8.2%)	41 (9.5%)

TABLE 26 - Comparison of primary and secondary outcomes at 6- and 12-month end points

SE, standard error.

Adjusted for care home as a random effect and baseline score, type of care home and centre as fixed effects.

Tukey–Kramer-adjusted CIs and p-values.

Participants who died before follow-up are given a BI score of zero.

EQ-5D-3L calculated using the UK tariff.

Outcome	Follow-up	Randomisation arm				Difference in adjusted means (95% CI)a	p-valueb	Group × time interaction
		Intervention		Control
		Adjusted mean (SE)a	n	Adjusted mean (SE)a	n
BIc	6 months	4.78 (0.20)	525	4.78 (0.22)	448	0.004 (–0.52 to 0.53)	0.99	0.35
	12 months	3.93 (0.21)	512	3.77 (0.22)	430	0.16 (–0.40 to 0.72)	0.58
RMI	6 months	2.64 (0.11)	421	2.67 (0.12)	346	–0.03 (–0.33 to 0.27)	0.84	0.23
	12 months	2.19 (0.13)	354	2.46 (0.14)	271	–0.26 (–0.62 to 0.09)	0.15
GDS-15	6 months	6.20 (0.21)	338	6.68 (0.22)	284	–0.48 (–1.04 to 0.09)	0.10	0.57
	12 months	6.22 (0.22)	297	6.40 (0.25)	219	–0.18 (–0.80 to 0.43)	0.56
EQ-5D-3Ld	6 months	0.218 (0.017)	363	0.226 (0.017)	315	–0.008 (–0.052 to 0.036)	0.72	0.56
	12 months	0.202 (0.018)	316	0.184 (0.019)	244	0.018 (–0.031 to 0.067)	0.48

Subgroup analysis

Subgroup analyses were performed according to a predefined list of variables to further evaluate potential influences of the OT intervention. Analyses considered participants’ age, the type of care home in which participants’ resided, the severity rating of participants’ BI scores, the level of cognitive impairment (derived from the MMSE screening measure), and whether the measures were completed by the participant or a consultee. The forest plot (Figure 4) shows no significant findings across all subgroup interactions (Table 30).

FIGURE 4.

Subgroup analysis: comparison of BI at 3 months. Participant numbers for age and MMSE reflect data missing at baseline.

TABLE 30 - Subgroup analysis: comparison of BI at 3 months

Type of care home means were adjusted for baseline BI score and site as fixed effects and care home as a random effect. All other subgroup means were adjusted for baseline BI score, TACs and type of care home as fixed effects and care home as a random effect.

Outcome	Randomisation arm				Difference in adjusted means (95% CI)	p-value (Interaction)
	Intervention		Control
	Adjusted mean (SE)a	n	Adjusted mean (SE)a	n
Respondent
Consultee	5.27 (0.30)	193	5.30 (0.32)	144	–0.04 (–0.85 to 0.77)	0.48
Resident	5.58 (0.23)	346	5.28 (0.33)	292	0.30 (–0.30 to 0.89)
Level of BI
Very severe (0–4)	5.21 (0.44)	256	5.15 (0.44)	222	0.06 (–0.63 to 0.75)	0.29
Severe (5–9)	5.03 (0.33)	126	5.31 (0.37)	94	–0.28 (–1.21 to 0.65)
Moderate (10–14)	5.42 (0.57)	87	5.15 (0.59)	72	0.27 (–0.81 to 1.35)
Mild/independent (15–20)	7.35 (0.91)	70	6.12 (0.96)	48	1.23 (–0.06 to 2.52)
Level of MMSE
Very cognitively impaired (0–9)	4.48 (0.36)	133	4.50 (0.36)	125	–0.02 (–0.92 to 0.88)	0.13
Rest (10–30)	5.49 (0.31)	175	4.65 (0.32)	155	0.84 (0.06 to 1.62)
Type of care home
Residential	5.70 (0.31)	197	5.30 (0.33)	153	0.40 (–0.41 to 1.22)	0.50
Nursing	5.28 (0.24)	342	5.23 (0.26)	283	0.05 (–0.61 to 0.70)
Age group (years)
< 75	5.55 (0.38)	88	5.97 (0.44)	66	–0.42 (–1.53 to 0.69)	0.53
75–84	5.41 (0.28)	189	5.44 (0.30)	145	–0.02 (–0.79 to 0.75)
85–94	5.55 (0.26)	228	5.09 (0.28)	188	0.46 (–0.25 to 1.17)
≥ 95	5.29 (0.59)	34	5.22 (0.57)	35	0.07 (–1.53 to 1.66)

Adverse events

There were no reported treatment-related adverse events.

Falls

Unadjusted data regarding the number of falls experienced by participants between randomisation arms are presented in Table 32. Comparison of the frequency of falls per resident showed a significantly higher (adjusted) fall rate per resident was reported in the intervention arm [rate ratio 1.74, 95% CI 1.09 to 2.77; p = 0.02] (Table 33). When adjusted for care home, TACs and type of care home there was a suggestion of greater odds of falling in the OT arm (OR 1.55, 95% CI 0.96 to 2.53; p = 0.07) (Table 34).

Process evaluation: summary

A more detailed report of the process evaluation is reported elsewhere. 88 Qualitative evaluation of intervention fidelity was also investigated in semistructured interviews (n = 17) with all occupational therapists, and critical incident reports from the trial (n = 20). These data demonstrated four identified mechanisms through which intervention fidelity was maintained:

1. Occupational therapists described managing a shift from their earlier roles as therapists to their ‘new’ role as experimental occupational therapists within a clinical trial. This was characterised as balancing work in which, over time, they had developed a sense of equilibrium between their professional responsibility as therapists and the tightly defined therapy outlined within the study protocol.

2. A considerable amount of time was spent by some therapists building rapport with care home staff, developing an organisational context more conducive to rehabilitation in general and the trial intervention in particular. This was felt to be important in establishing the foundations on which trial interventions could then be established. One strategy through which this mechanism operated was through therapists ‘mucking in’ and working as a team with members of care home staff.

3. The work focused on re-engineering the personal environments of care home residents. The work that therapists undertook to re-engineer the care home environment around their rehabilitation plans for patients was completed at different degrees of scale, depending on local circumstances. Generally, therapists were careful to propose realistic intervention plans that could be carried out within available resources such as funding, skills availability and access to external services. Through careful assessment and the provision of aids and minor environmental adaptations, the trial interventions intended to provide patients with an environment that best matched their needs.

4. The data clearly demonstrate that therapists were not passive participants in the delivery of the OTCH interventions. Therapists appeared to have learned through their experience of the interventions over time with individual patients and across waves of recruitment within the trial. Therapists modified the way they worked with care homes and patients, adapting to new situations and learning how to carry out their work more effectively. This was important in enabling therapists’ confidence in their work within the trial to grow with time.

These findings from the OTCH process evaluation characterise the real-world nature of fidelity within the OTCH rehabilitation intervention, and specifically the negotiated nature of implementation within clinical settings, and around individual patients’ needs.

Overview

To assess the economic viability of delivering OT to stroke survivors in care homes, we conducted a within-trial cost–utility analysis. Costs and quality-adjusted life-years (QALYs) for individuals in the OT intervention group were compared with equivalent data for those receiving usual care. For our base case, we conducted intention-to-treat analysis using ‘complete case’ data. This included participants who died during follow-up, provided that they had returned all costs and outcomes data in the intervention period before their death. As all trial follow-up was completed within 12 months, no discounting was applied to either costs or outcomes.

Estimating costs

In line with the National Institute for Health and Care Excellence methods guide,89 we sought to estimate costs from an NHS and Personal Social Services perspective. This included the cost of the intervention, hospital contacts and community health and social services, and any ADL equipment costs incurred within the intervention or as part of usual care. In deriving costs for the OT intervention, we considered staff training (for therapists who provided the intervention), care home staff training (workshops), participant contact and non-contact time, and equipment and travel costs. Each therapist in the study team completed a ‘Treatment Log’ timesheet (see Appendix 10) to record the number of minutes spent on specific aspects of therapy (such as communication, adaptive equipment or cognitive training) at each session with each participant. To estimate non-contact time spent on the intervention, we also interviewed by telephone two senior occupational therapists who delivered the intervention and training for all aspects of the study. Cost assumptions used here were based on information retrieved in these discussions and confirmed with the principal investigator.

Across all sites, 29 occupational therapists delivered the OTCH intervention. All study staff attended a group training day at the central site (Birmingham). Two such days took place, each run by three senior occupational therapists. Training included adapting OT for the care home environment. Costs were apportioned equally to all participants in the intervention arm. This was estimated as a full working day (7.5 hours) for each of the 29 OT staff (NHS band 6), with average travel to Birmingham estimated as 100 miles. Six days of senior occupational therapist time was also included.

For the workshops attended by nursing and care staff in each care home, it was estimated that the three course co-ordinators (senior occupational therapists) spent one week developing training documents. Although some care homes may have had repeat training (e.g. because of high staff turnover) it was assumed that all care staff from each care home in the intervention arm attended the workshop only once. Seminars lasted 2 hours and were delivered by one senior occupational therapist. Costs were apportioned equally to all participants in the intervention arm. In an additional sensitivity analysis, we included the opportunity cost of time spent by care home staff attending these workshops (assuming workshops would be delivered once every 3 years).

The OTCH intervention was delivered by NHS community occupational therapists, with level of experience ranging from band 5 to 7. In deriving costs, we assumed that the intervention was delivered only by occupational therapists at NHS band 6. 90 For every intervention session spent with each individual participant, the approximate time (in minutes) spent on specific tasks was recorded in the treatment log. As the treatment log recorded only face-to-face time with participants, we ascribed additional time per task for non-contact time – including initial risk assessment for each participant, general administration, liaising with care home staff and collecting and demonstrating the equipment to care home staff. Based on estimates from both of the therapists interviewed and estimates reported in Personal Social Services Research Unit (PSSRU) costs,91 it was estimated that for every hour of contact time there was an additional 40 minutes (67% of contact time) of non-contact time.

Measuring outcomes

Our principal measure of benefit was expressed as QALYs. Health utility was estimated using the EQ-5D-3L,96 a generic measure of HRQoL that generates a single index score. This outcome measure has been successfully applied in previous studies with nursing home residents. 97,98 We applied the Measurement and Valuation of Health Group’s A1 tariff, which used time trade-off data obtained from a sample of approximately 3000 members of the general population in UK. 99 Participants were asked to complete (self-complete or by proxy) the EQ-5D-3L questionnaire at baseline and at 3, 6 and 12 months. Life-years were weighted by these utility scores and linear interpolation was used to generate QALYs per patient using area under the curve methods. If a participant died during the follow-up period, EQ-5D-3L values (and costs) were set to zero at the point of death.

Analysis

The use of regression analysis is advocated to account for potential baseline differences and/or confounders when comparing costs and outcomes between treatment arms, and is essential to formally take account of the cluster-randomised design. 100 We regressed (1) costs and (2) QALYs against treatment arm with covariate adjustment for sex, age and baseline EQ-5D-3L, with cluster-adjustment based on care home. The regression coefficients for costs and outcomes were then used to estimate the incremental cost-effectiveness ratio (ICER), dividing the mean difference in costs by the mean difference in QALYs between the groups. In line with National Institute for Health and Care Excellence guidance, we compared ICERs with a cost-effectiveness threshold (λ) of £20,000 and £30,000 per QALY. 89

Economic evaluation results

All treatment logs were returned and intervention costs were therefore available for all participants in the intervention arm (Table 36). The estimated cost of training occupational therapists to deliver the intervention was £28.62 per participant, while providing the training workshop to care home staff was estimated to cost £36.99 (or £69.83 per participant when also including cost of care home staff who received the training). In total, 2538 therapy sessions were recorded in the OTCH intervention, with an average duration of 41 minutes (SD 34.0 minutes) each. The mean number of sessions per participant was 5.1 (SD 3.0 sessions). The cost of the OTCH intervention per participant varied considerably based on the number of therapy sessions, whether or not other participants were also treated in their care home on the same day and how long each session lasted. Seventy (12.3%) participants in the intervention arm did not receive any treatment, whereas the most intensive treatment involved 18 visits in the 3-month period. A large cost contributor in the intervention was travel costs by therapists to each care home; this ranged from £0.00 to £371.62 per participant. In total, we estimated the mean intervention cost per participant in the active treatment arm to be £398.98 (SD £347.00).

Cost–utility analysis

Excluding participants for whom we did not have complete paired costs and outcomes data, the remaining ‘complete case’ data set, which was used in subsequent regression analyses, consisted of 581 out of 1042 (55.8%) participants (329 out of 568 in the intervention arm; 252 out of 474 in the control arm) in 177 out of 228 care homes (77.6%). The full data set (1042 participants) was considered in subsequent sensitivity analysis, following multiple imputation.

Total costs and QALYs were assessed using ordinary least squares regression, with cluster option to account for randomisation at care home level. This produced a mean incremental cost of £438.78 (95% CI –£360.89 to £1238.46). The mean incremental QALY gain was 0.009, with wide CIs (95% CI –0.030 to 0.048). Neither the mean incremental cost nor the mean incremental QALYs reached an arbitrary 5% significance level. The ICER was estimated to be £49,825.

A summary of costs and effects at base-case and sensitivity analyses, along with cost-effectiveness ratios, is shown in Table 42. In sensitivity analysis, removal of high-cost participants did not change overall cost-effectiveness results, with both incremental costs and incremental QALYs reduced very slightly. Inclusion of care worker time increased intervention costs slightly without affecting outcomes; therefore, the cost-effectiveness of intervention was less favourable.

When we repeated our regression using imputed data sets, incremental costs were slightly lower at £339 (95% CI –£133.60 to £811.98). Incremental QALYs were slightly higher at 0.013 (95% CI –0.015 to 0.041), leading to a slightly more favourable ICER of £26,770. However, this is still higher than the current threshold, and our conclusions regarding cost-effectiveness remained unchanged. In all sensitivity analyses incremental cost and QALY differences were not significantly different between arms.

The bootstrapped replications for costs and QALYs are shown on a cost-effectiveness plane in Figure 5. The majority are located on the north-east quadrant, suggesting the OTCH intervention is associated with better outcomes and higher costs than usual care. However, most of the dots lie very close to both axes, indicating that the cost and outcome differences between groups are very small. The CEAC (Figure 6) shows the probability of the intervention being more cost-effective than usual care at various willingness-to-pay thresholds. At £20,000 per QALY, this probability was 29.2% and at a threshold of £30,000 the probability was 39.2%. The incremental cost is more than £20,000/QALY in all analyses; therefore, based on current cost-effectiveness thresholds,89 we would not endorse the OTCH programme.

FIGURE 5.

Cost-effectiveness plane (base case).

FIGURE 6.

Cost-effectiveness acceptability curve (base case).

Occupational Therapy intervention for residents with stroke living in UK Care Homes trial design summary

The OTCH study was a Phase III pragmatic parallel-group cluster RCT with an economic evaluation. The evaluation assessed the clinical efficacy and economic impact of providing an OT service for stroke survivors living in care homes. The primary research objective was to assess the influence of a 3-month course of OT, according to whether or not it helped participants maintain levels of independence in ADL compared with usual care. Secondary outcome measures assessed the influence of the OT intervention on participants’ mobility, mood and HRQoL.

More care homes were recruited than originally planned because of a larger than expected number of small clusters. The number of eligible residents within each care home with a history of stroke was lower than expected. 11,15–17 All baseline characteristics were similar across randomisation arms in regards to age, sex, ethnicity and comorbidities. A large proportion of the participant population had significant physical disabilities, substantial limitations on activity (see Figures 2 and 3) and increased dependence for personal ADL. Moreover, the assistance of a consultee to consent on the participant’s behalf was needed in 61% of cases,68 indicative of reduced autonomy. There was a high prevalence of cognitive and language impairment: 57% of participants scored below 15 points on the Sheffield screening test, indicative of significant language impairment, and 70% of participants scored in the range signifying cognitive impairment on the MMSE. The participant profile overall is suggestive of significant frailty.

Principal findings

Further exploratory analyses

These additional exploratory analyses considered the potential dose effect of OT on depression scores, and the relationship between baseline RMI score and survival. No evidence of an association between OT dose and GDS-15 score was observed. However, there was strong evidence that mobility is associated with survival. Participants with a mobility score of 2 or below at baseline were 1.57 times as likely to die than those with greater mobility over the course of the 12-month trial duration (hazard ratio 1.57, 95% CI 1.15 to 2.15).

Economic evaluation discussion

Based on the mean incremental cost per QALY gain, it is unlikely that the OTCH intervention is cost-effective when compared with usual treatment. Although outcomes were virtually equivalent in both arms, costs were higher in the intervention arm and the intervention did not lead to a reduction in health resource use from other sources.

We acknowledge some limitations in this economic analysis. We requested information on equipment on the assumption that this would have been purchased by the NHS or personal social services. However, it became apparent that some of this equipment may have been purchased by the care home (although who purchased it was not systematically recorded), and that this could be considered outside of the NHS and personal social services perspective. In addition, apart from the training of care home staff, we did not consider all wider social costs, although we considered that, as all participants were in residential care, the burden on community care workers, friends and family would be likely to be less than if the participants lived in their own homes.

Death rates were similar between study arms. We tested the robustness of our estimates by performing multiple imputation for missing data and also running a sensitivity analysis excluding participants who died during follow-up (not reported); our cost-effectiveness conclusions were consistent. 103

Despite these limitations, the cost-effectiveness results were conclusive. At £20,000 per QALY, this probability was 29.2% and at a threshold of £30,000 the probability was 39.2%. Based on the results of this RCT, OT was not a cost-effective routine intervention for stroke survivors living in care homes.

Strengths and weaknesses

This was the largest cluster randomised trial conducted in care homes to date. The potential pitfalls of the cluster design associated with the provision of informed consent, identification of the unit of inference and methods of stratification were addressed in the trial protocol. 71,104 The trial was sufficiently powered to detect a clinically significant change in the BI measure following a 3-month course of individualised OT,82 involving task-related ADL practice, provision of adaptive equipment, adaptations to the individual’s environment and caregiver training. The unadjusted ICC for BI scores was 0.36 at baseline; however, for the change in scores from baseline to 3 months, allowing for the effect of treatment, TACs and type of care home, it decreased to 0.09. The mixed modelling approach has been shown to be a reasonable method of analysis when some cluster sizes are small, allowing for appropriate inferences to be made in relation to fixed effects. 105 There is additional evidence that even with small cluster sizes, mixed modelling is a better approach than an analysis that ignores clustering. 106 Furthermore, results of the sensitivity analysis showed that the conclusions were robust when small clusters were excluded. Results from the complete case analysis, where BI scores were not imputed for participants who died before the primary end point revealed similar neutral results to the primary analysis.

The OT administered to participants was similar to a standard NHS intervention,88 shown to be of benefit to stroke survivors living in their own homes. 34 Compliance over the course of the trial was good, resulting in high completion rates among survivors for all assessments at each end point. On average, participants in the intervention arm were visited on five occasions with visits lasting a median of 30 minutes. Retention of care home participation was good throughout the study, with 204 (89%) homes providing data at the final 12-month end point. The main reason for care home withdrawal was the death of all participating residents. A total of 313 out of 1042 (30%) participants died during the trial. Overall, we regard the findings to be robust and conclusive.

We acknowledge several potential limitations. The percentage of care home residents affected by stroke was less than expected, which resulted in a larger number of small clusters than was originally expected. Previous research suggested the incidence of stroke in care home residents to be approximately 20–40%. 11,15–17 However, this trial observed incidence of stroke in care home residents to be approximately 16% (1556 out of 9840; see Table 6). This figure concurs with results from a recent census of care homes in the UK. 107 In addition, GP surgeries’ response rate to trial correspondence requesting from confirmation of participants’ stroke was low despite multiple attempts.

There were a high proportion of participants with cognitive impairment, and GDS scores were indicative of moderate and severe depression. If these participants had been excluded during recruitment, it would have reduced the external validity of the trial. However, we acknowledge that these participants may have had potentially limited engagement in therapy, as demonstrated by the observed distribution of therapy content (see Table 20). Approximately half of all therapy hours were spent on communication with participants, carers and consultees, as opposed to ADL training (see Table 20). Therapy time attributed to this category was spent on providing information and guidance to residents, staff or relatives on personal ADL, initiating referrals to other agencies and ordering equipment. In the pilot phase the duration of the intervention was also 3 months. 48 The median number of visits per resident per month was 2.7 (interquartile range 1–4.2), and the median duration of therapy time per resident per month was 4.5 hours (interquartile range 2–6.9 hours). 48 This represents a significant difference in the amount of therapy received by residents in the two trials. Residents received more therapy in the pilot phase. In the main trial, the median number of visits was 5 (interquartile range 3–7 visits) over 3 months and the median duration of therapy per resident was 145 minutes (interquartile range 85–255 minutes). The mean duration of therapy listed in Table 20 is 208 minutes (SD 208 minutes; minimum 10 minutes, maximum 1380 minutes). The difference in the amount of therapy administered between the pilot trial and the main trial indicates that the high levels of disability among participants may have prevented engagement in personal ADL training and reduced therapy time as a result.

When viewing these figures it is also important to acknowledge differences in eligibility criteria between the two trials. The pilot trial included 118 residents with a moderate to severe BI score that was between 4 and 15 out of 20. The Phase III definitive trial reported here used a pragmatic approach and included residents with a BI score between 0 and 20. As a consequence the mean baseline BI scores differed between the two trials. In the pilot trial, the mean baseline BI score in the intervention arm was 10.1 (SD 5.7) and 9.5 (SD 5.2) in the control arm. The mean baseline BI scores in both arms in the pilot trial are classed as moderate. The mean baseline scores in the main trial were in the severe range across both arms. Table 16 displays the baseline BI scores. Overall, 48% of participants in the intervention arm had a BI score between 0 and 4, indicating very severe limitations when engaging in personal ADL. It is also important to reflect on the conclusions from the pilot trial. 48 After receiving a higher dose of therapy, participants in the intervention arm showed a tendency for improvement between baseline and the primary outcome time point at 3 months for the Barthel composite measures only (see Table 21 and Table 25 for similar results from this trial). However, participants in the intervention arm deteriorated in a similar way to the control group at all subsequent follow-up time points. 48 Therefore, despite the difference in the amount of therapy administered between the two trials potentially being viewed as a limitation, the main trial has provided a realistic portrayal of the high level of impairment exhibited by care home residents living with stroke-related disabilities throughout England and Wales (see Table 16 and Figures 2 and 3). A realistic conclusion is that the majority of participants may have lacked the capacity to directly engage with structured, repetitious activity-based therapy.

Safety

The OT provided to residents was not an experimental intervention. Therapy given to residents was similar to what stroke survivors living in the community would receive from the NHS. Overall, tolerance of the OT provided to participants was very good, resulting in no adverse events occurring that were attributable to the intervention. There was a higher incidence of falls among participants in the intervention arm. It is a possibility that, by facilitating an increase in functional activity, it may have led to an increased risk of falling. However, according to recently published figures,108–110 the average fall rate of residents in long-term institutional care varies from 1.45 to 2.50 per annum. This suggests the quarterly fall rate of 0.18 recorded in the intervention arm is below the previously observed parameters.

Research in context

The neutral results observed in the OTCH trial concur with results observed in other RCTs conducted in a care home setting. These trials assessed the benefit of activity-based interventions at reducing levels of depression, and incidence of falls. 111,112 Collectively, these neutral findings suggest that we may be expecting too much from this predominantly frail population with a high prevalence of dementia and depression and with low autonomy to respond to activity-based therapies (see Figures 2 and 3 in relation to levels of severe disability). It seems more appropriate to seek alternative care approaches to ameliorate disability for this very inactive patient group. The established OT intervention has good evidence of efficacy when administered to patients in their own homes. 34 This suggests that the barriers to success are more attributable to the care home environment and the care home population. Furthermore, the suitability and application of a patient-centred goal-setting approach in this population may require further scrutiny. Currently, patient-centred care does not have a universally accepted definition or accepted methods of how it can be measured for adherence. 113

A changing role of care homes is acknowledged in a report from the Centre for Policy and Aging. 16 Residents are being admitted to care homes with a higher level of dependency and more complex care needs than ever before. 26 In the census of UK Bupa (The British United Provident Association Limited) care homes in 2012, 87% of residents were classed as having high-support needs and total dependency. 16 Providing care for residents with high-support needs requires careful consideration of the care home environment. The therapists delivering the OTCH intervention in the participating care homes reported that the use of adaptive equipment and environmental modifications (e.g. the installation of bed levers, grab rails or raised toilet seats) was highly variable.

Future work

In order to advance the level of care currently provided to care home residents, future research should identify applicable criteria to promote an enabling environment where possible. The emphasis in this population is suggested to be about providing care for residents with high support needs for a short period towards the end of life. Further research is needed to demonstrate how the care home environment can be suitably modified to tackle these needs. A key factor in promoting an enabling environment is minimising health-related complications caused by inactivity (such as urinary incontinence and pressure sores). In addition, further work is needed in measuring health-related complications in this population in order to identify problems early.

Conclusions

The neutral results from this Phase III cluster randomised trial are deemed robust and conclusive. The trial was geared towards evaluating whether or not there is sufficient evidence to support an improved NHS provision of OT for care home residents with stroke-related disabilities. The clinical and health economic evidence presented here does not support commissioning a routine OT service in this population of care home residents. There may be benefit for residents with less severe limitations on functional activity; however, these residents were in the minority of the sampled population of stroke survivors. It appears justified to suggest that referrals of care home residents with stroke-related disabilities for OT may be of benefit on an individual basis if left in the hands of the health professional initiating the referral. OT as a service for stroke survivors still able to live in their own home has good evidence of success. 34,37 However, the benefit of this style of therapy as a routine service was not evident for stroke survivors resident in UK care homes.

Participating centres and collaborative group members

Occupational therapists

A Lake, S Kilmister, V Blakemore, R Parker, B Jenkins, H Nicholls, L Whitfield, J Mackenzie, K Wood, B Lang, S Baker, S Rundle, A McMichael, S Evans, J Shirley, E Paterson, L Hinds, M Sensier, J Cowling, B Jones and M March.

Research networks

Central England Primary Care Research Network, South West Stroke Research Network, West Midlands Stroke Research Network.

Care homes

We are grateful to the staff and residents for their co-operation and enthusiastic participation in the OTCH trial from the following care homes: Abbey Park, Coventry; Abelard, Nottingham; Acacia Care Centre, Nottingham; Acocks Green Nursing Home, Birmingham; Alexandra House Nursing Home, Nottingham; Alexandra House, Solihull; Alexandra Rose Residential Care Home, Portsmouth; Allambie House Care Home, Coventry; Allerton Court, Sandwell; Alston View Nursing & Residential Home, Preston; Alton Manor Ltd, Southsea; Alverstock House Nursing Home, Gosport; Anville Court Nursing Home, Wolverhampton; Arboretum Nursing Home, Walsall; Arborough House Ltd, Southsea; Arden Park Care Home, Leamington Spa; Ardenlea Court, Solihull; Ash Hall Care Home, Stoke-on-Trent; Ash Lodge, Sandwell; Ashleigh Manor, Plymouth; Ashley Lodge Care Home, Birmingham; Atholl House, Compton; Autumn House Nursing Home, Stone; Bearwood Nursing Home, Sandwell; Beaumaris Care Home, Newport; Beech Hill Grange Nursing Home, Sutton Coldfield; Beech Lodge Nursing & Residential Home, Cheadle; Beechcroft Residential Home, Sandwell; Belmont Residential Care Home, Preston; Bentley Court Care Home, Wolverhampton; Birch Green Care Centre, Skelmersdale; Birds Hill Nursing Home, Poole; Birds Hill Starling, Poole; Bloomfield Court, Tipton; Bluebell Nursing Home, Southsea; Bournville Grange, Birmingham; Bridgewood Mews, Tipton; Broadgate Care Home, Nottingham; Brompton Lodge, Rhos on Sea; Brookvale Care Home, Solihull; Burkitt Care Home, Nottingham; Cams Ridge Nursing & Residential Care Home, Fareham; Cann House, Plymouth; Carpenter Place Care Home, Birmingham; Chalgrove Care Home (Edwardian), Poole; Chalgrove Care Home (Tudor), Poole; Chelston Park Nursing and Residential Home, Wellington; Chorley Lodge Residential Care Home, Chorley; Churchfield Christian Care Centre, Nottingham; Churchfield Court Care Home, West Bromwich; Cole Valley Nursing Home, Birmingham; Compton Manor Care Home, Coventry; Cordelia Court Care Home, Coventry; Cosham Court Nursing Home, Portsmouth; Courtfield Lodge Nursing & Residential Home, Ormskirk; Croft Manor Residential Care Home, Fareham; Crossways, Walsall; Crown Meadow Nursing Home, Sandwell; Crystal Hall Residential & Nursing Home, Preston; Cuerden Grange Nursing Home, Preston; Cuerden Grange Rest Home, Preston; Davlyn House, Stoke-on-Trent; Deerwood Grange Nursing Home, Sutton Coldfield; Digby Manor Care Home, Erdington; Down House, Plymouth; Druids Meadow Residential Care, Birmingham; Dunkirk Memorial Home, Taunton; Eastleigh Nursing Home, Minehead; Edgbaston Beaumont Care Home, Birmingham; Edward House, Nottingham; Elizabeth House Care South, Poole; Elm House, Nottingham; Eryl Fryn Nursing & Residential Home, Llandudno; Evedale Care Home, Coventry; Eversleigh, Wolverhampton; Farehaven Lodge, Fareham; Frethey House Care Home, Taunton; Giltbrook House Nursing Home, Nottingham; Godiva Lodge Care Home, Coventry; Greenfield Nursing Home, Preston; Greville House Care Home, Sutton Coldfield; Hallcroft, Nottingham; Hamilton House Care Centre, Portsmouth; Haulfre, Beaumaris; Hawthorne Lodge, Nottingham; Hazel House Nursing Home, Leyland; Heartlands Care Home, Birmingham; Heathlands Care Home, Poole; High Pastures Nursing Home, Deganwy; Holmpark Care Home, Birmingham; Ivy House Care Home, Birmingham; Kelvedon House, Wednesbury; Keresley Wood Care Centre, Coventry; Kingland House Residential Home, Poole; Kinmel Lodge, Kinmel Bay; Knoll House, Wolverhampton; Lake View Nursing Home, Chorley; Landermeads Nursing Home, Nottingham; Langdale Nursing Home, Gosport; Leabrook House Nursing Home, Sandwell; Lilliput House Care Home, Poole; Llys Eilian, Colwyn Bay; Longdean Lodge, Portsmouth; Lucton House Care Home, Birmingham; Lyme Valley House, Newcastle-under-Lyme; Maple Dene Care Home, Birmingham; Marian House Nursing Home, Sutton Coldfield; Mary Street Care Home, Birmingham; Meadow House Residential Home, Portsmouth; Meadowbank Nursing Home, Bamber Bridge; Meadowside/St Francis, Plymouth; Melbourne House Care Home, Coventry; Merafield View, Plymouth; Merry Hall Nursing and Residential Care Home, Fareham; Moorcroft, Llandudno; Moorhaven Care Home, Taunton; Moorlands Nursing Home, Nottingham; Moundsley Hall Nursing and Residential Home, Birmingham; Mountbatten Nursing Home, Taunton; Neville Williams House, Birmingham; New Day Nursing Home, Birmingham; New Milton Nursing Home, Stoke-on-Trent; Northcott House Nursing & Residential Care Home, Gosport; Oakland Grange, Southsea; Oaklands Care Centre, Birmingham; Oakwood Rest Home, Erdington; Olivet Christadelphian Care Homes, Birmingham; Orchard House Nursing Home, Sutton Coldfield; Osborn Manor, Fareham; Park House Nursing Home, Nottingham; Park Manor Care Home, Poole; Parkfields, Wolverhampton; Parkwood House Nursing Home, Plymouth; Parkwood House Residential Home, Plymouth; Patricia Venton Centre, Plymouth; Peel House Nursing and Residential Home, Fareham; Pelsall Hall, Walsall; Perry Locks Nursing Home, Birmingham; Pine Meadows Care Centre, Leek; Plas Gogarth, Llandudno; Plas Gwilym, Caernarfon; Plas Gwynfa Nursing Home, Abergele; Plas y Bryn Nursing Home, Caernarfon; Priestly Rose Care Home, Erdington; Prince of Wales Care Home, Shirley; Priory Grange, Rhos on Sea; Queen Anne Lodge, Southsea; Queenswood Methodist Homes for the Aged, Nottingham; Radnor House Care Home, Birmingham; Rayner House, Solihull; Richmond Hall, Walsall; Robert Harvey House Care Home, Birmingham; Rosemary Lodge Nursing Home, Birmingham; Roxton Nursing Home, Sutton Coldfield; Ruksar, Wolverhampton; Russell Churcher Court, Gosport; Rylands, Newport; Sant Tysilio Nursing Home, Llanfairpwll; Seaview Residential Care Home, Southsea; Selbourne Court Care Home, Coventry; Selly Park Care Centre, Birmingham; Sherwood Court, Preston; Silverbirches, Birmingham; Silverdale Nursing Home, Newcastle-under-Lyme; Solent Cliffs Nursing & Residential Care Home, Fareham; Somerset Abbeyfield, Taunton; South View Lodge, Hesketh Bank; Southern House, Abergele; Sovereign House Nursing Home, Coventry; Springbank Nursing Home, Stoke-on-Trent; Springfield Court Nursing Home, Aughton; Springfield House Nursing Home, Codsall; Springfield Nursing Home, Chorley; Springfield Residential Home, Nottingham; Springholme Care (Anglesey) Ltd, Pentraeth; St Agnes’s Care Home, Erdington; St Bernards Residential Care Home, Solihull; St Catherine’s Residential Home, Sutton Coldfield; St Joseph’s Home, Birmingham; St Martin’s Nursing Home, Sutton Coldfield; St Paul’s Convent, Birmingham; St Ronan’s Nursing and Residential Care Home, Southsea; St Vincent House, Gosport; Staddon Lodge Residential Care Home, Poole; Stennards Leisure Home, Birmingham; Stratford Court Care Home, Birmingham; Sunrise Operations Tettenhall Ltd (Assisted Living), Wolverhampton; Sunrise Senior Living Solihull, Solihull; Sycamore House Care Home, Nottingham; Talbot View, Bournemouth; Tamar House, Plymouth; Templeman House, Bournemouth; Thalassa Nursing Home, Gosport; The Beaufort Care Home, Coventry; The Carlton Care Home, Nottingham; The Field House Care Home, Birmingham; The Firs Nursing Home, Nottingham; The Friendly Inn, Birmingham; The Gables Care Home, Sutton Coldfield; The Grange, Nottingham; The Green Nursing Home, Birmingham; The Grove, Solihull; The Haven Nursing Home, Coventry; The Herons, Nottingham; The Manor House, Stone; The Old Vicarage, Long Eaton; The Orchard, Wednesbury; The Poplars Nursing Home, Sandwell; The Priory Nursing and Residential Home, Shirley; The Priory, Telford; The Regency Nursing Home, Southsea; The Ridings, Birmingham; The Shrubbery, Codsall; Torr Home Nursing, Plymouth; Torr Home Residential, Plymouth; Tudor House Care Home, Birmingham; Uplands Nursing Home, Birmingham; Valley View, Plymouth; Victoria Gardens Care Home, Coventry; Wellesley House Nursing Home, Wolverhampton; Winsor Nursing Home, Minehead; Woodcote Hall, Newport; Woodcroft, Colwyn Bay; Woodland Villa Nursing Care Home, Plymouth; Woodland Villa Residential Care Home, Plymouth; Woodside Grange Care Home, Rhos on Sea; Wrottesley Park House Nursing Home, Wolverhampton; and Wyndeley Grange Nursing, Sutton Coldfield.

Contributions of authors

Catherine M Sackley (Professor of Rehabilitation, University of East Anglia) was the chief investigator, was responsible for developing the intervention and contributed significantly towards the development of the study protocol and drafting the report.

Marion F Walker (Professor of Stroke Rehabilitation, University of Nottingham) was a co-applicant for funding, was responsible for developing the intervention and contributed significantly towards the development of the study protocol.

Christopher R Burton (Senior Research Fellow, Bangor University) was a co-applicant for funding and contributed significantly towards the development of the study protocol.

Caroline L Watkins (Professor of Stroke and Older People’s Care, University of Central Lancashire) was a co-applicant for funding and contributed significantly towards the development of the study protocol.

Jonathan Mant (Professor of Primary Care Research, Institute of Public Health) was a co-applicant for funding and contributed significantly towards the development of the study protocol.

Andrea K Roalfe (Senior Lecturer in Medical Statistics, University of Birmingham) was the trial statistician. She wrote the statistical analysis plan and carried out the statistical analysis.

Keith Wheatley (Professor of Medical Statistics, University of Birmingham) was a co-applicant for funding and contributed towards the development of the study protocol and the final report.

Bart Sheehan (Consultant in Liaison Psychiatry, John Radliffe Hospital, Oxford) was a co-applicant for funding and contributed significantly towards the development of the study protocol.

Leslie Sharp (Dissemination Officer, University of East Anglia) was a co-applicant for funding, a service user and patient and public involvement representative, and contributed towards the development of the study protocol.

Katie E Stant (Study Co-ordinator, University of Birmingham) was the study co-ordinator and was involved significantly in the conduct of the trial.

Joanna Fletcher-Smith (Research Occupational Therapist, University of Nottingham) was involved in the development of the intervention and recruitment for the study and contributed to the final report.

Kerry Steel (Occupational Therapist, University of Birmingham) was involved in the development of the intervention, recruitment for the study, and contributed to the final report.

Garry R Barton (Reader in Health Economics, University of East Anglia) completed the health economic analysis and drafted the health economic evaluation in the report.

Lisa Irvine (Senior Research Associate, University of East Anglia) completed the health economic analysis and drafted the Health Economic evaluation in the report and contributed to the final report overall.

Guy Peryer (Senior Research Associate, University of East Anglia) was involved in data analysis, the creation of figures, wrote the final report and addressed reviewer comments on the final manuscript.

Other members of the trial team

K Lett (University of Birmingham), J Williams (Portsmouth Hospitals NHS Trust), F Rashid (University of Birmingham) and P Masterson-Algar (Bangor University).

Trial Steering Committee (independent members)

Professor H Dawes (independent chairperson, Oxford Brookes University), Professor R Harwood (Nottingham University Hospitals NHS Trust), Dr P Logan (Queen’s Medical Centre, Nottingham), N Phillips (Patient Representative) and Professor M Underwood (University of Warwick).

Data Monitoring Committee (independent members)

Professor D Barer (Newcastle University), Professor G Mountain (University of Sheffield) and Professor J Norrie (University of Aberdeen).

Publications

Fletcher-Smith JC, Walker MF, Cobley CS, Steultjens EM, Sackley CM. Occupational therapy for care home residents with stroke. Cochrane Database Syst Rev 2013;6:CD010116.

Fletcher-Smith J, Walker MF, Drummond A, Sackley CM. Occupational therapy for care home residents with stroke: what is routine practice? Br J Occup Ther 2013;76(Suppl. 1).

Masterson-Algar P, Burton CR, Rycroft-Malone J, Sackley CM, Walker MF. Towards a programme theory for fidelity in the evaluation of complex interventions. J Eval Clin Pract 2014;20:445–52.

Sackley CM, Burton C, Herron-Marx S, Lett K, Mant J, Roalfe A, et al. A cluster randomised controlled trial of an occupational therapy intervention for residents with stroke living in UK care homes (OTCH): study protocol. BMC Neurol 2012;12:52.

Sackley CM, Walker MF, Burton C, Watkins C, Mant J, Roalfe AK, et al. A cluster randomised controlled trial of an occupational therapy intervention for residents with stroke-related disabilities living in UK care homes (OTCH). BMJ 2015;350:h468.

Sands G, Kelly D, Birt L, Fletcher-Smith J, Sackley CM. An occupational therapy intervention for residents with stroke living in UK care homes: a content analysis of occupational therapy records from the OTCH trial. Br J Occup Ther 2015;78:422–30.

The treatment of stroke patients in UK care homes: the potential for occupational therapy. Stroke Matters. The Stroke Association; 2010.

Data sharing statement

Data are available from the chief investigator, Professor Catherine Sackley, at catherine.sackley@kcl.ac.uk

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health.

What is Occupational Therapy?

Occupational Therapy helps a person to achieve health and well-being through taking part in everyday tasks, such as dressing.

The therapist’s knowledge of adapting activities and surroundings is used to help people achieve the things they want to do and promote safety.

What will it involve?

The Occupational Therapist will be working with residents to help them carry out day-to-day activities such as:

• Moving around the home

• Dressing and grooming

• Washing, bathing or showering

• Eating and drinking

The OT in Care Homes study

At the start of the study homes will be divided into 2 groups, one group will receive the services of an occupational therapist for about 10 weeks. The other group will continue as before during the study but will receive staff training at the end.

Full details of the study are available from

[contact details removed]

What would happen if I take part?

A plan will be agreed between the resident and the occupational therapist/s. The aim is to help the resident achieve what they want to do. It will not involve anything they do not want to work on. It may include any of the following:

• Information and advice

  • Ways of managing both for the resident and their carers

• Activity and treatment

  • Relearning ways of doing activities or trying new methods

  • Activities aimed at improving the residents’ abilities, they may be asked to practice between sessions

  • Techniques to Increase/maintain mobility

• Equipment to help them manage

  • Walking aids

  • Equipment to help the resident with activities such as washing and eating.

• Home adaptations

  • Advice on layout of room furniture for ease of use

  • Advice to reduce hazards

  • Supply grab rails

• Wheelchairs

  • Advice on mobility, posture and comfort

• Seating

  • Dining room chair

  • Armchair

• Health and safety promotion

  • Reducing the risk of falls

  • Increasing/maintaining activity levels

• Referrals

  • Referrals can be made to other specialists according to the residents’ wishes and needs

Benefits of taking part

It is hoped that the time spent with the therapists will help the residents’ mobility and give them more choice over their day to day activities, in a safe and knowledgeable way.

For further information about Occupational Therapy please contact:

[contact details removed].

Study Title ‘Cluster randomised controlled trial of an occupational therapy intervention for residents with stroke in UK care-homes’

Study Number: REC09/H1210/88

Study Title ‘Cluster randomised controlled trial of an occupational therapy intervention for residents with stroke in UK care-homes (acronym – OTCH).’

What does the study involve?

This study aims to look at the effects and value of a targeted course of occupational therapy on people living with stroke in nursing and residential homes.

What will my assessment consist of?

Your assessment will be the same as you would conduct on any other patient. You will have access to the baseline outcome measures (Barthel index and Rivermead mobility score) to help inform your assessment and develop a patient specific treatment plan. You can use information from family and carers to inform your assessment if the patient has difficulties in communicating.

What treatments will I be expected to offer?

You will be offering all the usual interventions that you provide as an occupational therapist. The interventions you will be offering will address the performance of a task, the environment in which the task is conducted, and help address any impairments that may limit the performance of the ADL’s.

These interventions could involve:

• task-specific practice

• supplying aids/adaptations to the environment/help to simplify the task

• delivering specific therapeutic interventions

• educating carers in encouraging/assisting the patient.

Is there anything I shouldn’t do?

The aim of this study is to look at the role of occupational therapy in improving function for people post stroke so some interventions such as reminiscent therapy and relaxation groups are not within the remit of this study.

The outcome measures will highlight the main functional difficulties of the patients so treatment goals can focus on these key areas of functional difficulties.

Does education of carers/family count as a treatment?

Yes, any time spent on this activity counts as part of an intervention. There is space to record this in the treatment log. Discussions with other members of staff or agencies (such as the GP) are also to be included in the treatment log.

How do I record my interventions?

All interventions are split into types and times spent on each intervention needs to be recorded for each contact with the patient.

What if the patients’ needs change during treatment?

The aim is to provide the patients with standard occupational therapy care so, as the patients’ needs change, you can adapt you treatments to address patient specific needs as you would do with any other patient.

What equipment can I supply?

You can supply the adaptive equipment you would supply as an occupational therapist if it will benefit the patient i.e. grab rails, toilet raises, cutlery, mobility aids, seating, etc. There is the possibility of funding small pieces of equipment through the study if it is not available through the normal routes.

Where do I write my notes?

You may be required to write in the patients’ notes at the nursing home.

You will also have your profession specific notes that you will need to keep securely in a locked cabinet at your normal place of work.

How much time can I spend with each person?

This is dependent on the individual goals you have set with the patient and what interventions will best suit their needs. As mentioned previously, the aim of this study is to provide a typical occupational therapy intervention.

What happens if a patient gets hurt?

If a patient falls during a treatment or if the equipment provided fails it is described as an adverse event. All adverse events need to be reported to the study co-ordinator so they can be logged. You may also be required to complete the documentation used by the care home for reporting falls/injuries.

Mini-Mental State Examination

The MMSE was used to assess the cognitive function of patients. 75,87 The evaluation uses a series of 12 questions and tests in which patients are assessed in terms of their memory, language, attention and orientation. The test is scored out of 30, 30 being the maximum. A score between 0 and 20 is indicative of cognitive impairment.

Sheffield Screening Test for Acquired Language Disorders

The Sheffield Screening Test for Acquired Language Disorders was used to assess the severity of communication problems in study participants and to identify those with aphasia. 74 The test involves a 10-item questionnaire which is divided into two parts: one assessing receptive skills and one assessing expressive skills. The test is scored from 0 to 20, with 20 signifying no impairment. Participants scoring < 15 are considered to have a language disorder.

Barthel Index of Activities of Daily Living

The BI consists of a questionnaire which scores the patients’ ability to complete 10 activities. 76,77 The assessment consists of questions related to self-care activities (feeding, grooming, bathing, dressing, bowel and bladder care, and toilet use), and mobility-based activities (ambulation, transfers and stair climbing). The results of the questionnaire are added to provide a measure of independence ranging from 0 to 4 (very severe), 5 to 9 (severe), 10 to 14 (moderate), 15 to 19 (mild) and 20 (completely independent).

Rivermead Mobility Index

The RMI was used to evaluate the effectiveness of occupational therapy on the mobility of patients. 79,83 The index is delivered using a 15-item yes/no questionnaire which focuses on ambulation but also assesses the participant’s ability to run, stand unsupported, use stairs, and bathe unsupervised. Positive responses to each question are scored with 1 point, up to a maximum of 15.

Geriatric Depression Scale

The GDS is a screening test for depression in the elderly in which patients answer ‘yes’ or ‘no’ to questions which refer to how they felt over the preceding week, and can be delivered using either a long-form (30 questions) or a short-form (15 questions) questionnaire. 84,85 For the short version with 15 questions, scores of 0 to 4 are considered normal; 5 to 8 indicate mild depression; 9 to 11 indicate moderate depression; and 12 to 15 indicate severe depression.

EQ-5D-3L

The European Quality of Life EQ-5D (EQ-5D-3L) is a standardised measure of HRQoL, developed by the EuroQol Group in order to provide a simple, generic measure of health for clinical and economic appraisal. 95,99,114 The EQ-5D descriptive system comprises the following five dimensions: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. Each dimension has three levels of scoring (3L; no problems, some problems, extreme problems) and is expressed as a digit (1, 2 or 3). The digits for the five dimensions are combined in a five-digit permutation describing the respondent’s health state (e.g. 11121 moderate pain and discomfort, 13111 extreme problems with self-care). The study used the EQ-5D-3L measure to conduct an assessment of HRQoL and an economic evaluation.

Data table for Figure 2

Data table for Figure 3