Assessment of Driving-related Performance in Chronic Whiplash Using

Accident Analysis and Prevention 60 (2013) 5 14

Contents lists available at ScienceDirect

Accident Analysis and Prevention

journa l h om epage: www.elsev ier .co

Assessment of driving-related performance in chan advanced driving simulator

Hiroshi T ndAndrew a NHMRC Centr rapy, The University b Centre for Acc , Brisb

a r t i c l e i n f o

Article history:Received 15 JuAccepted 30 Ju

Keywords:AttentionAutomobile drChronic painComputer simTask performaWhiplash inju

a b s t r a c t

Driving is often nominated as problematic by individuals with chronic whiplash associated disorders

1. Introdu

Individucommonly Pereira et aciated with

This studyThe University

CorresponHealth and Rehland 4072, Aus

E-mail add(H. Takasaki).

0001-4575/$ http://dx.doi.oly 2012ly 2013

iving

ulationnceries

(WAD), yet driving-related performance has not been evaluated objectively. The purpose of this studywas to test driving-related performance in persons with chronic WAD against healthy controls of sim-ilar age, gender and driving experience to determine if driving-related performance in the WAD groupwas sufciently impaired to recommend tness to drive assessment. Driving-related performance wasassessed using an advanced driving simulator during three driving scenarios; freeway, residential anda central business district (CBD). Total driving duration was approximately 15 min. Five driving taskswhich could cause a collision (critical events) were included in the scenarios. In addition, the effect ofdivided attention (identify red dots projected onto side or rear view mirrors) was assessed three times ineach scenario. Driving performance was measured using the simulator performance index (SPI) which iscalculated from 12 measures. z-Scores for all SPI measures were calculated for each WAD subject basedon mean values of the control subjects. The z-scores were then averaged for the WAD group. A z-scoreof 2 indicated a driving failing grade in the simulator. The number of collisions over the ve criticalevents was compared between the WAD and control groups as was reaction time and missed responseratio in identifying the red dots.

Seventeen WAD and 26 control subjects commenced the driving assessment. Demographic data werecomparable between the groups. All subjects completed the freeway scenario but four withdrew dur-ing the residential and eight during the CBD scenario because of motion sickness. All scenarios werecompleted by 14 WAD and 17 control subjects. Mean z-scores for the SPI over the three scenarios wasstatistically lower in the WAD group (0.3 0.3; P < 0.05) but the score was not below the cut-off pointfor safe driving. There were no differences in the reaction time and missed response ratio in dividedattention tasks between the groups (All P > 0.05). Assessment of driving in an advanced driving simulatorfor approximately 15 min revealed that driving-related performance in chronic WAD was not sufcientlyimpaired to recommend the need for tness to drive assessment.

2013 Elsevier Ltd. All rights reserved.

ction

als with chronic whiplash associated disorders (WAD)report difculties in driving (Hoving et al., 2003;l., 2008; Takasaki et al., 2011). Chronic WAD is asso-

a variety of physical, psychological and cognitive

was cleared by the institutional human medical ethics committees of of Queensland and The Queensland University of Technology.ding author at: CCRE Spine, Division of Physiotherapy, School ofabilitation Science, The University of Queensland, Brisbane, Queens-tralia. Tel.: +61 7 3365 2275; fax: +61 7 3365 1622.resses: [email protected], [email protected]

symptoms (Radanov et al., 1995; Dallalba et al., 2001; hberg et al.,2003; Treleaven et al., 2005; Pereira et al., 2008; Takasaki et al.,2012) that may directly or indirectly negatively affect the motorskills, visual perception and cognitive skills required for safe driv-ing (Austroads, 2003). It is possible that individuals with chronicWAD may have impaired driving-related performance and on-roadsafety. However, no studies have objectively quantied driving-related performance in individuals with chronic WAD to determineif they are t to drive safely.

The gold standard for objective assessment of driving-relatedperformance is an on-road assessment. On-road assessments areless than ideal for research purposes. Firstly, the outcomes ofan on-road driving test cannot be compared between individualsbecause road conditions cannot be standardized for all subjects.

see front matter 2013 Elsevier Ltd. All rights reserved.rg/10.1016/j.aap.2013.07.033akasakia,, Julia Treleavena, Venerina Johnstona, AHainesb, Gwendolen Jull a

e of Clinical Research Excellence Spinal Pain, Injury and Health, Division of Physiotheof Queensland, Brisbane, Queensland, Australiaident Research & Road Safety Queensland, The Queensland University of Technologym/locate /aap

ronic whiplash using

ry Rakotonirainyb,

School of Health and Rehabilitation Science,

ane, Queensland, Australia

6 H. Takasaki et al. / Accident Analysis and Prevention 60 (2013) 5 14

Secondly, challenging trafc situations are needed to investigatedriving safety comprehensively. Inherent in providing challengingon-road trafc situations is the risk of being involved in a car crash(Rizzo, 2004). An alternative is the use of driving simulators whichallow maniized, repeacontrolled simulators et al., 2002;and the useto the on-roIt is thus prsimulator tochronic WA

1.1. Purpos

The purprelated pesufcientlydrive assessan advancewith chroni

2. Method

2.1. Study d

This croand asymptand drivinghuman meland and Tprovided w

2.2. Subject

Subjectssubjects weclinic. Genebetween 20of cervical forders (e.g.affecting drdiovascularprone to mgroup wereyears duraton the Neclesser scoreFurther inctory of whidiagnosed p

2.3. Driving

This stuFrance), hoAustralia. Aplatform (Edegrees of three dimeate to the generated uonto three tal and 40

irro 9 cms rele mi

righl noiNeR

rateas

SENotionetersz.

riving

s studnario. Theed blectele th

on tfor 5 f scey scio. Thss. Baas learivinsted otherh driving task (e.g., changing lanes and turning at intersec-were also given by pre-recorded instructions. Crash soundsroduced, without any car motion reactions, when there wasion with any vehicle (car, cyclist and pedestrian).

Freeway scenario road map for this study was a two-way freeway with twon each direction. In keeping with local road conditions, lane

were 4 m (leftmost lane) and 4.55 mm (center lane) andeed limit was 100 km/h throughout the scenario. The roadrved slightly to the right (Fig. 2) but the radius of curve was

m and therefore was considered as a straight road. Three tasks were included; merging onto the freeway, changingnd sudden braking as these were identied as troublesomeevious study of chronic WAD (Takasaki et al., 2011).pulation of driving environments to deliver standard-table and challenging driving scenarios in a safe andenvironment. A number of studies have used drivingfor the assessment of driving-related performance (Ku

Lew et al., 2005; Yuen et al., 2007; Crizzle et al., 2012) of a driving simulator is regarded as a valid alternativead assessment (Lew et al., 2005; Shechtman et al., 2009).eferable, at least for research purposes, to use a driving

study driving-related performance in individuals withD.

e

ose of this study was to investigate whether driving-rformance in individuals with chronic WAD was

impaired to require future consideration of tness toment. Driving-related performance was assessed usingd driving simulator and compared between individualsc WAD and asymptomatic healthy controls.

s

esign

ss-sectional study included persons with chronic WADomatic healthy control subjects of similar ages, gender

experience. This study was cleared by the institutionaldical ethics committees of The University of Queens-he Queensland University of Technology. All subjectsritten informed consent prior to data collection.

s

were recruited via community advertisements. WADre also recruited from a university whiplash researchral inclusion criteria were; (1) current drivers aged

and 60 years, (2) residing in Brisbane, (3) no historyracture or dislocation, concussion or neurological dis-, multiple sclerosis, stroke), (4) no medical problemsiving (e.g., upper or lower limb fractures/injuries, car-

problems, respiratory and visual disorders), and (5) nototion sickness. Further inclusion criteria for the WAD

ongoing neck pain of between three months and sixion related to a whiplash injury and a score of 8/100k Disability Index (NDI) (Vernon and Mior, 1991) as a

is regarded as recovered (Sterling et al., 2003a,b, 2006).lusion criteria for the control group were; (1) no his-plash, (2) no current headache or neck pain, and (3) nosychological problems.

simulator

dy used an advanced driving simulator (OKTAL, Paris,used at The Queensland University of Technology,

real car without an engine was mounted on a motionmotion 1500, REXROTH, Boxtel, Netherlands) with six-freedom, which allowed the car to move and twist innsions in order to provide simulated motion appropri-driving situation (Fig. 1). A virtual environment wassing eight computers. Virtual sceneries were projectedat screens (4 m wide 3 m high) with 180 horizon-vertical forward eld of view, and onto side and rear

view m15 cm mirrorleft sidfor thementaby SCArefreshback wSD-LC,and mparamof 20 H

2.4. D

Thiing sce(CBD))precednot colpossibdriventimed order ofreewascenarsicknesick weach dthe poto do for eactions) were pa collis

2.4.1. The

lanes iwidthsthe spmap cu10,000drivinglanes ain a prFig. 1. An advanced driving simulator.

rs, which were replaced by LCD monitors (side mirrors,; rear view mirror, 24.5 cm 8 cm). The angles of the

ative to the straight line ahead were left 61.7 for therror, left 32.5 for the rear view mirror and right 35.5

t side mirror. Surround sound for engine and environ-se were also generated. All computers were controlledTMstudio ver.1.0 software (OKTAL, Paris, France). The

of the visual virtual environment was 60 Hz. Feed-provided by a force feedback system (SENSO-wheelSODRIVE, Weling, Germany) on the steering wheel

platform to provide a realistic driving experience. All while driving were recorded at a sampling frequency

scenarios

y examined driving-related performance in three driv-s (freeway, residential area and central business district

CBD was a replicate of Brisbane, Australia. These werey a 5-min-familiarization scenario, in which data wered. Each scenario was designed to replicate as closely ase local road conditions and environment. Vehicles arehe left side of the road in Australia. Each scenario wasmin to minimize the possibility for motion sickness. Thenarios was not randomized and all subjects drove the

enario rst, followed by the residential and then CBDis order was chosen to reduce drop-outs due to motionsed on our pilot trials, the probability of being motionst when driving on a straight road. Before commencingg scenario, subjects were instructed to remain withinspeed limits and within the one lane unless instructedwise by pre-recorded voice instructions. Instructions

H. Takasaki et al. / Accident Analysis and Prevention 60 (2013) 5 14 7

Fig. 2. Road n(2) The rst dr(a red circle on(5) The third ddivided attentattention task

Box 1: Pscenario

The sufreewa

A car ispeed a

There aon the

Cars B subjectdrive w

After beby the ject {the par

Car A sject reathe sub

A voice

The rstthe freewayconditions bof subjects

The secoing simulatto simulatetrolled basechange. Fremaximum ing lanes ancater for parother cars ithis speed rin the mainencouragedlation was tBox 1.

The thirdcritical eveAn accidentject (Cars Bbetween th

Box 2: Procedures for the critical event of suddenbraking in the freeway scenario.

hen the distance between the subject and Cars B and Ccomes 60 m, Cars B and C drive at the same speed as thebject, maintaining a 60 m separation.rs B and C brake to 20 km/h over 5 s when the distancetween Car B and an accident site on the left hand lane is2 m.e second later, Car B stops before the accident site andr C accelerates to 50 km/h over 5 s.ve seconds later, Car C accelerates at 120 km/h over 20 s.

on requiring sudden braking from pilot trials (see Box 2 for).

Residential scenario road map used in the residential scenario was a two-waywith one lane (5 m width) in each direction, with the excep-

the fourth driving task where the road became two lanes.local road conditions, the speed limit was 60 km/h through-s scenario. The scenario included straight and curved roadso intersections (Fig. 3). The proportion of straight or curvedwas equal. The curved roads were gentle (radius = 70 m or

as sharp curves do not have sufcient validity when drivingetwork of the freeway scenario. (1) The start point of this scenario.iving task, merging onto freeway. (3): The rst divided attention task

the right side mirror). (4) The second driving task, changing lanes.riving task, sudden braking (the rst critical event). (6) The secondion task (a red circle on the rear view mirror). (7) The third divided(a red circle on the left side mirror). (8) The end point of this scenario.

rocedures for changing lanes in the freeway.

bject drives in the left lane after merging onto they.

Wbesu

Cabe16

OnCa

Fi

situatidetails

2.4.2. The

street tion ofAs per out thiand twroads 300 m)n the right lane (Car A) begins to drive at the sames the subject.re two cars in each lane (Car B on the left and Car Cright) in front of the subject.and C are activated when the distance between the

and Car C becomes 350 m and Car B and C begin toith a speed of 40 km/h.ing activated, cars B and C travel at a speed calculatedfollowing formula: Speed [km/h] = Speed of the sub-the maximum value of: (5) or (distance [m] betweenticipant and Car B)/7}.lows down to 50 km/h over 20 seconds when the sub-ches 500 m from a certain point to make a space forjects car to change lanes.

instruction change lanes to the right is given.

driving task of merging onto the freeway had all cars on in the right hand lane. This ensured consistent drivingetween subjects and avoided differences in the timing

merging.nd driving task was changing lanes (Video 1). The driv-or permitted subjects to drive at their self-paced speed

their usual driving. The speed of other cars was con-d on the subjects speed to simulate a realistic laneeway speed limits are 100 km/h. In our pilot trials, thespeed driven by a participant was 120 km/h. Chang-d sudden breaking on the freeway were programed toticipants driving speeds between 120 and 70 km/h, Then the programed scenario reacted appropriately withinange. If the subject drove


the residential scenario.

of the roadApproximalarge multi-a realistic aprogramedbraking waD.

The secothird criticaand adult wThe child bbetween thwas requiresection slowintersectionuals both wthem avoidselected in

The thirintersectionhad the righthe other la

The nexevent. The each direct3.35 m (censafely, the sright lane, lright hand l

2.4.3. CBD sThe CBD

map of the ity of road The CBD scesignals andcal event w(Video 6). Ation 2 s afteA mood disapproachinintersectionFig. 4. Turning left at the rst intersection in

to lessen subjects anticipation of this critical event.tely 200 m further on, Car D was parked in front of apassenger vehicle (Car E), which hid Car D to generatend challenging trafc situation. The oncoming lane was

to have seven cars following each other closely so thats the subjects only choice to avoid a collision with Car

nd task was turning left at the rst intersection, thel event. The trafc signal was always green. A childere placed on the sidewalk at the intersection (Fig. 4).egins to run across the intersection when the distancee subject and the child was 8 m (Video 4). The subjectd to anticipate the potential hazard, turn at the inter-ly and brake suddenly when the child ran across the

to avoid hitting the child. Pilot studies with individ-ith and without neck pain revealed that only half of

ed a collision at a distance of 8 m, thus this distance wasorder to simulate a very challenging trafc situation.d task was turning right at a second intersection (T-) without trafc signals. Vehicles in the subjects lanet of way, indicated by a stop sign and lane marking onne.

t task was changing lanes (Video 5), the fourth criticalroad map was a two-way street with two lanes in theion. The lane widths were 3.15 m (leftmost lane) andter lane). To change from the left to the right hand laneubject needed to appreciate the cars and space in theet the car on the right go ahead and then change to theane. Box 3 presents the details of this task.

cenario scenario was developed to replicate a real trafc roadBrisbane CBD. The width of one lane was 3.5 m (major-network) or 3.35 m and the speed limit was 40 km/h.nario included turning at four intersections with trafc

a sharp curve (radius = 11.8 m) (Fig. 5). The fth criti-as programed when turning left at the rst intersection

cyclist riding on the sidewalk crossed the intersec-r the trafc light changed from red to green (Box 4).turbing event, where the subject was honked by a carg from the behind (Box 5), was programed at the nal

(Video 7).

Fig. 5. Road network of the Brisbane CBD scenario. The course is presented withwhite arrors and numbers. The simulated Mary Street is a two-way street withtwo lanes in each direction, the simulated Albert Street is a two-way street withtwo lanes in one direction and one lane in the other direction, and the simulatedMargaret, Edward and Alice Streets are all one way streets with four lanes. (1) Thesubject drives on Mary Street, stops at an intersection between Mary street andAlbert street, and then turns left into Albert Street (the fth critical event) (Box 4).(2) On Albert street, a stationary car with ashing hazard signals is located in theright lane. The subject is required to drive in the left lane in Albert Street. (3) Thesubject is instructed to turn left at the intersection of Albert and Margaret Streetsand drive in the second lane from the right. The trafc signal is green. On Mar-garet Street, cars are stationary in the rst, third and fourth lanes and therefore thesubject can drive only in the second lane from the right. (4) At the intersection ofMargaret and Edward Streets, the subject is instructed to turn right into EdwardStreet and keep to the second lane from the right. The trafc signal is green. (5)Edward Street changes to Alice Street after a sharp curve (radius = 11.8 m). (6) Thesubject is instructed to turn right at the intersection of Alice and Albert Streets. Thetrafc signal is green. On Albert Street, the subject is instructed to go straight fortwo blocks. At the intersection of Albert and Margaret Streets, a mood disturbingevent is programed where the participant is honked by a car approaching from thebehind. The rst red circle-dot of the divided attention task is generated on the rightside mirror in Mary Street, the second on the left side mirror in Margaret Street andthe third on the rear view mirror in Alice Street.


Box 3: Procedures for changing lanes in the residentialsecenario.

A voiceing lan

A voice One lan The su A car b

and drias the smirror.

A parkeCar G asubject

When tCar F a

The suahead.

Then, t

Box 4: Pat an int

The su A cyclis

the cyc A voice

is given The tra

red whtion is

The su The cyc

by the fbetweefor thegreen p

The cycthe traf

When tto stop

Box 5: PCBD sec

Car H is The su The tra

the dist After 2

aching subject

Car H sthe intebehindthe inte

2.5. Outcom

A questigender, yeadriven per of neck pain

(NDI) was recorded for the WAD subjects. The validity and reli-ability of the NDI have been established in the chronic whiplashpopulation ( = 0.87) (Hoving et al., 2003; Nieto et al., 2008). Anysymptoms related to motion sickness during testing was measured

h groonnaving

Indemberer dr instruction, drive on the left hand lane until chang-es is necessary is given.

instruction, Keep left is given.e changes to two lanes.

bject is driving in the left lane.ehind in the right lane (Car F) catches up the subject

in botQuesti

Drimancethe nuconsidves beside the subject (4 m behind) at the same speed

ubject so that Car F cannot be seen via the rear view

d car is located in the left lane, after a gentle left curveppears and buildings are placed on the left so that the

cannot see Car G before the gentle curve.he distance between the subject and Car G is 30 m,ccelerates to 100 km/h.bject stops behind Car G and needs to let Car F go

he subject can change lanes to the right hand lane.

rocedures for the critical event with turning leftersection in the CBD secenario.

bject is driving in the left lane on Mary Street.t is riding on the left sidewalk and the subject passeslist.

instruction, turn left at the following intersection.fc signal at the intersection changes from green toen the distance between the subject and the intersec-80 m.bject stops at the intersection.lists approaches the intersection at a speed calculatedollowing formula: Speed [km/h] = 3.6 {(distance [m]n the cyclist and the trafc signal stop line)/(time (s)

trafc light to go through a full cycle and return tolus 2 s (19 s total))}.list crosses the intersection without braking 2 s afterc signal changes to green.he trafc signal changes to green, the subject needs

and let the cyclist go across the intersection.

rocedures for the mood disturbing event in theenario.

following the subject 20 m behind at the same speed.bject is instructed to go straight at the intersection.fc signal at the intersection changes to amber whenance between the subject and the trafc light is 40 m.s, Car H begins honking their horn for 4 s and is appro-within 6 m behind, accelerating 10 km/h faster than the.tops 6 m behind the subject if the subject stops atrsection with red trafc signal, or Car H drives 6 m

the participant and turns right if the subject crossesrsection and ignores the red trafc signal.

e measures

onnaire was used to record subject characteristics; age,rs holding a driver license, self-reported kilometersweek and days driven per week. The length of history

related to WAD, self-reported neck pain and disability

disturbing ewere record

2.5.1. The SThe SPI

et al., 2005)trol (ve mspeed contrthe posted of acceleratof red-lightmean absollane positiolane positioability on cof collisionwere calculMathWorks(e.g., merginturning intenumber of and the ovetrol group. 2005). Gootrol domainoverall scor

2.5.2. DividThree d

three drivinappeared owas instruction time wwhen the suwithin 5 s, recorded asresponse ra

2.6. Statisti

This studformance wassessmentfore based et al., 2005)sample sizedetect the fa larger samis more impstudy. Our sickness wthese two fain each gro

Descriptvariables. D(ShapiroWups with the Modied Motion Sickness Assessmentire (M-MSAQ) (Brooks et al., 2010).performance was evaluated with the Simulator Perfor-x (SPI). Responses in the divided attention tasks and

of collisions in the ve critical events were collated toiving safety. In addition, subjects reactions to the moodvent in the CBD scenario (car behind honking the horn)ed.

PIis an established measure of driving performance (Lew. It includes 12 measures with two domains; speed con-easures) and direction control (seven measures). Theol domain includes; speed (percent of time exceedinglimit); standard deviation (SD) of speed variability; SDion variability; SD of the throttle speed; and the number

violations. The direction control domain includes; theute value of lane position error on straight roads; SD ofn variability on straight roads; mean absolute value ofn variability on curved roads; SD of lane position vari-urved roads; SD of steering wheel speed; the numbers; and the number of deviations off-road. All measuresated with a custom developed MATLAB program (The

Inc., Natick, MA, USA). Data from specic driving tasksg onto the freeway, changing lanes, sudden braking andrsections) were not included in SPI calculations but thecollisions were included. The z-scores for each domainrall SPI score were calculated and compared to the con-A z-score of 2.0 indicates a failing grade (Lew et al.,d internal consistency in each domain (the speed con-, = 0.9; the direction control domain, = 0.8) and thee ( = 0.9) was reported by Lew et al. (2005).

ed attention tasksivided attention tasks were included in each of theg scenarios (Figs. 24). A red dot (7.5 cm diameter)

n a side or rear view mirror while driving. The subjectted to ash high beam as soon as they saw the dot. Reac-as measured in milliseconds and the dot disappearedbject responded. If the subject failed to ash high beamthe dot automatically disappeared, reaction time was

5 s and a miss was noted. Reaction time and missedtio were computed for each divided attention task.

cs

ys aim was to investigate whether driving-related per-as sufciently impaired in chronic WAD to warrant

of tness to drive. The minimum sample size was there-on the failing grade of the SPI (i.e., z-score = 2) (Lew. G*Power 3.1.3 (Faul et al., 2007) demonstrated that a

of 16 (eight in each group) would provide 95% power toailing grade of the SPI with an level of 0.05. However,ple size reduces the chance of a type-II error, whichortant than the type-I error given the purpose of this

pilot studies indicated that withdrawals due to motionere likely in the simulator in this study. Consideringctors, we aimed to recruit approximately 25 individuals

up.ive statistics (mean SD) were used to summarizeata for the SPI were tested for normal distributionilk tests). Speed variability and throttle speed data were


Table 1Characteristics of subjects who commenced the study and subjects who completed all scenarios.

Variables Commenced the study Completed all scenarios

WAD (n = 17) Control (n = 26) WAD (n = 14) Control (n = 17)

Age (years) 35.1 12.0 36.7 10.4 33.4 10.8 36.9 11.1Female (n [%]) 11 [65] 19 [73] 8 [57] 10 [59]Driving experience (years) 12.2 9.7 16.8 12.5 12.1 9.3 17.7 12.6Driving frequency per week (days) 5.6 2.2 5.5 2.2 5.3 2.4 6.1 1.6Total kilometers driven per week (km) 165.4 185.5 178.3 135.7 173.7 201.9 231.9 130.8Symptom duration (months) 32.1 23.9 NA 35.6 24.8 NANDI (%) 28.8 13.6 NA 29.0 13.7 NA

No signicant differences in measures between and within the WAD and control groups who commenced the study and who completed all scenarios (All P > 0.05).

normally distributed and therefore the raw data were used tocalculate z-scores. All other SPI data were normalized using log-arithmic transformations (Lew et al., 2005), allowing parametricstatistical tests. Independent samples t-tests were used to com-pare mean z-scores in all SPI measures between the groups (Lewet al., 2005). Negative z-scores meant below-normal performance.In each scenario, z-scores were computed for subjects who com-pleted the scenario. Fifteen SPI measures, including the 12 SPIparameters, speed control and direction control domains and over-all SPI score, were compared between the WAD and control groups.For assessment of overall driving-related performance, z-scores ofthe SPI in each scenario were averaged and compared between thegroups using Independent samples t-tests. Independent samplesMannWhiof other meAll statisticCorporation

3. Results

Subjectsplays the with chronmenced thesubjects annarios beca

Table 1 pthe 26 cont17 controls

differences in demographics between the WAD and control groups(All P > 0.05). There was no signicant difference in withdrawalrates in each scenario between the groups (P > 0.05). There was arelatively equal distribution of WAD subjects with milder (NDI < 30)or moderate to severe (NDI > 30) self-reported pain and disability(Vernon and Mior, 1991; Sterling et al., 2003a,b, 2006) within theWAD group who commenced (milder WAD = 47%, moderate/severeWAD = 53%) and who subsequently completed all scenarios (milderWAD = 50%, moderate/severe WAD = 50%). The freeway scenariowas restarted for three subjects in each group as they drove 0.05).

Table 2 presents the mean SPI z-scores for the WAD group.werece, l grol driv

perfcena

speenarf thele 3 n eance and ed thl subant.

Table 2Mean z-scores

Measures scena

Overall SPI Speed contrSpeed (timeSpeed variabAccelerationSpeed jerk Red-light vioDirection coLane positioLane positioLane positioLane positioSteering jerkCollisionsa

Deviations o

Abbreviation: Npresented in b

a Data is trab Values arec Values ared Values aree z-Scores otney U tests and Fisher tests were used for comparisonsasures including demographic data between the groups.al analyses were performed by SPSS version 20.0 (IBM, New York, USA). Signicance level was set at P < 0.05.

were recruited from April 2011 to May 2012. Fig. 6 dis-ow of subjects through the study. Seventeen subjectsic WAD and 26 controls entered the study and com-

assessment in the driving simulator but three WADd nine control subjects failed to complete the three sce-use of motion sickness.resents demographic data for the 17 WAD subjects androls who commenced the study and the 14 WAD and

who completed all scenarios. There were no signicant

There formancontrooveralpoorerthree spoorerCBD scgrade o

Tabratio idiffereWAD detectcontrosignic

of the SPI in the WAD group.

Freeway scenariob (n = 16) Residential

0.1 0.4 0.2 0.5 ol 0.1 0.7 0.3 0.8

over posted limit)a 0.0 1.0 0.3 0.9 ility 0.2 1.4 0.2 1.1

variabilitya 0.4 1.5 0.4 1.3 0.5 1.4 0.4 1.3

lations NA 0.0 0.0 ntrol 0.1 0.4 0.1 0.6 n (straight road)a 0.6 0.7 0.2 1.8 n (curved road)a NA 0.5 1.1

an variability (straight road) 0.0 0.9 0.0 1.1 n variability (curved road)a NA 0.1 1.3 a 0.3 1.7 0.1 1.0

0.1 0.9 0.4 1.1 ff roada 0.4 1.0 0.2 1.4 A, not applicable. Values are normalized z-scores relative to the control group; negative old.nsformed using logarithmic transformation.

normalized z-scores relative to the control group (n = 26). Statistical signicance is by t-t normalized z-scores relative to the control group (n = 23). Statistical signicance is by t-t normalized z-scores relative to the control group (n = 17). Statistical signicance is by t-tf the WAD group who completed all scenarios (n = 13) are averaged. several negative values indicating poorer driving per-but few reached signicance when compared to theup. The WAD group had statistically (P < 0.05), poorering performance (overall SPI). There was statisticallyormance in speed control, and speed variability over therios, poorer lane position in the freeway scenario, anded control domain, speed, and speed variability in theio. Notably, no measure approached or met the failing

SPI (2.0).presents the mean reaction time and missed responsech divided attention task. There was no signicantin any measure of divided attention between thecontrol groups (All P > 0.05). However, WAD subjectse red dot in the left mirror more frequently thanjects in all scenarios, but the difference was not

rioc (n = 15) CBD scenariod (n = 13) Averagee (n = 13)

0.3 0.5 0.3 0.30.6 0.8 0.4 0.50.8 0.7 0.5 0.61.0 1.5 0.6 0.70.9 1.7 0.5 1.00.6 1.4 0.6 1.20.1 1.1 0.0 0.60.1 0.6 0.2 0.40.3 1.1 0.1 0.80.2 0.8 0.2 0.80.3 1.4 0.3 0.90.3 1.1 0.1 1.00.5 1.1 0.4 0.90.4 0.0 0.0 0.5

0.6 0.9 0.5 0.6values indicate below-normal performance. z-Scores with P < 0.05 are

est (40 degrees of freedom).est (36 degrees of freedom).est (28 degrees of freedom).


WAD-Interest in participating

(n=65)

Teste d

(n=17)

Excluded (n=48)

Not meeting inclusion criteria (n=10)

NDI0 (n=1)

Prone to motion sickness (n=3)

Declined to participate (n=2)

No response after providing infor mation (n=2)

Control-Interest in participating

(n=32)

Fig. 6. Flowchbusiness distri

Table 4 at each critsignicant P > 0.05). Thpants had t60% of both

Table 3Reaction time

Scenario

Freeway sce

Residential s

CBD scenario

Abbreviations:No signicant

a 17 WAD sub 16 WAD suc 14 WAD suResidential scenario

(n=17)

Completed re sidential scenario (n=16 )

Withdrew due to motion sickness (n=1)

M-MSAQ=3.8

Com

With

M-MSCBD scenario

(n=16)

Completed CBD scenario (n=14)

Withdrew due to motion sickness (n=2)

M-MSAQ=6.2 4.3

Completed

Withdrew

M-MSAQ=

art of subjects. Abbreviations: WAD, whiplash associated disorders; M-MSAQ, Modiedct.

presents the number of subjects who had a collisionical event. Collisions in most events were rare with nodifferences between the WAD and control groups (Alle exception was in the residential area where partici-o turn left and confronted a pedestrian. Approximately

WAD and control subjects hit the pedestrian.

Review othe CBD sce28 of the 3one controllight changthat their sp

and incorrect response ratio in divided attention tasks.

Location of a dot WAD

Reaction time (s) Incorrect response

narioa Left 3.3 1.6 6 [35] Center 2.3 1.6 4 [24] Right 1.8 1.3 2 [12]

cenariob Left 3.8 1.6 8 [50] Center 2.0 1.6 3 [19] Right 1.8 1.2 1 [6]

c Left 4.3 1.4 11 [79] Center 2.7 1.8 4 [29] Right 2.0 1.6 3 [21]

Left, left side mirror; Center, rear view mirror; Right, right side mirror.differences in each measure between the WAD and control groups (All P > 0.05).bjects and 26 control subjects.bjects and 23 control subjects.bjects and 17 control subjects.sidential scenario

(n=26)

re sidential scenario (n=23 )

due to motion sickness (n=3)

5.5 1.1CBD scenario

(n=23)

CBD scenario (n=17)

due to motion sickness (n=6)

3.6 2.8

Motion Sickness Assessment Questionnaire; CBD, Brisbane central

f participants reactions to the mood disturbing event innario (a car behind honking their horn), revealed that1 subjects stopped at the intersection. One WAD and

subject crossed the intersection even though the trafced from amber to red. They reported that they thoughteed was too fast to stop at the intersection and sudden

Control

ratio (n [%]) Reaction time (s) Incorrect responseratio (n [%])

3.4 1.7 13 [50]2.0 1.4 2 [8]1.4 0.8 1 [4]4.1 1.4 15 [65]1.9 1.4 3 [13]1.3 0.8 1 [4]5.0 0.1 16 [94]3.3 1.9 9 [53]1.7 1.3 2 [12]


Table 4The number of collisions in each critical event.

Scenario Critical event WAD Control

Freeway sce

Residential s

CBD scenari

No signicantP > 0.05).

braking waclosely. Anoa collision w

We quesinuence dures averaresponse ragroups (milthe milder three of theand speed cferences in groups (All

4. Discussi

This stuindividualsthat the maand 12% of acating poorno SPI measindicating tThe averagewhich contmined by Lein persons wdifferencesmissed respof collisiondriving-relaimpaired totheir tness

Pain in icar crash (Laimpact of c2005; Veld2011; Fan etive skills (Ket al., 2010;whether Wof pain andmance or pwith milderdid not havattention pmilder WADlyzed. Thussubstantivecohort. Nec

1990; Marottoli et al., 1998; Barry et al., 2003) but the intensityof neck pain is not always associated with the magnitude of lim-itation in neck rotation (Howell, 2011), which could explain our

ations. preed r

mots retisticnariotiousovend h

rateg safeical

scey inted s it iThe crseclativerials,o crias n

moodistu

studnd cHowollisiay relashubjeio whterpnd pimitebe anin peped en thd thio wferen

nereverwithceler

as feporteaki es studnario Sudden braking 1/17 2/26

cenarioSudden braking 0/16 2/23Hitting a childrunning out ontothe road at anintersection

10/16 14/23

Changing lanes 2/16 0/23

o Turning left at anintersection

1/14 3/17

differences between the WAD and control groups in any events (All

s dangerous with the car behind honking and followingther WAD subject moved to the opposite lane to avoidhen the car honked.tioned, post hoc, whether higher disability level couldriving ability and therefore compared the 15 SPI meas-ged across all scenarios, reaction time and missedtio in the divided attention tasks between the two WADder WAD vs moderate/severe WAD). It was found thatWAD group had statistically poorer performances in

15 SPI measures; variance speed, variance accelerationontrol domain (P < 0.05). There were no signicant dif-any divided attention measures between the two WADP > 0.05).

on

dy compared driving-related performance between with chronic WAD and healthy controls and determinedjority (74%) of WAD groups SPI z-scores were negativell possible SPI measures were statistically inferior, indi-er driving performance in the WAD group. However,ure met the established failing criteria (z-score 2.0),hat driving impairments were negligible or at least mild.

overall SPI z-score for the WAD group was 0.3 0.3,rasts markedly to the overall SPI of 4.6 4.7 deter-w et al. (2005) when investigating driving performanceith traumatic brain injury. In addition, there were no

between WAD and control groups in reaction times,onse ratios in divided attention tasks and the number

s in each critical event. Thus, this study indicates thatted performance of persons with chronic WAD is not

the extent that would require specic consideration of to drive.

tself may be an important predictor of involvement in agarde et al., 2005) and many studies report the negative

hronic pain on driving (Jones et al., 1991; Duong et al.,huijzen et al., 2006; Pereira et al., 2008; Nilsen et al.,t al., 2012). Pain can also impact negatively on cogni-

observOur

perceivdrivingsubjecnot staall sceor cauhead mWAD ation stdriving

Critdrivingsome battempculty aWAD. an inteour repilot twere nthere wto the mood of thisWAD aevent. had a cThis ma whipWAD sscenarless, insmall ator is lmight safety develo

Whrevealescenarthe difety andwere sciated and acinatedself-re(Takas

Thi

uhajda et al., 2002; Pais-Vieira et al., 2009; Thompson

Takasaki et al., 2012). We therefore explored post hoc,AD subjects with self-reported moderate/severe levels

disability had either poorer driving-related perfor-oorer abilities in the divided attention tasks than those

symptoms. However, the moderate/severe WAD groupe poorer performance in either SPI measures or dividederformances. Unexpectedly poorer performance in the

group was identied in three of 15 SPI measures ana- the magnitude of pain and disability did not have a

impact on driving-related performance in our WADk rotation is important for driving (Hunter-Zaworski,

chronic WAment constsubjects (Wdue to motscreened ansickness. It negligible iTable 1 wereach scenarthe WAD grof subjects imum sampvious study found that 50% of WAD patients witheduction of driving ability after a whiplash reported nowre cautiously (Takasaki et al., 2011). Interestingly, WADsponded more frequently than control subjects (albeitally) to the divided attention task on the left mirror ins, which could support the notion of hyper-vigilanceness in driving. Further research is needed to comparement while driving between individuals with chronicealthy controls to better understand any compensa-ies used by persons with chronic WAD to maintain theirty.and mood disturbing events were programed into thenarios based on driving tasks nominated as trouble-dividuals with chronic WAD (Takasaki et al., 2011). Weto program critical events with different levels of dif-s unknown what level of difculty is critical in chronicritical event where the child ran out onto the road attion was designed as the most challenging task duringly short period of driving time. As anticipated by our

over half of the subjects (60%) had a collision. Theretical events where all subjects crashed and converselyone without a crash. There were three overt reactionsd disturbing event. This suggests that the critical andrbing events were sufciently realistic for the purposesy. There were no signicant differences between theontrol groups in the numbers of collision in any criticalever, no WAD subject, but rather two control subjects,on in the sudden braking task in the residential scenario.ect self-reported modication of driving behavior after

injury (e.g., more cautious driving). Interestingly twocts had a collision when changing lanes in the residentialile the count for the control group was nil. Neverthe-

retation is limited as the sample size in this study wasrecise prediction of accident risk using a driving simula-d (Rudin-Brown et al., 2009). However, changing lanes

important task in future research investigating drivingople with neck pain. Other critical events could also beto test the driving safety of persons with WAD.e SPI measures were considered more closely, it wasat three of ve measures of speed control in the CBDere statistically lower for the WAD group. Reasons force are not clear. There could be some increase in anxi-vousness while driving in the CBD scenario, where thereal parked cars and potential hazards. This may be asso-

the nding of increased speed variability (i.e., brakingation). Certainly, anxiety and nervousness were nom-atures by 54% of the cohort in our previous study ofd reduction of driving ability after a whiplash injuryt al., 2011).y attempted to recruit approximately 25 subjects withD but only 17 subjects entered the study due to recruit-raints and 14 completed all scenarios. Considering allAD and control), 28% (12/43) withdrew at various stagesion sickness (Fig. 6) even though all participants wered excluded from the study if they were prone to motionis considered that these withdrawals during testing hadmpact on results as; (1) characteristics summarized ine comparable between the WAD and control groups inio, (2) the control group had a larger sample size thanoup, giving stable reference values, and (3) the numberwho completed all scenarios was greater than the min-le size required to accomplish the primary aim of the


study. Mullen et al. (2010) demonstrated no association betweendriving performance and susceptibility to motion sickness, whichfurther supports the suggestion of a lack of substantive impact ofwithdrawals due to motion sickness on ndings of this study.

4.1. Limitat

This studrelated perfchronic WAin concentrities in dividriving timdriving-relaimportant rwho experinological pronly ve critions due tohave been wand healthyfor technicadriving taskA totally diperformancadequate bu(e.g., betweculty and with furtheWAD.

5. Conclus

This stuindividualsresidential asimulator. Tin individuahealthy conmend a neewhiplash as

Acknowled

The authDr. Dion Scoware.

Appendix A

Supplemthe online v

References

Austroads, 20Drivers: MGuidelinesSydney.

Barry, C.J., SmM., Trayneperforman

Brooks, J.O., GoJr., W.C., Oing simula

Crizzle, A.M., CAssociatiosons with

Dallalba, P.T., Sterling, M.M., Treleaven, J.M., Edwards, S.L., Jull, G.A., 2001. Cervicalrange of motion discriminates between asymptomatic persons and those withwhiplash. Spine (Phila Pa 1976) 26 (19), 20902094.

Duong, B.D., Kerns, R.D., Towle, V., Reid, M.C., 2005. Identifying the activities affectedby chronic nonmalignant pain in older veterans receiving primary care. J. Am.

atr. Soilson

rted iErdfelowerv. Resck, K.arwoong MeTexasJ.L., O

disabon tecrders.E.R., 2cervic221.

Zaworical li., Mcc

3613 Jang, ationv. 5 (2, M.C.

on at, E., Ch

treatmdle-ag., Pooity of

iminarli, R.A.tti, M.elf-reN.W.,ormanp. Th, Miro

of the.K., Laing fug and

F., Grilash ous heira, Mt of pr. NeurM.J., Ju

persiholog, B.P.lash

somat297.., 200

and Pdelphrown,redictarch Can, Oeen o

c Inj. M., Ju-term

M., Juor sys

M., Kal chai, H., Cty of tle wit

, 1774i, H., Jnic w. Medon, Deen c

pain.ions

y has limitations. This study did not investigate driving-ormance with prolonged driving, which patients withD report as problematic. They also report a reductionation while driving (Takasaki et al., 2011). Poorer abil-ded attention tasks might be more evident over longeres. Nevertheless, it would be difcult to investigateted performance over longer times especially in theesidential or CBD areas in light of the number of subjectsenced motion sickness in these scenarios. Further tech-ogress will be needed to counter this problem. We usedtical events from potentially innumerable critical situa-

limited testing time, and more or alternate events mayarranted to detect differences between WAD subjects

controls. Notably, the skill of reversing was not testedl reasons, despite it being one of the most troublesomes for persons with chronic WAD (Takasaki et al., 2011).fferent experimental setting is required to investigatee in reversing tasks. Finally, the sample size tested wast a larger sample size could allow multiple comparisonsen control subjects, WAD subjects without driving dif-WAD subjects with driving difculty), and provide usr understanding of driving in individuals with chronic

ion

dy compared driving-related performance between with chronic WAD and healthy individuals in freeway,nd CBD scenarios over 15 min of driving in an advancedhis study determined that driving-related performancels with chronic WAD was not substantially different totrol subjects and there appears to be no basis to recom-d for tness to drive assessment in persons with chronicsociated disorders.

gments

ors wish to acknowledge Mr. Wolbert Van den Hoorn,tt and Dr. Gregoire Larue for programing analyses soft-

. Supplementary data

entary material related to this article can be found, inersion, at http://dx.doi.org/10.1016/j.aap.2013.07.033.

03. Assessing Fitness to Drive for Commercial and Private Vehicleedical Standards for Licensing and Clinical Management Guidelines:

and Standards for health Professionals in Australia, 3rd ed. Austroads,

ith, D., Lennarson, P., Jermeland, J., Darling, W., Stierman, L., Rizzo,lis, V.C., 2003. The effect of wearing a restrictive neck brace on driverce. Neurosurgery 53 (1), 98101.odenough, R.R., Crisler, M.C., Klein, N.D., Alley, R.L., Koon, B.L., Logan

gle, J.H., Tyrrell, R.A., Wills, R.F., 2010. Simulator sickness during driv-tion studies. Accid. Anal. Prev. 42 (3), 788796.lassen, S., Winter, S.M., Silver, W., Lafranca, C., Eisenschenk, S., 2012.ns between clinical tests and simulated driving performance in per-epilepsy. Epilepsy Behav. 23 (3), 241246.

GeriFan, A., W

repoFaul, F.,

cal pBeha

FitzpatriR., HRatiThe

Hoving, neckitatidiso

Howell, and 211

Hunter-phys

Jones, J.G(5),

Ku, J.H.,validBeha

Kuhajdapain

Lagardepainmid

Lew, H.Lvalidprel

MarottoTinefor s

Mullen, perfOccu

Nieto, R.ness

Nilsen, HDrivusin

hberg,whiptane

Pais-Viemenpain

Pereira, withpsyc

Radanovwhipand 281

Rizzo, MplesPhila

Rudin-Bto PRese

ShechtmbetwTraf

Sterling,long

Sterling,mot

Sterling,logic

Takasakabilipeop(10)

TakasakchroPhys

Thompsbetwneckc. 53 (4), 687694., K.G., Acharya, M., Cranney, A., Buenger, U., Marshall, S., 2012. Self-

ssues with driving in patients with chronic pain. PM&R 4 (2), 8795.der, E., Lang, A.G., Buchner, A., 2007. G*power 3: a exible statisti-

analysis program for the social, behavioral, and biomedical sciences.. Methods 39 (2), 175191., Wooldridge, M., Tsimhoni, O., Collins, J., Green, P., Parma, K., Koppa,d, D., Krammes, R., Poggioli, B., 2000. Alternative Design Consistencythods for Two-Lane Rural Highways. Texas Transportation Institute,

A&M University System, College Station.leary, E.F., Niere, K.R., Green, S., Buchbinder, R., 2003. Validity of theility index, northwick park neck pain questionnaire, and problem elic-hnique for measuring disability associated with whiplash-associated

Pain 102 (3), 273281.011. The association between neck pain, the neck disability indexal ranges of motion: a narrative review. J. Can. Chiropr. Assoc. 55 (3),

ski, K.M., 1990. T-intersection simulator performance of drivers withmitations. Transport. Res. Rec. 1281, 1115.ann, J., Lassere, M.N., 1991. Driving and arthritis. Br. J. Rheumatol. 3064.D.P., Lee, B.S., Lee, J.H., Kim, I.Y., Kim, S.I., 2002. Development and

of virtual driving simulator for the spinal injury patient. Cyberpsychol.), 151156.

, Thorn, B.E., Klinger, M.R., Rubin, N.J., 2002. The effect of headachetention (encoding) and memory (recognition). Pain 97 (3), 213221.astang, J.F., Lafont, S., Coeuret-Pellicer, M., Chiron, M., 2005. Pain andent were associated with trafc accident involvement in a cohort of

ed workers. J. Clin. Epidemiol. 58 (5), 524531.le, J.H., Lee, E.H., Jaffe, D.L., Huang, H.C., Brodd, E., 2005. Predictive

driving-simulator assessments following traumatic brain injury: ay study. Brain Inj. 19 (3), 177188., Richardson, E.D., Stowe, M.H., Miller, E.G., Brass, L.M., Cooney Jr., L.M.,E., 1998. Development of a test battery to identify older drivers at riskported adverse driving events. J. Am. Geriatr. Soc. 46 (5), 562568.

Weaver, B., Riendeau, J.A., Morrison, L.E., Bdard, M., 2010. Drivingce and susceptibility to simulator sickness: are they related? Am. J.er. 64 (2), 288295., J., Huguet, A., 2008. Disability in subacute whiplash patients: useful-

neck disability index. Spine (Phila Pa 1976) 33 (18), E630E635.ndr, N.I., Kaasa, S., Jenssen, G.D., Fayers, P., Borchgrevink, P.C., 2011.nctions in a video simulator in chronic non-malignant pain patientsnot using codeine. Eur. J. Pain 15 (4), 409415.p, H., Wiklund, U., Sterner, Y., Karlsson, J.S., Gerdle, B., 2003. Chronicassociated disorders and neck movement measurements: an instan-lical axis approach. IEEE Trans. Inf. Technol. Biomed. 7 (4), 274282.

., Mendes-Pinto, M.M., Lima, D., Galhardo, V., 2009. Cognitive impair-efrontal-dependent decision-making in rats after the onset of chronicoscience 161 (3), 671679.ll, G.A., Treleaven, J.M., 2008. Self-reported driving habits in subjects

stent whiplash-associated disorder: relationship to sensorimotor andic features. Arch. Phys. Med. Rehabil. 89 (6), 10971102., Sturzenegger, M., Di Stefano, G., 1995. Long-term outcome afterinjury. A 2-year follow-up considering features of injury mechanismic, radiologic, and psychosocial ndings. Medicine (Baltimore) 74 (5),

4. Safe and undafe driving. In: Rizzo, M., Eslinger, P. (Eds.), Princi-ractice of Behavioral Neurology and Neuropsychology. W.B. Saunders,ia, pp. 197220.

C., Williamson, A., Lenne, M., 2009. Can Driving Simulation be Used Changes in Real-World Crash Risk? Monash University Accidententre, Clayton.

., Classen, S., Awadzi, K., Mann, W., 2009. Comparison of driving errorsn-the-road and simulated driving assessment: a validation study.

Prev. 10 (4), 379385.ll, G., Kenardy, J., 2006. Physical and psychological factors maintain

predictive capacity post-whiplash injury. Pain 122 (12), 102108.ll, G., Vicenzino, B., Kenardy, J., Darnell, R., 2003a. Development of

tem dysfunction following whiplash injury. Pain 103 (12), 6573.enardy, J., Jull, G., Vicenzino, B., 2003b. The development of psycho-nges following whiplash injury. Pain 106 (3), 481489.hien, C.-W., Johnston, V., Treleaven, J., Jull, G., 2012. Validity and reli-he perceived decit questionnaire to assess cognitive symptoms inh chronic whiplash associated disorders. Arch. Phys. Med. Rehabil. 931781.ohnston, V., Treleaven, J., Pereira, M., Jull, G., 2011. Driving with ahiplash-associated disorder: a review of patients perspectives. Arch.. Rehabil. 92 (1), 106110..P., Urmston, M., Oldham, J.A., Woby, S.R., 2010. The associationognitive factors, pain and disability in patients with idiopathic chronic

Disabil. Rehabil. 32 (21), 17581767.


Treleaven, J., Jull, G., Lowchoy, N., 2005. Smooth pursuit neck torsion test inwhiplash-associated disorders: relationship to self-reports of neck pain anddisability, dizziness and anxiety. J. Rehabil. Med. 37 (4), 219223.

Veldhuijzen, D.S., Van Wijck, A.J.M., Wille, F., Verster, J.C., Kenemans, J.L., Kalk-man, C.J., Olivier, B., Volkerts, E.R., 2006. Effect of chronic nonmalignant painon highway driving performance. Pain 122 (12), 2835.

Vernon, H., Mior, S., 1991. The neck disability index: a study of reliability and validity.J. Manipulative Physiol. Ther. 14 (7), 409415.

Yuen, H.K., Gillespie, M.B., Barkley, R.A., Day, T.A., Bandyopadhyay, D., Sharma, A.K.,2007. Driving performance in patients with cancer in the head and neck region:a pilot study. Arch. Otolaryngol. Head Neck Surg. 133 (9), 904909.

Assessment of driving-related performance in chronic whiplash using an advanced driving simulator1 Introduction1.1 Purpose

2 Methods2.1 Study design2.2 Subjects2.3 Driving simulator2.4 Driving scenarios2.4.1 Freeway scenario2.4.2 Residential scenario2.4.3 CBD scenario

2.5 Outcome measures2.5.1 The SPI2.5.2 Divided attention tasks

2.6 Statistics

3 Results4 Discussion4.1 Limitations

5 ConclusionAcknowledgmentsAppendix A Supplementary dataAppendix A Supplementary data

Assessment of Driving-related Performance in Chronic Whiplash Using

Documents

Transcript of Assessment of Driving-related Performance in Chronic Whiplash Using