Neurodevelopmental and/or intellectual disabilities/disorders (NDID) consist of a range of diagnoses and functional impairments of a neurological origin that can be experienced as functional delays in developmental milestones such as language, communication, social skills, intellect, executive functioning, and motor development (1–4). Examples of NDIDs include autism spectrum disorders (ASD), attention deficit/hyperactivity disorder (ADHD), Down syndrome, and fetal alcohol spectrum disorder (FASD). The prevalence of NDID across developed countries in children and youth 18 years of age and younger ranges from 8% to 15% (5–7).

Because of their limitations, children with NDID are often excluded from participating with their peers in typical school and community activities. This may lead to isolation and loneliness which can result in limiting their learning abilities and development potential, with long term consequences for their physical and mental health (8). Adapted community-based physical activity programs may help to address the challenge of isolation by providing a safe and playful environment to facilitate integration, socialization and learning.

The literature on the effects of physical activity programs (PAP) on children and youth with NDID has shown the positive impacts of physical activity on a variety of outcomes such as physical literacy (9), social integration (10, 11), friendship development (12–14) and cognitive functioning (15–17). However, the majority of these studies are limited due to the high likelihood of the Hawthorne effect as well as observational biases in context of controlled trials and quasi-experimental studies. Other limitations include short follow-up periods and a limited range of outcomes. For example, in the studies conducted by Ketcheson et al. 2021 (18) and Zhao & Chen 2018 (19), follow-up measures were taken one-week after the intervention. Furthermore, studies such as Sotoodeh et al. 2017, Zachor et el. 2017 and Caputo et al. 2018 (20–22) did not have any follow-up measures, which raises concerns regarding the results reported. With respect to limited outcomes, many studies measured only one outcome, potentially overlooking other outcomes that could have been impacted. For instance, many studies collected data on just one type of outcome measures among motor development, social skills, quality of life (QoL), community activities, leisure or home participation. Others only collected specific measures of motor progress (23), or did not consider how social communication may have been impacted (24).

In 2015, we published a retrospective study of a gymnastic-based movement therapy program for children with NDID called Empowering Steps Movement Therapy (ESMT™) (25). Following this study, we conducted an exploratory prospective cohort study that is described in this current paper using standardized measurement tools to confirm previous findings and develop more complex analyses that will be reported in this paper. We also attempted to avoid the weaknesses of previous studies by studying children in a more natural environment (preventing Hawthorn effect), without observational biases and minimizing selection biases.

The aim of this study was to investigate the possible effects of the ESMT™ program on children and youth with NDID. The effects of the program were assessed for numerous outcomes, including their impact on children's motor skills, social adaptive skills, leisure adaptive skills, community participation, and their QoL. When changes were observed, we sought to determine the reasons, identify the main factors involved, and understand the dynamics of the change process.

The study was reviewed and approved by the University of British Columbia/Children's and Women's Health Center of British Columbia Research Ethics Board (Certificate numbers: H14-02258).

Baseline characteristics of children and their families have been summarized in Table 1.

A high percentage of the participants presented with ASD: 31/42 (73.8%), with either single diagnosis (26/31) or in combination with another diagnosis (5/31). One study participant did not have a precise diagnosis, and another one did not report the diagnosis. The diagnoses were based on de-identified information provided by the parents in the demographic questionnaire, without medical verification. The mean age across the 42 participants was 107.1 (SD: 35.7) months at baseline; median age: 99.5 months (IQR: 55.8). Out of the 42 subjects, 34 (81%) were males and 8 were females.

Our findings show trends towards improved motor skills, as well as in home and community participation among children with NDID (mostly with ASD) in context of attending the ESMT™ program, over a 12-month follow-up. Although the observed possible gains could be the result of a child's normal development, we believe that the observed positive trend is mostly related to the specific precise scaffolding offered by the one-to-one personalized approach offered by the ESMT™ program. It is also possible that other contextual factors played an important role for some children (such as attending another programs), but it is not likely to influence most of the children in our study. A smaller trend of positive change was also observed in Social Adaptive skills. Overall, these findings are consistent with the ones observed in the retrospective study of the ESMT™ program published in 2019 (25). The discussion addresses results for each outcomes separately.

Motor skills showed a trend toward improvement (0.17 score point in BOT-2 SF for each month in the program) with even larger positive trend in children with ASD (0.20 score point). This improvement is supported by other studies. Najafabadi et al. 2018 (43) showed similar results when conducting a quasi-experimental study to investigate the effects of a 12-week SPARK exercise program (three 40-minute sessions per week), on the social and motor skills of 26 children with ASD between the ages of five and twelve. Similarly, Ketcheson et al. 2017 (44) conducted a non-RCT study of 20 children with ASD aged 4–6 years with a control group (n = 9) that did not receive the intervention. The intervention group participated in an 8-week intervention consisting of motor skill instruction for 4 h/day, 5 days/week. Statistically significant differences were found between groups on all three motor outcomes including locomotor, object control raw scores, and the total gross quotient (p ≤ 0.01). Finally, Sowa et al. 2012 (45) conducted a meta-analysis of 16 papers with a total of 133 participants with ASD, 4–14 years of age on the effects of physical exercise on three core symptom areas of ASD, motor, social, and communication skills. The results showed benefits of physical exercise on the participants’ motor and social functioning. Notably, individual-based interventions resulted in a greater improvement than group-based interventions.

Our findings with regard to the motor skills support a child-centered approach in which the training plans are consistently individualized to each child's unique needs, including their specific learning styles, weakness and strengths; this method works to maximize each child's potential for success through the ESMT™ program. This is also important to ensure the child remains motivated in the learning process. Children attending the ESMT™ program are assigned to a personal therapist and a set time for the school year to encourage consistent attendance and allow for relationship development with their therapist and other children/staff attending at the same time.

Social adaptive skills showed minimal change during the follow-up (0.04 score point). However, children with ASD progressed better (0.13 score point) than children with other diagnoses. This supports that the personalized approach in the ESMT™ program might have led to more direct communication and interaction between the child and the instructor resulting in the development of communication skills, and a better response to a child's special needs and improvements in a child's sense of trust, especially important for children with ASD. This result is particularly interesting because the ESMT™ program does not offer a structured interactional strategy to promote social development among children. Yet, their personalized approach led to the progress in social adaptive skills, especially in children with ASD. It is also important to note that social adaptive skills maybe affected by other factors such as socioeconomic status or parental education.

Other studies have shown positive impacts of personalized adapted programs to improve social adaptive skills among children with ASD. Zanobini and Solari 2019 (46) found positive results in social outcomes from a one-to-one swimming program using a quasi-experimental design including pre-test, post-test and after 6 months follow up. They recruited 25 children with ASD, 3–8 years old. The analysis revealed a significant Time x Group interaction for social outcome where the experimental group showed higher improvement compared to the control group. In contrast, Bremer et al. 2014 (47) studied the effectiveness of a 12-week fundamental motor skill intervention for 9 children aged 4 years old with ASD using a 1:1 or 1:2 ratio and RCT design. There were statistically significant changes between the control and the experimental groups in motor improvements but not in adaptive behavior or social skills.

Community participation indicated possible gains over time (0.16 score point). This positive change was larger in children with ASD (0.21 score point). Similar positive trends have been shown in several studies. For instance, Ajzenman et al. 2013 (48) in a pre-post pilot study aimed at determining the effect of 12 weeks hippotherapy (HPOT) on motor control, adaptive behaviors, and participation among children with ASD. They included a small sample size of 7 children aged 6–12 years, with no control group. Based on these interviews they found significant increases in participation in daily activities, self-care, low-demand leisure, and social interaction. However, Zwicker et al. 2015 (49) conducted another pre-test/post-test study to investigate the effects of a summer camp intervention for 11 children with Developmental Coordination Disorder (DCD) who attended a 2-week camp twice, one year after the other, with 2:1 child to staff ratio. They found no significant changes in community participation. Additionally, Cleary et al. 2017 (50) conducted a two-group single-blinded phase I randomized controlled trial on a 9-week aerobic exercise program for 19 young people aged 8–18 with cerebral palsy using a 1:1 or 1:2 staffing ratio. Results showed no difference between groups in community participation and preference. The possible positive change in community participation after attending the ESMT™ program may be due to its participant-centered, multidimensional approach and the sense of trust that progressively develop over time between the child and their personal coach. This approach promotes more effective communication with the instructors, builds self confidence, and supports the ability to speak and share opinions. Together, these results may contribute to increased community participation.

However, the change in this outcome could also be connected to other factors such as family interest in participating in community activities, family culture, income, etc. We also observed a positive trend in home participation overtime similar to community participation (0.16 score point). This positive change was higher in children with ASD (0.25 score point). This positive trend is likely supported by the same factors that support community participation to enhance socialization.

A negative trend for QoL was observed during the course of the study (−0.22 and −0.10 score points) for all participants and children with ASD, respectively. This negative trend was not expected, but it has been observed in other studies. In a randomized controlled trial of a 9-week aerobic exercise program in a young population of children with cerebral palsy (n = 19), Cleary et al. 2017 (50) failed to show a significant difference in the QoL outcome measure within the intervention group. Oriel et al. 2012 (51), studied the benefits of an inclusive community-based aquatics program which ran for 8 weeks with 1:2 adult to child ratio using a quasi-experimental design. They included N = 23 children with and without disabilities. There was a large age discrepancy between the two groups: disability group mean 12.3 years, without disability group mean 6.77 years. Measures included QoL, self-concept, and peer acceptance. They observed a significant decrease in happiness in children with disabilities. Children with NDID also demonstrated a significant decrease in the School subsection of the PedsQL 4.0 compared to neurotypical children. Some studies, however, did show a positive change in QoL. As an example, Ansa et al. 2021 (52) studied the effect of an 8-week community-based functional aerobic training on the motor performance and QoL for 10 children with Cerebral Palsy 12–16 years of age. Unlike the other studies, they found significant improvement in QoL. Changes in QoL score may be difficult to interpret because it is affected by multiple factors around the child that were not considered in this study. For instance, a model proposed to understand the factors that influence the QoL in ambulatory children with Cerebral Palsy, adapted from the framework of the International Classification of Functioning (ICF), emphasizes the importance of activity capacity, activity performance, body structure and function, health conditions, participation, and personal factors such as age or satisfaction (53). For example, the child's motor function could improve while the QoL regresses because of a problem with the school experience, or a personal. negative experience. Factors related to the family such as parental illness, conflict with siblings or divorce could also impact the child's QoL negatively as per family transactional interaction (54). In addition, the change in QoL is not expected to follow at the same pace as the changes in motor skills; observation for periods longer than one year may be required to captured possible delayed improvement. Another factor to consider is the difficulty for some parents to complete the KIDSCREEN-27 to describe their child's quality of life. For instance, parents' scoring may be affected by some disappointment when they consider their child to have made little progress. Finally, the KIDSCREEN-27 and similar tools were developed mainly for neurotypical children and may not be sensitive enough to capture changes in children with NDIDs. It was not fully validated to reflect what represents quality of life for children with NDIDs.

Our analysis also investigated the impact of children's and families' baseline characteristics, as well as the child's participation in other physical or not-physical activities on observed changes in measured outcomes. However, all findings regarding the possible confounders and effect modifiers are to be considered with caution because of the small sample size and selected population. Table 3 shows the list of potential confounder and effect modifiers identified for each outcome. Attending other one-to-one PAP has been shown to decrease the monthly motor skills score. This effect seemed to be opposite for attending other group PAP. Those differential effects might reflect the fact that children with more severe condition may attend more one-to-one programs than group programs and at the same time, these children have more difficulty to improve their skills, despite attending more programs. Regarding the positive effect of attending other group PAP, there maybe two possible explanations: one is that attending more group PAPs may stimulate more participation, hence more engagement in the PAP and more progress. The second reason maybe that families of children who were attending more activities might be generally more involved in their child's progress, which might result in an improvement in their outcome scores. These hypotheses should be verified in future studies with larger sample sizes and mixed method to better capture the process of improvement and related factors. Observing the positive impact of a younger age for community participation and QoL outcomes, if verified, may highlight the importance of early interventions. Finally, the positive impact of shorter previous attendance on QoL maybe the result of the child not reaching the plateau effect yet. Gymnastics is recognized as a physical literacy program which helps children to develop fundamental movement skills such as jumping, climbing, running, balancing and coordination. In early stages of physical development, the emphasis is on physical literacy improvements where there is generally faster development than in later stages where the emphasis is on skill development and the refinement of more difficult skills. This can lead to a plateau effect which is associated with a period of stagnation or leveling off in progress and is a natural phenomenon in various aspects of life, including learning, exercise and personal growth.

In summary, our investigation suggests possible benefits of the ESMT™ program for physical abilities and participation in both community and home settings, though these results should be interpreted with caution in context of an exploratory study. The positive changes may reflect the program's inclusive atmosphere, where children of all abilities learn similar skills at their pace. The activities are carefully scaffolded to each participant's developmental level, ensuring that all participants are in their zone of proximal development, whether that be learning to walk or mastering a double back salto. This extremely large range in learning has one thing in common: all children are being reinforced by their effort rather than their skill level. Children of all abilities share laughter, tears, and cheers of success, while cheering on those that are struggling. That is what allows all the participants, regardless of ability and differences among them, to feel a sense of camaraderie in the gym. The common thread is that all participants are learning at their own unique pace and based on their specific needs, as well as the personal and family related factors. This process is in line with the sociocultural learning theory of child development that emphasizes the paramount importance of the social context, community involvement and the use of a large range of artefacts and language as tools for optimal learning (55). The learning process is continuous due to the stimulation of increasingly more complex psychological functions such as, e.g., curiosity, logical thinking, problem solving and complex cognition. Success at completing tasks is a source of self-esteem and self-confidence; this promotes further motivation to acquire new knowledge along with physical and social skills.

The findings of this study should be considered in light of several limitations. First, bias from loss to follow-up may have influenced the results, as many participants dropped out for reasons like personal health and family issues, surgery, or study demands (e.g., completing multiple questionnaires). We compared baseline and follow-up 1 scores, as well as the percentage change between the two assessments, between the dropout group (those who left after the first follow-up) (Supplementary Table S1) and the non-dropout group (those who remained) (Supplementary Table S2). This showed that the dropout group made greater progress while they were in the study, meaning our analysis focused on children with slower progress, likely underestimating improvements. However, for community and home participation outcomes, the results might be biased toward being overly optimistic since the dropout group showed less progress.

A second limitation is the absence of a control group not receiving the personalized intervention. It is then difficult to assess the specific effects of attending the program and interpret the meaning of an improvement of 0.17 or 0.20 score point in the BOT-2 scale. Selecting a control group identical to the study participants and not receiving any interventions is almost impossible; from our experience, most children with NDID receive some kind of interventions in different settings, special schools, etc. making the comparison difficult and not valid. Further, comparison with the progression of children with neurotypical development is irrelevant. Finally for the same reason of huge heterogeneity among children attending ESMT™, it is not possible to determine a minimal clinically important difference (MCID) that could be used as valid standard comparator. Consequently, we used the change in score as the value that helped determining the progress, and we observed that children with ASD improved faster than the group that had other diagnosis.

The small sample size is another limitation that limited the ability to detect statistical significance and restricted the development of multivariate models to only include covariates one at the time to assess their impact on the main association. The lack of blinding, due to the study's nature, could also be a limitation, although data collection was conducted rigorously by a single assistant within the child's usual program. Another limitation is the limited description of the concomitant interventions that children may receive when they attend other community programs. We only collected the information regarding attendance of such programs but no information regarding the specific features of the programs. We also did not collect any information regarding the possible practice of physical activity at home. Practicing at home or attending other programs may have a positive effect regarding child learning and development; however, attending several programs may also reflect the family anxiety regarding their child's development. One limitation is a possible “parental report bias”, as questionnaires were completed by parents instead of the children, due to the children's limited capacity to respond as a result of their NDID. Lastly, while most findings were positive, previous research indicates optimal physical activity outcomes require participation three times a week (56), while most children in the ESMT™ program participated 1–2 times a week. The cost of one-to-one programming, not covered by Canada's medical system, also limits the feasibility of more frequent sessions for many families.

Despite these limitations, our findings show in a reliable way the positive impact of the one-to-one PAP on motor skills development and social participation at home and in the community. Further investigations are still needed to include a larger sample size and collect and monitor the child's and family's personal factors in more detail. A comparison with a control group should also be included to account for the effects of normal development.

Finally, it may be possible to enhance the exposure to the inclusive and motivating environment of the ESMT™ intervention by including other professionals to the team such as a Speech and Language Pathologist, Behavior Consultant and Physiotherapist, which may lead to even greater results. It would be the beginning of a Hub model for the treatment of the neurodiverse population with all professionals working as a team. The effects of this intervention model on children and adolescents with ASD should be considered in future studies.

One final point of attending the ESMT™ intervention is the high cost that may limit access to the program. Although access to the facility and use of various equipment are free of charge, as they are covered by the general gym program, the sessions' fees reflect the program's expenses for management and paying professionals for individualized one-to-one coaching. We hope this article promotes interest for further development of one-to-one programs and help advocating for public financial support for such programs.

This exploratory cohort study was conducted with children and youth with NDID (mostly ASD) while attending the ESMT™ program to assess changes in motor and social adaptive skills and changes in a wide range of other outcomes. Despite the major limitations, this prospective cohort study suggests that the training program may support improvements in motor skills, community and home participation. These findings also suggest that the benefits of attending this community-based PAP may go beyond the motor skills development. Children with ASD may benefit the most from this personalized program, especially among those with higher levels of severity. ASD is a well-documented disease that can be generally characterized by limited social communication and “restricted, repetitive sensory-motor behaviours”, difficulty in relationship building, and over- or under-reaction to various environmental stimuli (1, 57). For children with ASD, the one-to-one approach maybe the preferred intervention as it provides opportunity for more effective scaffolded communication with the instructors in an environment that can be better adapted to their needs (e.g., less stimulation by sound or light). Finally, the observed changes were associated and affected by several other child and family related factors. Health care providers may consider those factors when suggesting participation in community-based PAPs for children with NDID. They may also use the results of the present study to organize the care plan and data collection for children and youth with NDID. Families are also encouraged to have their children and youth with NDID participate in community-based PAPs. Finally, our study suggests that a multisystem approach embedded in social contexts could be included within the standard-of-care treatment for ASD (22).