Attention deficit hyperactivity disorder (ADHD) belongs to the most common neurodevelopmental childhood disorders worldwide (Tamm, 2009). Affected individuals typically present symptoms such as hyperactivity, impulsiveness, and inattention oftentimes leading to reduced quality of social functioning (Tamm, 2009). Additionally, studies have shown that many children with ADHD perform inferiorly in tests of fine and gross motor skills compared to age-matched neurotypical controls (Pitcher et al., 2003; Tseng et al., 2004).

Motoric ability, especially postural control and balance in ADHD patients have previously been investigated (Raberger and Wimmer, 2003; Ghanizadeh, 2011; Mao et al., 2014; Goetz et al., 2017). Difficulties in static and dynamic balance tasks among affected children likely stem from executive control problems and suggest a common cerebellar dysfunction (Raberger and Wimmer, 2003; Goetz et al., 2017). Interestingly, Schlee et al. (2012) did not find any deficiencies in ADHD patients compared to neurotypical children measured in a single and double-legged static balance test, which may, however be owed to the easiness of the tasks.

Balance ability is essential for many aspects of daily life, like riding a bike or engaging in physical activity. Balance, or more so postural control, can be defined as the act of maintaining, achieving, or restoring a state of balance during any posture or activity, and therefore constitutes a fundamental motor function (Pollock et al., 2000). Integrating inputs from the somatosensory, visual, and vestibular systems and using these sensory signals to generate coordinated motor actions and maintain body equilibrium is a process that may be impaired in ADHD patients (Kandel et al., 2000).

While balance ability of ADHD-affected children has been extensively studied in the past (Pitcher et al., 2003; Wang et al., 2003; Shum and Pang, 2009; Konicarova et al., 2014; Mao et al., 2014; Christ et al., 2024), little is known about postural strategies used by the affected group to maintain balance. The hip and ankle postural strategies have previously been identified and explored in the literature among a healthy population (Nashner and McCollum, 1985; Runge et al., 1999; Guihard and Gorce, 2002; Boonstra et al., 2013). While the ankle strategy is commonly described as a single-segment inverted pendulum and is predominantly used during activities that require little balance regulation, the hip strategy is characterized as a double-segment inverted pendulum and is observed during more complex balance tasks involving unstable and narrow surfaces or backward translation (Horak and Nashner, 1986; Runge et al., 1999).

In addition, a so-called upper body strategy, with regards to movements of the head, arms, and trunk, seems to play an essential role in maintaining balance (Hsu et al., 2007; Pinter et al., 2008; Kilby et al., 2015; Boström et al., 2018). This is not surprising as moving or stretching out the arms to remain or restore equilibrium is a much-employed tactic in challenging balance tasks, which is supported by the literature (Milosevic et al., 2011; Shafeie et al., 2012; Cheng et al., 2014; Patel et al., 2014; Boström et al., 2018; Objero et al., 2019). Arm raises have been associated with an elevation of the center of mass (COM) position (Roos et al., 2008) as well as with a forward shift of the COM after its displacement due to a fall (Marigold et al., 2003) suggesting that movements of the upper body play a vital role for example in the design of fall prevention programs for the elderly (Allum et al., 2002; Marigold et al., 2003; Roos et al., 2008; Objero et al., 2019).

Postural strategies are typically defined by joint torques, ground reaction forces (GRF), and electromyography (EMG) activations (Runge et al., 1999; Clifford and Holder-Powell, 2010). Especially joint torques seem to be a valuable measure, as they can indicate how body movements are generated (Runge et al., 1999). While most studies assessing postural control and balance ability in children with ADHD draw conclusions based on postural sway (Wang et al., 2003; Shum and Pang, 2009; Shorer et al., 2012; Amini et al., 2018), a measure based on body kinematics, this study analyzes a less frequently used kinetic measure based on body forces, namely joint torque.

Studying the use of hip and ankle postural strategies, supported by an upper-body strategy, as postulated by Boström et al. (2018), may not only provide important awareness and knowledge for fall risk reduction in older adults, but may also contribute to the understanding of balance strategies used by children with and without ADHD. Recent meta-analysis underline that children with ADHD have an increased risk of injury, which is likely owed to deficits in motor control (Amiri et al., 2017; Ruiz-Goikoetxea et al., 2018). The aim of this study was to investigate the differences in postural strategies used by children with ADHD and typically developing children of the same age during static and dynamic balance tasks. To the best of our knowledge, this is the first attempt to do so, and thus the findings may be of great practical relevance for injury prevention and wellbeing within the affected group. To quantify differences in balance strategies between the two groups, the following two-tailed hypotheses were tested.

ADHD-affected children generated joint torque amplitudes different from those of their typically developing peers. This is shown by the statistically significant main effect ADHD for both, the one-leg stand (F = 5.49, p = 0.02, ηp2 = 0.02) and the balance beam task (F = 6.39, p = 0.01, ηp2 = 0.009). The analysis yielded no significant differences between children with and without ADHD in terms of joint torque variance for both tasks. Strongly significant main effects for the JointGroup variable across all conditions were expected and indicate the dissimilarity in torque generation for the joint groups analyzed. Furthermore, the covariate Age had a significant effect on the torque amplitude measures, but not on the torque variation for each task (Table 1).

Significant interaction effects on the torque amplitude in the one-leg stand (F = 10.01, p < 0.001, ηp2 = 0.08), on the torque variation in the one-leg stand (F = 3.79, p = 0.02, ηp2 = 0.03), and on the torque variation in the balance beam task (F = 5.39, p = 0.005, ηp2 = 0.02), show that the two groups were affected differently in terms of torque generation in the various joint groups (Table 1; Figures 1, 2). A subsequent post-hoc test revealed that for the torque amplitude, only ankle and upper body joint torques but not hip joint torques differed significantly between groups, with higher torque amplitudes found for the ADHD group in all four instances. Specifically, group differences were found for the static balance task in the ankle joint (F = 20.7, p < 0.001, ηp2 = 0.08) and the upper body joints (F = 5.49, p = 0.04, ηp2 = 0.02), and for the dynamic balance task in the ankle joint (F = 5.2, p = 0.05, ηp2 = 0.007) and the upper body joints (F = 6.39, p = 0.04, ηp2 = 0.009; Table 2; Figure 1). Descriptive statistics are provided in Table 3.

The aim of this study was to investigate the differences in postural strategies between children with and without ADHD during static and dynamic balance tasks. The results can at least partially confirm our hypothesis that there is a significant difference in torque amplitude and variation between children with ADHD and typically developing children in the tested ankle, hip, and upper body joint groups, during static and dynamic balance tasks. While significant differences between the two groups were found, these are not present across all joint groups and balance tasks. An investigation of general differences in task performance and balance ability between children with and without ADHD, preceded this analysis, yielding significant group differences in the task of dynamic balance (Christ et al., 2024).

For the assessment of static balance through a 25-s one-leg stand, group differences were found for torque amplitude, i.e., the amount of torque applied to the ankle and upper body joints differed significantly between the groups. While the groups differed by only 7.5%, with higher upper body joint torques produced by children with ADHD, the group differences were more pronounced for the ankle joint torque during the same task. The group of children with ADHD thereby showed a 20% higher joint torque amplitude than the neurotypical children. Correspondingly, the evaluation of our dynamic balance task in the form of an eight-meter balance walk over a narrow beam revealed significant group differences for the ankle and upper body joints in terms of torque amplitude. For this more challenging balance task, the group difference was accordingly marginally higher with a 22% higher ankle joint torque and 11% higher upper body joint torque amplitude in the ADHD group compared to the neurotypical children.

The ankle strategy is an activation pattern that effectively restores equilibrium by shifting the body's COM predominantly around the ankle joints (Horak and Nashner, 1986). Increased ankle torque generation during tasks of postural stability is associated with reduced neuromuscular control or inefficient movement patterns, such as corrective use of certain muscles to compensate for insufficient strength, flexibility, or coordination (Pozzi et al., 2015; Fok et al., 2021). Deficits in proprioception and neuromuscular control are common among children with ADHD (Hassan and Azzam, 2012; Shum and Pang, 2009; Izawa et al., 2012; Iglesias et al., 2014; Stray et al., 2015). A weakness in neuromuscular control among ADHD-affected children may manifest as reduced precision in muscle activation (Stray et al., 2015), which could in turn interfere with the adjustment of joint torques involved in maintaining balance and stability. Likewise, impaired proprioception or processing of sensory inputs within the patient group is likely owed to the compromised interaction among inputs from the vestibular, visual, and somatosensory systems, which feed the postural control system with orientation information and maintain body equilibrium (Guskiewicz and Perrin, 1996). However, it is also crucial to emphasize that high joint torques do not inevitably imply a shortcoming. In fact, most sports require high torque generation for efficient movement control and injury prevention.

Our findings of increased torque production in the upper body joints in the ADHD group suggest that affected children exhibit significantly more head, arm, and trunk movements during static and dynamic balance than typically developing children. Moving the upper body, especially the arms is a characteristic strategy to maintain or regain balance (Milosevic et al., 2011; Shafeie et al., 2012; Cheng et al., 2014; Patel et al., 2014; Boström et al., 2018; Objero et al., 2019). However, this strategy seems to be used to some extent by both, the neurotypical and ADHD-affected children as the group difference of 7.5% in the one-leg stand task is not as large as might be expected. A group difference of more than 11% on the balance beam task suggests that as task difficulty increases, the groups diverge further in their need for compensatory arm movements. Difficulties in long-term postural control have previously been observed in adult patients with ADHD, whose postural deficits were linked to symptoms of hyperactivity and impulsivity (Jansen et al., 2019). Similarly, attention regulation is often affected, so it is conceivable that the patient group may have difficulties in consciously controlling movements due to the nature of these symptoms, resulting in increased or inconsistent joint torques.

Within the 25-s single-leg standing task, no significant group differences were found for joint torque variation, our measure of the amount of change of torque for any joint group. This is somewhat surprising as the often uncoordinated and imprecise movements presented by children with ADHD would suggest that especially the measure of torque variation would reflect this. In contrast to that, it is the group of typically developing children that presents a 34% higher torque variation around the hip during the task of static balance. These findings are not in line with Mao et al. (2014) who noticed that children with ADHD, unlike their healthy controls, continuously changed their patterns of movement to find the correct balance strategy in a dynamic horseback riding task. Although not significant, this rather large group difference is interesting as it suggests distinctive tactics to restore equilibrium between the two groups. A similar observation to the static balance task was made in dynamic balancing, where the group difference was in fact significant with a 25% higher hip torque variation in the neurotypical group. Moreover, the 9% higher hip torque amplitude of the neurotypical group during static balance is noteworthy as this difference is trending toward significance.

The discovery of restricted hip movement is not completely novel and has already been made by Stray et al. (2009) where 80% of the ADHD patients displayed limited hip movement during a passive movement test, while this was not shown by the group of typically developing children. The authors conclude that this restricted hip movement may be a direct consequence of the balance problems linked to ADHD, as these are related to difficulties in keeping the torso in an upright position. They argue that the observed hip stiffness results from the extreme use of gross movement muscles like the erector spinae and psoas major muscles to compensate suboptimal proximal stabilization (Stray et al., 2009). Therefore, many ADHD patients seem to have induced biomechanical adaptations leading to strategies that involve more ankle torque to counterweigh the difficulty to create hip torque. This adaptation, however, is rather undesirable as the hip strategy seems to be more efficient than the ankle strategy for the restoration of balance. Regarding our findings, it is therefore thinkable that only the neurotypical subjects were able to generate sufficiently large compensatory hip torques as a way to regain stability. The hip strategy specifically involves activating the large muscles of the thighs and trunk to compensate for loss of balance (Horak and Nashner, 1986). Early activation of hip and trunk muscles during challenged translation on a narrow surface has previously been reported in healthy adults (Henry et al., 1998). Although hip and ankle strategy are two distinctive and similarly important methods, it is known that especially for more demanding balancing tasks and faster perturbations the hip strategy seems to be more efficient in stabilizing the body in terms of balance recovery and muscle stress compared to the ankle and upper body strategies (Runge et al., 1999).

Another interesting theory for hip stiffness and thus reduced torque generation in ADHD subjects is related to the brain's dopamine deficiency. Dopamine is an important neurotransmitter strongly involved in movement, motivation, and attention inter alia (Ungerstedt et al., 1982). Parkinson's disease, another condition in which low dopamine concentration in the brain plays an essential role, is characterized by symptoms like rigidity or stiffness (Heisters, 2011). Low brain dopamine levels may hence be another reasonable explanation for the observed hip stiffness in ADHD. Understanding why this stiffness manifests around the hip more than in other body regions requires further research.

Pertaining to brain regions involved in ADHD, not only disturbances in the prefrontal cortex and basal ganglia, but notably also impairments of the cerebellum are known to contribute to motor deficits such as balance or muscle tone regulation in the subject group (Goetz et al., 2014). Numerous neuroimaging studies report a reduction in cerebellar volume in ADHD patients compared to typically developing subjects (Castellanos et al., 1996; Krain and Castellanos, 2006; Mackie et al., 2007; Bledsoe et al., 2009). Since the cerebellum has a substantial function in postural control, sensory integration, and motor coordination, differences in balance strategies and thus differences in resulting torque generation between the groups of ADHD affected and neurotypical children during tasks of static and dynamic balance are plausible. Besides, excessive, unregulated movements may be owed to weakened impulse regulation in connection with cerebellar deficits (Jansen et al., 2019).

Moreover, a notable finding in our data is that age turned out to have a small but significant effect on torque amplitude in both balance tasks, while it did not significantly affect the torque variation. Many children with ADHD display a delay in motor development and compromised mechanisms of somatosensory integration (Rosa Neto et al., 2015). Yet, there seem to be age-related differences in torque generation among children aged 8 to 12 years.

The present study has several limitations. There is currently no standard data or uniform calculation method for the use of ankle, hip, or upper body strategies during balance tasks in children. Conclusions on increased or reduced joint torque amplitudes and variations can therefore only be drawn with respect to our relatively small control group of neurotypical children, of which we also lack standard data. Another limitation is that small faults in performance, such as losing balance and slipping of the balance beam, certainly create larger torques, which may be a confounding variable. However, such faults happened only rarely and were observed among both groups. A further limitation concerning the methodology is that the inverse dynamics simulation used to model the joint torques has to make specific assumptions to overcome the motor redundancies of muscular activation patterns. These assumptions, however, are potentially tailored to a neurotypical population and may or may not differ between neurotypical and non-neurotypical subjects. The small sample sizes and unequal gender distribution of the two groups are not necessarily representative of a broader ADHD population. Literature suggests that balance performance may differ among boys and girls (Nolan et al., 2005; Mickle et al., 2011). Although this evidence was shown for postural sway it is not unlikely that the same may apply to postural strategies in terms of joint torques, thus the findings must be interpreted with care. However, our results are useful to make deductions on boys and girls with and without ADHD on a broader age spectrum of 8 to 12 years. Another potential limitation is that the study did not differentiate between subtypes of the disorder, which could be improved by adding this element as a statistical covariate. However, postural stability in young ADHD patients is more dependent on the comorbid psychiatric disorders than on subtypes (Ghanizadeh, 2011). In summary, future studies would need to include a larger number of subjects and, if possible, collect detailed information on the demographic and clinical characteristics of the patient group to strengthen the applicability of the findings. A strength of our study is that all children in the ADHD group were medication-naïve, thereby excluding the possibility that postural stability is confounded by symptom-reducing medication.

Although during static and dynamic balance tasks various postural strategies are typically engaged simultaneously, our findings indicate that in a direct group comparison, children with ADHD tend to create torque predominantly around the ankle joints to stabilize their body, while they may not be able to rely on proximal stabilization from the hip, possibly due to high muscle tone and stiffness in this area. The larger torques observed when shifting the body COM around the ankle joints may result from reduced neuromuscular control, compromised integration of proprioceptive inputs, and inefficient movement patterns, commonly present in ADHD patients as reported by related literature. However, further research is warranted to confirm this connection. Likewise, increased movements of head, arms, and trunk during balancing may be related to symptoms like hyperactivity and impulsivity. These deficits in postural control are likely associated with a reduced cerebellar volume commonly reported in ADHD patients. While the amplitude of ankle and upper body torques varied notably between the two groups, no differences were observed for torque variation for these two joint groups. For the hip joint, lower torque variations were presented by the ADHD group compared to neurotypical group, implying very little hip movement during both balance tasks. Children with ADHD may have adopted postural strategies with a high involvement of ankle torque to compensate for their potential hip stiffness. Comparable to Parkinson's disease, hip rigidity in ADHD patients may be linked to dopamine deficiency in the brain, which, however, remains to be confirmed by future research.

Differences in joint torques during balance tasks between children with and without ADHD are the result of several factors, ranging from motor to sensory and even cognitive challenges. Creating joint torque to stabilize the body and control movements is beneficial and fundamental in many sports and daily activities. Children with ADHD may have difficulties efficiently employing and combining postural strategies to the requirements of a given situation. Therefore, assessing postural control within the affected group is an essential step in creating therapeutic interventions and training programs tailored to the precise needs of each individual so that postural stability is supported, whereby injury risk is minimized and wellbeing sustained.