The Armed4Stroke study is designed as a multicentre, observer-blinded, phase II, randomized controlled superiority trial with two parallel groups. Patient-caregiver dyads will be randomly allocated at a 1:1 ratio to 8 weeks of CME and telerehabilitation (Armed4Stroke program) in addition to usual care, or usual care alone. The trial is registered since 11 December 2018 in the Netherlands Trial Register as NL7422. The study protocol was approved by the Medical Ethics Review Committee of VU Medical Centre on April 30th 2019 and is registered with trial number 2019.081 - NL67357.029.18.
The multicentre study is conducted in the in- and outpatient clinics of 4 rehabilitation centres in the Netherlands: 1) Reade Rehabilitation, Amsterdam; 2) Roessingh, Centre for Rehabilitation, Enschede; 3) Sint Maartenskliniek, Nijmegen; and 4) Vogellanden, Zwolle. The Amsterdam UMC, location VU Medical Centre, is the initiator of this study in collaboration with Amsterdam Rehabilitation Research Centre | Reade and Roessingh Research and Development, Enschede.
Seventy-two patients with stroke, and 1 or 2 of their informal caregivers will be enrolled during in- or outpatient rehabilitation. The caregiver can be a partner, family member, or other person close to the patient. They are not healthcare professionals, nor are they paid for their efforts. Stroke is defined by the World Health Organization as “a clinical syndrome typified by rapidly developing signs of focal or global disturbance of cerebral functions, lasting more than 24 hours or leading to death, with no apparent causes other than of vascular origin” . Subarachnoid haemorrhage, and stroke resulting from a brain tumour or traumatic brain injury are excluded as these subtypes differ in their prognosis.
Inclusion criteria for both patient and caregiver are: 1) 18 years or older; 2) written informed consent; 3) able to understand the Dutch language on sufficient level to understand instructions and complete the questionnaires; and 4) motivated for CME. Inclusion criteria for the patient are: 1) < 3 months poststroke; 2) living independently prior to stroke; 3) discharged or planned to be discharged home; 4) able to follow instructions (Montreal Cognitive Assessment [MoCA] score > 21 points); and 5) able and willing to appoint an informal caregiver. An additional inclusion criterion for the caregiver is: being medically stable and able to support the patient. A trained physiotherapist will judge safety and determine if caregivers are physically and mentally able to support the patient during an intake exercise session. Exclusion criteria for both patient and caregiver will be a serious comorbidity that interferes with participation, e.g., premorbid restrictions in mobility as a result of a neurological disease, congestive heart failure or fractures of the lower extremity. Patients will not be enrolled in other clinical trials during the study period.
Demographic and social characteristics of patients and caregivers will be recorded at baseline including: age, sex, country of birth, level of education, type and duration of relationship, living arrangement, work status, and comorbidities following the Cumulative Illness Rating Scale . In addition, we will document stroke characteristics: type and location, stroke severity following the Bamford classification system , date of stroke, previous stroke, hemiplegic side, presence of sensory deficits, hemianopia, neglect or aphasia, communicative ability following item 19 of the Utrecht Communication Observation , and cognition following the MoCA .
Every patient with stroke, starting in- or outpatient rehabilitation is screened for eligibility according to the inclusion and exclusion criteria by their rehabilitation physician and/or physiotherapist. Eligible dyads receive verbal and written information about the study. If a dyad wants to participate after a reflection period of 1 week, the intake exercise session is scheduled. Upon enrolment, informed consent will be signed by patient and caregiver(s) (Additional file 2: Informed consent form) and primary and secondary endpoints are measured by an independent assessor (MM, CN). Following the assessments, an activity monitor is worn by the patient for 1 week and dyads are randomized to the intervention or control group at a 1:1 ratio using an online randomization module with a minimization algorithm to prevent unequal group sizes. The module will be managed by a researcher (EW, GK) from the initiating institute, not involved in inclusion, assessment or intervention delivery. Factors in the minimization include: 1) centre; 2) inpatient versus outpatient; 3) ≥70 years versus < 70; and 4) Functional Ambulation Category (FAC) 5 versus FAC < 5. Primary and secondary endpoints are repeated by a blinded assessor (MM) at the end of treatment and 6 months follow-up (Fig. 1: Study design). During the study period, serious adverse events will be recorded and reported to the Medical Ethics Review Committee of VU Medical Centre. Data collection is done using electronic Case Report Forms (eCRFs) within the Good Clinical Practice-proof, cloud-based clinical data management platform Castor EDC . The study is monitored by the independent clinical research bureau from VU medical centre. Univariate checks, e.g., range checks, and conditional variables are built-in the eCRFs. The first participant was enrolled on September 2nd 2019 and participants are currently being recruited and enrolled.
Armed4Stroke intervention program
The Armed4Stroke program consists of 8 weeks of complementary exercises executed with a caregiver (CME), in addition to usual care. A trained physiotherapist compiles a tailor-made, progressive training program, containing personal goals and task-specific exercises focusing on balance, gait, outdoor activities and physical training. A total of 31 goals and 80 exercises according to evidence-based physical therapy guidelines  were developed to achieve important milestones for community ambulation, based on expert’s opinion and items of the hierarchical Mini-BESTest  and Rivermead Mobility Index (RMI) . Dyads are asked to perform the exercises minimally 5 times a week for 30 min per session. This intensity is in line with current exercise guidelines  and was shown to be feasible in previous CME studies [17, 18]. Dyads are advised to execute their program outside usual training hours and during weekends, when patients are mostly inactive [31, 32].
Dyads are supported by a trained physiotherapist during at least 4 face-to-face sessions, scheduled every other week. During the first session, the program is explained in detail and explicit attention is given to information about the influence of physical activity and exercise on mood and health-related quality of life . In addition, the role of caregivers and their benefits will be discussed. During subsequent sessions, the program is adapted according to the progress and personal goals of dyads, barriers are identified and addressed, and dyads are motivated to continue their exercise program. During the program, dyads are supported by exercise videos which are built into a web-based telerehabilitation system. Telerehabilitation poststroke should not be restricted to a time, place or device . Therefore, the web-based system can be accessed on all devices with an internet connection including a computer, tablet or smartphone. In addition, dyads can communicate a-synchronically (i.e., exchange messages independent of time) with their physiotherapist. Therapists can use the telerehabilitation system to set-up goals and exercises, monitor compliance and progress, give feedback and motivate participants. Physiotherapists were extensively trained in applying the program to optimize standardization.
All participants will receive usual in- or outpatient physical therapy according to the Royal Dutch Guidelines of Physical Therapy . Physical therapy sessions are designed to improve control of standing balance, physical condition, and walking competency. There are no restrictions with respect to content, time, or duration of therapy. Patients randomized to the control group will not have access to the Armed4Stroke telerehabilitation module and no explicit attention is given to the involvement of caregivers in goal-setting and rehabilitation. An 8-week self-reported exercise diary is kept by all the participants to monitor compliance, record fall incidents and telerehabilitation use in a usual care context.
The primary outcome is the mobility domain of the SIS version 3.0 , a stroke-specific questionnaire that evaluates self-reported health on 8 domains and a Visual Analogue Scale (VAS). Items are scored on a 5-point Likert scale and domain scores are calculated ranging from 0 to 100 . The SIS has excellent clinimetric properties in English [34, 35] and Dutch . The mobility domain consists of 9 questions about perceived ability to maintain balance, make transfers or walk in the community. The other 7 domains, the VAS and a composite physical domain are evaluated as secondary endpoints.
Secondary endpoints patient and caregiver
General Self-efficacy Scale (GSES)
The GSES is a valid questionnaire to evaluate general self-efficacy beliefs [37, 38]. The 10 items are rated on a 4-point scale and summed to produce a total score.
Preparedness for Caregiving Scale (PCS) and Transition Preparedness Scale (TPS)
The PCS is a reliable and valid tool to assess transition preparedness in caregivers, to assume their new role and to handle the stresses of caregiving . The PCS was translated to Dutch using forward- and back-translations and a modified version was developed to assess transition preparedness in patients, the TPS (Additional file 3: Transition Preparedness Scale). Psychometric evaluation of both scales is currently being conducted by the authors.
Hospital Anxiety and Depression Scale (HADS) and Computer Adaptive Tests (CAT)
The HADS is a psychometrically robust measure to assess mood [40, 41]. The scale consists of two 7-item subscales for anxiety and depression. In addition, anxiety and depression items of the Dutch-Flemish PROMIS item bank are administered with CAT [42,43,44]. CAT dynamically selects the optimal number of questions and the best next question, based on responses to previous questions.
Fatigue Severity Scale (FSS)
The 9-item FSS is a reliable and valid instrument to assess fatigue. A mean item score is calculated ranging from 1 to 7 .
General functioning subscale of the McMaster Family Assessment Device (FAD-GF)
The FAD-GF is a reliable and valid questionnaire to determine general family functioning [46, 47]. A mean item score is calculated ranging from 1 to 4, based on 12 items. Perceptions of family functioning can differ and therefore the FAD-GF is administered to both patients and caregivers .
Secondary endpoints patient
Length of Stay (LOS)
LOS is defined as the number of days between admission and discharge from the rehabilitation centre and will only be considered as a secondary endpoint for patients enrolled during inpatient rehabilitation.
Stroke Self-efficacy Questionnaire (SSEQ)
The SSEQ is a reliable and valid measure of self-efficacy in relevant domains of functioning poststroke . The SSEQ consists of 13 items with scores ranging from 0 to 10. Rasch analysis identified 2 dimensions: 1) Activity; and 2) Self-management ; and we will report these separate dimensions.
The RMI measures self-reported mobility poststroke with 14 questions and 1 observation . A total score of 0 to 15 is calculated. Psychometric properties of the English and Dutch version [51,52,53] have been established.
Functional walking ability is evaluated with the 0 to 5 FAC score, a reliable, valid and responsive scale poststroke . An assessor rates the level of independence, regardless of the use of assistive devices.
Six-minute Walking Test (6MWT) and Five-meter Walk Test (5MWT)
The 6MWT [55, 56] and 5MWT [57, 58] are psychometrically robust tests to determine walking capacity poststroke. Distance walked during 6 min is recorded in meters. Mean gait speed is calculated in meters per second based on 3 repetitions. The 5MWT at a comfortable speed is recommended as the most responsive and convenient method of evaluating gait speed, when compared to fast speed or a 10-m distance .
Leg section of the Motricity Index (MI-leg)
The MI-leg is a reliable and valid tool to assess strength and range of motion of the paretic leg [60, 61]. The test consists of 3 joint movements, i.e., ankle dorsiflexion, knee extension, and hip flexion which are rated on a 0 to 33 scale and summed. When the maximum score of 99 is achieved, 1 point is added.
Berg Balance Scale (BBS)
The BBS measures balance during 14 tasks . Items are rated on a 0 to 4 scale and a total score is calculated ranging from 0 to 56. Reliability, validity and responsiveness are established poststroke [62, 63].
MOX-2 activity monitor
Physical activity will be measured for 7 days during wake hours using the MOX-2. The MOX-2 is a small, lightweight device and will be worn on the non-paretic thigh using an adhesive plaster. It measures 3D-accelerations during daily activities and mean daily physical activity is calculated over 1 week.
Nottingham Extended Activities of Daily Living index (NEADL)
The NEADL is a 22-item reliable and valid test of extended ADL poststroke [30, 64, 65]. Rasch analysis did not support a unidimensional structure or polytomous scoring . Therefore, the original dichotomous scoring method and domain scores: 1) Mobility; 2) Kitchen; 3) Household; and 4) Leisure will be used.
Community Ambulation Questionnaire (CAQ)
The CAQ measures the level of independent walking outside the home, i.e., without physical assistance or supervision . Patients are categorized as: 1) Unable to walk outside unassisted; 2) Only ambulant around the house (e.g., as far as the letterbox); 3) Ambulant in the immediate environment (e.g., around the block); or 4) Independent community ambulation (e.g., to visit a friend or shop).
The EuroQol 5D (EQ-5D)
The 3-level version of the EQ-5D is a valid and reliable, generic measure of health-related quality of life poststroke [67, 68]. The instrument consists of 5 dimensions: 1) Mobility; 2) Self-care; 3) Usual activities; 4) Pain/Discomfort; and 5) Anxiety/Depression; and a VAS to assess perceived health.
Modified Rankin Scale (mRS)
The mRS is a widely used 0 to 5 global rating scale of disability poststroke. An assessor rates the level of independence in pre-stroke activities. The scale has good reliability and validity poststroke .
Secondary endpoints caregiver
Caregiver Strain Index (CSI) and Care-related Quality of Life instrument (CarerQoL)
Both the CSI [70,71,72] and the CarerQoL [73,74,75] are reliable and valid instruments to evaluate caregiver burden. The CSI consists of 13 items which are summed to produce a total score. The CarerQoL consists of seven 3-level burden dimensions and a 0 to 100 VAS measuring happiness.
We expect a significant improvement of 10 points (10%) on the SIS mobility domain (mean 79.4, SD 14) in favour of the intervention group. We expect that minimally 30 patients are required per trial arm . Including 20% drop-outs (for patients and caregivers), a sample size of 72 patient-caregiver dyads is needed to achieve sufficient statistical power of 80% using a two-tailed significance level alpha of p < 0.05. Currently, 26 participants have been enrolled in the ongoing trial.
Baseline characteristics will be presented in mean and standard deviation or median and interquartile range, depending on the normality of data distributions judged by visual plot and Shapiro-Wilk tests. Between-group differences will be studied to determine whether groups are comparable at baseline using non-parametric Wilcoxon signed rank sum tests when the data is not normally distributed, or student t-tests for independent samples in the presence of normally distributed data. The study endpoints will be compared between the intervention and control group at different time points using linear mixed model analysis. Time since stroke, group allocation, minimization factors and baseline value of the outcome measure will be added to the longitudinal model. Intention-to-treat analysis will be applied and all hypotheses will be tested two-sided using a critical value of < 0.05.