Dose and staffing comparison study of upper limb device-assisted therapy

Automated patient-robot assignment for a robotic rehabilitation gym: a simplified simulation model

Background

A robotic rehabilitation gym can be defined as multiple patients training with multiple robots or passive sensorized devices in a group setting. Recent work with such gyms has shown positive rehabilitation outcomes; furthermore, such gyms allow a single therapist to supervise more than one patient, increasing cost-effectiveness. To allow more effective multipatient supervision in future robotic rehabilitation gyms, we propose an automated system that could dynamically assign patients to different robots within a session in order to optimize rehabilitation outcome.

Methods

As a first step toward implementing a practical patient-robot assignment system, we present a simplified mathematical model of a robotic rehabilitation gym. Mixed-integer nonlinear programming algorithms are used to find effective assignment and training solutions for multiple evaluation scenarios involving different numbers of patients and robots (5 patients and 5 robots, 6 patients and 5 robots, 5 patients and 7 robots), different training durations (7 or 12 time steps) and different complexity levels (whether different patients have different skill acquisition curves, whether robots have exit times associated with them). In all cases, the goal is to maximize total skill gain across all patients and skills within a session.

Results

Analyses of variance across different scenarios show that disjunctive and time-indexed optimization models significantly outperform two baseline schedules: staying on one robot throughout a session and switching robots halfway through a session. The disjunctive model results in higher skill gain than the time-indexed model in the given scenarios, and the optimization duration increases as the number of patients, robots and time steps increases. Additionally, we discuss how different model simplifications (e.g., perfectly known and predictable patient skill level) could be addressed in the future and how such software may eventually be used in practice.

Conclusions

Though it involves unrealistically simple scenarios, our study shows that intelligently moving patients between different rehabilitation robots can improve overall skill acquisition in a multi-patient multi-robot environment. While robotic rehabilitation gyms are not yet commonplace in clinical practice, prototypes of them already exist, and our study presents a way to use intelligent decision support to potentially enable more efficient delivery of technologically aided rehabilitation.

The effect of time spent in rehabilitation on activity limitation and impairment after stroke

BACKGROUND

Stroke affects millions of people every year and is a leading cause of disability, resulting in significant financial cost and reduction in quality of life. Rehabilitation after stroke aims to reduce disability by facilitating recovery of impairment, activity, or participation. One aspect of stroke rehabilitation that may affect outcomes is the amount of time spent in rehabilitation, including minutes provided, frequency (i.e. days per week of rehabilitation), and duration (i.e. time period over which rehabilitation is provided). Effect of time spent in rehabilitation after stroke has been explored extensively in the literature, but findings are inconsistent. Previous systematic reviews with meta-analyses have included studies that differ not only in the amount provided, but also type of rehabilitation.

OBJECTIVES

To assess the effect of 1. more time spent in the same type of rehabilitation on activity measures in people with stroke; 2. difference in total rehabilitation time (in minutes) on recovery of activity in people with stroke; and 3. rehabilitation schedule on activity in terms of: a. average time (minutes) per week undergoing rehabilitation, b. frequency (number of sessions per week) of rehabilitation, and c. total duration of rehabilitation.

SEARCH METHODS

We searched the Cochrane Stroke Group trials register, CENTRAL, MEDLINE, Embase, eight other databases, and five trials registers to June 2021. We searched reference lists of identified studies, contacted key authors, and undertook reference searching using Web of Science Cited Reference Search.

SELECTION CRITERIA

We included randomised controlled trials (RCTs) of adults with stroke that compared different amounts of time spent, greater than zero, in rehabilitation (any non-pharmacological, non-surgical intervention aimed to improve activity after stroke). Studies varied only in the amount of time in rehabilitation between experimental and control conditions. Primary outcome was activities of daily living (ADLs); secondary outcomes were activity measures of upper and lower limbs, motor impairment measures of upper and lower limbs, and serious adverse events (SAE)/death.

DATA COLLECTION AND ANALYSIS

Two review authors independently screened studies, extracted data, assessed methodological quality using the Cochrane RoB 2 tool, and assessed certainty of the evidence using GRADE. For continuous outcomes using different scales, we calculated pooled standardised mean difference (SMDs) and 95% confidence intervals (CIs). We expressed dichotomous outcomes as risk ratios (RR) with 95% CIs.

MAIN RESULTS

The quantitative synthesis of this review comprised 21 parallel RCTs, involving analysed data from 1412 participants.  Time in rehabilitation varied between studies. Minutes provided per week were 90 to 1288. Days per week of rehabilitation were three to seven. Duration of rehabilitation was two weeks to six months. Thirteen studies provided upper limb rehabilitation, five general rehabilitation, two mobilisation training, and one lower limb training. Sixteen studies examined participants in the first six months following stroke; the remaining five included participants more than six months poststroke. Comparison of stroke severity or level of impairment was limited due to variations in measurement. The risk of bias assessment suggests there were issues with the methodological quality of the included studies. There were 76 outcome-level risk of bias assessments: 15 low risk, 37 some concerns, and 24 high risk. When comparing groups that spent more time versus less time in rehabilitation immediately after intervention, we found no difference in rehabilitation for ADL outcomes (SMD 0.13, 95% CI -0.02 to 0.28; P = 0.09; I2 = 7%; 14 studies, 864 participants; very low-certainty evidence), activity measures of the upper limb (SMD 0.09, 95% CI -0.11 to 0.29; P = 0.36; I2 = 0%; 12 studies, 426 participants; very low-certainty evidence), and activity measures of the lower limb (SMD 0.25, 95% CI -0.03 to 0.53; P = 0.08; I2 = 48%; 5 studies, 425 participants; very low-certainty evidence). We found an effect in favour of more time in rehabilitation for motor impairment measures of the upper limb (SMD 0.32, 95% CI 0.06 to 0.58; P = 0.01; I2 = 10%; 9 studies, 287 participants; low-certainty evidence) and of the lower limb (SMD 0.71, 95% CI 0.15 to 1.28; P = 0.01; 1 study, 51 participants; very low-certainty evidence). There were no intervention-related SAEs. More time in rehabilitation did not affect the risk of SAEs/death (RR 1.20, 95% CI 0.51 to 2.85; P = 0.68; I2 = 0%; 2 studies, 379 participants; low-certainty evidence), but few studies measured these outcomes. Predefined subgroup analyses comparing studies with a larger difference of total time spent in rehabilitation between intervention groups to studies with a smaller difference found greater improvements for studies with a larger difference. This was statistically significant for ADL outcomes (P = 0.02) and activity measures of the upper limb (P = 0.04), but not for activity measures of the lower limb (P = 0.41) or motor impairment measures of the upper limb (P = 0.06).

AUTHORS' CONCLUSIONS

An increase in time spent in the same type of rehabilitation after stroke results in little to no difference in meaningful activities such as activities of daily living and activities of the upper and lower limb but a small benefit in measures of motor impairment (low- to very low-certainty evidence for all findings). If the increase in time spent in rehabilitation exceeds a threshold, this may lead to improved outcomes. There is currently insufficient evidence to recommend a minimum beneficial daily amount in clinical practice. The findings of this study are limited by a lack of studies with a significant contrast in amount of additional rehabilitation provided between control and intervention groups. Large, well-designed, high-quality RCTs that measure time spent in all rehabilitation activities (not just interventional) and provide a large contrast (minimum of 1000 minutes) in amount of rehabilitation between groups would provide further evidence for effect of time spent in rehabilitation.

Comparing a Novel Neuroanimation Experience to Conventional Therapy for High-Dose Intensive Upper-Limb Training in Subacute Stroke: The SMARTS2 Randomized Trial

BACKGROUND

Evidence from animal studies suggests that greater reductions in poststroke motor impairment can be attained with significantly higher doses and intensities of therapy focused on movement quality. These studies also indicate a dose-timing interaction, with more pronounced effects if high-intensity therapy is delivered in the acute/subacute, rather than chronic, poststroke period.

OBJECTIVE

To compare 2 approaches of delivering high-intensity, high-dose upper-limb therapy in patients with subacute stroke: a novel exploratory neuroanimation therapy (NAT) and modified conventional occupational therapy (COT).

METHODS

A total of 24 patients were randomized to NAT or COT and underwent 30 sessions of 60 minutes time-on-task in addition to standard care. The primary outcome was the Fugl-Meyer Upper Extremity motor score (FM-UE). Secondary outcomes included Action Research Arm Test (ARAT), grip strength, Stroke Impact Scale hand domain, and upper-limb kinematics. Outcomes were assessed at baseline, and days 3, 90, and 180 posttraining. Both groups were compared to a matched historical cohort (HC), which received only 30 minutes of upper-limb therapy per day.

RESULTS

There were no significant between-group differences in FM-UE change or any of the secondary outcomes at any timepoint. Both high-dose groups showed greater recovery on the ARAT (7.3 ± 2.9 points; P = .011) but not the FM-UE (1.4 ± 2.6 points; P = .564) when compared with the HC.

CONCLUSIONS

Neuroanimation may offer a new, enjoyable, efficient, and scalable way to deliver high-dose and intensive upper-limb therapy.