A Cross-Sectional Study to Measure Physical Activity with Accelerometry in ADHD Children according to Presentations

(1) Background: Attention deficit hyperactivity disorder (ADHD) is a common mental disorder affecting 5–7% of school-aged children. Previous studies have looked at the effects of physical activity interventions on the symptoms of ADHD, although few have compared the motor behavior of children with ADHD versus those without. This exploratory study provides detailed information on the patterns and intensity of physical activity and sedentary behavior in children with ADHD as measured by Actigraph GT3X accelerometry, as well as the differences in physical activity in the different presentations of ADHD; (2) Methods: A cross-sectional design was used with a sample of 75 children, aged 6 to 12 years, with and without ADHD. The ADHD group had a previous diagnosis, determined by clinical assessment based on DSM-5 criteria; (3) Results: Physical activity levels were higher in children with ADHD compared to children without ADHD, but there was no difference in sedentary time between groups during weekdays or weekends. Physical activity decreased with age, with significant differences in the ADHD group, who exhibited more minutes of moderate Physical activity in 6–7 year-olds than 10–11 year-olds during weekdays and weekends; (4) Conclusions: Sedentary time increased by age in children without ADHD, and there was a decrease in moderate-intensity physical activity time in children with ADHD by age.


Introduction
Attention Deficit Hyperactivity Disorder (ADHD) is one of the most common neurodevelopmental disorders, with an incidence of 5-7% of the school-age population, and is more common in boys than in girls [1]. In Spain, the prevalence is 6.8% [2]. ADHD, with childhood-onset that in most cases persists into adolescence and adulthood, corresponds to a symptomatic triad, marked by a combination of signs or patterns of steady indicators such as hyperactivity, impulsivity, and attention deficit which can have negative impacts on household, school, social, and occupational functioning [3]. In addition, the DSM 5 lists 18 symptoms in two dimensions: hyperactivity-impulsivity and inattention. Depending on the presenting symptoms, there are three main possible presentations: predominantly hyperactive/impulsive presentation, predominantly inattentive presentation, and combined presentation. For a diagnosis, at least six signs of hyperactivity-impulsivity and/or inattention must be present in infancy and young adult life, and a minimum of five signs in activities of ADHD children. However, most previous studies have been interventions focused on increasing levels of moderate or vigorous physical activity with a range of activities that require different physical and motor skills rather than measuring the amount of physical activity on a typical day [26,27]. Along these lines, other authors, such as Villalba-Heredia et al. [28], assessed the relationship between physical activity, sleep, and academic performance, reporting results indicating that ADHD has an effect on the amount of sleep during non-school days, waking time at the weekend, sleep efficiency during the week, and academic performance. On the other hand, authors such as Best [29] have stated that aerobic exercise seems to improve executive function, and it is possible that activity requiring greater cognitive commitment may have a greater impact on executive function than simpler exercises that require limited cognitive involvement. This is similar to a study from Michigan State University [30], where they concluded that a few minutes of exercise could help ADHD children to improve their academic performance. The study showed for the first time that children with ADHD were better able to suppress or manage distractions and better at concentrating on a task after an exercise session. These and other authors reinforce the importance of observing PA and sedentary time in relation to improving executive functions and academic performance [31]. Motor deficits are very common in children with ADHD, with a prevalence of 30% to 50% of children with the disorder [32]. According to some studies, locomotor hyperactivity is a feature of ADHD that remains into adulthood, with these motor control deficits being more prevalent in children aged 6 to 9 and seeming to attenuate with increasing age [33]. Recently, a study with an African child population looked at the movement patterns of children with ADHD and found mean values of approximately 20 min/day of moderate physical activity, 4 min/day of vigorous physical activity, and 296 min/day of sedentary activity [34].
Measurement of physical activity is as important as it is complex, and there are many ways it can be quantified. As Aparicio-Ugarriza et al. [35] noted, recent years have seen the development of various objective measurement devices, such as accelerometers, that are appropriate for field studies, in contrast to subjective methods such as questionnaires and physical activity diaries [36]. The present study is a pilot cross-sectional study that examined the amount and intensity of physical activity and the degree of sedentary lifestyles in children with ADHD using accelerometry, assessing the different presentations and making a comparison to children without ADHD. It may provide data that will help to increase knowledge about the characteristics of ADHD children. Understanding patterns of PA in children with ADHD may help us better understand the diagnosis in relation to movement.

Participants
The sample comprised 75 children aged 6 to 12 years old, with an average age of 9.47 (SD = 1.73), from public and private elementary schools in northern Spain. A control group of 28 children without ADHD and a group of 47 ADHD children were included.

Procedure and Exclusion/Inclusion Criteria
The study complied with the ethical regulations of the Declaration of Helsinki [37] and was approved by the Ethics Committee of the Principality of Asturias (CPMP/ICH/135/95, code: TDAH-Oviedo). Participation in the study was voluntary, and the anonymity and ethical treatment of the data were ensured. The STROBE guidelines for Strengthening the Reporting of Observational Studies in Epidemiology were used in this study.
The sample was split into two groups: the "ADHD" group of students diagnosed with ADHD, attending state-funded and private schools in the Principality of Asturias, and the "Non-ADHD" group of students in a state school. The sample of children with ADHD was provided by two different hospitals dealing with ADHD in a single health authority region. The evaluation was carried out by a clinician, with the families' approval, in order to verify the inclusion of the participants in the study. The participants came from different schools selected for ease of accessibility, all from the same area of the Principality of Asturias, and all following the same teaching system. There were no differences between the schools since Spanish schools follow a state curriculum, with a school schedule of 25 h per week and the same number of hours of physical education for all participants-2 h per week. Schools were not analyzed separately to determine whether there were differences in intensity between them.
To be included in the study, subjects in the ADHD group had to meet the following requirements (based on the DSM-5 criteria [3]): (1) must exhibit an impulsive pattern of inattention and/or hyperactivity lasting at least 6 months; (2) inattention and/or hyperactivity-impulsivity symptoms must impair performance or developmental levels; (3) these symptoms must be apparent before the subject is 12 years old; (4) multiple symptoms must be present in at least two or more contexts (e.g., home or school, with friends or family, or other activities); (5) there must be clear evidence that symptoms impair or impair functioning in social, school, or work settings. Additionally, participants must not be receiving any treatment. Data collection was conducted prior to the COVID-19 lockdown between March 2019 and June 2019.
In order to define the groups and to determine the relationship between the variables in relation to the different presentations of ADHD, clinicians confirmed the DSM-5 criteria based on the results from the Farré and Narbona test [38], Evaluation of Attention Deficit Hyperactivity Disorder (ADHD). The "Inattentive" group was made up of students who scored above the 90th percentile in the AD (attention deficit) items. The "Hyperactive/Impulsive" group was made up of students with the Hyperactive/Impulsive presentation, scoring above the 90th percentile in the H (Hyperactivity/Impulsivity) items. The "Combined" group, with the presentation combining Attention Deficit and Hyperactivity, were those scoring above the 90th percentile in the total of the items from the H + AD (Hyperactivity/Impulsivity and Inattention) items.

Instruments
A variety of data collection tools were used in the study based on the variables to be included. This information was provided by the school, the parents, or the students.

Physical Activity (PA) and Sedentary Time
The physical activity level of the students was objectively measured with ActiGraph-GT3X accelerometers (ActiGraphTM, LLC, Fort Walton Beach, FL, USA) [39]. Participating students wore accelerometers 24 hours a day for a week (24/7). Participants were instructed to wear the accelerometer on the right hip, secured by an elastic belt, at all times. As the monitors were not waterproof, participants were asked to remove them during showers or baths or if swimming or diving; these times were considered non-wearing times. The sampling interval (epoch) was set at 10 s [40]. The minimum number of days to be included in the sample was 5 days, including at least one weekend day [41]. Non-wear was defined as any sequence of 10 min of consecutive zero counts per minute [36]. Data were collected during the second quarter of the academic year. Using these accelerometers made it possible to objectively quantify both the frequency and intensity of the participant's physical activity (light, moderate, and vigorous), in addition to providing information about rest time (hours of sleep) and hours of inactivity (sedentary), which is one of the main strengths of this study. The sleep variable was collected and analyzed in a previous study [28]. Data were quantified using the cut-off points established by the Freedson et al. [42] algorithm for 6-12 year-olds, provided as part of the ActiLife software. The frequency and intensity of physical activity during one week were objectively measured and recorded for 24 h a day. The accelerometers were placed on the hip with a flexible band since this is the location that has been proven to provide the most reliable data [43].
The categories of intensity are given in Table 1, based on steps per minute, also known as counts per minute (cpm) [42].

ADHD
The Assessment of Attention Deficit with Hyperactivity Scale (EDAH) by Farré and Narbona [38] measures the main characteristics of ADHD and behavioral disorders coexisting with ADHD. The scale consists of 20 items, 5 for Attention Deficit, 5 for Hyperactivity/Impulsivity, and 10 for Behavior Disorders. The questionnaire is completed by students' parents or teachers. Responses are given using a 4-point Likert format with values ranging from 0 to 3. Cronbach's alpha was high for the full scale (α = 0.929) and the subscales: AD (α = 0.898), H (α = 0.849), and CD (α = 0.899). For the purposes of the present study, only the AD and H subscales were considered. Scores above 90% indicate inattention deficit and/or hyperactivity-impulsivity. In the present study, a high-reliability index was obtained, α = 0.97.

Statistical Analysis
All of the collected data was stored in an SPSS 25.0 database. To analyze whether there were statistically significant differences between the groups, univariate and multivariate analyses of covariance (ANCOVA and MANCOVA (with covariates age and sex)) were performed. The independent variable was first the group (ADHD and control) and then the ADHD presentation groups (as established by the "Statistical Manual of Mental Disorders" DSM-5 [3]). The dependent variables were the scores in the physical activity variables (weekday and weekend counts, minutes of sedentary activity, and light, moderate, or vigorous physical activity on weekdays and weekends). To determine between-group differences, we used the multiple comparisons test with the Scheffé test. The sample followed a normal distribution according to Gravetter and Walnau [44], with the exception of vigorous PA). An indicator of the magnitude of the effect [45] was included: Cohen's (1988) delta was used as a criterion, according to which the effect is small when ηp 2 = 0.01 (d = 0.20), medium when ηp 2 = 0.059 (d = 0.50), and high when ηp 2 = 0.138 (d = 0.80).

Preliminary Results
The descriptive body composition and socio-educational variables of the 75 children participating in the study are shown in Table 2. No significant differences between groups were found by age (p = 0.231) or BMI (p = 0.161). The groups were not offset by sex since ADHD is more frequent in boys than in girls (χ2(1) = 4.587); (p = 0.032). Table 2 provides the main descriptive statistics for the sample in relation to the variables analyzed in this study.

Relationship between Counts, Minutes of Physical Activity Intensity, Sedentary Minutes and ADHD
The variables sex and age in the MANCOVA analysis were not statistically significant as covariates (p = 0.703), where nonparametric tests were used in vigorous PA (Table 3. The ADHD group had higher weekday counts than the non-ADHD group, with the difference being statistically significant (F(1.75) = 27.514; p < 0.000; d = 1.25). The ADHD group also spent more minutes during weekdays than the non-ADHD Similarly, during the weekends, the ADHD group had higher counts than the Non-ADHD group, with the differences being statistically significant (F(1.75) = 20.085; p < 0.000; d = 1.07). The ADHD group also spent more weekend minutes than the Non-ADHD  19). There were also statistically significant differences between the groups during weekends in sedentary time (F(1.75) = 6.939; p < 0.05; d = 0.63) with more minutes of sedentary time in the ADHD group than the Non-ADHD group. As with weekdays, medication had no statistically significant effect on weekends p = 0.667. Weekday counts and minutes are the totals over five days, and weekend counts and minutes are the totals over two days.
Differences between the sexes were found for weekend PA variables  Table 4 shows the ANOVA results of the interaction between weekend and weekday physical activity and the different ADHD presentations. Both during the week and at weekends, the hyperactive/impulsive and combined groups of students had statistically significantly higher counts than the control group. They also spent statistically significantly more time than the control group in light, moderate, and vigorous physical activity, both during the week and at weekends. The only statistically significant difference between the control group and the inattentive group was in the amounts of moderate physical activity, both during the week and at weekends; the inattentive group had higher values than the control group. There were statistically significant differences in the weekday count means of the control group compared to the Hyperactive/Impulsive and Combined groups (p = 0.002, p = 0.000), as well as in the weekend count means of the control group compared to the Hyperactive/Impulsive and Combined groups (p = 0.002, p = 0.009), the control group exhibited lower values in both cases.

Physical Activity Parameters and the Different Presentations of ADHD
There were also similar differences in the mean amount of light physical activity over a weekday between the control, Hyperactive/Impulsive, and Combined groups (p = 0.000, p = 0.000) and in the mean amounts of weekend light physical activity between the control, Hyperactive/Impulsive and Combined groups (p = 0.005, p = 0.000). There were also statistically significant differences in mean weekday minutes of moderate activity between the control group and all three ADHD presentations, with lower values for the control group compared to the Inattention, Hyperactive/Impulsive and Combined groups (p = 0.022, p = 0.000, p = 0.000), with a similar pattern for weekend minutes of moderate activity (Table 4).
A similar pattern was evident in the mean amounts of weekday vigorous activity between the control, Hyperactive/Impulsive, and Combined groups (p = 0.004, p = 0.002), which was repeated with the mean amounts of weekend vigorous activity between the control group and the Hyperactive/Impulsive and Combined presentations (p = 0.016, p = 0.002).
Similarly, statistically significant differences were found in mean weekend minutes of vigorous physical activity between the control group and the Hyperactive/Impulsive and Combined presentations (p = 0.016, p = 0.002), respectively, with higher values for the Hyperactive/Impulsive and Combined groups compared to the non-ADHD group.
There were no statistically significant differences in the means of sedentary time between the groups examined, either during the week or at weekends.

Physical Activity Intensity and Sedentary Behavior Changes by Age
Looking more deeply at the age range using Scheffé's post hoc test for the ADHD group in Table 5, we found a decrease in the number of minutes of moderate physical activity by age. This difference was statistically significant for both weekdays and weekends in the ADHD group (p < 0.05), with significantly more minutes of moderate PA in 6-7 year-olds than 10-11 year-olds (d = 5.37, MD = 412.33, p = 0.03) on weekdays.
We did find age-related differences in the non-ADHD group for minutes of light physical activity on weekdays (d = 4.53, MD = −328.18, p = 0.014) with an increase in minutes in these variables as a function of age. Children aged 8-9 had more sedentary minutes on weekdays than those aged 6-7 (d = 4.81, MD = 1807.31, p = 0.008), and children aged 10-11 had more sedentary minutes than those aged 6-7 (d = 6.63, MD = 2421.08, p = 0.000). On weekends, sedentary minutes also increased as a function of age, with 10-11 year-olds having more than 6-7 year-olds (d = −4.72, MD = −875.96, p = 0.010). There were no differences between ages in the other physical activity variables for the Non-ADHD group (p > 0.05).

Discussion
This study produced valuable findings about the amounts of moderate physical activity in children with and without ADHD, additionally demonstrating whether the participants did the amount of physical activity that the WHO recommends for children. Children with ADHD did more PA than non-ADHD children, but about 92% of the sample in both groups complied with the WHO guidelines for PA in children [12], with a mean of 60 daily minutes of moderate-to-vigorous intensity, mainly aerobic, physical activity during the week. The amount of sedentary time did not seem to correlate with ADHD or any of the presentations of ADHD. The findings of the current study are important because we identified associations that help us to understand the importance of ADHD presentations when it comes to higher levels of physical activity. We did not really know the relationship between ADHD and the different intensities of physical activity measured by accelerometry previously due to a lack of studies on the subject, although it is important to highlight authors such as Neudecker et al. [46] who suggest that PA may help to decrease the negative effects of ADHD contributing to long-term health benefits in children and adolescents with ADHD.
In this sample of children with and without ADHD, there was no difference in the objective measurements of physical activity between boys and girls, although differences were found within the non-ADHD group by sex. Our results are consistent with previous studies, which found sex differences in the amount and intensity of PA in non-ADHD children [47]. We did, however, see notable differences in the frequency and intensity of physical activity that could be attributed to the presence of ADHD. A previous study using a child population aged 6 to 11 years, with PA monitoring for 14 h a day, seven days a week, found higher levels of vigorous and moderate physical activity in children with ADHD compared to those without ADHD [23].
Sedentary time has been shown to be more important than moderate or high activity in predicting ADHD symptoms as a diagnostic measure of activity [48]. However, we saw no significant differences in sedentary patterns. Boonstra et al. [49] reported that sedentary time and sedentary prevalence did not differ between children and adolescents with and without ADHD, whereas higher physical activity patterns have been found in children with ADHD compared to children without ADHD [50]. Various studies have provided supporting evidence for this line of thinking, and this variable could be used as a predictor of ADHD. In terms of the amounts of physical activity in relation to the different presentations of ADHD, we found higher activity in the hyperactive/impulsive and combined presentations compared to the control group, in line with at least one other study [48]. In contrast, other studies have not found any differences in physical activity patterns based on ADHD presentations [51].
Other studies using objective measurements to analyze movement, via infrared [33,52,53] or Kinect, with the ADHD Movements program [4,11] also reported greater physical activity in children with ADHD than children without. Sempere-Tortosa et al. [4] showed a higher concentration of activity in the afternoon compared to the morning, when no differences between groups were observed. Our study showed higher activity during weekends than during weekdays, something that could be explained by previous research, given that children are at school during weekday mornings and the highest motor activity occurs in the afternoons. On weekdays, students are at school in the morning, mostly sitting down, except for physical education. This reason, as well as the fact that in Spain, some after-school sports competitions are on weekends, may explain why there is more physical activity during weekends.
In line with Volkmar's [25] results, children with ADHD had a generally higher level of PA than the control group and tended to spend more time in moderate PA on weekdays and at weekends. Our data suggested that children with ADHD were more hyperactive at weekends than on weekdays.
Previous research has reported similar mean effect sizes in children and in adolescents [33], although some studies have reported differences by age [54]. In line with Kofler et al. [54], in our study, we found differences in the ADHD group in moderate physical activity, with a decrease in moderate physical activity as children got older. On the other hand, we found no differences in this group in sedentary time by age, which may be due to the motor load that is partly explained by the disorder itself, along with other variables that were not studied. In line with Kontostoli et al. [55], which showed how sedentary time increases by 21 min each year starting in childhood, our study indicated age-related differences in sedentary time and light physical activity in children without ADHD. Our study is consistent with the recent study by Santiago-Rodriguez et al. [34], which reported similar values for sedentary physical activity, although our sample exhibited slightly lower values in moderate and vigorous physical activity than African children of the same age with ADHD. Regarding age, similarly to Murillo et al. [33], the levels of moderate and vigorous physical activity decreased as children aged. There are no cut-off points or algorithms for relating children's physical activity to ADHD or ADHD presentations. There is evidence that the amount and intensity of physical activity differ between children with and without ADHD [56], using the limited research to date with accelerometry as an identifier of ADHD through activity [8].
This study has some limitations. It did not consider other comorbid disorders, nor were the use of stimulant treatments or their effects on ADHD subjects' locomotor activity examined. The main limitation of our study was the sample size of the subgroups for the different presentations of ADHD. A lack of previous studies providing information on amounts of different intensities of activity or activity related to ADHD presentations on both weekdays and weekends makes it difficult to compare the patterns and motor behavior based on the intensity of physical activity of children with ADHD versus those without ADHD. It would be interesting to record the intensity of physical education classes, as well as to record whether children engage in active transportation. Other limitations of the study are the size and accessibility of the sample-our sample size was small, and it was a pilot study-as well as not examining the influence of variables such as ethnicity/race or socioeconomic status. ADHD is a disorder with a higher prevalence in boys than in girls, which may imply a limitation in the homogeneity of the sample. One of the strengths of this study is the use of GT3X accelerometry to measure physical activity and sedentary time patterns in children with and without ADHD. Future studies might find it interesting to determine the diagnostic utility of accelerometry in patients with this disorder, looking to see whether there is variability when analyzing other ethnicities and lifestyles and the interaction of age and sex on motor symptoms based on a larger sample.

Conclusions
In conclusion, greater physical activity was associated with ADHD and ADHD presentations in children with ADHD compared to Non-ADHD children. We found that 92% of our 6-12-year-old population did sufficient levels of physical activity, although children with ADHD had higher levels of MVPA. Children with ADHD did not present differences in the amount or intensity of physical activity by sex, while in children without ADHD, girls demonstrated more activity and a greater intensity of physical activity than boys. We reported an increase in sedentary time by age in Non-ADHD children and a decrease in moderate-intensity physical activity time in ADHD children by age. There is a need for additional work to investigate these topics more fully. A metanalysis might be helpful in terms of conceptualizing how/why increased gross motor movements may manifest in children with ADHD [57].