An investigation of gait in children with Attention Deficit HyperactivityDisorder: A case controlled study

Nicole Papadopoulos a,d,n, Jennifer L. McGinley b,c, John L. Bradshaw a, Nicole J. Rinehart a,d

a Centre for Developmental Psychiatry & Psychology, School of Psychology and Psychiatry, Monash University, Australiab The University of Melbourne, Australiac Clinical Research Centre for Movement and Gait Disorders, Southern Health, Australiad Deakin Child Study Centre, School of Psychology, Deakin University, Australia

a r t i c l e i n f o

Article history:Received 28 February 2013Received in revised form17 April 2014Accepted 26 April 2014

Keywords:ADHDAutismGaitSocial-communication disturbanceInattention

a b s t r a c t

This study aimed to compare the gait of children with ADHD – Combined Type (ADHD-CT) to typicallydeveloping (TD) children. Children with ADHD-CT (n¼14; mean age 10 years 4 months) and a TD group(n¼13; mean age 10 years 9 months) walked at self-selected slow, preferred and fast speed on anelectronic walkway system. Participants completed a total of 15 walking trials; 5 trials per walkingcondition. Groups were matched on age, intellectual functioning, height and weight. In the preferredwalking condition, there was no difference in spatio-temporal gait variables between the ADHD-CT andTD control groups. At self-selected fast speed, children with ADHD-CT were faster and walked with ahigher cadence. The subtle alterations in gait pattern that may reflect a timing deficit is consistent withprevious ADHD motor studies. In addition, this study extends previous studies in characterising theunique gait profile of non-medicated children with ADHD-CT where a diagnosis of autism spectrumdisorder has been ruled out.

& 2014 Published by Elsevier Ireland Ltd.

1. Introduction

Attention Deficit Hyperactivity Disorder (ADHD) is the mostprevalent childhood onset psychiatric disorder characterised byclinically significant symptoms of inattention, hyperactivity andimpulsiveness that are present before 7 years of age (AmericanPsychological Association, 2000). The Diagnostic and StatisticalManual 4th Edition Revised (DSM-IV-TR) defines three subtypes ofADHD; ADHD – Predominantly Inattentive subtype (ADHD-PI),ADHD – Predominantly Hyperactive-Impulsive (ADHD-HI) andADHD – Combined Type (ADHD-CT) (APA, 2000). In addition tocore clinical symptoms, motor disturbance is common in ADHD-CT(Harvey and Reid, 1997; Piek et al.,1999; Reiersen et al., 2008).Indeed, up to 50% of children with ADHD meet diagnostic criteriafor Developmental Coordination Disorder (DCD), and individualswith a co-morbid diagnosis of ADHD and DCD experience greatermotor difficulties (Pitcher et al., 2003). However, the specificpattern of motor disturbance that characterises ADHD is onlyemerging in the literature, far more is known about the cognitiveprofile of the disorder.

Children with ADHD present with an uneven executive func-tion profile, characterised by difficulties on tasks measuringinhibition and sustained attention (Barkley, 2001; Johnson et al.,2007), which occur in the context of relatively intact planningability and cognitive flexibility (Ozonoff and Jensen, 1999). Giventhe close overlap in neural networks that underpin cognitive andmotor abilities (Diamond, 2000), it is not surprising that the ADHDmotor profile is also characterised by specific areas of motorimpairment which occur in the context of areas of preservedmotor function. For example, the nature of motor problemscommonly reported for individuals with ADHD range from poorermotor performance on standardized measures of fundamentalmovement skills and fitness (Harvey and Reid, 2003) to mildbalance problems on posturography tasks (Buderath et al., 2009),and it has been proposed that these motor problems may beassociated with core ADHD symptoms.

Pitcher et al., (2003) reported that a greater proportion ofindividuals with ADHD-PI subtype experienced motor difficulties(58% of n¼50) compared to ADHD-HI (49% of n¼16) and ADHD-CT (47% of n¼38) subtypes. More specifically, Piek et al. (1999)found that children with ADHD-PI (N¼16) had greater fine motorskill difficulties on the manual dexterity subscale of the MovementAssessment Battery for Children (MABC) compared to individualswith ADHD-CT (N¼16), whereas children with ADHD-CT experi-enced greater difficulties with gross motor tasks (balance subscaleof the MABC) compared to the ADHD-PI group. Furthermore a

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/psychres

Psychiatry Research

http://dx.doi.org/10.1016/j.psychres.2014.04.0370165-1781/& 2014 Published by Elsevier Ireland Ltd.

n Corresponding author at: School of Psychology, Deakin University BurwoodCampus, Melbourne Burwood Campus, 221 Burwood Highway, Burwood, VIC 3125,Australia. Tel.: þ61 3 9244 5295.

E-mail address: Nicole.Papadopoulos@deakin.edu.au (N. Papadopoulos).

Please cite this article as: Papadopoulos, N., et al., An investigation of gait in children with Attention Deficit Hyperactivity Disorder: Acase controlled study. Psychiatry Research (2014), http://dx.doi.org/10.1016/j.psychres.2014.04.037i

Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎

study conducted by Reiersen et al. (2008) reported an associationbetween inattentive symptoms and parent reported motor pro-blems measured on the Child Behaviour Checklist. Despite this,there are very few studies that have investigated the associationbetween ADHD symptomatology (inattention, hyperactivity andimpulsivity) and motor performance in ADHD.

In addition to the association between inattentive symptoms andmotor features, autistic features have also been shown to be associatedwith greater motor impairment in childrenwith ADHD (Reiersen et al.,2008). Indeed, this pattern compliments the association foundbetween the severity of core social-communicative features and motorimpairment in Autism Spectrum Disorders (ASD) (Qiu et al., 2010;Papadopoulos et al., 2012b). It has been suggested that overlap insocial-communicative and motor difficulties in ASD may reflectdisruption to common underlying brain pathways thought to developin parallel (Qiu et al., 2010). It is therefore important for future ADHDmotor research to consider co-occurring developmental factors suchas DCD and ASD (Reiersen et al., 2008).

There is some overlap in the literature reporting disruption tomotor brain regions, specifically fronto-striatal-cerebellar braincircuitry in ASD and ADHD (Bradshaw, 2001). Upper limb motorstudies have provided insight in understanding fronto-striatal andcerebellar type motor impairment in these disorders using a rangeof tasks (Rinehart et al., 2001, 2006a; Papadopoulos et al., 2012a,2012b). Numerous gait studies have also considered basal gangliaand cerebellar type motor anomalies in children with ASD(Ambrosini et al., 1998; Rinehart et al., 2006b, 2006c; Calhounet al., 2011; Nobile et al., 2011; Nayate et al., 2012). Quantitativegait analysis techniques report that the gait pattern of childrenwith autism at a preferred walking pace is characterised by a widebase of support (Nayate et al., 2012; Nobile et al., 2011), variation(both increases and decreases) in stride length and stride timevariability (Rinehart et al., 2006a, 2006b; Nobile et al., 2011) aswell as increased cadence (Calhoun et al., 2011). Investigationsusing qualitative gait methods in ASD have also reported a widebase of support (Ambrosini et al., 1998) and reduced smoothnessof gait and postural abnormalities of the head and trunk (Rinehartet al., 2006c).

Unlike children with ASD who may show characteristic gaitdisturbances from an early age (Teitelbaum et al., 2004), therehave been relatively fewer investigations of gait in children withADHD. To our knowledge only two studies have measured gaitvariables in children with ADHD (Leitner et al., 2007; Buderathet al., 2009) using instrumented gait analysis techniques. In astudy by Buderath et al. (2009) the gait and postural abnormalitiesof children with ADHD-CT (on medication; n¼10) were comparedto a cerebellar lesion and typically developing (TD) group. Nodifferences between the ADHD group and TD controls were foundin stride length, cadence or stride timing on a treadmill walkingtask, however ADHD participants scored significantly lower thanTD controls when walking backwards on a beam. In a pacedstationary stepping task with an external cue (metronome), theADHD group was slightly slower than TD controls in the fast pacedcondition. This pattern was consistent with the cerebellar lesiongroup, providing preliminary support that ADHD may be asso-ciated with cerebellar dysfunction. Another study by Leitner et al.(2007) investigated the effect of dual tasking on gait function inchildren with ADHD without significant motor impairment (9–16years; n¼16), off methylphenidate. They reported no differencebetween the ADHD and TD groups in the preferred walkingcondition (single task condition; off medication), although a trendto higher stride time variability was noted in the ADHD group(p¼0.09). In the dual task condition, stride time variabilitysignificantly reduced in the ADHD group compared to the baseline(single task) condition. Stride time variability also reduced whenADHD individuals were medicated compared to baseline. It was

concluded that methylphenidate had a positive effect on gait,reducing the mildly increased stride time variability in theADHD group.

The aim of the current study is to characterise the gait ofchildren with ADHD using the same protocol used in Nayate et al.'s(2012) study of gait in children diagnosed with autism andAsperger's disorder to further inform our understanding of dis-rupted underlying neural circuitry in ADHD. It was hypothesisedthat there would be no difference in gait variables (speed, stridelength, cadence, base of support and double support time)between the ADHD group and the TD group in the preferredbaseline walking condition. Further based on the gait studiesconducted by Leitner et al. (2007) and Buderath et al. (2009) itwas hypothesised that the ADHD group may display subtle timinganomalies. In addition, based on previous research in ASD thatindicates a significant relationship between motor disturbance andsocial-communicative disturbance (Qiu et al., 2010; Papadopouloset al., 2012b) we predicted a positive association between gaitvariables characteristic of ASD such as a wider base of support andincreased stride length variability and social-communicationsymptoms in children with ADHD-CT. Lastly, based on the findingthat children with ADHD-inattentive type may experience moremotor difficulties than other subtypes (Pitcher et al., 2003), the asso-ciation between inattentive symptoms measured on the Conner'sRating Scale and spatio-temporal gait variables was explored.

2. Method

2.1. Participants

Informed consent was obtained from parents/guardians of all participants, inaccordance with the Declaration of Helsinki. Ethical approval was obtained fromthe Human Research Ethics committee at Southern Health and Monash University,Melbourne Australia.

Fourteen boys diagnosed with ADHD – Combined type (ADHD-CT) agedbetween 7 and 13 years were recruited from Private Paediatricians in Melbourne.The paediatricians specialised in ADHD with 10–20 years of clinical experience inthe field, and undertook further assessment and diagnosis of ADHD-CT. Thesechildren fulfilled DSM-IV-TR (APA, 2000) criteria for ADHD – Combined type(ADHD-CT). Diagnosis of ADHD-CT was further confirmed by a doctoral-leveltrained graduate student (author N.P.) under the supervision of a clinical psychol-ogist (author N.R.) using the Conners Rating Scale (Conners, 2001), parent inter-views, direct child observations and information from teachers and therapists.Paediatricians also confirmed that children with a diagnosis of ADHD-CT did nothave a co-morbid diagnosis of autistic disorder, Asperger's disorder or pervasivedevelopmental disorder not otherwise specified based on their assessment andmedical records. Participants who entered the study were further screened for apossible diagnosis of ASD using the Autism Diagnostic Observation Scale (ADOS) bya qualified researcher. No participants were excluded based on elevated ADOSscores (ASD cut off score¼7).

Additional exclusion criteria included co-morbid medical (e.g., tuberoussclerosis), hearing or visual impairments, or genetic (e.g., Fragile X syndrome)disorders. The majority of participants in the study (12/14) was on stimulantmedication such as methylphenidate (Ritalin). Participants on medication discon-tinued medication at least 24 h before testing commenced.

A reference sample of 13 typically developing (TD) boys aged between 7 and 14years were recruited from local schools and the community. The TD children hadno prior history of psychological, neurological or psychiatric diagnosis. This wasconfirmed by researchers via interview with children's parents, from whominformation on developmental, medical and psychiatric history was obtained (i.e.whether children had any previous diagnoses or had received any intervention). Allparticipants in the TD group were also screened for ADHD-CT symptoms using theConner's Rating Scale (Conners, 2001) and for autistic symptoms using the SocialResponsiveness Scale (Constantino and Gruber, 2005). One TD child was excludedfor having Social Responsiveness Scale (SRS) scores indicative of clinically elevated(t score476) social responsiveness difficulties.

The intellectual functioning of TD boys was assessed using the WechslerAbbreviated Scale of Intelligence (WASI) (Wechsler, 1999) or the Wechsler Intelli-gence Scale for children 4th edition (WISC-IV) (Wechsler, 2005), and the intellec-tual functioning of children with ADHD-CT was assessed using the WISC-IV. TheWASI and WISC-IV are highly compatible with.87 correlation between their fullscale IQ scores (Wechsler, 1999). Motor proficiency was assessed and childrenwere screened for DCD using the Movement Assessment Battery for Children

N. Papadopoulos et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

Please cite this article as: Papadopoulos, N., et al., An investigation of gait in children with Attention Deficit Hyperactivity Disorder: Acase controlled study. Psychiatry Research (2014), http://dx.doi.org/10.1016/j.psychres.2014.04.037i

(2nd edition) with a cut-off of o15th percentile (Henderson et al., 2007). Noparticipants were excluded for experiencing significant motor impairment.

Independent measures t-tests were conducted to investigate the effect of age,intelligence, height and weight between groups. There was no significant differencein age (t(25)¼0.056, p¼0.956), intellectual functioning; (Full Scale IQ (FSIQ) t(25)¼1.36, p¼0.188, Perceptual Reasoning Index (PRI) t(25)¼�0.829, p¼0.415, VerbalComprehension Index (VCI) t(25)¼�1.717, p¼0.098), height t(25)¼�0.482,p¼0.634 and weight t(25)¼�0.527, p¼0.603 between groups (see Table 1 forparticipant characteristics).

2.2. Measures

2.2.1. Conners Rating ScaleThe inattention and hyperactivity-impulsivity subscales from the parent

reported Conners Rating Scale (CRS – long form) (Conners, 2001) were utilised inthis study. The CRS is a commonly used standardized screening instrument thatmeasures DSM-IV-TR ADHD symptomatology. The parent long form consists of 80items, measuring indices of oppositional behaviour problems, cognitive/inattentivebehaviour, hyperactivity, anxious-shy behaviour, perfectionism, social problems,psychosomatic problems for children aged 3–17 years. Test–retest reliability rangesfrom 0.47 to 0.85 and internal consistency ranges from 0.73 to 0.91 (Conners, 2001).

2.2.2. Developmental Behavioural ChecklistCommunication disturbance was assessed using the Developmental Behavioural

Checklist –Parent Report (DBC-P) (Einfeld and Tonge, 2002). The DBC is a 96 itemquantitative measure of behavioural and emotional disturbance in young peopleaged 4–18 years, is comprised of five subscales, and has demonstrated good validityand reliability (Einfeld and Tonge, 1995). The behavioural subscales include dis-ruptive/antisocial behaviour, self absorbed, communication disturbance (a measureof communication deviance), anxiety and social relating. Items are scored on a Likertscale ranging from 0—“not true as far as you know” to 3–“often true or very true”.The DBC generates a rating of overall behavioural disturbance, as well as a rating onthe five behavioural subscales mentioned above.

2.3. Procedure

Participants attended a single 60 min session at the Clinical Research Centre forMovement Disorders & Gait, Kingston Centre Cheltenham. Gait was measuredusing the GAITRites system (CIR system Inc, Clifton, N.J): an electronic walkway830 cm�89 cm, with pressure sensors spaced 1.27 cm apart embedded in ahorizontal grid. A 2 m non-recordable zone was added on each end of the walkwayto minimise the effects of acceleration and deceleration. After each walk, data wereanalysed using GAITRite software.

Anthropometric measures (height, weight and leg length) were taken beforethe commencement of walking conditions. All participants completed five trials ofthe three walking conditions in the following order: (1) Preferred walking-participants were instructed to walk at their preferred speed, (2) fast walking-participants were instructed to walk at a faster pace than their preferred speed, butnot to run, and (3) slow walking- participants were instructed to walk at a paceslower than their preferred speed. Participants were given one demonstration trialand a practice trial prior to the commencement of each different walking conditionto make sure that they understood the task.

2.4. Data analysis

For each walk the GAITRite software generated step-to-step values for a rangeof gait variables. Spatial-temporal parameters (average of left and right) wereextracted which included: speed (cm walked per second), cadence (steps taken perminute), stride length (combined left and right footfalls in full gait cycle), doublesupport (percentage of time both feet are grounded in a complete gait cycle), heel-to-heel base of support (perpendicular distance from the heel point in one footfall

to the line of progression of the opposite foot). A Y-axis range (maximum distancetravelled laterally in the horizontal plane of the mat during a walk) was alsocalculated. Mean values for each walk were calculated and averaged across all trialsin that condition, to produce a global mean for each gait variable in each condition.Step to step variability was calculated as the mean standard deviation from eachtrial for speed, cadence, stride length, double support and base of support.

Data were analysed using independent samples t-tests to examine groupdifferences. The data for each of these variables were tested for normal distribu-tions using box plots and the Kolmogorov–Smirnov statistical test. Assumptions ofnormality were met for all variables except cadence in the fast walking condition.A Mann–Whitney U test was therefore conducted to investigate group differencesfor this condition. To control for anthropometric differences, data were normalisedfor each walk in accordance with the methods described in Hof (1996). Effect sizesfor gait variables were calculated using Cohen's d. Spearman's correlations wereconducted to investigate any significant relationships between communicationdisturbance measured on the DBC and the five subscale scores as well as inattentivesymptoms measured on the CRS and the five gait variables.

3. Results

There was no significant difference between the ADHD-CTgroup and the TD group in the baseline preferred walking or slowcondition for speed, cadence, stride length, double support or baseof support. However, large and moderate effect sizes werereported for cadence (0.84) and y-axis range (0.65) respectively.At self selected fast speed, the ADHD-CT group was faster (t(25)¼�2.31, p¼0.030) than the TD group, due to increased cadence(U¼33, p¼0.005) with large effect sizes (0.92) reported (seeTable 2 for all effect sizes). Notably, in all conditions, children withADHD walked with comparable stride length but higher cadencethan the TD group, with the increased cadence in the ADHD groupalso nearing significance in the preferred (p¼0.052) and slowconditions (p¼0.066). Normalisation of the data according to Hof(1996) did not alter any of the findings. Variability did not differbetween groups in any of the walking conditions, although astrong trend for increased variability in cadence was evident in thepreferred (p¼0.07) and fast condition (p¼0.058) in the ADHDgroup. Moderate effect sizes were reported for cadence in thepreferred walking condition and moderate effect sizes werereported for variability in most spatio-temporal gait variables(speed, cadence, stride length, and double support) in the fastwalking condition (see Table 3 for all effect sizes).

Table 1Demographic information for ADHD-CT and typically developing (TD) controlgroups.

ADHD-CT (n¼14) Typically developingcontrols (n¼13)

Age (months) 129.8 (24.5) 129.2 (33.7)Height (cms) 147.6 (10.4) 150.7 (21.4)Leg length (cms) 76.1 (6.4) 77.9 (12.2)Weight (kg) 43.0 (14.8) 46.9 (23.0)FSIQ 97.5 (12.5) 105.4 (17.5)VCI 94.8 (13.4) 105.0 (17.4)PRI 102.5 (14.5) 107.6 (17.5)

Table 2Spatio-temporal gait parameters.

Condition ADHD-CT(n¼14)

TD controls(n¼13)

P-value Effectsize

PreferredSpeed (cm/s) 130.7 (13.9) 126.9 (17.3) 0.531 0.254Cadence (steps/min) 122.3 (7.4) 115.0 (10.4) 0.052 0.843Stride Length (cm) 128.4 (10.3) 133.4 (20.8) 0.428 �0.322Double Support % gait cycle 23.5 (3.1) 24.4 (3.0) 0.447 �0.309Base of support (cm) 8.7 (2.1) 8.6 (2.5) 0.926 0.038Y-axis range (cm) 23.3 (3.4) 21.6 (1.6) 0.114 0.656

SlowSpeed (cm/s) 95.8 (19.5) 86.5 (24.8) 0.288 0.435Cadence (steps/min) 104.0 (11.2) 93.95 (15.81) 0.066 0.769Stride Length (cm) 109.6 (12.7) 109.4 (23.1) 0.985 0.007Double support % gait cycle 26.7 (3.3) 29.1 (5.3) 0.175 �0.559Base support (cm) 8.8 (2.2) 8.3 (2.4) 0.619 0.202Y-axis range (cm) 23.7 (3.9) 22.1 (3.3) 0.283 0.440

FastSpeed (cm/s) 180.6 (18.6) 157.1 (33.0) 0.030n 0.923Cadence (steps/min) 147.0 (12.1) 129.9 (19.7) 0.005n 1.094Stride length (cm) 147.7 (12.6) 146.0 (26.7) 0.830 0.089Double support % gait cycle 19.5 (2.8) 21.9 (3.4) 0.052 �0.816Base support (cm) 9.5 (2.1) 8.2 (2.2) 0.140 0.610Y-axis range (cm) 22.9 (2.6) 21.5 (2.6) 0.165 0.572

n Po0.05.

N. Papadopoulos et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Please cite this article as: Papadopoulos, N., et al., An investigation of gait in children with Attention Deficit Hyperactivity Disorder: Acase controlled study. Psychiatry Research (2014), http://dx.doi.org/10.1016/j.psychres.2014.04.037i

3.1. Association between social-communication and gait variables

A series of Spearman's correlations were conducted to investi-gate the relationship between gait variables in all walking con-ditons and emotional behavioural disturbance (total DBC score) aswell as the DBC subscales for the ADHD-CT group. Only commu-nication disturbance (a measure of social communicationdeviance) was significantly positively correlated with base ofsupport (rs¼0.56, p¼0.037). There were no other significantcorrelations between any of the gait variables and DBC subscales.

3.2. Association between ADHD symptoms and gait variables

The severity of Inattentive symptoms on the CRS was signifi-cantly correlated with speed (rS¼0.55, p¼0.04) and cadence(rS¼0.60, p¼0.02) in the fast walking condition, although notin the slow (speed; rS¼�0.06, p¼0.753, cadence; rS¼�0.07,p¼0.727) or preferred (speed; rS¼0.05, p¼0.808, cadencerS¼0.19, p¼0.353) walking conditions. There were no other sig-nificant correlations between inattentive symptoms and any othergait variables. There were also no significant correlations betweenthe hyperactivity-impulsivity subscales of the CRS and any of thegait variables.

4. Discussion

This study aimed to investigate the gait profile of a sample ofchildren diagnosed with ADHD-CT who had been screened forDCD and ASD. The finding of no significant difference in spatio-temporal gait variables between the ADHD and TD groups in thepreferred walking condition is largely consistent with previousgait studies (Leitner et al., 2007; Buderath et al., 2009). In addition,differences in timing regulation with reported large effect sizes forthe preferred, slow and fast walking conditions support previousmotor studies (see Hart et al., 2012) that consistently report timingdeficits in children (Luman et al., 2009) and adults (Valera et al.,2010) with ADHD. Difficulties in timing regulation have long beenattributed to cerebellar dysfunction (Ivry and Keele, 1989); a brainregion consistently implicated in ADHD (Berquin et al., 1998).However, the current study extends previous ADHD motor

research in defining the gait profile characteristic of children withADHD without clinically significant symptoms of ASD and DCD.

It has previously been documented that up to 30% of indivi-duals with ADHD experience clinically significant autistic symp-toms (Reiersen et al., 2007). Further, in our previous study ofmotor functioning that comprised a sub-set of participants fromthis study (Papadopoulos et al., 2012c), social responsivenessmeasured on the Social Responsiveness Scale (SRS) was found tobe related to error variability on a repetitive upper limb Fitts'aiming task. The current study found a moderate positive correla-tion between increased levels of communication disturbancemeasured on the DBC and a wider base of support in ADHD.

Inattentive symptoms were also found to be associated withincreased speed and cadence in the fast walking condition. Thisfinding partially supports previous parent- reported motor studies(Reiersen et al., 2008) as well as motor studies using standardizedassessment batteries (Piek et al., 1999;Pitcher et al., 2003). The linkbetween inattention and motor problems is also apparent inADHD studies that report an improvement in motor performancewhen children are on medication (e.g., methylphenidate) (Leitneret al., 2007; Brossard-Racine et al., 2012). Despite this, it iscurrently unclear whether a particular ADHD subtype (e.g.,ADHD-PI) is associated with a different gait profile. More broadly,larger scale gait studies including well-defined groups of childrenwith neurodevelopmental disorders will be necessary to see if gaitcharacteristic differ according to diagnosis (e.g., ADHD, ASD), anddegree of comorbidity.

The current gait study is a cross sectional study, and although itallows an investigation of motor performance at a time point itdoes not provide a picture of motor performance across thedevelopmental trajectory. Future research should therefore endea-vour to investigate motor performance in ADHD longitudinally.Secondly, the small sample size, as well as multiple statistical testslimits the power of the current study and our ability to generaliseour findings. Inadequate power may have limited our ability todetect significant differences in variability of some gait variables.Specifically, moderate effect sizes were reported for speed,cadence, stride length and double support time in the fast walkingcondition which may provide some evidence of a more variablegait pattern at fast walking speed for the ADHD-CT group. Largeeffect sizes for cadence across all walking conditions (preferred,slow and fast) may also indicate that cadence may be affected in allwalking conditions and requires further exploration. Futureresearch is necessary to investigate the gait profile of clearlydefined groups of children diagnosed with autism, AD and ADHDat the same time point. Replication of the current study usinglarger sample sizes to increase power is also necessary. Lastly, theuse of blinded investigators conducting the gait trials and sensitivegait paradigms such as 3D motion analysis may allow a morecomprehensive unbiased investigation of whole body movementacross time and space and enable greater detection of gait anomalies.

These preliminary findings indicate that the gait profile of non-medicated children with ADHD without ASD and DCD reflects adeficit in timing regulation when children are asked to walk at afast pace, consistent with cerebellar disruption. Future researchshould investigate whether the unique profile of gait disturbancecharacteristic of individuals with ADHD may be useful in objec-tively distinguishing between children with autism, Asperger'sdisorder, and other neurodevelopmental conditions.

Appendix A. Supplementary materials

Supplementary data associated with this article can be found in theonline version at http://dx.doi.org/10.1016/j.psychres.2014.04.037.

Table 3Variability (S.D.) for gait parameters.

Condition ADHD-CT(n¼14)

TD controls(n¼13)

P-value Effectsize

PreferredSpeed (cm/s) 7.5 (5.2) 5.7 (3.3) 0.318 0.408Cadence (steps/min) 4.5 (4.3) 1.9 (1.61) 0.070 0.701Stride Length (cm) 3.9 (1.8) 4.1 (2.2) 0.685 -0.164Double Support % gait cycle 1.0 (0.4) 1.0 (0.3) 0.985 -0.007Base of support (cm) 1.1 (0.8) 1.1 (0.3) 0.735 -0.137

SlowSpeed (cm/s) 8.2 (4.9) 6.6 (3.6) 0.336 0.393Cadence (steps/min) 5.2 (4.2) 3.3 (2.3) 0.148 0.484Stride Length (cm) 6.2 (3.2) 5.8 (2.5) 0.734 0.138Double Support % gait cycle 1.6 (1.0) 1.6 (1.1) 0.881 0.060Base support (cm) 1.4 (0.5) 1.2 (0.6) 0.290 0.433

FastSpeed (cm/s) 14.4 (15.2) 7.1 (5.3) 0.106 0.665Cadence (steps/min) 9.4 (8.2) 4.1 (5.0) 0.058 0.759Stride Length (cm) 5.7 (4.4) 3.4 (1.1) 0.088 0.708Double Support % gait cycle 1.7 (1.3) 1.1 (0.5) 0.089 0.702Base support (cm) 1.4 (0.7) 1.3 (0.7) 0.591 0.217

N. Papadopoulos et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

Please cite this article as: Papadopoulos, N., et al., An investigation of gait in children with Attention Deficit Hyperactivity Disorder: Acase controlled study. Psychiatry Research (2014), http://dx.doi.org/10.1016/j.psychres.2014.04.037i

References

Ambrosini, D., Courchesne, E., Kaufman, K., 1998. Motion analysis of patients withinfantile autism. Gait and Posture 7, 188.

American Psychiatric Association, 2000. Diagnostic and Statistical Manual ofMental Disorders IV-TR, 4th ed. American Psychiatric Association Press,Washington, DC.

Barkley, R.A., 2001. The executive functions and self regulation: an Evolutionaryneuropsychological perspective. Neuropsychological Review 11, 1–29.

Bradshaw, J., 2001. Developmental Disorders of the Frontostriatal System: Neu-ropsychological, Neuropsychiatric, and Evolutionary Perspectives. PsychologyPress, Philadelphia.

Berquin, P.C., Giedd, J.N., Jacobsen, L.K., Hamburger, S.D., Krain, A.L., Rapoport, J.L.,Castellanos, F.X., 1998. Cerebellum in attention deficit hyperactivity disorder: amorphometric MRI study. Neurology 50 (4), 1087–1093.

Brossard-Racine, M., Shevell, M., Snider, L., Belanger, S.A., Majnemer, A., 2012.Motor skills in children newly diagnosed with attention deficit hyperactivitydisorder prior to and following treatment with stimulant medication. Researchin Developmental Disabilities 33, 2080–2087.

Buderath, P., Gartner, K., Frings, M., Christiansen, H., Schoch, B., Konczak, J.,Timmann, D., 2009. Postural and gait performance in children with attention-deficit/hyperactivity disorder. Gait and Posture 29 (2), 249–254.

Calhoun, M., Longworth, M., Chester, V.L., 2011. Gait patterns in children withautism. Clinical Biomechanics 26, 200–206.

Conners, C., 2001. Conners Rating Scales – Revised (CRS-R): Technical Manual.Multi-Health Systems Inc., North Tonawanda, N.Y.

Constantino, J., Gruber, C.P., 2005. Social Responsive Scale (SRS) Manual. WesternPsychological Service, Los Angeles.

Diamond, A., 2000. Close interrelation of motor development and of the cerebellumand prefrontal cortex. Child Development 71, 44–56.

Einfeld, S.L., Tonge, B., 1995. The Developmental Behaviour Checklist: the develop-ment and validation of an instrument to assess behavioral and emotionaldisturbance in children and adolescents with mental retardation. Journal ofAutism and Developmental Disorders 25 (2), 81–104.

Einfeld, S.L., Tonge, B.J., 2002. Manual for the Developmental Behaviour Checklist:Primary Carer Version (DBC-P) and Teacher Version (DBC-T), 2nd ed. MonashUniversity Centre for Developmental Psychiatry and Psychology, Clayton,Melbourne.

Hart, H., Radua, J., Mataix-Cols, D., Rubia, K., 2012. Meta-analysis of fMRI studies oftiming in attention deficit hyperactivity disorder (ADHD). Neuroscience andBiobehavioural Reviews 36 (10), 2248–2256.

Harvey, W.J., Reid, G., 1997. Motor performance of children with attention deficithyperactivity disorder: a preliminary investigation. Adapted Physical ActivityQuarterly 14, 190–202.

Harvey, W.J., Reid, G., 2003. Attention-Deficit/Hyperactivity Disorder: a review ofresearch on movement skill performance and physical fitness. Adapted PhysicalActivity Quarterly 20 (1), 1–25.

Henderson, S.E., Sugden, D.A., Barnett, A.C., 2007. Movement Assessment Batteryfor children, 2nd edition Pearson Assessment, London.

Hof, A.L., 1996. Scaling gait data to body size. Gait and Posture 4, 222–223.Ivry, R.B., Keele, S.W., 1989. Timing functions of the cerebellum. Journal of Cognitive

Neuroscience 1 (2), 136–152.Johnson, K.A., Robertson, I.H., Kelly, S.P., Silk, T.J., Barry, E., Daibhis, A., Watchorn, A.,

Keavey, M., Fitzgerald, M., Gallagher, L., Gill, M., Bellgrove, M.A., 2007.Dissociation in performance of children with ADHD and high-functioningautism on a task of sustained attention. Neuropsychologia 45, 2234–2245.

Leitner, Y., Barak, R., Giladi, N., Peretz, C., Eshel, R., Gruendlinger, L., Hausdorff, J.M.,2007. Gait in attention deficit hyperactivity disorder: effects of methylpheni-date and dual tasking. Journal of Neurology 254, 1330–1338.

Luman, M., van Noesel, S.J.P., Papanikolau, A., Van Oostenbruggen-Scheffer, J.,Veugelers, D., Sergeant, J.A, Oosterlaan, J., 2009. Inhibition, reinforcementsensitivity and temporal information processing in ADHD and ADHDþOCD:evidence of a separate entity? Journal of Abnormal Child Psychology 37,1123–1135.

Nayate, A., Tonge, B.J., Bradshaw, J., McGinley, J., Iansek, R., Rinehart, N., 2012.Differentiation of high-functioning autism and Asperger's disorder based on

neuromotor function. Journal of Autism & Developmental Disorders 42 (5),707–717.

Nobile, M., Perego, P., Piccinini, L., Mani, E., Rossi, A., Bellina, M., Molteni, M., 2011.Further evidence of complex motor dysfunction in drug naïve children withautism using automatic motion analysis of gait. Autism 15 (3), 263–283.

Ozonoff, S., Jensen, J., 1999. Brief report: specific executive function profiles in threeneurodevelopmental disorders. Journal of Autism & Developmental Disorders29, 171–177.

Papadopoulos, N., McGinley, J.L., Tonge, B.J., Bradshaw, J.L., Saunders, K., Rinehart, N.J.,2012a. An investigation of upper limb motor function in high functioning autismand Asperger's disorder using a repetitive Fitts' aiming task. Research in AutismSpectrum Disorders 6 (1), 286–292.

Papadopoulos, N., McGinley, J., Tonge, B.J., Bradshaw, J., Saunders, K., Murphy, A.,Rinehart, N.J., 2012b. Motor proficiency and emotional-behavioural disturbancein autism and Asperger's disorder: another piece of the Neurological puzzle?Autism 16 (6), 627–640.

Papadopoulos, N., Rinehart, N., Bradshaw, J.L., Taffe, J., McGinley, J., 2012c. Is therea link between motor performance variability and social-communicativeimpairment in children with ADHD-CT: a kinematic study using an upperlimb Fitts’ aiming task. Journal of Attention Disorders, http://dx.doi.org/1087054712454569.

Piek, J., Pitcher, T., Hay, D., 1999. Motor coordination and kinaesthesis in boys withattention deficit-hyperactivity disorder. Developmental Medicine & ChildNeurology 41, 159–165.

Pitcher, T., Piek, J., Hay, D., 2003. Fine and gross motor ablity in males with ADHD.Developmental Medicine and Child Neurology 45, 525–535.

Qiu, A., Adler, M., Crocetti, D., Miller, M., Mostofsky, S., 2010. Basal ganglia shapespredict social, communication, and motor dysfunctions in boys with autismspectrum disorder. Journal of the American Academy of Child and AdolescentPsychiatry 49 (6), 539–551.

Reiersen, A.M., Constantino, J.N., Volk, H.E., Todd, R.D., 2007. Autistic traits in apopulation based ADHD twin sample. Journal of Child Psychology and Psy-chiatry 48, 464–472.

Reiersen, A., Constantino, M., Todd, R., 2008. Co-occurrence of motor problems andautistic symptoms in Attention-Deficit/Hyperactivity Disorder. Journal ofAmerican Academy of Child Adolescent Psychiatry 47 (6), 662–672.

Rinehart, N.J., Bradshaw, J.L., Brereton, A.V., Tonge, B.J., 2001. Movement Prepara-tion in high-functioning autism and Asperger's disorder: a serial choice-reaction time task involving motor reprogramming. Journal of Autism andDevelopmental Disorders 31 (1), 79–88.

Rinehart, N.J., Bellgrove, M.A., Tonge, B.J., Brereton, A.V., Howells-Rankin, D.,Bradshaw, J.L., 2006a. An examination of movement kinematics in youngpeople with high-functioning autism and Asperger's disorder: further evidencefor a motor planning deficit. Journal of Autism Developmental Disorders 36,757–767.

Rinehart, N.J., Tonge, B., Bradshaw, J.L., Iansek, R., Enticott, P., McGinley, J., 2006b.Gait function in chldren with high-functioning autism and Asperger's disorder:evidence for basal ganglia and cerebellar involvement? European Journal ofChild and Adolescent Psychiatry 15 (5), 256–264.

Rinehart, N.J., Tonge, B.J., Iansek, R., McGinley, J., Brereton, A.V., Enticott, P.,Bradshaw, J.L., 2006c. Gait function in newly diagnosed children with autism:cerebellar and basal ganglia related motor disorder. Developmental Medicineand Child Neurology 48, 819–824.

Teitelbaum, O., Benton, T., Shah, P.K., Prince, A., Kelly, J.L., Teitelbaum, P., 2004.Eshkol Wachman movement notation in diagnosis: the early detection ofAsperger's syndrome. Proceedings of the National Academy of Sciences of theUnited States of America 101 (32), 11909–11914.

Valera, E.M., Spencer, R.M., Zeffiro, T.A., Makris, N., Spencer, T.J., Faraone, S.V.,Biederman, J., Seidman, L.J., 2010. Neural substrates of impaired sensorimotortiming in adult attention deficit hyperactivity disorder. Biological Psychiatry 68,359–367.

Wechsler, D., 1999. WASI – Wechsler Abbreviated Scale of Intelligence Manual.Harcourt Assessment, USA.

Wechsler, D., 2005. Manual for the Wechsler Intelligence Scale for Children –

Fourth edition – Australian Adaptation. The Psychological Corporation, SanAntonio, TX.

N. Papadopoulos et al. / Psychiatry Research ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Please cite this article as: Papadopoulos, N., et al., An investigation of gait in children with Attention Deficit Hyperactivity Disorder: Acase controlled study. Psychiatry Research (2014), http://dx.doi.org/10.1016/j.psychres.2014.04.037i