Unsupervised Therapy After Stroke in the Home Setting With a Hand Rehabilitation Device (ReHandyBot)

Feasibility of Unsupervised Therapy After Stroke in the Home Setting With a Hand Rehabilitation Device (ReHandyBot): an Exploratory Study

The ReHandyBot is a robot for hand rehabilitation after stroke. The aims of this study are (1) to investigate the feasibility of unsupervised therapy with the ReHandyBot with stroke inpatients, first in a rehabilitation clinic and then at participants' home, (2) to evaluate the usability of the ReHandyBot (user interface, implemented exercises, and gaming environment, which were adapted for independent usage), and (3) to quantify the dose of additional robotic therapy that patients perform without supervision. The study consists of two primary phases. The first is a familiarization phase performed at the clinic, where therapists teach to the participants how to perform the exercises with the robot. Then, if capable of training with the robot safely, after discharge from the clinic participants can bring the robot home and autonomously train with it. The hypothesis is that unsupervised therapy might be a possible way to increase therapy dose for stroke patients, with the potential to further improve recovery of hand function, with minimal additional burden for therapists and for the healthcare system.

This group will perform therapy with the ReHandyBot first with the supervision of a therapist at the rehabilitation clinic. Then, after discharge, if the participant learnt how to safely train with the device, they can bring the device home and train with it unsupervised. On the other hand, if participants are not capable of training without supervision, after discharge they perform unsupervised therapy at home with a booklet of exercises (i.e., without the robot).

During the familiarization phase at the rehabilitation clinic, participants perform one week of supervised and one week of minimally-supervised therapy with the ReHandyBot under the supervision of a supervisor (i.e., therapist or researcher). Minimally-supervised therapy means that participants try to perform therapy with the device by themselves, while the supervisor is still present but helps only in case participants encounter problems or if they have any questions. During the supervised and minimally-supervised phases (first 2 weeks), the intervention dose is 5 sessions of approximately 45 minutes per week. These sessions are performed in addition to the conventional therapy plan. During each session, the robot proposes a set of 3 exercises, each lasting between 10 and 15 minutes.

After the familiarization phase, participants train without supervision at the clinic until discharge and then at home for two weeks. If the therapist thinks that the participant have properly learnt how to use the device and can train with it safely, the participant can keep training with the device unsupervised (both at the clinic and at home). If participants are not ready for unsupervised therapy with the device, they receive a booklet of exercises to perform without supervision (both at the clinic and at home). These exercises do not imply the use of the robot but are exercises discussed with the therapists and meaningful for the specific patient.

Daily dose in minutes of self-administered robot-assisted therapy performed by subjects in the home setting without supervision.

This is measured during the intervention (i.e., during the two weeks of unsupervised therapy at home).

Daily dose in number of task repetitions of self-administered robot-assisted therapy performed by subjects in the home setting without supervision.

This is measured during the intervention (i.e., during the two weeks of unsupervised therapy at home).

Percentage increase in therapy time due to the unsupervised robotic therapy compared to the case where the participants would perform conventional (i.e., normally prescribed) therapy only.

This is measured during the intervention (i.e., during the two weeks of unsupervised therapy at home).

Feasibility of therapy with the ReHandyBot as assessed by the number of adverse events occurring during this study.

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Feasibility of therapy with the ReHandyBot as assessed by the number of device deficiencies occurring during this study.

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Feasibility of unsupervised therapy with the ReHandyBot as assessed by the number of subjects who can transition to unsupervised robot-assisted therapy at home out of the total number of tested subjects.

Feasibility of unsupervised therapy with the ReHandyBot as assessed by attendance during the unsupervised phase at home. Attendance is measured as the percentage of days where the subject trains at least once without supervision out of the 14 days offered for unsupervised training at home.

Usability of the ReHandyBot measured with the System Usability Scale. The possible response options vary between 1 (strongly disagree) and 5 (strongly agree).

Usability of the ReHandyBot measured with the System Usability Scale. The possible response options vary between 1 (strongly disagree) and 5 (strongly agree).

Usability is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

Usability is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

Usability of the ReHandyBot measured with the Post-Study System Usability Questionnaire (version 3). This questionnaire consists of 16 questions and the possible response options vary between 1 (strongly agree) and 7 (strongly disagree).

Usability of the ReHandyBot measured with the Post-Study System Usability Questionnaire (version 3). This questionnaire consists of 16 questions and the possible response options vary between 1 (strongly agree) and 7 (strongly disagree).

Usability is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

A 5-point scale is presented by the robot at the end of each therapy session to measure patient's satisfaction with robot-assisted therapy.

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Usability of the gaming environment is measured with the Intrinsic Motivation Inventory (items: Interest/Enjoyment). The possible response options vary between 1 (not true at all) and 7 (completely true).

Usability of the gaming environment is measured with the Intrinsic Motivation Inventory (items: Interest/Enjoyment). The possible response options vary between 1 (not true at all) and 7 (completely true).

User experience is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in usability metrics (i.e., System Usability Scale, NASA Task Load Index, Post-Study System Usability Questionnaire, and Intrinsic Motivation Inventory)

Change in the metrics (i.e., System Usability Scale, NASA Task Load Index, Post-Study System Usability Questionnaire, and Intrinsic Motivation Inventory) used to assess usability between the first and second time when they are measured. This change is used to define how the absence of the therapist during robot assisted therapy impacts the perceived usability of the device.

This is calculated at the end of the study (i.e., at the end of the two weeks of unsupervised therapy at home).

The Customer Satisfaction Score consists of the question "Overall, how satisfied are you with the therapy with ReHandyBot?". The possible response options vary between 1 (extremely dissatisfied) and 5 (extremely satisfied).

The Customer Satisfaction Score consists of the question "Overall, how satisfied are you with the therapy with ReHandyBot?". The possible response options vary between 1 (extremely dissatisfied) and 5 (extremely satisfied).

User experience is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

The Net Promoter Score consists of the question "How likely would you be to recommend therapy with ReHandyBot to another patient?". The possible response options vary between 1 (extremely unlikely) and 10 (extremely likely).

The Net Promoter Score consists of the question "How likely would you be to recommend therapy with ReHandyBot to another patient?". The possible response options vary between 1 (extremely unlikely) and 10 (extremely likely).

User experience is measured again at the end of the study protocol (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in the metrics used to assess user experience (i.e., Customer Satisfaction Score and Net Promoter Score) between the first and second time when they are measured. This change is used to define how the absence of the therapist during robot assisted therapy impacts user experience when training with the device.

This is calculated at the end of the study (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in upper limb functions as assessed by the Fugl-Meyer Assessment of Upper Extremities (FMA-UE)

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) upper limb functions as assessed by the FMA-UE.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) upper limb functions as assessed by the ABILHAND scale.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) upper limb functions as assessed by the BBT test.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in upper limb functions as assessed by the Motor Evaluation Scale for Upper Extremities in Stroke Patients (MESUPES)

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) upper limb functions as assessed by the MESUPES. The scores can vary between 0 and 58, and a higher score means a better outcome.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) upper limb functions as assessed by the mAS.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in the active Range of Motion (aROM) for grasping measured in millimeters as assessed by the custom robotic assessment "aROM - Hand"

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) aROM for grasping, i.e. for hand opening and closing. This is measured by the robot with a custom assessment.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in the active Range of Motion (aROM) for forearm rotation measured in degrees as assessed by the custom robotic assessment "aROM - Forearm"

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) aROM for forearm pronation and supination. This is measured by the robot with a custom assessment.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in proprioception as assessed by the custom robotic assessment "JND" measuring the minimum difference in length (mm) and/or angle (degrees) that a patient can perceive

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) proprioception, defined as the minimum difference in length and/or angle that a patient can perceive (i.e. just noticeable difference). This is measured by the robot with a custom assessment.

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

Change in haptic perception as assessed by the custom robotic assessment "Weber Fraction" measuring the minimum difference in stiffness that a patient can perceive in percentage

Change between baseline (i.e. beginning of the study) and final (i.e. end of the study) haptic perception, defined as the minimum difference in stiffness that a patient can perceive expressed in percentage according to the Weber Fraction law . This is measured by the robot with a custom assessment named "Weber fraction".

This is calculated at the end of the study, after performing the final measurements (i.e., at the end of the two weeks of unsupervised therapy at home).

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Task performance (i.e., correct trials over total trials) during robotic therapy throughout the study.

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Ratio of effective (i.e., net total therapy time without breaks) over total duration of the therapy session during robotic therapy throughout the study.

This is monitored over the whole protocol duration (i.e., from beginning of the first week to the end of the two weeks of unsupervised therapy at home).

Correlation between achieved unsupervised robot-assisted therapy dose and different parameters (i.e., cognitive assessments (Goodglass-Kaplan Scale, LCF-R, Montreal Cognitive Assessment (MoCA)) performed during the screening visit, the clinical assessments, the robotic assessments, demographics and the collected medical data).

This is calculated at the end of the study (i.e., at the end of the two weeks of unsupervised therapy at home).

Inclusion Criteria: Informed Consent signed by the subject; female and male patients between 18 and 90 years old; acute/subacute stroke (recruitment within 12 weeks from stroke onset); pre-stroke modified Rankin score ≤ 1; National Institutes of Health Stroke Scale (NIHSS) ≥ 1 in at least one of the items regarding motor functions, sensory functions and ataxia; possibility (e.g., enough space) to set up the ReHandyBot at home. Exclusion Criteria: modified Ashworth Scale > 2 for one or more of the following muscles: shoulder adductors, forearm pronator and supinator, flexors and extensors of elbow, wrist and fingers; moderate to severe aphasia: Goodglass-Kaplan's scale < 3; moderate to severe cognitive deficits: levels of cognitive functioning-revised (LCF-R) < 9; functional impairment of the upper limb due to other pathologies; severe pain in the affected arm: visual analogue scale for pain (VASp) ≥ 5; other pathologies which may interfere with the study; pacemakers and other active implants; after discharge the patient will go to an assisted living facility (e.g., care home).