Tb

VAa

b

c

d

e

f

g

Mh

i

a

ARRA

KOIDEMSI

1

t

c(J

0d

Accident Analysis and Prevention 49 (2012) 486– 492

Contents lists available at SciVerse ScienceDirect

Accident Analysis and Prevention

jo ur n al hom ep a ge: www.elsev ier .com/ locate /aap

he effect of a low and high resource intervention on older drivers’ knowledge,ehaviors and risky driving

. Jonesa,∗, A. Gielenb, M. Baileyc, G. Rebokd, C. Agnesse, C. Soderstromf, J. Abendschoen-Milanig,

. Liebnoh, J. Gaines i, J. Parrish i

Johns Hopkins Bloomberg School of Public Health, 624 N. Broadway, Suite 544, Baltimore, MD 21205-1996, United StatesJohns Hopkins Center for Injury Research and Policy, 624 N. Broadway, Room 554, Baltimore, MD 21205, United StatesJohns Hopkins Bloomberg School of Public Health, Center for Injury Research & Policy, 624 N. Broadway, Room 529, Baltimore, MD 21205, United StatesThe Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University, 624 N. Broadway, Baltimore, MD 21205, United StatesUniversity of Maryland School of Pharmacy, Department of Pharmacy Practice and Science, 20 North Pine Street, Room 446 South, Baltimore, MD 21201, United StatesMaryland Motor Vehicle Administration, 6601 Ritchie Highway, NE, Room 155, Glen Burnie, MD 21062, United StatesShock, Trauma and Anesthesiology Research – Organized Research Center, Charles “McC” Mathias National Study Center for Trauma and EMS, University of Maryland School ofedicine, 110 South Paca Street, Rm 4-S-128, Baltimore, MD 21201, United StatesMd. Police and Correctional Training Commissions, 6852 4th Street, Sykesville, MD 21784, United StatesThe Erickson Living Foundation, 701 Maiden Choice Drive, Catonsville, MD 21228, United States

r t i c l e i n f o

rticle history:eceived 21 October 2011eceived in revised form 12 March 2012ccepted 15 March 2012

eywords:ld Driver

nterventionsriverlderlyotor vehicle

afetynjury prevention

a b s t r a c t

This study aimed to compare an in-class Seniors on the MOVE (Mature Operators Vehicular Education)interactive multi-session driving curriculum with a self-guided MOVE curriculum for older adults. Usinga two group randomized design, we sought to determine if there are between-group differences in olderdrivers’ knowledge and safety behaviors among participants. Forty-four participants with an averageage of 79 years (SD = 7.1) were randomly assigned to the original MOVE program (SOM-A) or a lowerresource (SOM-B) self-guided intervention. SOM-A is a four session program designed to improve olderdrivers safety knowledge and better understand skills for safer driving. SOM-B is a self-guided programwith one required in-class session and one optional session. Subsequent to completion of both curricula,participants were offered CarFit, a comprehensive check of how well a senior driver and their vehiclework together. Baseline, post-intervention and 6-month follow up questionnaires were completed byparticipants. We found significant differences (p = .01) in the mean driving safety knowledge scores whencomparing participants in SOM-A (3.7, SD 2.0) to those in SOM-B (0.87, SD 2.6). With regard to behavioraloutcomes, we focused on always wearing a seatbelt, talking with a health care provider about drivingability, and sitting 10–12 inches from the steering wheel. The vast majority of participants reportedalways wearing their seat belts (SOM-A 100%, SOM-B 92%, p = 1.0), and very few reported talking with theirdoctors (SOM-A Baseline – 0%, Follow up 1 – 0%, p = n/a). Mean behavior change scores for participants

sitting 10–12 inches from the steering wheel were significantly more likely among SOM-A (mean = .65,SD = .5) participants than those in SOM-B (mean = .29, SD = .5, p = .01) at first follow-up. Taken together,these findings suggest that the more intensive program is more effective and that driving safety programsfocused on behaviors to self evaluate driving abilities continue to be needed to help older drivers remain

ge. Th

safer on the road as they apotential.

. Introduction

An increase in life expectancy has led to substantial growth inhe number of older Americans (60+ years), and thus the number

∗ Corresponding author. Tel.: +1 410 502 9932; fax: +1 410 955 7241.E-mail addresses: vjones@jhsph.edu (V. Jones), agielen@jhsph.edu (A. Gielen),

soderstrom@mdot.state.md.us (C. Soderstrom), jmilani@som.umaryland.eduJ. Abendschoen-Milani), jean.gaines@erickson.com (J. Gaines),ohn.Parrish@erickson.com (J. Parrish).

001-4575/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.oi:10.1016/j.aap.2012.03.021

e involvement of health care providers in such efforts may be an untapped

© 2012 Elsevier Ltd. All rights reserved.

who drive on our roadways (National Center for Health Statistics,2007). Of all drivers involved in fatal crashes in 2008, approximately27% were aged 55 years or more (NSC, 2009). Older driver crashestend to be related to driver inattention or slowed perception andresponse, lack of recognition of these changes (often as the resultof decline in cognitive function), prescription medication use, poorvision, and other medical conditions (National Center for Health

Statistics, 2007).

While programs exist to sustain driving privileges for olderdrivers, many of them focus on only one factor such as vision(Subzwari et al., 2009; Staplin and Dinh-Zarr, 2006; Eby et al.,

s and

2ddBirt

ereBPdpaaVeiee

spifoes2tspid

coTwcovAcioarcst(b

2

2

rcml

V. Jones et al. / Accident Analysi

000b). Interventions with the strongest evidence for improvingriving performance include both an educational and an on-roadriving component (Korner-Bitensky et al., 2009; Kua et al., 2007).ecause on-road programs are costly and not feasible to implement

n support of large cohorts of older drivers for all communities, aemaining challenge for older driver programs is to find interven-ions that are effective but less resource intensive.

Strategies to help older drivers assess their driving abilities aressential in preserving safe driving abilities. In a recent systematiceview of the effectiveness of older driver training programs littlevidence was found to support educational only programs (Korner-itensky et al., 2009). Also, two different evaluations of the 55 Aliverogram created by the AARP found that there was no impact onriving behavior, both when there was an on road-training com-onent (Bedard et al., 2008) and when there was not (Nasvadind Varvik, 2007). These authors suggest that their reported neg-tive findings may result from sample selection bias (Nasvadi andarvik, 2007) and issues with program implementation (Bedardt al., 2004). While an on-road driving component is optimal, its relatively costly and may not be feasible in all situations. Anducation only program that is comprehensive, effective, and costfficient would be ideal. Yet, no such program currently exists.

A curriculum that is comprehensive and improves knowledge,kills and performance by addressing perceptions, current drivingractices and environmental influences is important for chang-

ng driving behavior (Lindstrom-Forneri et al., 2010). Research hasound that older drivers commonly modify their driving behaviorn their own (Ackerman et al., 2011; Donorfio et al., 2009; Rudmant al., 2006), however, modifications vary by sex, age and healthtatus (Sullivan et al., 2011; Donorfio et al., 2009; D’Ambrosio et al.,008). Strategies to help older drivers assess their driving abili-ies are essential in preserving safe driving abilities. The presenttudy sought to improve upon the existing published educationalrograms by evaluating an intervention focused specifically driv-

ng related knowledge and behaviors with a sample of current olderrivers who are not found to be at high risk for driving impairment.

The Seniors on the M.O.V.E (Mature Operators Vehicular Edu-ation) program was designed by a multi-disciplinary committeef traffic safety experts experienced in working with older drivers.heir knowledge of older drivers’ and contributory crash factorsas integrated into the design of the structure and content of the

urriculum. The program incorporated sessions that the committeeriginated, along with selected components of pre-existing inter-entions (e.g., Roadwise Review and CarFit, both programs fromAA – American Automobile Association) developed to provideomprehensive education that would help older drivers to mod-fy their driving behaviors for improved safety. The program wasriginally pilot tested in partnership with selected senior centersnd senior residential properties, with some promising short-termesults (Jones et al., 2011). The specific aim of this study was toompare the impact of a multi-session interactive, expert-led ver-ion of the training program (Seniors on the MOVE – Version-A)o a self-guided and less resource intensive version of the programSeniors on the Move – Version-B) on older drivers’ knowledge andehavior pertaining to driving.

. Methods

.1. Participants

Between January and May 2009, drivers 60 years and older were

ecruited from a residential retirement community and a seniorenter in Baltimore County, Maryland. To be eligible for enroll-ent in this study, drivers needed to have a valid Maryland driver’s

icense, report driving at least once a week, and be available to

Prevention 49 (2012) 486– 492 487

attend the program sessions. Individuals who required adaptivedevices to operate their motor vehicles were excluded from partic-ipating in this study.

Participants who initially screened eligible for the study werethen administered the Mini Mental State Exam (Folstein et al., 1975)(MMSE) to assess their competency to provide informed consent.No one was excluded from the study based on his or her MMSEresults.

Eighty people were initially screened for eligibility into thestudy, of which 67 went on to complete the full screening battery.Of the 13 people who did not complete the full battery, 4 werefound to be ineligible because they did not drive at least once aweek; 8 refused to participate; and 1 did not want to completethe screening. Of the 67 who completed the screening battery, 11withdrew due to various reasons (see Fig. 1), and 9 were deter-mined to be at the high-risk impairment level and were thereforenot eligible for the intervention. Forty-seven remained and wererandomized into either SOM-A or B. Three participants withdrewfrom the study after randomization for various reasons (see Fig. 1),providing a final sample size of 44 (SOM-A, n = 20; SOM-B, n = 24).All but one participant (in the SOM-B) completed both follow upsurveys.

2.2. Design

We used a randomized design to test the efficacy of two pro-grams. Immediately following informed consent, individuals wereadministered a risk impairment assessment to determine “high”risk participants as described elsewhere (Jones et al., 2010) by iden-tifying those who had serious decrements on the Motor-Free VisualPerception Test (MVPT) (Colarusso and Hammill, 1972), Trail Mak-ing Test B and the Useful Field of View (UFOV) (Owsley et al., 1998,2004; McGwin et al., 2000; Ball et al., 2006). Those scoring abovethe cut points for each test were excluded from the study, were pro-vided information about driving rehabilitation services and advisedto consult with their primary health care provider (Jones et al.,2010). The remaining participants were then randomly assigned toSeniors on the MOVE Version A or B (SOM-A or SOM-B) by randomnumber generator. There were three people randomly assigned toSOM-A who had to switch to SOM-B because they had schedulingconflicts for participating in SOM-A, but not SOM-B. One personwas randomly assigned to SOM-B, but had to be switched to SOM-Abecause of scheduling conflicts.

2.3. The intervention

2.3.1. SOM-AThe SOM-A program, taught by experts in highway safety, occu-

pant protection, and medication management, consisted of fourtwo-hour sessions (Jones et al., 2011) and CarFIT (AAA, 2009).Content was informed by an assessment of current available pro-grams for mature drivers and contributing factors for mature drivercrashes as identified in the literature (Edwards et al., 2008; Ebyet al., 2000a,b; Ball et al., 1998) and revised based on the initialpilot test (Jones et al., 2011). In our previous study, we found that(a) driving exposure (driving distance and driving conditions) is oneoutcome that is more difficult to change and (b) more intermediatebehaviors to improve driving safety and address driving abilitiesshould also be considered (Jones et al., 2011). Such intermedi-ate behaviors include using necessary safety devices (seat belts),proper seating positions to reduce air bag related injuries, and con-sulting with medical providers to understand the impact of aging

and medications on driving.

The curriculum provided an interactive, multi-session trainingthat is limited to 40 participants to maximize interaction and learn-ing. Classes were held sequentially, for a total of 4 weeks with

488 V. Jones et al. / Accident Analysis and Prevention 49 (2012) 486– 492

tudy

odvoslicwdtmsscaoa

Fig. 1. S

ne 2 h training session being held each week. All sessions wereesigned to complement and build upon another. Materials pro-ided to the participants during the sessions included a CD ROMf the “Roadwise Review” and paper copies of the PowerPoint pre-entation. Because some participants could not attend all of theive class sessions, each one was videotaped, and participants werenvited to watch the sessions they missed. For participants whoould not attend a session in person they were given the option ofatching the videotaped session in a conference room in the resi-ential community center with a facilitator, who also attended allhe live classes and was present to answer questions and review

aterial. Those who watched a video instead of attending live ses-ion for SOM-A ranged from 10% to 25% of participants for eachession. Of the thirteen people who watched a video session, two

ould not attend the facilitated view class and viewed the videotapet home. One participant was unable to participate in the live classr review the video (session 4 SOM-A). Descriptions of the sessionsre listed below.

design.

The four sessions were:

(1) “Roadwise Review” – a driving abilities assessment CD ROMprogram – developed and presented by the AAA Foundation forTraffic Safety (Staplin and Dinh-Zarr, 2006);

(2) “Road Smart” – developed by a special committee from the Cen-tral Maryland Regional Safe Communities Center, presented bytrained driving educators;

(3) “Safer Driving: Me and My Vehicle” – developed and presentedby personnel of the Maryland Department of Health and MentalHygiene’s child passenger safety program: Kids In Safety Seats(KISS); and

(4) “Being Medwise to Stay Roadwise” Medication Management –developed and presented by the Peter Lamy Center for Drug

Therapy and Aging at the University of Maryland School ofPharmacy.

Participants were also offered CarFIT (described below) at theconclusion of the in-class sessions.

s and

2

prCwoaasiBe(dwtww

2

etaevtmiipido

2

qtpbp

Ttattsiadsai

2

pFspo

V. Jones et al. / Accident Analysi

.3.2. SOM-BThe SOM-B curriculum was designed as a self-guided and com-

ressed version of SOM-A. It consisted of three sessions (oneequired, and two optional of which one was in-class and the otherarFit). The first session, which was required, provided participantsith a two-hour class highlighting the importance of road safety,

ccupant protection, maneuvering, and medication managements it relates to operating a vehicle. Participants were also introducednd encouraged to use the CD ROM “Roadwise Review.” The secondession (an optional session) was for the participants to reconvenef they had questions about the first session or “Roadwise Review.”oth the first and second sessions were delivered by driving safetyducators who also taught the SOM-A sessions. The third sessionoptional) was the CarFit (AAA, 2009) program, which was con-ucted by the same trained technicians and health professionalsho provided it for the SOM-A participants. For SOM-B, all par-

icipants attended the required first session in person. One personho wanted to attend the optional second session and could not,atched the video in their home (4%).

.3.3. CarFitCarFit is an educational program created by the American Soci-

ty on Aging and developed in collaboration with AAA, AARP andhe American Occupational Therapy Association (AAA, 2009). It isn in-car assessment tool to help older drivers understand how theffects of aging change the way they fit in their vehicle. It also pro-ides information and materials on community-specific resourceshat could enhance their safety as drivers, and/or increase their

obility in the community. The driver’s vehicle is stationary dur-ng this assessment, and there is no on-the-road driving evaluationnvolved. CarFit was an optional in-vehicle educational activity forarticipants in both SOM-A and SOM-B and was offered after the

n-class sessions were completed. While CarFit is not an on-roadriving test, it does provide participants with an objective reviewf their placement in their vehicle (AAA, 2009).

.4. Data collection

At the initial session for each intervention group, a baselineuestionnaire was completed. At a “graduation” ceremony afterhe completion of the SOM (A and B) programs, both groups com-leted their first post-intervention assessment. Six months after theaseline, the second post-intervention assessments were mailed toarticipants to complete and return in an enclosed envelope.

All data were collected using paper and pencil questionnaires.he variables assessed included participants’ demographic charac-eristics, driving behaviors, knowledge of material taught in class,nd opinions about the training sessions (including evaluation ofhe presenters, location of the class, and overall satisfaction withhe program). Demographic variables used in this research includedex, age and marital status. Three driving behaviors – sitting 10–12nches from the steering wheel, talking to health care providersbout medication effects, and wearing seatbelts – were assessedichotomously (having done the behavior versus did not). To mea-ure driving knowledge, 15 items were developed by the authors tossess specific details taught during M.O.V.E. sessions. The specifictems used for these measures are provided in Section 3.

.5. Data analysis

Data were entered and verified for accuracy using a passwordrotected computer, and analyses were completed using SPSS 17.0.

irst, the results of the screening and enrollment are presented. Chi-quared tests were used to determine differences between SOM-Aarticipants and SOM-B participants on demographic factors. Sec-nd, results from measurements taken at baseline, follow up 1 and

Prevention 49 (2012) 486– 492 489

follow up 2 for self reported driving knowledge and behaviors arecompared for each study group using paired t-tests. Next, we com-pared changes in scores for each study group (differences betweenthe baseline and first and second evaluations) using t-tests. Theseanalyses were controlled for CarFit participation. The Johns HopkinsInstitutional Review Board reviewed and approved the project.

3. Results

Participants in both SOM-A and SOM-B were well educated,with everyone having completed high school, 43% completing col-lege, and 30% having a graduate degree. The intervention groupswere fairly evenly divided in terms of gender and marital status,with 57% being female and 55% married. The mean age of partici-pants in both groups at baseline was 79 years (SD = 7.1). A majorityreported transporting others at least once per week (61%). Anal-ysis of demographic data on the participants in SOM-A (n = 20)and B (n = 24) revealed that there were no statistically significantdifferences between the groups on any of these variables.

3.1. Driving knowledge and behaviors at baseline, T1, and T2

Responses to the individual knowledge items are presented inTable 1, and there were significant differences in SOM-A partici-pants between the baseline and follow up 1 for items #4, #7, #9 and#12 and between baseline and follow up 2 on item #9. Participantsin SOM-B demonstrated significant differences between baselineand follow up 2 only on item #12. Mean total correct knowledgescores were computed (possible score range 0–15) and change vari-ables were calculated as follow up 1 score minus the baseline scoreand follow up 2 score minus the baseline score.

When comparing the changes within each group, there weresignificant improvements between the SOM-A group baselineresponses (8.4, SD = 1.7) and follow up 1 (12.2, SD = 1.8) responses(p = .01), and between baseline and follow up 2 (12.1, SD = 1.7)responses (p = .01). The SOM-B group did not have significantchanges (p = .11) in their knowledge scores comparing baseline (9.4,SD = 2.1) to follow up 1 (10.3, SD = 3.0); however, there were signifi-cant improvements between their knowledge scores from baselineto follow up 2 (11.9, SD = 1.8) responses (p = .01).

Table 2 provides a comparison of the mean total knowledgescores by intervention group. When comparing baseline to followup 1, the change in the SOM-A group (3.7, SD = 2.0) was signifi-cantly greater than that in the SOM-B group (.87, SD = 2.6) (p = .01).However, there was no significance difference in the change whencomparing the difference between SOM-A and SOM-B from base-line to follow up 2, although the SOM-A score remained the sameindicating no diminution in the knowledge gains from T1 to T2.

Changes in responses at baseline vs. follow up 1 of self-reporteddriving behaviors of sitting 10–12 inches from the steering wheelwas the only significant behavior change detected and both groupsincreased (SOM-A Baseline – 20%, Follow up 1 – 80%; p = .001) (SOM-B Baseline – 42%, Follow up 1 – 71%; p = .01). Sitting 10–12 inchesfrom the steering wheel remained statistically significant whencomparing baseline to follow up 2 for both groups (SOM-A Baseline– 20%, Follow up 2 – 80%; p.001) (SOM-B Baseline – 42%, Follow up2 – 91%; p = .01). Talking with a health care provider about med-ications (SOM-A Baseline – 0%, Follow up 1 – 0%, p = n/a) (SOM-BBaseline – 2%, Follow up 1 – 2%; p = 1.0) and wearing a seatbeltwhen driving (SOM-A Baseline – 100%, Follow up 1 – 100%; p = 1.0)(SOM-B Baseline – 92%, Follow up 1 – 92%; p = 1) did not change in

either group.

When comparing the SOM-A group behavior change scores toSOM-B behavior change scores, there was a significant differencefor sitting 10–12 inches from the steering wheel at baseline vs.

490 V. Jones et al. / Accident Analysis and Prevention 49 (2012) 486– 492

Table 1Self-reported driving knowledge comparisons by intervention group comparing Baseline to Follow up at Follow up 1 and Baseline to Follow up at Follow up 2 (n = 44) (numberof total correct answers for each question).

Knowledge items SOM-A SOM-B

No Baselinen = 20

Follow up 1n = 20

p valuea Follow up 2n = 20

p valueb Baselinen = 24

Follow up 1n = 24

p valuea Follow up 2n = 23

p valueb

1 Have good eye sight 19 20 n/ac 20 n/ac 24 23 n/a 22 n/ac

2 Strong legs 9 13 .29 10 1.0 17 17 1.0 14 .513 Head and neck flexibility 15 18 .40 20 n/a 17 20 .71 22 .124 The proper placement of the head

restraint13 19 .03 17 .13 16 20 .13 17 .68

5 Which driving maneuvers are mostlikely to lead to a crash for older adults

12 16 .28 14 .69 17 17 1.0 17 1.0

6 Safe stopping distance in case of anemergency

4 7 .45 2 .70 8 12 .35 6 1.0

7 The time for checking tire pressure 9 16 .04 14 .65 7 8 1.0 10 .208 How often a health care provider

should review your medications15 17 .60 18 .41 22 20 .73 21 1.0

9 Definition of moderate drinking forolder adults

7 18 .01 14 .04 14 13 1.0 15 1.0

10 Individuals can monitor the impact ofdrinking on their behavior

14 17 .51 17 .53 17 19 .61 16 .61

11 OTC and alcohol can be combined 11 17 .11 16 .10 12 15 .42 17 .4212 Muscle relaxers do not affect driving 12 20 .01 18 .11 16 14 .80 19 .0313 Diabetes is linked to increased crashes 5 10 .20 11 .12 8 8 1.0 12 .2314 Properly adjusted mirrors will

eliminate blind spots16 17 1.0 17 1.0 21 20 1.0 19 1.0

15 It’s okay to put the should strap of aseat belt behind you

20 18 1.0 20 1.0 22 22 1.0 23 1.0

a McNemar’s Test – Baseline vs. Follow up 1 comparing percent mean correct.b McNemar’s Test – Baseline vs. Follow up 2 comparing percent mean correct.c Two by two tables with cells that have a value of zero do not yield a p value.

Table 2Comparing changes in mean total knowledge scores between intervention groups.

SOM-AMean (SD) SOM-BMean (SD) p valuea

Change in Knowledge ScoreBaseline vs. Follow up 1

3.7 (2.0) .87 (2.6) .01

Change in Knowledge ScoreBaseline vs. Follow up 2

3.7 (2.6) 2.5 (2.3) .11

a Independent sample t-test.

Table 3Comparing changes in mean behavior scores between intervention groups.

Behaviors SOM-AMean (SD)

SOM-BMean (SD)

p valuea

Sit 10–12 inches from steering wheelBaseline vs. Follow up 1

.65 (.5) .29 (.5) .01

Sit 10–12 inches from steering wheelBaseline vs. Follow up 2

1.05 (.6) 1.29 (.7) .17

Talk with health care provider about medication effectsBaseline vs. Follow up 1

.00 (.0) −.04 (.5) .58

Talk with health care provider about medication effects .05 (.2) .17 (.4) .38

fwliS

4

oedci

Baseline vs. Follow up 2

a Independent sample t-test.

ollow up 1 (p = .01); however, this difference was not significanthen comparing the differences in the groups for baseline vs. fol-

ow up 2 (p = .17). There was no statistically significance differencen the other driving behavior scores when comparing SOM-A andOM-B (Table 3).

. Discussion

This study offers some important new insights into the impactf a driving education program for older drivers. Increasing knowl-

dge has been used as a common outcome in evaluating olderriver educational interventions, suggesting that older drivers areapable of learning how to maintain or improve their driving abil-ties to prevent declines in driving performance associated with

age-related changes (Korner-Bitensky et al., 2009; Kua et al., 2007;Owsley et al., 2003). In summary, our results suggest that, amongSOM-A participants, knowledge gains were observed at both fol-low up 1 and follow up 2 relative to baseline scores. These findingsare generally consistent with those generated by previous research.Among SOM-B participants, knowledge gains were observed onlyat follow up 2. Comparisons between SOM-A to SOM-B suggest thatthere was a statistically significant difference in knowledge gains, infavor of SOM-A, from baseline to follow up 1. SOM-A participantsincreased their knowledge scores an average 3.7 points, whereas

SOM-B participants increased their knowledge scores by less thanone point (0.87). Overall, correct knowledge scores were modestlyhigh at baseline for both groups (8.4 in SOM-A and 9.4 in SOM-B,out of 15 possible points), suggesting that there continues to be a

s and

niatp

t(rItsrbssfifpwif

etttfnhtatprWipmiaudtd(ndmtlN2laafgmrAfiaaw

V. Jones et al. / Accident Analysi

eed to communicate important messages in support of safe driv-ng by older adults. Items that most participants did not know, evenfter the interventions (e.g., safe stopping distances, when to checkires) can suggest areas that need to be better addressed in futurerograms.

Our findings did not support previous research that suggestedhere were differences in driving behaviors based on sex or ageSullivan et al., 2011; D’Ambrosio et al., 2008). These findings couldeflect the sample selection or the sample size for the current study.nclusion requirements of individuals participating required themo drive one or more times a week and they were disqualified fromtudy participation if they were found at screening to be at “high”isk for unsafe driving. This selectivity in our sampling was felt toe necessary to increase the probability that our results are repre-entative of older adult drivers most likely to be on the road. Suchelectivity may be one alternative explanation for these discrepantndings. The limited study sample size also impacted our options

or statistical comparisons and reduced our ability to examine thesychometric properties of the measures. Although our findingsere modest, the utility of assessing driving knowledge and behav-

or should be further investigated with larger samples and testedor reliability and validity.

With regard to behavioral outcomes, we chose to focus thisvaluation on three key safety related issues: wearing a seatbelt,alking with a health care provider, and sitting 10–12 inches fromhe steering wheel. Both SOM-A and SOM-B increased the propor-ion of older drivers who improved their seating position at bothollow up 1 and follow up 2. However, the difference at T1 was sig-ificantly greater in SOM-A relative to SOM-B. SOM-A participantsad a 60 percentage point increase in the proportion who sat inhe correct position whereas SOM-B participants had a 29 percent-ge point increase. No differences were observed with regard tohe other two behavioral outcomes in either group at either timeeriod. For seat belts, this is due to the fact all but two participantseported at baseline that they wore their seat belt all of the time.

hether they in fact wear the seat belt properly and on every trips unknown. However, it has often been found that a sizable pro-ortion of the general population does wear seat belts; so, perhapsore attention should be allocated to other important safe driv-

ng related behaviors (such as talking with providers about drivingbilities) when working with a highly educated, older driving pop-lation. Although, a few participants in both SOM-A and SOM-Biscussed talking with health care providers about how medica-ions could affect their driving, it is not known why more did noto so as a result of the interventions. At baseline, only 4 participantsall in SOM-B) reported having talked to their provider. One expla-ation for not speaking to a medical provider is that participantsid not have a medical visit during the time between the assess-ents or if they did, there were more pressing medical concerns

hat took precedence. Another consideration is the relatively highevel of education attainment of this cohort. According the to 2003ational Assessment of Adult Literacy (NAAL) survey (Kutner et al.,006), adults with at least a high school education also had higher

evels of health literacy compared to adults with less education. In survey of community dwelling older adults, respondents with nowareness of “potentially driver impairing” medications (PDI) hadewer years of education (MacLennan et al., 2009). These data sug-est that MOVE participants with higher educational attainmentay believe they are able to make safer decisions regarding health-

elated driving behaviors without consulting a healthcare provider.lthough we do not have data to verify these possible explanations

or our results, the utility of discussing driving during routine med-

cal care is something that should be emphasized with both olderdults as well as with their care providers (Carr et al., 2010). Inddition, future educational interventions that involve older adultsith less educational attainment, health literacy and/or use of

Prevention 49 (2012) 486– 492 491

potentially driver impairing (PDI) medication may have anincreased impact. The importance of PDI’s as risk factors for crashesmakes it important that this topic receive more emphasis in thefuture.

The results of this study also confirm previous work that olderadults’ level of baseline knowledge about many aspects of driv-ing safety is very high (Jones et al., 2011). Nevertheless, gaps werefound in areas of safe driving knowledge, such as optimal sit-ting distance from the steering wheel and the potential effect ofmedication on driving abilities. Using the baseline and follow upmeasures, there is some indication that the original SOM-A inter-vention had a positive impact on senior’s knowledge and drivingbehavior when compared to the lower resource SOM-B interven-tion. Although SOM-B demonstrated knowledge and behavioralgains from baseline to first follow-up, the observed gains weregreater among participants in SOM-A relative to those in SOM-B.These results underscore the value of the more intensive ver-sion of the program relative to a less intensive version based onthe same material. It is also important to address the increase inmean knowledge scores for SOM-B at follow up 2, while therewas not a large difference at follow up 1. These changes could berelated to the timing of CarFit, between follow-up one and two.For example, the participation in CarFit may have provided SOM-B participants with new information when compared to SOM-Aparticipants.

Methodological limitations should be noted when interpret-ing study results. The sample was small and included those whoself-selected and were interested in participating in the interven-tion that was being offered. Results may not be generalizable toolder drivers in other settings. However, one strength of the studyincludes the randomized design and the high follow up rate withparticipants. The study also utilized an intervention that had beenmodified based on an earlier evaluation (Jones et al., 2011), whichallowed us to improve the quality of the program by revising mate-rial to be more pertinent to older drivers, as well as to expand theevaluation time frame. Another consideration in the design of thisstudy was the ability for participants to watch a video of an inter-vention session with a study staff member present in a communitysetting or alone in their home. While this provided a convenientstrategy to expose participants to the material taught in the ses-sions, we cannot be assured of the fidelity of the intervention for thefew participants who reviewed the sessions alone in their homes.

Future iterations of the SOM program may want to considerincluding more about on-the-road driving tips and less about childpassenger safety, which is a topic that does not apply to all olderadult drivers. The child passenger safety information was helpfuland well received for those to whom it applied, which suggests thatperhaps it be an optional session and advertised widely to thosewho drive young children. Future SOM iterations may also want toconsider the role of the family and the health care provider. Pre-vious research has found that older drivers consider the opinionsof others in making driving decisions (Coughlin et al., 2004). Veryfew older adults reported talking with anyone about their drivinghabits, but in our qualitative interviews with participants identifiedas high risk, many mentioned discussing the results with an adultchild or a spouse (Jones et al., in press). In addition, few of our par-ticipants, regardless of intervention or risk level, reported that theyhad considered talking with their health care provider about theirdriving. Based on the level of credibility that health care providershave and the frequency with which older individuals visit theirdoctors, the potential for them to influence an older individual’sdriving behavior holds substantial promise. As an enhancement to

SOM, engagement of health care providers should be explored. Inaddition, collecting more information related to medication man-agement practice may provide more insight into how older adultsview the role of their healthcare provider’s in safely managing their

4 s and

mSmt

5

eoioetiltamnttesostmghc

A

iHHgRMoaaCDsFMVP

R

AA

B

B

B

B

92 V. Jones et al. / Accident Analysi

edications. To develop these new and revised components of anyOM program would require formative research to determine theost effective means of communicating and engaging older drivers,

heir families, and their health care providers.

. Conclusion

The programs tested here drew on theory and practical experi-nce to create engaging safe driving educational opportunities forlder adults. Although this study had a small sample size which lim-ted the options for statistical analysis, the impact of the programsn knowledge was found to be statistically significant. However, thevaluated effects on reported behaviors were minimal. Only sittinghe appropriate distance from the steering wheel increased signif-cantly more in the intensive original SOM-A program than in theess intensive SOM-B program at immediate follow-up. Moreover,he SOM-A intervention utilized a highly skilled intervention staffnd required a large time commitment from participants, whichay reduce its future feasibility and sustainability in some commu-

ities. The self-guided intervention (SOM-B) showed less promisehan the SOM-A at increasing driving knowledge. Taken together,hese findings suggest that the more intensive program is moreffective. Yet, there continues to be an unmet critical need for driverafety programs that can demonstrate sustained impacts both onlder drivers’ knowledge and driving safety practices. Lastly, thistudy finds that a highly educated sample of older drivers is notalking with their health care provider about driving abilities and

edication implication for continued driving. Educational pro-rams need to improve an older person’s ability to discuss theirealth as it relates to driving safety and duration with their healthare provider.

cknowledgements

Funding for the evaluation was provided by the Erickson Liv-ng Foundation, the Maryland Department of Transportation, Stateighway Administration, Highway Safety Office, and the Johnsopkins Center for Injury Research and Policy (JHCIRP). We arerateful to these individuals for sharing their time and expertise:ichard Anderson, Ernest Lehr, Michelle Atwell, Tracy Whitman,yra Wieman, Claire Myer and Peter Moe. For their commitment

f time and energy we would like to thank to Katherine A. Marxnd Kasey L. Burke. We owe a special thanks for the resourcesnd expertise provided by the following organizations: Baltimoreounty Community Traffic Safety Program, Baltimore County Policeepartment, Catonsville Senior Center, Erickson Foundation, Erick-

on Living’s Charlestown Retirement Community, Mid-Atlanticoundation for Safety & Education, a non-profit affiliate of AAAid-Atlantic, Maryland Kids in Safety Seats (KISS), Maryland Motorehicle Administration and the University of Maryland School ofharmacy, Peter Lamy Center.

eferences

AA, 2009. CarFit: Helping Mature Drivers Find Their Safest Fit.ckerman, M., Crowe, M., Vance, D., Wadley, V., Owsley, C., Ball, K., 2011. The impact

of feedback on self-rated driving ability and driving self-regulation among olderadults. The Gerontologist 51, 367–378.

edard, M., Porter, M., Marshall, S., Isherwood, I., Riendeau, J., Weaver, B., Tuokko, H.,Molnar, F., Miller-Pogar, J., 2008. The combination of two training approachesto improve older adults’ driving safety. Traffic Injury Prevention 9, 70–76.

edard, M., Isherwood, I., Moore, E., Gibbons, C., Lindstrom, W., 2004. Evaluation ofa re-training program for older drivers. Canadian Journal of Public Health 95,259–299.

all, K., Owsley, C., Stalvey, B., Roenker, D., Sloane, M., Graves, M., 1998. Drivingavoidance and functional impairment in older drivers. Accident Analysis & Pre-vention 30, 313–322.

all, K.K., Roenker, D.L., Wadley, V.G., Edwards, J.D., Roth, D.L., McGwin, G., Raleigh,R., Joyce, J.J., Cissell, G.M., Dube, T., 2006. Can high-risk older drivers be

Prevention 49 (2012) 486– 492

identified through performance-based measures in a department of motor vehi-cles setting? Journal of the American Geriatrics Society 54, 77–84.

Carr, D.B., Schwartzberg, J.G., Manning, L., Sempek, J., 2010. Physician’s Guide toAssessing and Counseling Older Drivers, 2nd ed. National Highway Traffic SafetyAdministration, Washington, DC.

Colarusso, R., Hammill, D., 1972. Motor-free Test and Visual Perception. AcademicTherapy Publication, San Rafael, CA.

Coughlin, J., Mohyde, M., D’Ambrosio, L., Gilbert, J., 2004. Who drives the olderdriver decisions? Available at http://web.mit.edu/agelab/news events/pdfs/AgeLab driver decision.pdf. Accessed 1 April, 2011.

D’Ambrosio, L., Donorfio, L., Coughlin, J., Mohyde, M., Meyer, J., 2008. Gender dif-ferences in self-regulation patterns and attitudes towards driving among olderadults. Journal of Women and Aging 20, 256–282.

Donorfio, L., D’Ambrosio, L., Coughlin, J., Mohyde, M., 2009. To drive or not to drive,that isn’t the question—the meaning of self-regulation among older drivers.Journal of Safety Research 40, 221–226.

Eby, D.W., Shope, J.T., Molnar, L.J., 2000a. Driver Decision Workbook. University ofMichigan Transportation Research Institute.

Eby, D.W., Shope, J.T., Molnar, L.J., Vivoda, J.M., Fordyce, T.A., 2000b. Development ofan Older Driver Assessment (Rep. UMTRI No. 2000-04). University of MichiganTransportation Institute.

Edwards, J., Leonard, K., Lunsman, M., Dodson, J., Bradely, S., Myers, C., Hubble, B.,2008. Acceptability and validity of older driver screening with the Driving HealthInventory. Accident Analysis & Prevention 40, 1157–1163.

Folstein, M., Folstein, S., McHugh, P., 1975. “Mini-mental state.” A practical methodfor grading the cognitive state of patients for the clinician. Journal of PsychiatricResearch 12, 189–198.

Jones, V., Bailey, M., Rebok, G., Gielen, A., 2010. High versus low resource seniors onthe MOVE pilot study final evaluation report to the Erickson Foundation.

Jones, V., Cho, J., Abendschoen-Milani, J., Gielen, A., 2011. Driving habits and riskexposure in older drivers: lessons learned from the implementation of a selfregulation curriculum. Traffic Injury Prevention 5, 468–474.

Jones, V., Gielen, A., Bailey, M., Rebok, G., Gaines, J., Joyce, J., Parrish, J., in press. One ofmy fears is that physically or mentally, the time will come where I’ll be unable todrive anymore. And I’m not looking forward to that: a mixed methods feasibilitystudy to assess older driver’s risk impairment. Journal of Applied Gerontology.

Korner-Bitensky, N., Kua, A., von Zweck, C., Van Benthem, K., 2009. Older driverretraining: an update systematic review of evidence of effectiveness. Journal ofSafety Research 40, 105–111.

Kua, A., Korner-Bitensky, N., Desrosier, J., Man-Song-Hing, M., Marshall, S., 2007.Older driver retraining: a systematic review of evidence of effectiveness. Journalof Safety Research 39, 81–90.

Kutner, M., Greenberg, E., Jin, Y., Paulsen, C., 2006. The Health Literacy of Amer-ica’s Adults: Results From the 2003 National Assessment of Adult Literacy (NCES2006-483). U.S. Department of Education, National Center for Education Statis-tics, Washington, DC.

Lindstrom-Forneri, W., Tuokko, H., Garrett, D., Molnar, F., 2010. Driving as aneveryday competence: a model of driving competence and behavior. ClinicalGerontologist 33, 283–397.

MacLennan, P., Owsley, C., Rue, L., McGwin, G., 2009. Older adults’ knowledge aboutmedications that can impact driving. AAA Foundation for Traffic Safety.

McGwin, G., Sims, R., Pulley, L., Roseman, J., 2000. Relations among chronicmedical conditions, medications, and automobile crashes in the elderly: apopulation-based case-control study. American Journal of Epidemiology 52,424–432.

Nasvadi, G., Varvik, J., 2007. Crash risk of older drivers after attending a mature drivereducation program. Accident Analysis & Prevention. 39, 1073–1079.

National Center for Health Statistics, Trends in Health and Aging. Retrieved from:http://www.cdc.gov/nchs/agingact.htm (April, 2007).

National Study Center for Trauma and EMS, University of Maryland Schoolof Medicine, 2009. Traffic Safety Fact Book 2008. Retrieved from:http://medschool.umaryland.edu/Research-Centers/Shock,-Trauma-and-Anesthesiology-Research/NSC-Research—Traffic-Safety.asp (August, 2011).

Owsley, C., Ball, K., McGwin Jr., G., Sloane, M.E., Roenker, D.L., White, M.F.,Overly, E.T., 1998. Visual processing impairment and risk of motor vehiclecrash among older adults. Journal of the American Medical Association 279,1083–1088.

Owsley, C., Stalvey, B., Phillips, J., 2003. The efficacy of an educational interventionin promoting self-regulation among high-risk older drivers. Accident Analysisand Prevention 33, 393–400.

Owsley, C., McGwin Jr., G., Phillips, J.M., McNeal, S.F., Stalvey, B.T., 2004.Impact of an educational program on the safety of high-risk, visu-ally impaired older drivers. American Journal of Preventive Medicine 26,222–229.

Rudman, D., Friedland, J., Chipman, M., Sciortino, P., 2006. Holding on and lettinggo: the perspective of pre-seniors and seniors on driving self-regulation in laterlife. Canadian Journal on Aging 25, 65–76.

Subzwari, S., Desapriya, E., Babul-Wellar, S., Pike, I., Turcotte, K., Rajabali, F., Kinney,J., 2009. Vision screening of older drivers for preventing road traffic injuries andfatalities. Cochrane Database of Systematic Reviews 1.

Staplin, L., Dinh-Zarr, T., 2006. Promoting rehabilitation of safe driving abilities

through computer-based clinical and personal screening techniques. Topics inGeriatric Rehabilitation 22, 129–138.

Sullivan, K., Smith, S., Horswill, M., Lurie-Beck, J., 2011. Older adults’ safety percep-tions of driving situations: towards anew driving self-regulation scale. AccidentAnalysis & Prevention 43, 1003–1009.