preventing crashes among novice teenage drivers · naturalistic driving research driving...

Preventing Crashes Among

Novice Teenage Drivers: Research on Risk and Prevention

Dr. Bruce Simons-Morton

Senior Investigator, Health Behavior Branch &

Associate Director for Prevention

National Institute of Child Health &

Human Development

National Institutes of Health

YOUNG DRIVER PROBLEMDriver Fatal Crash Involvement/Million Miles

THE YOUNG DRIVER PROBLEMInexperienced Drivers of All Ages Have High Crash Rates

Twisk & Stacy, 2007

CONTRIBUTING FACTORS

Exposure- Miles- Condiitons

Individual Variability- Errors- Attention- Riskiness

Maturity/AgeInexperience

Crash Risk

Driving Performance

NATURALISTIC DRIVING RESEARCH DRIVINGNaturalistic Teenage Driving Study

A. Purpose: examine variability in teen driving performance

B. Overview

� N = 42 teens and 54 parents, 18-months of driving

� Continuous data collection

� Instrumentation: accelerometers, cameras, GPS

C. Surveys at 0, 6, 12, 18 months

Driving Clips

7

8

IRR=3.91

Simons-Morton, et al., AJPH, 2011

Naturalistic Teen Driving Study

Crash/Near Crash – Teens and Parents

Variability in CNC Rate by Group

Guo, Simons-Morton, et al., J of Pediatrics, 2013

Variability in CNC Rate Over Time

2.0 2.1

1.8

2.2

0.4 1.0

8.8

9.2

4.8

0.8

3.2

5.1

9.2

7.9

5.9 5.8

4.2

5.8

0

2

4

6

8

10

12

14

16

18

20

Months 1-3 Months 4-6 Months 7-9 Months 10-12 Months 13-15 Months 16-18

CN

C r

ate

(C

NC

pe

r 1

0K

KM

T)

Low-Risk Drivers (n=13)

Moderate-Risk Drivers (n=16)

All Drivers (n=42)

Kinematic Risky Driving

Acceleration Event Gravitational Force Correlation w/ CNC

Hard braking > -0.45 g 0.76

Rapid starts > 0.35 g 0.75

Sharp turns > 0.50 g 0.60

Yaw + 6 degrees w/i 3 seconds

0.46

Composite 0.60

Simons-Morton, et al., American Journal of Public Health, 2011; Simons-Morton et al., American Journal of Epidemiology, 2012

*Notes: 1) High monthly KRD = 10 times greater likelihood of a CNC in next month2) KRD rate/month is 75% accurate in predicting CNC

Turns

.57**

Yaw

.31*

TEEN AND ADULT KRD

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

1 2 3 4 5 6

IRs

fo

r g

-fo

rce

ra

tes

/ 1

00

mil

es

Time since licensure (3-month time periods)

Parent driver

Teen driver with nopassengers

Simons-Morton et al., Journal of Adolescent Health, 2011

Kinematic Risky Driving Rates

IRR = 5.08

Turns

.57**

Yaw

.31*

TEEN AND ADULT KRD

With Adult Passengers

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

1 2 3 4 5 6

IRs

fo

r g

-fo

rce

ra

tes

/ 1

00

mil

es

Time since licensure (3-month time periods)

Parent driver

Teen driver with nopassengers

Teen driver withadult passengers

IRR = 5.08

Simons-Morton et al., Journal of Adolescent Health, 2011

Kinematic Risky Driving Rates

INATTENTION INCREASES CRASH RISKCNC Odds Ratios

NTDS

(Novice Drivers)

100-Car Study(Experienced Drivers)

Secondary Task OR 95% CI OR 95% CI

Texting 4.3 1.9/10.0 n/a n/a

Dialing 7.8 2.7/23.1 2.5 1.4/4.5

Talking - Phone 0.8 0.4/1.5 0.7 0.5/1.1

Reaching - Phone 4.7 1.8/11.7 1.4 0.3/6.1

Reaching - Object (not phone) 7.8 3.5/16.8 1.2 0.6/2.3

Looking – Roadside object 3.7 1.7/8.5 0.7 0.4-1.2

Eating 3.3 1.5/7.2 1.3 0.7/2.1

Vehicle Operations 2.5 0.9/7.3 0.6 0.2/2.7

Radio/HVAC 1.4 0.8/2.7 0.5 0.3/0.9

Klauer, Guo, Simons-Morton et al., New England Journal of Medicine, 2014

Figure 4: The Smart Road Intersection

TEST TRACK RESEARCH ON DISTRACTION

Dialing When Approaching Intersection

Test Track Phone Task:

Intersection Stopping Behavior (n=16 teens; 16 experience adults)

200' w/phone

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Teen Adult Teen Adult

T0 T6

Pe

rce

nta

ge

Wh

o S

top

pe

d

Olsen, Simons-Morton, Lee, 2006

Time 1 Time 2

Teen Passengers

Teen Passengers Increase Fatal Crash Risk

0

1

2

3

4

5

6

7

Number of passengers

0 1 2 3+

Ages 16-17

0 1 2 3+

Ages 18-19

0 1 2 3+

Ages 30-59

Chen, Baker, 2003NPTS & NASS/GES Fatal crashes/10,000 trips

Observing Teen Drivers Leaving High School

10 area high schools; 3000 observations

Compared teen drivers with usual traffic

Speed - radar gun

Close following - video

1

2

3

HS parking lot

Radar

Video camera

Observer

Video camera

Simons-Morton. Lerner, Singer, AAP, 2005

Teen Driver Speed by

Driver and Passenger Type

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

A ll T e e n s D rive r G e n d e r P a sse n g er G e nd e r

Tra

ns

form

ed

Sp

ee

d

Fem ale Driver

Ma le D riv e rM a le P assenge r

M ale P assenge r

F ema le P ass enge r

F ema le P ass enge r

No P ass enge r

N o P ass enge r

Simons-Morton et al, AAP, 2005

Mean for usual

traffic

Experimental Simulation Study on Effect of

Teen Passengers on Teen Risky Driving

Confederate Passenger

TEEN PASSENGER STUDY #1 (n=58)

Passenger Effects (ANOVA)

Outcome/

ConditionPassenger Solo p

Average Average

Failed to stop 30% 24% <0.01

% time in red 22% 17% <0.01

*Significant passenger presence * passenger type interaction (p <0.001)

Risky Intersection Behavior

Simons-Morton, Bingham, et al., Health Psychology, 2013

TPS#1 (n=58)

Risky Intersection Behavior

**p <0.001

Simons-Morton, Bingham, et al., Health Psychology, 2013

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Passenger Drive Solo Drive

Prob

ab

ilit

y o

f fa

iled

to s

top

0

5

10

15

20

25

30

35

Passenger Drive Solo Drive

Percen

t ti

me i

n R

ed

zInteractions Between Passenger Presence and Passenger Type

Failed to Stop** Percent Time in Red**

______ Risk-accepting passenger

- - - - - - Risk-averse passenger

SUMMARY

Teen Risk Factors

1. Young age & inexperience

2. Kinematic risky driving

3. Inattention due to 2nd tasks

4. Teenage passengers

Crashes Increase at Licensure

27

Safety Approaches ToThe Novice Young Driver Problem

SAFETY APPROACH GOALS

EVIDENCE OF

SAFETY BENEFIT

� GDL Limit exposure +++

� Driver education &

training

Vehicle management

and driving test

???

� Supervised practice Provide experience,

improve skills,

judgment, norms

?

� Improve parent

management

Limit exposure +

Effects of GDL in Michigan

Shope, Molnar, 2004

• Delays licensure

• Reduces exposure to

high-risk driving

GDL POLICY ANALYSES

Fatal Crash Rate Declines by Number of

Graduate Driver Licensing Components

Chen L et al. Pediatrics 2006;118:56-62

EVALUATIONS OF DRIVER TRAINING

Approach Objective Safety Benefits

DeKalb Study, 1973 Intensive/usual DE No safety benefits

Vernick et al, 1999; Mayhew & Simpson, 2002; Elvik, 2004; Peck, 2011; Lenero & Mayhew,2015

Reviews (pre-openlicensure training)

No safety benefits

Beanland et al., 2013; Washington et al., 2011; Mynttinen et al., 2010

Reviews (post-drive training)

No safety benefits of post-drive training (2-stage)

LSEDE, 2015 Oregon, Manitoba Probably no benefits

Senserrick, Ivers, et al., 2009 NSW DRIVE risk orientation/resilience

44% reduced crash RR

Thomas et al., 2016 Hazard skills training Possible crash reduction

“More skillful drivers do not necessarily crash less; regardless of skill drivers must actually drive more safely to minimize risk.” (Lonero, Meyhew, 2015).

SUPERVISED PRACTICED NATURALISTIC DRIVING STUDY

Driving Errors Improve With Practice

Driving Errors by Type

First 10 Hours

N=90 participants#

9,823 observations

Last 10 hours

N=78 participants#

8,523 observations

N Percentage (SD) N Percentage (SD)

Driving Errors - Total 1,138 13.4 (19.4) 786 8.8* (13.0)

Recognition 90 0.9 (1.2) 34 0.4 (0.9)

Decision 352 3.5 (2.9) 124 1.4 (2.4)

Performance 638 8.4 (19.3) 605 6.8 (11.3)

Other 58 0.6 (1.0) 23 0.2 (0.8)

High Risk Secondary Tasks 554 5.8 (4.6) 878 9.9* (7.6)

SUPERVISED PRACTICE NATURALISTIC DRIVING STUDY

Parent Driving Instruction Topics

Topic†Proximal

Instruction#

Higher Order

Instruction*

Navigation 94% 4%

Warning/Detect Hazard 75% 16%

Vehicle Handling or

Operation85% 7%

Remark on Driving Behavior 74% 17%

Asks Question - Driving Task 80% 19%

Rules of the Road 78% 15%#Proximal relates to the present driving task or immediate future*Higher order relates to principles of driving

N=76

PRACTICE DRIVING AND INDEPENDENT

RISKY DRIVING MEASURES

Correlations

MeasuresKinematic Risky

Driving

Crash/Near Crash

3-Months Total 3-Months Total

Miles -.13 -.14 -.10 -.12

Road Type Diversity

.01 .02 -.04 -.17

RESEARCH ON PARENTAL

MANAGEMENT OF TEENAGE DRIVERS

Authoritative Parents are Demanding and Responsive

Parental Restrictions on Trip and Risk Conditions

Never

allowed

Always

allowed

1

2

3

4

5

6

7

1 month <3 mos 3-5 mos 6-8 mos 9-11 mos >11 mos

Trip limits Risk Limits

Hartos, Simons-Morton. 2001

CHECKPOINTS PARENT

MANAGEMENT PROGRAM

Parent Management

of Teen Driving

- Limits on teen driving

Mediators1

- restriction norms

- expectations

�

Persuasive

Communications

� �

1Protection motivation theory

The Checkpoints Parent-Teen

Driving Agreement

Teen driver will: Parent will:

□□□□ Always obey all traffic laws

□□□□ Never speed, tailgate, or cut others off

□□□□ Always wear a seat belt and require all passengers to wear seat belts

□□□□ Never drive after taking any drugs or alcohol or ride with a driver who has

taken any drugs or alcohol

□□□□ Always tell parent/guardian where going and with whom

□□□□ Always call home if going to be late

□□□□ Always call home if for any reason it is not safe to drive or ride

□□□□ Be a good role model behind the wheel

□□□□ Point out and discuss safe and dangerous

driving situations and practices

□□□□ Apply rules fairly and consistently

□□□□ Consider necessary exceptions to driving

limits

□□□□ Provide a safe ride home (no questions

asked at that time)

PART II:PART II:PART II:PART II: DRIVING PRIVILEGES: These need to be tailored to your teen's driving progressThese need to be tailored to your teen's driving progressThese need to be tailored to your teen's driving progressThese need to be tailored to your teen's driving progress

DRIVING PRIVILEGESNighttime

Teen

passengersWeather Road types Review date

We agree

Initials

Checkpoint 1

Month 18 pm None Dry Local ____ ____

Checkpoint 2

Months 2-69 pm None Moderate No high speed ____ ____

Checkpoint 3

Months 7-1211 pm 1 Most Most ____ ____

WE AGREE (sign) __________________________ ______________________________PARENT TEEN

PART I:PART I:PART I:PART I: DRIVING RULES: These are absolutes These are absolutes These are absolutes These are absolutes ————ones that apply to every trip, every timeones that apply to every trip, every timeones that apply to every trip, every timeones that apply to every trip, every time

Checkpoints in Driver Education

Percentage of families with a completed

agreement

70.6

29.4

0

20

40

60

80

100

Intervention Control

%

1

1 Intervention greater than control (x = 42.54; p < .0001)

Zakrajsek, Simons-Morton, Shope, F&CH, 2009.

CHECKPOINTS PROGRAM Tx Group Improvements in Driving Outcomes

Intervention m (sd)

Control m (sd)

p

Overall High Risk Driving (past week) – 19 items 0.50 (0.5) 0.82 (0.9) .04 Sped in residential or school zone 1.51 (1.7) 2.20 (2.3) .09 Drove 10-19 mph over limit 0.31 (0.1) 0.80 (1.8) .10

Drove 20+ mph over limit 0.02 (0.1) 0.28 (0.7) .02 Tailgated 0.08 (0.3) 0.37 (1.0) .07 Went through intersection on yellow 1.79 (2.2) 3.15 (3.9) .04 Raced another vehicle 0.05 (0.2) 0.24 (0.7) .07 Drove to show off 0.03 (0.2) 0.15 (0.4) .08

TECHNOLOGY AND PARENT MANAGEMENT

Event Recorders Provide Feedback

and Enable Parent Monitoring

name

name

name

DriveCam TeenSafe Driver Feedback

Simons-Morton, Bingham, Shope, et al.,

Journal of Adolescent Health, 2012.

Randomized Trial:Group #1: Immediate Feedback to Teen (LO) Group #2: Lights+ Feedback to Family (DC)

IMPROVING NOVICE TEEN

DRIVING SAFETY

Summary

1. Goals:

A. Delay licensure

B. Limit exposure to high risk driving conditions

C. Create safe driving norms and practices

2. Improve training & supervision (Driver Ed & practice driving)

A. Higher order instruction

B. Safe driving expectations, judgment, self-control, norms

3. Parental management

A. Set strict limits on newly licensed teens

B. Monitor driving performance

Thank you!

UMass

Virginia Tech

DriveCam IncU Michigan

CDM, Inc

Collaborators

NICHD: Johnathon Ehsani, Kaigang Li, Fearghal O’Brien, Pnina Gershon, Paul Albert

Sherbrook U: Marie Claude Ouimet

VTTI: Sheila Klauer, Tom Dingus, Feng Guo, Suzie Lee,

UMTRI: Ray Bingham, Jean Shope, Anuj Pradhan,

U of PA: Emily Falk