AHA-BOOST group sessions are focused around the four key aspects (extensively described below): neurophysiology, sequences of reaching and grasping, de-weighting of the arm and orientation of the hand towards objects. Each of the sessions is tailored to the individual patient, based upon the ongoing assessment of the therapists, position within the JSU diagram (4), discussion within the group of therapists, standardized assessment [Model of Bobath Clinical Practice (MBCP)] (6) and individual treatment goals of the patient.

Group sessions are conducted with a maximum of four patients, supervised by two trained therapists (physiotherapist and occupational therapist). Sessions may begin with a brief mobilization to prepare for the exercises, followed by approximately eight exercises. These exercises are designed for the patient to perform with minimal support of a therapists. By adapting the environment, the four key aspects are integrated. Additionally, patients perform 1 h per week individual robot-assisted therapy. During the full session the amount of assistance provided by the devices is based on the patients’ active possibilities.

The material used in the AHA-BOOST group sessions include standard rehabilitation material such as different sizes of balls, cones, hoops, beads, and daily objects. For the individual robot-assisted therapy, several upper limb rehabilitation technologies can be used, such as the Armeo Power® (Hocoma, Switzeland), Myro® (Tyromotion, Austria), Amadeo® (Tyromotion, Austria), or similar, depending upon therapy goals, abilities of patients and availability of the systems. The therapy goals and abilities of the patients are determined together with the patient by a multidisciplinary team and are based upon recommended clinical outcome measures, such as Fugl-Meyer assessment for upper extremity (FMA-UE), action research arm test, stroke upper limb capacity scale or goal attainment scale, or similar, aligned with consensus-based core recommendations from the Stroke Recovery and Rehabilitation Roundtable (7). In the AHA-BOOST therapy program, we primarily focus on performance-based measures to obtain detailed, standardized, and objective assessments of motor function and movement quality under controlled conditions. Our aim is to evaluate potential therapy effects within the structured context of the inpatient rehabilitation setting. According to the ICF model, we include both impairment and activity (capacity) outcome measures, based on international consensus. We are aware that real-world upper limb use is a key outcome for persons with stroke and acknowledge that within-laboratory capacity measures alone may not fully capture spontaneous functional use or performance in daily life. Future trials should therefore consider to include a meaningful measure of real-world upper limb use such as for instance the Motor Activity Log (MAL) or wearable sensor technology.

The development of the AHA-BOOST program was based upon the principles of the Bobath concept (8) and this program and concept were used in the AHA-BOOST pilot study (9). To structure our clinical reasoning process, the Model of Bobath Clinical Practice (MBCP) is used (6). The MBCP is a structured tool designed to support clinical reasoning in neurorehabilitation. It was developed through expert consensus using focus groups of experienced neurorehabilitation clinicians and is grounded in the Bobath concept—a widely applied clinical framework in neurorehabilitation (6). The MBCP provides a standardized approach for guiding therapeutic decisions in patients with acquired brain injury, including stroke. This clinical reasoning process is patient centered, meaning that goals and therapy are set in communication with the patient. A working hypothesis is developed based on a task-related functional movement analysis, which is conducted by a multidisciplinary team using observational assessment, clinical outcome measures, and reflection on the patient’s reaction to manual, verbal, and environmental facilitation. This hypothesis considers the movement diagnosis (‘how is the patient deviating from efficient/typical movement?’) and identifies the potential of the patient and as such forms the basis for therapy. Assessing typical and efficient movement post stroke includes evaluating motor control, biomechanical patterns, movement performance, symmetry, etc. In addition to observational assessment, clinical outcome measures are used (10). The outcome of the treatment (dis-) confirms the hypothesis and continues the iterative process of clinical reasoning. The MBCP was selected for several reasons. First, it offers a structured framework for clinical reasoning to plan therapy based on a patient’s functional potential. Second, it promotes a holistic, patient-centered approach that considers multiple domains of functioning (motor, sensory, cognitive) and supports interdisciplinary collaboration. Lastly, the Bobath concept, on which the MBCP is based, is commonly used in clinical settings. Although its superiority over other rehabilitation models remains debated, it continues to be widely accepted and clinically relevant in stroke rehabilitation (6). And while it is not a classical outcome measure and has not yet undergone formal psychometric validation, its clinical relevance and structured use in stroke rehabilitation are well-known (11).

Skilled grasp is a sensorimotor process, requiring the brain to extract cues from the environment to shape a motor command (12). Therefore, reaching and grasping an object requires the processing of its precise location with respect to the hand, the integration of the object’s intrinsic properties (such as size and shape of the object) and the integration of the goal of reaching and grasping. This information needs to be transformed into guiding the hand towards the object and shaping the hand for efficient grasp of the object in relation to the required goal, whilst considering the biomechanical interactions all along the multi-articulate anatomical chain, linking the proximal arm to wrist, hand, and fingers (12).

As presented in Figure 1, in daily life a person is performing a task in a chosen environment. The motor activity of the person is defined by the characteristics of the environment, the characteristics and the goal of the task and the ability of the person. When looking at the skills and abilities of the person, aspects of action, perception, cognition, and emotion must be considered (13). All these aspects need to be assessed and addressed to influence the efficiency of the task: (1) Action: Am I able to recruit muscles/synergies in an efficient way (intensity and selectivity)? How about the role of movement planning and the activity of the medial and lateral descending system?; (2) Perception: Am I able to inform my central nervous system by using my senses and receptors (skin receptors, proprioceptors, graviceptors, vestibular system, vision, hearing, and their connection with relevant areas in the brain, such as thalamus, parietal cortex)?; (3) Cognition: Am I able to understand the task, how it is performed, do I recognize the environment, am I able to be attentive, can I execute a task? In other words, how am I coping with neuropsychological disorders connected to the associative area’s in the brain?; and (4) Emotion: Do I recognize aspects of fear, nervosity, (lack of) drive (13)?

Finally, to reach efficiency in a functional activity during treatment, the person needs to be properly aligned, oriented and activated: (i) Alignment: is related to a proper biomechanical relation of bones, joints and muscle length in relation to the upcoming task; (ii) Orientation: is related to informing the brain about the body and its relationship to the environment (How is my body related to the base of support and related to the upcoming task, where is the object in relation to my hand, where is my arm/hand, what is the aim of the task, …). This gives us information about the integrity of the sensory system (receptors, ascending pathways and processing in, among others, the thalamus, parietal cortex and cerebellum); and (iii) Action: Is related to recruiting the relevant muscles in an ordered way (such as posture before movement and activating synergies). Here we generate information about the integrity of, for example, the frontal cortex, the cerebellum and the medial and lateral descending tracts.

People with stroke have impaired postural control; for example, in sitting, these individuals have more displacements of the center of pressure in both the anterior–posterior and medial-lateral directions (14). People with stroke are lacking orientation due to an impaired body scheme or malorientation in relation to the base of support. Additionally, most patients experience difficulties in activating muscles due to paresis. Considering only disability derived from upper-limb (UL) impairment, more than 75% of people with stroke remain with upper-limb impairment at the chronic stage, which renders the rehabilitation of the ULs after stroke a challenge (15). Over fifty non-conventional interventions aimed to improve upper limb recovery after stroke were evaluated by Saikaley et al. (16). Among these, modified constraint-induced movement therapy showed the highest probability of being the most effective (16). However, The Cochrane review by Pollock et al. (17) reported only moderate-quality evidence for constraint-induced movement therapy (CIMT) on ADL outcomes and concluded that no intervention currently used as routine practice is supported by high-quality evidence (17). Additionally, implementing CIMT in clinical practice shows several barriers: it seems only suitable for a very specific subgroup of stroke survivors and the therapy requires intensive therapeutic resources (18). To address this treatment gap and provide effective intervention for a broader population of stroke survivors, the arm-hand BOOST program (AHA-BOOST) was developed.

The JSU diagram was developed based on the hypothesis that not every stroke patient needs the same approach in treating the upper limb (4). Stratifying patients based on their level of impairment is the prerequisite for providing patient-tailored treatment for the upper limb. The AHA-BOOST program focuses on the mild to moderately impaired patient after stroke for which cut-off scores in 2 items of the FMA-UE (19) are used. Stage two is referring to the synergistic movement of the arm whilst stage five refers to hand function. Patients are included in the AHA-BOOST program if a score of 8–17 is achieved on stage 2 (volitional movement within synergies) of part A, or a score of <8 on stage 2, part A, in combination with a score >6 on stage 5 (hand) of part C. This means that patients can join the program if they can use their upper limb in a flexor or extensor synergy, or when they show some hand function (4, 19).

The key aspects of the AHA-BOOST program include the integration of knowledge on:

In the following sections, we will explain these four key aspects, provide the theoretical background and translate each of them into clinical practice.

See materials and equipment.

The AHA-BOOST group sessions are conducted by a team of specifically trained therapists, including both occupational therapists and physiotherapists. Prior to delivering the intervention, they undergo extensive training in AHA-BOOST therapy.

The training covers the following topics:

The program consists of twenty, one-hour group sessions (spread over 4 weeks, 5 days a week) and 1 h of individual robot-assisted therapy per week. The group sessions are executed with a maximum of four participants, under supervision of two experienced therapists (both an occupational and physiotherapist).

The program is provided in an inpatient rehabilitation setting. No specific infrastructure is needed.

The AHA-BOOST intervention is developed for patients in the sub-acute phase post stroke and the program consists of a total of 24 h of additional therapy delivered over a period of 4 weeks. These 24 h are divided in twenty, one-hour group sessions and 4 h of robot-assisted hand therapy. This means that patients will receive every day 1 h of BOOST therapy and weekly 1 h of robot-assisted therapy. The dose was determined based on clinical feasibility (as it concerns additional therapy, on top of the usual care therapy dose) and current practice standards in an inpatient rehabilitation setting, while still offering sufficient intensity to potentially elicit functional improvement. This is further supported by the recent European Stroke Organisation (ESO) guideline on motor rehabilitation with a recommendation to provide an additional minimal dose of 20 h of repetitive upper limb practice to improve arm capacity (68).

AHA-BOOST is a patient-centered approach based on the individual abilities and treatment goals of the patient. Exercises are structured around the four key aspects of AHA-BOOST, with difficulty levels gradually increasing. Each of the interventions is tailored to the individual patient, based upon the ongoing assessment of the therapists by means of the Model of Bobath Clinical Practice (MBCP) (6), discussion within the group of therapists and individual treatment goals of the patient.