Research Bridges self-management program for people with stroke in the community: A feasibility randomized controlled trial Suzanne McKenna1, Fiona Jones2, Pauline Glenfield3, and Sheila Lennon1,4* Background Enabling self-management behaviors is considered important in order to develop coping strategies and confidence for managing life with a long-term condition. However, there is limited research into stroke-specific selfmanagement interventions. Aim The aim of this randomized controlled trial was to evaluate the feasibility of delivering the Bridges stroke selfmanagement program in addition to usual stroke rehabilitation compared with usual rehabilitation only. Methods Participants recruited from the referrals to a community stroke team were randomly allocated to the Bridges stroke self-management program, receiving either one session of up to one-hour per week over a six-week period in addition to usual stroke rehabilitation, or usual rehabilitation only. Feasibility was measured using a range of methods to determine recruitment and retention; adherence to the program; suitability and variability of outcome measures used; application and fidelity of the program; and acceptability of the program to patients, carers and professionals. Results Twenty-five people were recruited to the study over a 13-month period. Eight out of the 12 participants in the Bridges stroke self-management program received all six sessions; there was one withdrawal from the study. There were changes in outcomes between the two groups. Participants who received the Bridges stroke self-management program appeared to have a greater change in self-efficacy, functional activity, social integration and quality of life over the six-week intervention period and showed less decline in mood and quality of life at the three-month follow-up. Professionals found the program acceptable to use in practice, and feedback from participants was broadly positive. Conclusions The findings from this study appear promising, but questions remain regarding the feasibility of delivering the Bridges stroke self-management program in addition to usual rehabilitation. The dose response of receiving the program cannot be ruled out, and the next stage of research should Correspondence: Sheila Lennon*, Physiotherapy (Flinders University), Health Sciences Building, Repatriation General Hospital, Daws Road, Daw Park, Adelaide, SA 5041, Australia. E-mail: [email protected] 1 School of Health Sciences, Health and Rehabilitation Sciences Research Centre, University of Ulster, Newtownabbey, UK 2 Faculty of Health & Social Care, St George’s University of London and Kingston University, London, UK 3 Community Stroke Team, South Eastern Health and Social Care Trust, Bangor, UK 4 Physiotherapy, School of Medicine, Flinders University, Adelaide, Australia Received: 27 November 2012; Accepted: 5 August 2013; Published online 21 November 2013 Conflicts of interest: None declared. Bridges is to be used only for the purpose of stroke self-management. All intellectual property rights in Bridges stroke self-management are owned by St George’s University of London and Kingston University, and shall not be used without their prior written consent. DOI: 10.1111/ijs.12195 © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
explore the feasibility of an integrated program. Exploration of the reasons behind relatively low recruitment and of the sensitivity of outcome measures to detect a change are also required. Additional investigation of intervention fidelity is required to monitor if the program is being delivered as intended. Key words: behavioral intervention, chronic disease, clinical trial, feasibility, self-management, stroke
Introduction Stroke is a major cause of disability worldwide, and the number of people living with the consequences of stroke is increasing on a global scale (1). A number of recent studies in the United Kingdom have highlighted a range of unmet needs post-stroke, particularly after the period of acute care and rehabilitation has been completed. There are indications from this research that individuals can feel ill prepared to cope in the longer term and ‘abandoned’ by services (2,3). It has been suggested that selfmanagement initiatives used for people with other long-term conditions may also be applicable to stroke (4). Self-management has been defined in different ways, but is commonly described as ‘an individual’s ability to manage the symptoms, treatment, physical and psychosocial consequences, and life style changes inherent with living with a chronic disease’ (5). This includes a range of attitudes, behaviors and skills directed at managing life with a chronic disease. Self-management programs have been shown to improve psychological outcomes and also have modest effects on health and social care needs and costs (6). Stroke self-management programs (SSMPs) are relatively rare (7). Two group-based programs delivered to stroke participants have demonstrated feasibility and some influence on reducing functional decline post-stroke, but findings are mixed (8,9). As with other long-term conditions, stroke may benefit from having different types of programs available, including both group and individual programs. Individualized programs could run alongside or be embedded into stroke rehabilitation. They may also be more applicable for those people who do not want to or are unable to attend group-based programs. The Bridges SSMP is an individualized, theory-driven program based on self-efficacy principles. It was developed in 2006 and tested on different people post-stroke through a series of single case studies; significant changes in self-efficacy were found, which suggested preliminary proof of concept (10). Further development and consultation with stroke patients, carers and professionals has informed the program, which now comprises structured one-to-one sessions with a professional to support patients in setting goals, recording progress and planning activities; it incorporates a patient-held workbook to facilitate a personal record of goals, progress and helpful strategies (10). Vol 10, July 2015, 697–704
697
Research Aim The aim of this randomized controlled trial (RCT) was to evaluate the feasibility of delivering the Bridges SSMP in addition to usual stroke rehabilitation compared with usual rehabilitation only. The following objectives were set in order to explore feasibility of the Bridges SSMP and the study design adopted for this feasibility RCT: 1. Explore participant recruitment and retention 2. Evaluate participant adherence to the Bridges SSMP 3. Assess the suitability of the selected outcome measures 4. Explore changes between groups in quality of life [EuroQol (11), Stroke Specific Quality of Life (SSQOL) (12)], self-efficacy [Self-Efficacy Scale (SES) (13), Stroke Self-Efficacy Questionnaire (SSEQ) (14)], functional activity [Barthel Index (15), Nottingham Extended Activities of Daily Living (NEADL) (16)], mood [General Health Questionnaire (GHQ) (17)] and community integration [Subjective Index of Physical and Social Outcome (SIPSO) (18)] 5. Explore the application and fidelity of the Bridges SSMP by the community stroke team 6. Explore the acceptability of the Bridges SSMP to participants and the community stroke team
Methods Ethical approval for this trial was granted by the Office for Research Ethics Committee Northern Ireland (08/NIROI/67). Design This was the first attempt to test the feasibility of implementing the Bridges SSMP in a community stroke service. The emphasis of the study was on the feasibility of delivery, including recruitment and retention; adherence to the program; suitability of and changes in outcome measures used; application and fidelity of the program; and acceptability of the program to patients, carers and professionals. Participants Any new stroke referral to the community stroke team of a Health and Social Care Trust in Northern Ireland or any stroke survivor who was within four-weeks of commencing rehabilitation with the community stroke team was considered for inclusion against the predetermined eligibility criteria. Eligibility criteria included a clinical diagnosis of stroke and the ability to follow a two-stage command. Patients were excluded if they were confused [with a mini mental score of less than 6 out of 10 (19)]; demonstrated neglect [with a star cancellation score of less than 48/52 (20)]; had competing comorbidities influencing their rehabilitation, e.g. malignant cancer; had significant issues with their readiness to participate as judged by the clinical team, e.g. illness of a carer; did not require weekly multidisciplinary rehabilitation, e.g. one or two advice sessions or review only; or had hearing and visual impairments precluding their ability to fully participate in the Bridges SSMP.
698
Vol 10, July 2015, 697–704
S. McKenna et al. Randomization and allocation of control/intervention Opaque sealed envelopes concealing the group allocation were prepared using block randomization. After written informed consent and baseline assessment, participants were randomly assigned to receive the Bridges SSMP or to a control group. Health professionals delivering the Bridges SSMP were not involved in the randomization process, and the research associate responsible for conducting outcome assessments and data processing was unaware of the group allocation. Neither the health professionals delivering the Bridges SSMP nor the participants could be blinded to group allocation. Intervention Participants allocated to the control group received the usual multidisciplinary stroke rehabilitation delivered by the community stroke team. In Northern Ireland this commonly consists of information on stroke; treatment input from two or more professionals, including involvement in goal setting and discharge planning; and advice on services, benefits and allowances. The Bridges SSMP Each participant randomized to the Bridges SSMP received one session of up to one-hour per week over a six-week period in addition to their usual rehabilitation. Although the Bridges SSMP was designed as a six-week program, in line with previous research by Kendall et al. (9), we defined compliers as those individuals who engaged in four or more sessions and used these criteria to distinguish between attendees and nonattendees. The SSMP was delivered by one of three members of the community stroke team who had previously undergone training in the program. The Bridges SSMP is distinct from usual rehabilitation and includes structured one-to-one sessions using strategies to promote specific behaviors that exemplify the hallmarks of selfmanagement (21). These behaviors include enabling patients to work out ways of taking control of their daily lives by setting small targets, recording their progress and problem solving. The content of the sessions is determined by the patient and their personal goals rather than being professionally directed. Selfmanagement behaviors are facilitated and reinforced by a patientheld stroke workbook that includes space to record personal goals and progress after stroke; it avoids technical terms and includes vignettes from other stroke survivors to illustrate successful selfmanagement (21). All professionals completed a treatment log for each individual enrolled in the trial, including the number of visits by each healthcare professional on a weekly basis. In addition, for participants receiving the Bridges SSMP, professionals were asked to record which components they worked on at each visit and the individual targets decided by each patient. The aim of this record was to explore whether the program had been delivered as intended and any differences in practice between using the Bridges SSMP and delivering usual care. Training in Bridges SSMP Training for professionals who delivered the Bridges SSMP took place over two-and-a-half-days. Day 1 was an introduction to self-management and current research evidence relevant to © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research
S. McKenna et al. stroke, followed by demonstration and practice of how to use the SSMP using case-based scenarios. Each professional had a minimum of two practice sessions with stroke patients before commencement of the trial. FJ then conducted a half-day of observation with the professionals to ensure that they were delivering the Bridges SSMP according to the key principles agreed by the trial management group. Feedback was given by FJ and steps were put in place to ensure consistency and fidelity, including a checklist for professionals to use to enable alignment with the principles of the Bridges SSMP. A further training day took place three-months later to explore issues that might impact delivery and to confirm the focus on patient-led goals, use of the workbook, reflection on progress and problem solving, and ways to engage participants in these strategies. Outcome measures A research associate, blinded to group allocation, assessed primary and secondary outcomes for all participants at baseline (pre-randomization), at discharge from the Bridges SSMP and three-months post-program completion. A range of outcome measures were used to test their feasibility and explore whether they would be meaningful to use in a fully powered trial. Previous research into self-management programs for people with stroke has demonstrated significant improvement in quality of life for participants (8), so health-related quality of life was chosen as one of the primary outcome measures. It was measured using both the EuroQol (11) and the SSQOL (12). Self-efficacy was also chosen as a primary outcome measure, as the Bridges SSMP intervention directly focuses on enhancing the sources of self-efficacy researched by Bandura (22), Lorig and Holman (23), and Marks (24) in order to influence confidence to self-manage. Self-efficacy was measured using the SES developed by Lorig et al. (13) and the SSEQ developed by Jones et al. (14). Both generic and stroke-specific measures were used to enable us to determine the most responsive measure. The choice of secondary outcome measures was also informed from the findings of studies exploring self-management programs. Secondary outcomes were functional independence [Barthel Index (15)], activity [NEADL (16)], mood [GHQ-28 (17)] and community integration [SIPSO (18)]. Data analysis Analysis of quantitative data focused on testing the changes in primary and secondary outcomes between participants receiving the Bridges SSMP in comparison with the control group using mean change scores and confidence intervals. Primary endpoints were mean change scores between groups in the health-related quality of life and self-efficacy measures up to three-months postprogram completion. Baseline characteristics were described and any differences between the two groups compared using the chisquare test and Mann–Whitney test. When using Mann–Whitney, each test was separately evaluated for the missing values (25). Qualitative methods Interviews were conducted with patients, carers and professionals to explore any issues relating to the acceptability of the program and feasibility to use with stroke patients in a community setting. The aim was to explore factors that impacted on delivery, consis© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
tency and/or compliance with the program. All qualitative data were transcribed and analyzed using a content-thematic approach. A summary of key qualitative findings is presented, but more detailed analysis of the qualitative data will be presented in a future paper.
Results Recruitment and retention As this was a feasibility study, a power calculation was not conducted. A total sample size of 30 individuals was agreed to be feasible through consultation between the research and community stroke teams. However, given the time restrictions of this study and staffing constraints of the community stroke team, 25 participants were recruited and randomized into either the control group (n = 13) or Bridges SSMP group (n = 12) over a 13-month period. Of the 152 individuals referred to the community stroke team during this period, 13 referrals were missed when recruitment was suspended owing to staffing constraints of the community stroke team. Of the remaining 139 participants who were screened, 25 were recruited and enrolled in the trial (see Fig. 1). Of the 25 individuals recruited to the study, all but one, who was a participant in the Bridges SSMP group, demonstrated full compliance with the intervention and assessment schedule. The most significant reasons for exclusion were participants requiring less than six-weeks of treatment (n = 48 individuals excluded), memory/cognitive problems (n = 15) and individuals not wanting to participate in the research trial (n = 16). Other reasons for exclusion included competing comorbidities (n = 5), poor rehabilitation potential as determined by the community stroke team (n = 5), readmission to hospital (n = 4), being treated for residual problems arising from a previous stroke (n = 3), unusual diagnosis (n = 2), being readmitted to hospital (n = 1) and death (n = 1). Thirteen further patients were excluded, and further exploration determined that these exclusions were usually associated with a perception by the team regarding readiness to participate. There were no statistically significant differences between groups at baseline for any measure; however, given that there were small numbers in each allocation group, we could not ensure groups were balanced with regard to their characteristics, so caution needs to be taken when interpreting results. Of particular note is that the control group was a higher-functioning group as measured by the Barthel Index (Bridges SSMP = 14·09; control group = 17·08) and the NEADL Scale (Bridges SSMP = 26·00; control group = 34·38); they were also at a later stage post-stroke (Bridges SSMP = 7·00 weeks post-stroke; control group = 11·38 weeks post-stroke) (see Table 1). Participant adherence to the intervention Adherence to the Bridges SSMP was high, with all but one participant completing four or more sessions. One participant (n = 1/12) did not receive the Bridges SSMP as intended, withdrawing from the study at week three. This participant was a younger individual, aged 24 years, who did not think the SSMP would benefit him. Eight individuals (n = 8/11; 73%) received all Vol 10, July 2015, 697–704
699
Research
S. McKenna et al.
Fig. 1 Recruitment process.
six sessions, three individuals (n = 3/11; 27%) received five sessions, and one individual (n = 1/11; 9%) received four sessions. According to feedback from professionals the majority of participants engaged with the program at each session; they liked the workbook and used it to record their personal successes, goals and targets. Suitability of outcome measures An independent outcome assessor was asked to provide feedback on the feasibility of using each of the outcome measures. The main issue was the time required to assess a participant on all eight outcome measures, which could take up to two-hours depending on the participants’ ability to concentrate and mood state. Some of the questions within measures were of a sensitive nature, for example, sexual relationships (SSQOL) and suicidal thoughts (GHQ). Some participants objected to these items, provoking a negative response from individuals. Overall it was felt
700
Vol 10, July 2015, 697–704
that the number of outcome measures used was a potential burden to participants, and not all measures had the potential to yield meaningful data. Changes in outcomes At baseline two participants had incomplete data sets. One participant did not complete the SES, and another participant failed to complete the NEADL measure. At each time point, a number of participants chose not to answer an item on the SSQOL scale regarding sexual relationships. The differential outcomes between the Bridges SSMP group and the control group were explored to identify whether the Bridges SSMP had any impact on quality of life, self-efficacy, activity, mood and community integration in comparison with routine multidisciplinary team care alone in community dwellers post-stroke. Changes were noted in all outcome measures postintervention irrespective of group allocation (see Table 2). © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research
S. McKenna et al.
Table 1 Participant characteristics at baseline Bridges SSMP (n = 11) Age, mean (SD) Weeks post-stroke, mean (SD) Outcome measure score, mean/max (SD) EuroQol Index EuroQol VAS SSQOL SSEQ SES Barthel Index NEADL SIPSO GHQ 28 Carer, n (%) Family HSC None Gender, n (%) Female Male Previous CVA, n (%) Yes No TIA
Control (n = 13)
P value (Mann–Whitney)
62·18 (13·57) 7·00 (4·45)
67·38 (10·60) 11·38 (12·70)
0·31 0·28
0·42/1 (0·36) 0·50/1 (0·19) 13·22/20 (2·35) 6·68/10 (2·56) 7·51/10 (2·45) 14·09/20 (5·30) 26·00/63 (14·19) 19·82/40 (6·75) 24·09/84 (10·90)
0·53 (0·41) 0·56 (0·16) 14·62 (3·42) 7·94 (1·85) 8·14 (2·09) 17·08 (3·40) 34·38 (17·44) 23·69 (9·00) 23·60 (15·52)
0·31 0·37 0·16 0·20 0·56 0·10 0·31 0·20 0·66
8 (72·7) 1 (9·1) 1 (9·1)
9 (69·2) 2 (15·4) 1 (7·7)
0·90
4 (36·4) 7 (63·6)
7 (53·8) 6 (46·2)
0·33
2 (18·2) 8 (72·7) 1 (9·1)
2 (15·4) 10 (76·9) 1 (7·7)
0·97
CVA, cerebrovascular accident; GHQ, General Health Questionnaire; HSC, health and social care; Max, maximum; NEADL, Nottingham Extended Activities of Daily Living; SD, standard deviation; SES, Self-Efficacy Scale; SIPSO, Subjective Index of Physical and Social Outcome; SSEQ, Stroke Self-Efficacy Questionnaire; SSMP, stroke self-management program; SSQOL, Stroke-Specific Quality of Life; TIA, transient ischemic attack; VAS, visual analogue scale.
Table 2 Mean change scores for primary outcome measures
Primary outcome
Maximum score
Score change indicating improvement
EuroQol Index
1
+
EuroQol VAS
1
+
SSQOL
20
+
SSEQ
10
+
SES
10
+
Group
Baseline to program completion/six-weeks
Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control
0·09 0·15 0·13 0·12 1·11 1·94 1·04 0·65 0·79 0·36
CI
Program completion/six weeks to three-months follow-up
CI
−0·09, 0·30 −0·10, 0·35 0·05, 0·27 0·05, 0·21 0·15, 2·65 0·74, 3·09 0·05, 1·69 0·08, 0·99 −0·68, 1·69 −0·16, 0·89
−0·05 −0·09 0·02 −0·02 1·05 0·12 −0·39 −0·15 −0·51 0·22
−0·13, 0·08 −0·37, 0·18 −0·13, 0·22 −0·16, 0·14 0·46, 1·60 −1·35, 1·37 −0·90, 0·28 −1·01, 0·71 −1·21, 0·56 −0·95, 1·33
CI, confidence interval; SES, Self-Efficacy Scale; SSEQ, Stroke Self-Efficacy Questionnaire; SSMP, Stroke Self-Management Program; SSQOL, StrokeSpecific Quality of Life; VAS, visual analogue scale.
At program completion/six-weeks both groups had evolved similarly with regard to the EuroQol scores, but the control group demonstrated more change in the SSQOL measure. At follow-up both groups showed an overall decline in the EuroQol; however, participants receiving the Bridges SSMP continued to show improvement in their SSQOL scores (see Fig. 2). At program completion/six-weeks, the Bridges SSMP participants showed more change in their self-efficacy scores (SES and SSEQ) than the control group (see Table 2). At follow-up both © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
groups showed decline in SSEQ scores (see Fig. 3), with the control group showing some improvement in SES scores. Conflicting findings were observed among secondary measures. At program completion/six-weeks positive results were found favoring the Bridges SSMP participants in functional activity (Barthel Index and NEADL) and social integration (SIPSO), while changes in mood favored the control group (GHQ). At three-month follow-up, the control group showed more improvement in NEADL and SIPSO measures. Both groups showed a Vol 10, July 2015, 697–704
701
Research
S. McKenna et al.
decline in the GHQ measuring mood, with the Bridges SSMP group showing less decline (See Table 3). Application and fidelity of the Bridges SSMP All professionals were instructed to complete a treatment log, whether they were involved in delivering the Bridges SSMP or providing rehabilitation to those in the control group. Data recorded in the logs were limited in detail, which meant the time spent on different components of the Bridges SSMP could not be determined. Professionals stated that most time was spent assisting participants in setting small targets that were meaningful to them, while between sessions participants acted on these targets and used the workbook as a resource to help problem-solve
Fig. 2 Stroke-Specific Quality of Life (SSQOL) means and confidence intervals. BG, Bridges group; CG, control group.
common issues. This directly addresses mastery, which is thought to be a key source of self-efficacy (22). This information indicates that practices being delivered during the study in the intervention group were authentic to the principles of the Bridges SSMP, which focus on promoting self-efficacy. However, the limited detail captured in the treatment logs precludes detailed analysis of intervention fidelity. Goals were recorded for all participants (n = 25) in both groups in the intervention logs. Analysis of records showed there were subtle differences between groups in how goals were described. While goals often focused on issues relating to mobility, irrespective of group allocation, for participants receiving the Bridges SSMP goals were recorded using language and terms that were expressed in the participants’ own words and were linked to their everyday activities. Details of the professionals involved in their treatment were recorded for 12 (n = 12/13) participants in the control group. On average, participants in the control group were seen by three to four different professionals during their rehabilitation. The aim of the intervention logs was to provide a detailed comparison between the amount of therapy received by the control group and Bridges SSMP participants, but data were missing in five (n = 5/ 11) of the Bridges SSMP participants. Where details are provided, it appears that the amount of therapy was similar for both groups. The acceptability of the Bridges SSMP Feedback was positive, with the professionals saying they valued the training and that delivering the program highlighted how the Bridges SSMP differed from their usual practice. Some feasibility issues were identified, but these were generally associated with the constraints of running a RCT within clinical practice and not the Bridges SSMP itself, e.g. timing of intervention delivery and additional paperwork. Overall, patients appeared to value the program and aspects of delivery such as the workbook and one-to-one facilitation with a health-care professional.
Discussion
Fig. 3 Stroke Self-Efficacy Questionnaire (SSEQ) means and confidence intervals. BG, Bridges group; CG, control group.
This is the first attempt to test the feasibility of the Bridges SSMP in a community stroke rehabilitation service. This study has raised a number of questions that require further investigation
Table 3 Mean change scores for secondary outcome measures
Secondary outcome
Maximum score
Score change indicating improvement
Barthel Index
20
+
NEADL
63
+
SIPSO
40
+
GHQ 28
84
−
Group
Baseline to program completion/six-weeks
Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control
1·73 1·46 17·40 12·69 7·55 4·77 −8·45 −11·31
CI
Program completion/six weeks to three-months follow-up
CI
0·14, 2·86 −0·07, 2·40 7·70, 27·10 5·67, 20·66 2·36, 14·04 1·66, 8·00 −14·05, −3·95 −19·28, −3·39
0·73 −0·08 2·09 4·23 0·55 4·08 0·45 2·77
−0·27, 1·27 −1·37, 1·04 −3·97, 6·77 −3·00, 10·67 −1·79, 3·99 −1·33, 8·83 −4·26, 5·26 −8·60, 13·93
CI, confidence interval; GHQ, General Health Questionnaire; NEADL, Nottingham Extended Activities of Daily Living; SIPSO, Subjective Index of Physical and Social Outcome; SSMP, stroke self-management program; VAS, visual analogue scale.
702
Vol 10, July 2015, 697–704
© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
S. McKenna et al. before feasibility can be fully confirmed. One of the key issues of interest was in relation to recruitment. The results showed that only 17·9% (n = 25/139) of individuals referred to the community stroke team were eligible to participate. This raises questions about the inclusion criteria and also decision-making by the community team, as a number of potential participants were judged to be unsuitable mainly on the basis of poor rehabilitation potential or a perceived disinclination to participate in a selfmanagement program. However, the number of patients who did not require weekly multidisciplinary rehabilitation, e.g. one or two advice sessions, accounted for the largest reason for nonrecruitment. This program was designed to be delivered by professionals and ultimately as part of their usual practice; however, it may be pertinent to consider designing a program that can be delivered over a smaller number of sessions. In addition, the rationale for exclusion requires further exploration if the aim is to design a program to be feasible to use within stroke rehabilitation. Retention of participants once recruited to the study was found to be acceptable in the Bridges SSMP group, and once recruited, all but one of the 12 individuals randomized to the SSMP demonstrated full compliance with the program and assessment schedule. The high level of compliance demonstrated in this study is greater than that demonstrated in previous research exploring group self-management programs in the stroke population. Kendall et al. (9) found 64% of individuals in their intervention group (n = 37/58) demonstrated compliance, while Cadilhac et al. (8) reported that only 52% (n = 25/48) of individuals recruited to their SSMP completed the program. An individualized program does have the potential to be more flexible and responsive to individual needs and ultimately could be more cost-effective if integrated into usual rehabilitation practice. Any analysis of outcomes needs to take into consideration the baseline characteristics of each group, in particular that the control group were a higher-functioning group and at a later stage post-stroke, which may suggest that they had less opportunity to improve in a range of outcome measures. In this study sample size was appropriate for an exploration of feasibility but not for testing any differences between groups. However, the outcomes explored do show some promise, as participants who had received the Bridges SSMP appeared to demonstrate more favorable changes in self-efficacy, functional activity (Barthel Index and NEADL) and social integration (SIPSO) over the six-week program period and showed less decline in mood and some improvement in quality of life at three-month follow up in comparison with the control group. The finding that Bridges SSMP participants demonstrated more improvement in self-efficacy over the intervention period is not surprising given that the intervention was designed specifically to impact on self-efficacy and has previously demonstrated proof of concept (10). Change scores between groups were relatively small, and further research would need to be conducted producing comparable findings in a larger sample before any conclusions could be made with regard to the impact of the Bridges SSMP. The organization of stroke care and the acute sudden nature of stroke are critical when examining the efficacy of selfmanagement programs. It could be argued that all or some of the © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research strategies and skills can be addressed as part of usual stroke care and rehabilitation, and programs could be more effective if these strategies were already integrated into interactions with healthcare professionals (26). Kennedy and colleagues reviewed the evidence from self-management programs for other long-term conditions and highlighted areas that may cause a program to fail (27). These issues include not taking account of professional beliefs and attitudes, failing to consider the organizational structure in which the program is delivered and not having an understanding of the individual challenges of the target population (26). Limitations and future research Several issues were highlighted in this study that should be addressed in future research to test the utility of this and other SSMPs. The issue of recruitment requires careful consideration to minimize the risk of participants being wrongly excluded. Modifications to training are required to explore and address any potential barriers to using the program, thus increasing accessibility of the program to a broader range of stroke patients. Our previous research has shown the workbook can be a barrier if a participant is unable to read or follow a two-stage command, particularly if there is no carer who can help (28). In this study a substantial amount of time was required to complete the range of outcome measures, which resulted in a substantial burden for participants. It would seem necessary in future research to refine the number of measures used. The suggestion from the results of this study would be to retain the primary outcome measures related to quality of life (EuroQol and SSQOL) and self-efficacy (SES and SSEQ) but to reduce the number of or find alternative secondary measures that may be less time-intensive. Although the Bridges SSMP used in this study is somewhat standardized by employing a workbook structure, it would be advisable in future trials to keep more detailed records regarding the time spent on each component. This would enable investigation of whether there is any component or combination of components of the SSMP that is associated with more positive outcomes. It would also help ensure that the fidelity of the program is being monitored on an ongoing basis. In future, the application of the Bridges SSMP and fidelity to the program will require detailed scrutiny to determine the key differences from usual practice and ensure the program is being delivered as intended. The study design also meant transfer of knowledge between professionals treating control participants and those treating Bridges SSMP participants could not be ruled out. The use of a cluster design to minimize treatment contamination with an embedded process evaluation to explore contextual issues such as professional and patient experiences with larger sample sizes will be a natural next step in the development and testing of this intervention. The dose response of receiving the Bridges SSMP is a possibility within this feasibility RCT, and the next stage of research will also explore the feasibility of integrating the program into usual stroke rehabilitation. A larger study will also need to consider stratification of samples, especially with regard Vol 10, July 2015, 697–704
703
Research to functional ability, which varied considerably between groups in this study. Data collection to allow for covariate analysis would also help to determine under which conditions and to what sample of individuals the intervention should be delivered to optimize its effects. The success of any individualized SSMP also depends on the behaviors and enthusiasm of the professionals delivering the intervention, and further work is required to explore the feasibility of how self-management programs can become more universally acceptable and minimize provider burden. Integration of programs such as the Bridges SSMP into community stroke rehabilitation may lessen the demand on resources, but a full economic evaluation of the costs of providing the program in comparison to usual care is required.
Conclusion Questions remain regarding the feasibility of delivering the Bridges SSMP in addition to posthospital stroke rehabilitation; however, the Bridges SSMP can be delivered to both male and female participants at a range of ages with different levels of stroke severity. The program also enabled a patient-focused process for involving participants in their rehabilitation by promoting self-management practices. The following key feasibility issues were identified: low recruitment rate, small sample size, rigidity of eligibility criteria, small treatment effects, excessive number of outcome measures and inadequate recording of treatment fidelity. The dose response of receiving the program cannot be ruled out, and the next stage of research should explore the feasibility of an integrated program using a cluster-randomized controlled design to minimize treatment contamination with a full economic evaluation of the costs of providing this individualized program in comparison with usual care.
Acknowledgements The work of Suzanne McKenna was supported by the award of a PhD studentship from the Department of Employment and Learning. This feasibility study was funded by Northern Ireland Chest, Heart and Stroke.
References 1 WHO. Global Status Health Report on Noncommunicable Diseases 2010. Geneva, World Health Organization, 2010. 2 Care Quality Commission. Supporting Life after Stroke: A Review of Services for People Who Have Had a Stroke and Their Carers. London, Care Quality Commission, 2011. 3 McKevitt C, Fudge N, Redfern J et al. Self-reported long-term needs after stroke. Stroke 2011; 42:1398–403. 4 Jones F. Strategies to enhance chronic disease self-management: how can we apply this to stroke? Disabil Rehabil 2006; 28:841–7.
704
Vol 10, July 2015, 697–704
S. McKenna et al. 5 Barlow J, Sturt J, Hearnshaw H. Self-management interventions for people with chronic conditions in primary care: examples from arthritis, asthma and diabetes. Health Educ J 2002; 61:365–78. 6 de Silva D. Helping People Help Themselves: A Review of the Evidence Considering Whether it is Worthwhile to Support Self-Management. London, The Health Foundation, 2011. 7 Lennon S, McKenna S, Jones F. Self-management programmes for people post stroke: a systematic review. Clin Rehabil 2013; 27:867– 78. 8 Cadilhac DA, Hoffmann S, Kilkenny M et al. A phase II multicentered, single-blind, randomized, controlled trial of the stroke selfmanagement program. Stroke 2011; 42:1673–9. 9 Kendall E, Catalano T, Kuipers P, Posner N, Buys N, Charker J. Recovery following stroke: the role of self-management education. Soc Sci Med 2007; 64:735–46. 10 Jones F, Mandy A, Partridge C. Changing self-efficacy in individuals following first stroke: preliminary study of a novel self-management intervention. Clin Rehabil 2009; 23:522–33. 11 The EuroQol Group. EuroQol – a new facility for the measurement of health-related quality of life. Health Policy 1990; 16:199–208. 12 Williams L, Weinberger M, Harris L, Clark D, Biller J. Development of a stroke-specific quality of life scale. Stroke 1999; 30:1362–9. 13 Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome Measures for Health Education and Other Health Care Interventions. California, Sage Publications, 1996. 14 Jones F, Partridge C, Reid F. The Stroke Self-Efficacy Questionnaire: measuring individual confidence in functional performance after stroke. J Nurs Healthcare Chronic Dis 2008; 17(7b):244–52. 15 Mahoney FD, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J 1965; 14:61–3. 16 Nouri F, Lincoln NB. An extended activities of daily living scale for stroke patients. Clin Rehabil 1987; 1:301–5. 17 Goldberg DP, Hillier VF. A scaled version of the General Health Questionnaire. Psychol Med 1979; 9:139–45. 18 Trigg R, Wood D. The Subjective Index of Physical and Social Outcome (SIPSO): a new measure for use with stroke patients. Clin Rehabil 2000; 14:288–99. 19 Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing 1972; 1:233–8. 20 Bhavnani G, Cockburn J, Lincoln N. The reliability of the Rivermead Assessment and implications for some commonly used assessments of perception. Br J Occup Ther 1983; 46:17–9. 21 Jones F. Factors which influence the resumption of activity and participation following stroke. PhD thesis, University of Brighton, Brighton, UK, 2005. 22 Bandura A. The nature and structure of self-efficacy. In: Bandura A (ed): Self-Efficacy: the Exercise of Control. New York, W.H Freeman and Company, 1997. 23 Lorig K, Holman H. Self-management education: history, definition, outcomes and mechanisms. Ann Behav Med 2003; 26:1–7. 24 Marks R. Efficacy theory and its utility in arthritis rehabilitation: review and recommendations. Disabil Rehabil 2001; 23:271–80. 25 Nachar N. The Mann–Whitney U: a test for assessing whether two independent samples come from the same distribution. Tutorials Quant Meth Psych 2008; 4:13–20. 26 Jones F, Riazi A, Norris M. Self-management after stroke: time for some more questions. Disabil Rehabil 2012; 35:257–64. 27 Kennedy A, Rogers A, Bower P. Support for self care for patients with chronic disease. BMJ 2007; 335:968–70. 28 Jones F, Lennon S. A new stroke self-management programme: preliminary analysis of training for practitioners. Int J Stroke 2009; 4(s2):23.
© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
explore the feasibility of an integrated program. Exploration of the reasons behind relatively low recruitment and of the sensitivity of outcome measures to detect a change are also required. Additional investigation of intervention fidelity is required to monitor if the program is being delivered as intended. Key words: behavioral intervention, chronic disease, clinical trial, feasibility, self-management, stroke
Introduction Stroke is a major cause of disability worldwide, and the number of people living with the consequences of stroke is increasing on a global scale (1). A number of recent studies in the United Kingdom have highlighted a range of unmet needs post-stroke, particularly after the period of acute care and rehabilitation has been completed. There are indications from this research that individuals can feel ill prepared to cope in the longer term and ‘abandoned’ by services (2,3). It has been suggested that selfmanagement initiatives used for people with other long-term conditions may also be applicable to stroke (4). Self-management has been defined in different ways, but is commonly described as ‘an individual’s ability to manage the symptoms, treatment, physical and psychosocial consequences, and life style changes inherent with living with a chronic disease’ (5). This includes a range of attitudes, behaviors and skills directed at managing life with a chronic disease. Self-management programs have been shown to improve psychological outcomes and also have modest effects on health and social care needs and costs (6). Stroke self-management programs (SSMPs) are relatively rare (7). Two group-based programs delivered to stroke participants have demonstrated feasibility and some influence on reducing functional decline post-stroke, but findings are mixed (8,9). As with other long-term conditions, stroke may benefit from having different types of programs available, including both group and individual programs. Individualized programs could run alongside or be embedded into stroke rehabilitation. They may also be more applicable for those people who do not want to or are unable to attend group-based programs. The Bridges SSMP is an individualized, theory-driven program based on self-efficacy principles. It was developed in 2006 and tested on different people post-stroke through a series of single case studies; significant changes in self-efficacy were found, which suggested preliminary proof of concept (10). Further development and consultation with stroke patients, carers and professionals has informed the program, which now comprises structured one-to-one sessions with a professional to support patients in setting goals, recording progress and planning activities; it incorporates a patient-held workbook to facilitate a personal record of goals, progress and helpful strategies (10). Vol 10, July 2015, 697–704
697
Research Aim The aim of this randomized controlled trial (RCT) was to evaluate the feasibility of delivering the Bridges SSMP in addition to usual stroke rehabilitation compared with usual rehabilitation only. The following objectives were set in order to explore feasibility of the Bridges SSMP and the study design adopted for this feasibility RCT: 1. Explore participant recruitment and retention 2. Evaluate participant adherence to the Bridges SSMP 3. Assess the suitability of the selected outcome measures 4. Explore changes between groups in quality of life [EuroQol (11), Stroke Specific Quality of Life (SSQOL) (12)], self-efficacy [Self-Efficacy Scale (SES) (13), Stroke Self-Efficacy Questionnaire (SSEQ) (14)], functional activity [Barthel Index (15), Nottingham Extended Activities of Daily Living (NEADL) (16)], mood [General Health Questionnaire (GHQ) (17)] and community integration [Subjective Index of Physical and Social Outcome (SIPSO) (18)] 5. Explore the application and fidelity of the Bridges SSMP by the community stroke team 6. Explore the acceptability of the Bridges SSMP to participants and the community stroke team
Methods Ethical approval for this trial was granted by the Office for Research Ethics Committee Northern Ireland (08/NIROI/67). Design This was the first attempt to test the feasibility of implementing the Bridges SSMP in a community stroke service. The emphasis of the study was on the feasibility of delivery, including recruitment and retention; adherence to the program; suitability of and changes in outcome measures used; application and fidelity of the program; and acceptability of the program to patients, carers and professionals. Participants Any new stroke referral to the community stroke team of a Health and Social Care Trust in Northern Ireland or any stroke survivor who was within four-weeks of commencing rehabilitation with the community stroke team was considered for inclusion against the predetermined eligibility criteria. Eligibility criteria included a clinical diagnosis of stroke and the ability to follow a two-stage command. Patients were excluded if they were confused [with a mini mental score of less than 6 out of 10 (19)]; demonstrated neglect [with a star cancellation score of less than 48/52 (20)]; had competing comorbidities influencing their rehabilitation, e.g. malignant cancer; had significant issues with their readiness to participate as judged by the clinical team, e.g. illness of a carer; did not require weekly multidisciplinary rehabilitation, e.g. one or two advice sessions or review only; or had hearing and visual impairments precluding their ability to fully participate in the Bridges SSMP.
698
Vol 10, July 2015, 697–704
S. McKenna et al. Randomization and allocation of control/intervention Opaque sealed envelopes concealing the group allocation were prepared using block randomization. After written informed consent and baseline assessment, participants were randomly assigned to receive the Bridges SSMP or to a control group. Health professionals delivering the Bridges SSMP were not involved in the randomization process, and the research associate responsible for conducting outcome assessments and data processing was unaware of the group allocation. Neither the health professionals delivering the Bridges SSMP nor the participants could be blinded to group allocation. Intervention Participants allocated to the control group received the usual multidisciplinary stroke rehabilitation delivered by the community stroke team. In Northern Ireland this commonly consists of information on stroke; treatment input from two or more professionals, including involvement in goal setting and discharge planning; and advice on services, benefits and allowances. The Bridges SSMP Each participant randomized to the Bridges SSMP received one session of up to one-hour per week over a six-week period in addition to their usual rehabilitation. Although the Bridges SSMP was designed as a six-week program, in line with previous research by Kendall et al. (9), we defined compliers as those individuals who engaged in four or more sessions and used these criteria to distinguish between attendees and nonattendees. The SSMP was delivered by one of three members of the community stroke team who had previously undergone training in the program. The Bridges SSMP is distinct from usual rehabilitation and includes structured one-to-one sessions using strategies to promote specific behaviors that exemplify the hallmarks of selfmanagement (21). These behaviors include enabling patients to work out ways of taking control of their daily lives by setting small targets, recording their progress and problem solving. The content of the sessions is determined by the patient and their personal goals rather than being professionally directed. Selfmanagement behaviors are facilitated and reinforced by a patientheld stroke workbook that includes space to record personal goals and progress after stroke; it avoids technical terms and includes vignettes from other stroke survivors to illustrate successful selfmanagement (21). All professionals completed a treatment log for each individual enrolled in the trial, including the number of visits by each healthcare professional on a weekly basis. In addition, for participants receiving the Bridges SSMP, professionals were asked to record which components they worked on at each visit and the individual targets decided by each patient. The aim of this record was to explore whether the program had been delivered as intended and any differences in practice between using the Bridges SSMP and delivering usual care. Training in Bridges SSMP Training for professionals who delivered the Bridges SSMP took place over two-and-a-half-days. Day 1 was an introduction to self-management and current research evidence relevant to © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research
S. McKenna et al. stroke, followed by demonstration and practice of how to use the SSMP using case-based scenarios. Each professional had a minimum of two practice sessions with stroke patients before commencement of the trial. FJ then conducted a half-day of observation with the professionals to ensure that they were delivering the Bridges SSMP according to the key principles agreed by the trial management group. Feedback was given by FJ and steps were put in place to ensure consistency and fidelity, including a checklist for professionals to use to enable alignment with the principles of the Bridges SSMP. A further training day took place three-months later to explore issues that might impact delivery and to confirm the focus on patient-led goals, use of the workbook, reflection on progress and problem solving, and ways to engage participants in these strategies. Outcome measures A research associate, blinded to group allocation, assessed primary and secondary outcomes for all participants at baseline (pre-randomization), at discharge from the Bridges SSMP and three-months post-program completion. A range of outcome measures were used to test their feasibility and explore whether they would be meaningful to use in a fully powered trial. Previous research into self-management programs for people with stroke has demonstrated significant improvement in quality of life for participants (8), so health-related quality of life was chosen as one of the primary outcome measures. It was measured using both the EuroQol (11) and the SSQOL (12). Self-efficacy was also chosen as a primary outcome measure, as the Bridges SSMP intervention directly focuses on enhancing the sources of self-efficacy researched by Bandura (22), Lorig and Holman (23), and Marks (24) in order to influence confidence to self-manage. Self-efficacy was measured using the SES developed by Lorig et al. (13) and the SSEQ developed by Jones et al. (14). Both generic and stroke-specific measures were used to enable us to determine the most responsive measure. The choice of secondary outcome measures was also informed from the findings of studies exploring self-management programs. Secondary outcomes were functional independence [Barthel Index (15)], activity [NEADL (16)], mood [GHQ-28 (17)] and community integration [SIPSO (18)]. Data analysis Analysis of quantitative data focused on testing the changes in primary and secondary outcomes between participants receiving the Bridges SSMP in comparison with the control group using mean change scores and confidence intervals. Primary endpoints were mean change scores between groups in the health-related quality of life and self-efficacy measures up to three-months postprogram completion. Baseline characteristics were described and any differences between the two groups compared using the chisquare test and Mann–Whitney test. When using Mann–Whitney, each test was separately evaluated for the missing values (25). Qualitative methods Interviews were conducted with patients, carers and professionals to explore any issues relating to the acceptability of the program and feasibility to use with stroke patients in a community setting. The aim was to explore factors that impacted on delivery, consis© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
tency and/or compliance with the program. All qualitative data were transcribed and analyzed using a content-thematic approach. A summary of key qualitative findings is presented, but more detailed analysis of the qualitative data will be presented in a future paper.
Results Recruitment and retention As this was a feasibility study, a power calculation was not conducted. A total sample size of 30 individuals was agreed to be feasible through consultation between the research and community stroke teams. However, given the time restrictions of this study and staffing constraints of the community stroke team, 25 participants were recruited and randomized into either the control group (n = 13) or Bridges SSMP group (n = 12) over a 13-month period. Of the 152 individuals referred to the community stroke team during this period, 13 referrals were missed when recruitment was suspended owing to staffing constraints of the community stroke team. Of the remaining 139 participants who were screened, 25 were recruited and enrolled in the trial (see Fig. 1). Of the 25 individuals recruited to the study, all but one, who was a participant in the Bridges SSMP group, demonstrated full compliance with the intervention and assessment schedule. The most significant reasons for exclusion were participants requiring less than six-weeks of treatment (n = 48 individuals excluded), memory/cognitive problems (n = 15) and individuals not wanting to participate in the research trial (n = 16). Other reasons for exclusion included competing comorbidities (n = 5), poor rehabilitation potential as determined by the community stroke team (n = 5), readmission to hospital (n = 4), being treated for residual problems arising from a previous stroke (n = 3), unusual diagnosis (n = 2), being readmitted to hospital (n = 1) and death (n = 1). Thirteen further patients were excluded, and further exploration determined that these exclusions were usually associated with a perception by the team regarding readiness to participate. There were no statistically significant differences between groups at baseline for any measure; however, given that there were small numbers in each allocation group, we could not ensure groups were balanced with regard to their characteristics, so caution needs to be taken when interpreting results. Of particular note is that the control group was a higher-functioning group as measured by the Barthel Index (Bridges SSMP = 14·09; control group = 17·08) and the NEADL Scale (Bridges SSMP = 26·00; control group = 34·38); they were also at a later stage post-stroke (Bridges SSMP = 7·00 weeks post-stroke; control group = 11·38 weeks post-stroke) (see Table 1). Participant adherence to the intervention Adherence to the Bridges SSMP was high, with all but one participant completing four or more sessions. One participant (n = 1/12) did not receive the Bridges SSMP as intended, withdrawing from the study at week three. This participant was a younger individual, aged 24 years, who did not think the SSMP would benefit him. Eight individuals (n = 8/11; 73%) received all Vol 10, July 2015, 697–704
699
Research
S. McKenna et al.
Fig. 1 Recruitment process.
six sessions, three individuals (n = 3/11; 27%) received five sessions, and one individual (n = 1/11; 9%) received four sessions. According to feedback from professionals the majority of participants engaged with the program at each session; they liked the workbook and used it to record their personal successes, goals and targets. Suitability of outcome measures An independent outcome assessor was asked to provide feedback on the feasibility of using each of the outcome measures. The main issue was the time required to assess a participant on all eight outcome measures, which could take up to two-hours depending on the participants’ ability to concentrate and mood state. Some of the questions within measures were of a sensitive nature, for example, sexual relationships (SSQOL) and suicidal thoughts (GHQ). Some participants objected to these items, provoking a negative response from individuals. Overall it was felt
700
Vol 10, July 2015, 697–704
that the number of outcome measures used was a potential burden to participants, and not all measures had the potential to yield meaningful data. Changes in outcomes At baseline two participants had incomplete data sets. One participant did not complete the SES, and another participant failed to complete the NEADL measure. At each time point, a number of participants chose not to answer an item on the SSQOL scale regarding sexual relationships. The differential outcomes between the Bridges SSMP group and the control group were explored to identify whether the Bridges SSMP had any impact on quality of life, self-efficacy, activity, mood and community integration in comparison with routine multidisciplinary team care alone in community dwellers post-stroke. Changes were noted in all outcome measures postintervention irrespective of group allocation (see Table 2). © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research
S. McKenna et al.
Table 1 Participant characteristics at baseline Bridges SSMP (n = 11) Age, mean (SD) Weeks post-stroke, mean (SD) Outcome measure score, mean/max (SD) EuroQol Index EuroQol VAS SSQOL SSEQ SES Barthel Index NEADL SIPSO GHQ 28 Carer, n (%) Family HSC None Gender, n (%) Female Male Previous CVA, n (%) Yes No TIA
Control (n = 13)
P value (Mann–Whitney)
62·18 (13·57) 7·00 (4·45)
67·38 (10·60) 11·38 (12·70)
0·31 0·28
0·42/1 (0·36) 0·50/1 (0·19) 13·22/20 (2·35) 6·68/10 (2·56) 7·51/10 (2·45) 14·09/20 (5·30) 26·00/63 (14·19) 19·82/40 (6·75) 24·09/84 (10·90)
0·53 (0·41) 0·56 (0·16) 14·62 (3·42) 7·94 (1·85) 8·14 (2·09) 17·08 (3·40) 34·38 (17·44) 23·69 (9·00) 23·60 (15·52)
0·31 0·37 0·16 0·20 0·56 0·10 0·31 0·20 0·66
8 (72·7) 1 (9·1) 1 (9·1)
9 (69·2) 2 (15·4) 1 (7·7)
0·90
4 (36·4) 7 (63·6)
7 (53·8) 6 (46·2)
0·33
2 (18·2) 8 (72·7) 1 (9·1)
2 (15·4) 10 (76·9) 1 (7·7)
0·97
CVA, cerebrovascular accident; GHQ, General Health Questionnaire; HSC, health and social care; Max, maximum; NEADL, Nottingham Extended Activities of Daily Living; SD, standard deviation; SES, Self-Efficacy Scale; SIPSO, Subjective Index of Physical and Social Outcome; SSEQ, Stroke Self-Efficacy Questionnaire; SSMP, stroke self-management program; SSQOL, Stroke-Specific Quality of Life; TIA, transient ischemic attack; VAS, visual analogue scale.
Table 2 Mean change scores for primary outcome measures
Primary outcome
Maximum score
Score change indicating improvement
EuroQol Index
1
+
EuroQol VAS
1
+
SSQOL
20
+
SSEQ
10
+
SES
10
+
Group
Baseline to program completion/six-weeks
Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control
0·09 0·15 0·13 0·12 1·11 1·94 1·04 0·65 0·79 0·36
CI
Program completion/six weeks to three-months follow-up
CI
−0·09, 0·30 −0·10, 0·35 0·05, 0·27 0·05, 0·21 0·15, 2·65 0·74, 3·09 0·05, 1·69 0·08, 0·99 −0·68, 1·69 −0·16, 0·89
−0·05 −0·09 0·02 −0·02 1·05 0·12 −0·39 −0·15 −0·51 0·22
−0·13, 0·08 −0·37, 0·18 −0·13, 0·22 −0·16, 0·14 0·46, 1·60 −1·35, 1·37 −0·90, 0·28 −1·01, 0·71 −1·21, 0·56 −0·95, 1·33
CI, confidence interval; SES, Self-Efficacy Scale; SSEQ, Stroke Self-Efficacy Questionnaire; SSMP, Stroke Self-Management Program; SSQOL, StrokeSpecific Quality of Life; VAS, visual analogue scale.
At program completion/six-weeks both groups had evolved similarly with regard to the EuroQol scores, but the control group demonstrated more change in the SSQOL measure. At follow-up both groups showed an overall decline in the EuroQol; however, participants receiving the Bridges SSMP continued to show improvement in their SSQOL scores (see Fig. 2). At program completion/six-weeks, the Bridges SSMP participants showed more change in their self-efficacy scores (SES and SSEQ) than the control group (see Table 2). At follow-up both © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
groups showed decline in SSEQ scores (see Fig. 3), with the control group showing some improvement in SES scores. Conflicting findings were observed among secondary measures. At program completion/six-weeks positive results were found favoring the Bridges SSMP participants in functional activity (Barthel Index and NEADL) and social integration (SIPSO), while changes in mood favored the control group (GHQ). At three-month follow-up, the control group showed more improvement in NEADL and SIPSO measures. Both groups showed a Vol 10, July 2015, 697–704
701
Research
S. McKenna et al.
decline in the GHQ measuring mood, with the Bridges SSMP group showing less decline (See Table 3). Application and fidelity of the Bridges SSMP All professionals were instructed to complete a treatment log, whether they were involved in delivering the Bridges SSMP or providing rehabilitation to those in the control group. Data recorded in the logs were limited in detail, which meant the time spent on different components of the Bridges SSMP could not be determined. Professionals stated that most time was spent assisting participants in setting small targets that were meaningful to them, while between sessions participants acted on these targets and used the workbook as a resource to help problem-solve
Fig. 2 Stroke-Specific Quality of Life (SSQOL) means and confidence intervals. BG, Bridges group; CG, control group.
common issues. This directly addresses mastery, which is thought to be a key source of self-efficacy (22). This information indicates that practices being delivered during the study in the intervention group were authentic to the principles of the Bridges SSMP, which focus on promoting self-efficacy. However, the limited detail captured in the treatment logs precludes detailed analysis of intervention fidelity. Goals were recorded for all participants (n = 25) in both groups in the intervention logs. Analysis of records showed there were subtle differences between groups in how goals were described. While goals often focused on issues relating to mobility, irrespective of group allocation, for participants receiving the Bridges SSMP goals were recorded using language and terms that were expressed in the participants’ own words and were linked to their everyday activities. Details of the professionals involved in their treatment were recorded for 12 (n = 12/13) participants in the control group. On average, participants in the control group were seen by three to four different professionals during their rehabilitation. The aim of the intervention logs was to provide a detailed comparison between the amount of therapy received by the control group and Bridges SSMP participants, but data were missing in five (n = 5/ 11) of the Bridges SSMP participants. Where details are provided, it appears that the amount of therapy was similar for both groups. The acceptability of the Bridges SSMP Feedback was positive, with the professionals saying they valued the training and that delivering the program highlighted how the Bridges SSMP differed from their usual practice. Some feasibility issues were identified, but these were generally associated with the constraints of running a RCT within clinical practice and not the Bridges SSMP itself, e.g. timing of intervention delivery and additional paperwork. Overall, patients appeared to value the program and aspects of delivery such as the workbook and one-to-one facilitation with a health-care professional.
Discussion
Fig. 3 Stroke Self-Efficacy Questionnaire (SSEQ) means and confidence intervals. BG, Bridges group; CG, control group.
This is the first attempt to test the feasibility of the Bridges SSMP in a community stroke rehabilitation service. This study has raised a number of questions that require further investigation
Table 3 Mean change scores for secondary outcome measures
Secondary outcome
Maximum score
Score change indicating improvement
Barthel Index
20
+
NEADL
63
+
SIPSO
40
+
GHQ 28
84
−
Group
Baseline to program completion/six-weeks
Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control Bridges SSMP Control
1·73 1·46 17·40 12·69 7·55 4·77 −8·45 −11·31
CI
Program completion/six weeks to three-months follow-up
CI
0·14, 2·86 −0·07, 2·40 7·70, 27·10 5·67, 20·66 2·36, 14·04 1·66, 8·00 −14·05, −3·95 −19·28, −3·39
0·73 −0·08 2·09 4·23 0·55 4·08 0·45 2·77
−0·27, 1·27 −1·37, 1·04 −3·97, 6·77 −3·00, 10·67 −1·79, 3·99 −1·33, 8·83 −4·26, 5·26 −8·60, 13·93
CI, confidence interval; GHQ, General Health Questionnaire; NEADL, Nottingham Extended Activities of Daily Living; SIPSO, Subjective Index of Physical and Social Outcome; SSMP, stroke self-management program; VAS, visual analogue scale.
702
Vol 10, July 2015, 697–704
© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
S. McKenna et al. before feasibility can be fully confirmed. One of the key issues of interest was in relation to recruitment. The results showed that only 17·9% (n = 25/139) of individuals referred to the community stroke team were eligible to participate. This raises questions about the inclusion criteria and also decision-making by the community team, as a number of potential participants were judged to be unsuitable mainly on the basis of poor rehabilitation potential or a perceived disinclination to participate in a selfmanagement program. However, the number of patients who did not require weekly multidisciplinary rehabilitation, e.g. one or two advice sessions, accounted for the largest reason for nonrecruitment. This program was designed to be delivered by professionals and ultimately as part of their usual practice; however, it may be pertinent to consider designing a program that can be delivered over a smaller number of sessions. In addition, the rationale for exclusion requires further exploration if the aim is to design a program to be feasible to use within stroke rehabilitation. Retention of participants once recruited to the study was found to be acceptable in the Bridges SSMP group, and once recruited, all but one of the 12 individuals randomized to the SSMP demonstrated full compliance with the program and assessment schedule. The high level of compliance demonstrated in this study is greater than that demonstrated in previous research exploring group self-management programs in the stroke population. Kendall et al. (9) found 64% of individuals in their intervention group (n = 37/58) demonstrated compliance, while Cadilhac et al. (8) reported that only 52% (n = 25/48) of individuals recruited to their SSMP completed the program. An individualized program does have the potential to be more flexible and responsive to individual needs and ultimately could be more cost-effective if integrated into usual rehabilitation practice. Any analysis of outcomes needs to take into consideration the baseline characteristics of each group, in particular that the control group were a higher-functioning group and at a later stage post-stroke, which may suggest that they had less opportunity to improve in a range of outcome measures. In this study sample size was appropriate for an exploration of feasibility but not for testing any differences between groups. However, the outcomes explored do show some promise, as participants who had received the Bridges SSMP appeared to demonstrate more favorable changes in self-efficacy, functional activity (Barthel Index and NEADL) and social integration (SIPSO) over the six-week program period and showed less decline in mood and some improvement in quality of life at three-month follow up in comparison with the control group. The finding that Bridges SSMP participants demonstrated more improvement in self-efficacy over the intervention period is not surprising given that the intervention was designed specifically to impact on self-efficacy and has previously demonstrated proof of concept (10). Change scores between groups were relatively small, and further research would need to be conducted producing comparable findings in a larger sample before any conclusions could be made with regard to the impact of the Bridges SSMP. The organization of stroke care and the acute sudden nature of stroke are critical when examining the efficacy of selfmanagement programs. It could be argued that all or some of the © 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
Research strategies and skills can be addressed as part of usual stroke care and rehabilitation, and programs could be more effective if these strategies were already integrated into interactions with healthcare professionals (26). Kennedy and colleagues reviewed the evidence from self-management programs for other long-term conditions and highlighted areas that may cause a program to fail (27). These issues include not taking account of professional beliefs and attitudes, failing to consider the organizational structure in which the program is delivered and not having an understanding of the individual challenges of the target population (26). Limitations and future research Several issues were highlighted in this study that should be addressed in future research to test the utility of this and other SSMPs. The issue of recruitment requires careful consideration to minimize the risk of participants being wrongly excluded. Modifications to training are required to explore and address any potential barriers to using the program, thus increasing accessibility of the program to a broader range of stroke patients. Our previous research has shown the workbook can be a barrier if a participant is unable to read or follow a two-stage command, particularly if there is no carer who can help (28). In this study a substantial amount of time was required to complete the range of outcome measures, which resulted in a substantial burden for participants. It would seem necessary in future research to refine the number of measures used. The suggestion from the results of this study would be to retain the primary outcome measures related to quality of life (EuroQol and SSQOL) and self-efficacy (SES and SSEQ) but to reduce the number of or find alternative secondary measures that may be less time-intensive. Although the Bridges SSMP used in this study is somewhat standardized by employing a workbook structure, it would be advisable in future trials to keep more detailed records regarding the time spent on each component. This would enable investigation of whether there is any component or combination of components of the SSMP that is associated with more positive outcomes. It would also help ensure that the fidelity of the program is being monitored on an ongoing basis. In future, the application of the Bridges SSMP and fidelity to the program will require detailed scrutiny to determine the key differences from usual practice and ensure the program is being delivered as intended. The study design also meant transfer of knowledge between professionals treating control participants and those treating Bridges SSMP participants could not be ruled out. The use of a cluster design to minimize treatment contamination with an embedded process evaluation to explore contextual issues such as professional and patient experiences with larger sample sizes will be a natural next step in the development and testing of this intervention. The dose response of receiving the Bridges SSMP is a possibility within this feasibility RCT, and the next stage of research will also explore the feasibility of integrating the program into usual stroke rehabilitation. A larger study will also need to consider stratification of samples, especially with regard Vol 10, July 2015, 697–704
703
Research to functional ability, which varied considerably between groups in this study. Data collection to allow for covariate analysis would also help to determine under which conditions and to what sample of individuals the intervention should be delivered to optimize its effects. The success of any individualized SSMP also depends on the behaviors and enthusiasm of the professionals delivering the intervention, and further work is required to explore the feasibility of how self-management programs can become more universally acceptable and minimize provider burden. Integration of programs such as the Bridges SSMP into community stroke rehabilitation may lessen the demand on resources, but a full economic evaluation of the costs of providing the program in comparison to usual care is required.
Conclusion Questions remain regarding the feasibility of delivering the Bridges SSMP in addition to posthospital stroke rehabilitation; however, the Bridges SSMP can be delivered to both male and female participants at a range of ages with different levels of stroke severity. The program also enabled a patient-focused process for involving participants in their rehabilitation by promoting self-management practices. The following key feasibility issues were identified: low recruitment rate, small sample size, rigidity of eligibility criteria, small treatment effects, excessive number of outcome measures and inadequate recording of treatment fidelity. The dose response of receiving the program cannot be ruled out, and the next stage of research should explore the feasibility of an integrated program using a cluster-randomized controlled design to minimize treatment contamination with a full economic evaluation of the costs of providing this individualized program in comparison with usual care.
Acknowledgements The work of Suzanne McKenna was supported by the award of a PhD studentship from the Department of Employment and Learning. This feasibility study was funded by Northern Ireland Chest, Heart and Stroke.
References 1 WHO. Global Status Health Report on Noncommunicable Diseases 2010. Geneva, World Health Organization, 2010. 2 Care Quality Commission. Supporting Life after Stroke: A Review of Services for People Who Have Had a Stroke and Their Carers. London, Care Quality Commission, 2011. 3 McKevitt C, Fudge N, Redfern J et al. Self-reported long-term needs after stroke. Stroke 2011; 42:1398–403. 4 Jones F. Strategies to enhance chronic disease self-management: how can we apply this to stroke? Disabil Rehabil 2006; 28:841–7.
704
Vol 10, July 2015, 697–704
S. McKenna et al. 5 Barlow J, Sturt J, Hearnshaw H. Self-management interventions for people with chronic conditions in primary care: examples from arthritis, asthma and diabetes. Health Educ J 2002; 61:365–78. 6 de Silva D. Helping People Help Themselves: A Review of the Evidence Considering Whether it is Worthwhile to Support Self-Management. London, The Health Foundation, 2011. 7 Lennon S, McKenna S, Jones F. Self-management programmes for people post stroke: a systematic review. Clin Rehabil 2013; 27:867– 78. 8 Cadilhac DA, Hoffmann S, Kilkenny M et al. A phase II multicentered, single-blind, randomized, controlled trial of the stroke selfmanagement program. Stroke 2011; 42:1673–9. 9 Kendall E, Catalano T, Kuipers P, Posner N, Buys N, Charker J. Recovery following stroke: the role of self-management education. Soc Sci Med 2007; 64:735–46. 10 Jones F, Mandy A, Partridge C. Changing self-efficacy in individuals following first stroke: preliminary study of a novel self-management intervention. Clin Rehabil 2009; 23:522–33. 11 The EuroQol Group. EuroQol – a new facility for the measurement of health-related quality of life. Health Policy 1990; 16:199–208. 12 Williams L, Weinberger M, Harris L, Clark D, Biller J. Development of a stroke-specific quality of life scale. Stroke 1999; 30:1362–9. 13 Lorig K, Stewart A, Ritter P, Gonzalez V, Laurent D, Lynch J. Outcome Measures for Health Education and Other Health Care Interventions. California, Sage Publications, 1996. 14 Jones F, Partridge C, Reid F. The Stroke Self-Efficacy Questionnaire: measuring individual confidence in functional performance after stroke. J Nurs Healthcare Chronic Dis 2008; 17(7b):244–52. 15 Mahoney FD, Barthel DW. Functional evaluation: the Barthel Index. Md State Med J 1965; 14:61–3. 16 Nouri F, Lincoln NB. An extended activities of daily living scale for stroke patients. Clin Rehabil 1987; 1:301–5. 17 Goldberg DP, Hillier VF. A scaled version of the General Health Questionnaire. Psychol Med 1979; 9:139–45. 18 Trigg R, Wood D. The Subjective Index of Physical and Social Outcome (SIPSO): a new measure for use with stroke patients. Clin Rehabil 2000; 14:288–99. 19 Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing 1972; 1:233–8. 20 Bhavnani G, Cockburn J, Lincoln N. The reliability of the Rivermead Assessment and implications for some commonly used assessments of perception. Br J Occup Ther 1983; 46:17–9. 21 Jones F. Factors which influence the resumption of activity and participation following stroke. PhD thesis, University of Brighton, Brighton, UK, 2005. 22 Bandura A. The nature and structure of self-efficacy. In: Bandura A (ed): Self-Efficacy: the Exercise of Control. New York, W.H Freeman and Company, 1997. 23 Lorig K, Holman H. Self-management education: history, definition, outcomes and mechanisms. Ann Behav Med 2003; 26:1–7. 24 Marks R. Efficacy theory and its utility in arthritis rehabilitation: review and recommendations. Disabil Rehabil 2001; 23:271–80. 25 Nachar N. The Mann–Whitney U: a test for assessing whether two independent samples come from the same distribution. Tutorials Quant Meth Psych 2008; 4:13–20. 26 Jones F, Riazi A, Norris M. Self-management after stroke: time for some more questions. Disabil Rehabil 2012; 35:257–64. 27 Kennedy A, Rogers A, Bower P. Support for self care for patients with chronic disease. BMJ 2007; 335:968–70. 28 Jones F, Lennon S. A new stroke self-management programme: preliminary analysis of training for practitioners. Int J Stroke 2009; 4(s2):23.
© 2013 The Authors. International Journal of Stroke © 2013 World Stroke Organization
