People with neurological disorders (PwNDs) typically present limitations in performing functional tasks (1, 2). Therefore, a major focus of neurological rehabilitation is to maximize functional motor abilities, such as walking (3). It has been postulated that functional improvements occur due to a combination of two phenomena: (1) motor recovery via the relearning of premorbid motor skills and (2) motor compensation via learning new movement strategies (4).

Although both motor recovery and compensation can lead to functional improvements, the decision between the two approaches is not trivial. When motor recovery is prioritized, more emphasis is placed on movement quality, providing more feedback on movement performance, leading to the relearning of skilled movements. Conversely, when compensation is prioritized, physiotherapists train people to use assistive devices or other motor strategies. As a result, less attention is paid to the quality of the movement, and this could lead to different rehabilitation outcomes, as suggested by Krakauer (5).

The factors leading to the selection of motor recovery or compensation are unknown. The clinical decision-making process may be influenced by evidence-based practice recommendations and implicit, subjective theories based on clinical experience and contextual factors (6, 7). The physiotherapist’s expectation of good functional improvement may lead to the adoption of a restorative rather than a compensatory approach. Factors influencing the outcome of motor rehabilitation vary. These include people after stroke age, the severity of walking disabilities, and balance impairment at baseline (8, 9). For people with multiple sclerosis (MS), the impact of balance impairment, disease severity, and disease progression on gait rehabilitation outcomes is still under debate (10–12). In people with Parkinson’s disease (PD), age, motor and cognitive impairment, and disease severity were factors associated with gait improvements (13).

To date, no studies have examined the relationship between these factors and the clinical decision-making process in everyday clinical practice. Our hypothesis is that the restorative approach is chosen when a larger number of sessions are available and clinical characteristics suggest a good clinical outcome. The aim of this study was to investigate the factors influencing the choice between rehabilitation approaches. The secondary aim of this study was to investigate whether restorative or compensatory approaches consisted of different physiotherapy (PT) treatments in everyday clinical practice.

This unique observational study investigated the factors associated with the selection of restorative or compensatory approaches in PT clinical practice. We identified four main variables associated with the restorative rehabilitation approach: being in the subacute phase of the disease, having a low level of impairment at baseline with a high level of functional independence, and having lower cognitive performance.

The physiotherapists interviewed were all able to define their intervention as restorative or compensatory, indicating that the PT theoretical frameworks are well recognized by clinicians (4, 21). Both approaches were used to treat PwNDs, but the restorative approach was used more often than the compensatory approach in everyday clinical practice.

Our results suggest that specific clinical characteristics are associated with physiotherapists’ choice of approach. The factor that was most strongly associated with the restorative approach was the onset, with subacute subjects being 36 times more likely to receive restorative rehabilitation compared to chronic ones. The findings of this study support the a priori hypothesis that a physiotherapist’s decision-making process was influenced by the expectation of a good clinical outcome, and thus a restorative approach was more likely chosen.

Spontaneous neurological recovery is by far the largest contributor to behavioral restitution, and it is known that this phenomenon occurs during the first few weeks after stroke (subacute phase) (22, 23). Similarly, severe impairment was positively associated with worse clinical outcomes in longitudinal studies predicting walking recovery, and this association was also present when considering independence (mBI). Interestingly, we found a complex interaction between impairment and functional impendence. Severely impaired subjects were likely treated with a compensatory approach unless their independence level was high.

In other studies, preserved cognitive status was a predictor of good clinical outcome, which contrasts with our findings (24, 25). However, our results should be treated with caution, as the mean MMSE score was close to the highest score on the scale, meaning that the cognitive impairments measured by this test were rare in our sample. The MMSE has been widely used as a screening tool for dementia, but it has been criticized for not being sensitive enough to cognitive impairments specific to these populations, suggesting that other validated tests should be used in this population (26–28). Moreover, average MMSE scores were similar for subjects treated with a restorative and compensatory approach, and the association found between MMSE scores and the restorative approach was unexpected.

In contrast with our hypothesis, other variables associated with good clinical outcomes were not associated with the restorative approach. We anticipated that the age of individuals with PwNDs would influence the physiotherapist approach, with older subjects more likely to receive the compensatory approach. This expectation is based on previous reports indicating that older age is associated with poorer rehabilitation outcomes in individuals after stroke and in those with PD (13, 29). Similarly, balance performance and subjects’ motivation for PT were not considered in the choice of rehabilitation approach, although both factors have been shown to be associated with functional outcomes at the end of rehabilitation (9, 11, 12, 30).

In addition, we expected that a greater number of available sessions would influence the physiotherapist’s decision to select a restorative approach, as the time spent in rehabilitation should result in an increased therapy dosage (31, 32). The lack of association between the number of sessions and the restorative approach may be due to a lack of strong evidence supporting a dose–response effect in PwNDs (33–35).

Walking training, balance training, and proprioceptive exercises were frequently used to improve walking function in accordance with international guidelines for stroke, MS, and PD (35–37). It should be noted that a comprehensive data collection on the content of gait rehabilitation was beyond the scope of this study, as this topic has been investigated elsewhere (3). The treatments described in this study are limited to the everyday clinical practice of the participating rehabilitation centers. Other common PT approaches, such as dance or aqua therapy (38, 39), were not investigated.

Our findings indicated that certain interventions were exclusive to the restorative approach, such as manual therapy and motor imagery. However, most treatments used were similar for both approaches, resulting in a non-statistically significant difference. A more detailed description of the PT treatments is needed to understand the specifics between the restorative and compensatory approaches.

The study has certain limitations. The definitions of compensation and restitution in PwNDs, by Levin et al. (4), were used in this investigation. However, different interpretations of these concepts among physiotherapists represent a potential for subjective bias in their responses. Furthermore, we asked physiotherapists to classify their interventions on PwNDs at the conclusion of the PT program. Some interventions may have started with a restorative approach and then shifted to a compensatory approach or vice versa from session to session, depending on the needs of PwNDs. It was not possible to collect this type of information using our physiotherapist treatment report form.

No information was collected on the composition of the multidisciplinary rehabilitation team.

Finally, we did not consider other variables that have already been identified as predictors of good rehabilitation outcomes, such as the early presence of lower limb spasticity and muscle strength (40) or disease severity (12, 13, 41).

In this sample, a restorative approach was often used in neurological rehabilitation. The main factor that influenced the choice of a restorative approach was predicted clinical improvement.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving humans were approved by Ethics Committee of the Don Gnocchi Foundation. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

FMes: Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Writing – original draft, Writing – review & editing. TB: Data curation, Formal analysis, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. FMar: Data curation, Investigation, Methodology, Writing – original draft, Writing – review & editing. SS: Data curation, Investigation, Writing – original draft, Writing – review & editing. CA: Data curation, Investigation, Writing – original draft, Writing – review & editing. SB: Data curation, Investigation, Writing – original draft, Writing – review & editing. VB: Data curation, Investigation, Writing – original draft, Writing – review & editing. FMat: Data curation, Investigation, Writing – original draft, Writing – review & editing. EP: Data curation, Investigation, Writing – original draft, Writing – review & editing. MP: Data curation, Investigation, Writing – original draft, Writing – review & editing. AT: Data curation, Investigation, Methodology, Writing – original draft, Writing – review & editing. SM: Funding acquisition, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing. DC: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Writing – original draft, Writing – review & editing.