Page 1
Michigan’s Graduated Driver Michigan’s Graduated Driver Licensing (GDL) and BeyondLicensing (GDL) and Beyond
Jean T. Shope, MSPH, PhDMichigan Traffic Safety Summit
March 15, 2007
Support: NHTSA, NIH, CDC/NCIPC, NSCColleagues: Waller, Molnar, Zakrajsek, Bingham, Elliott, Simons-Morton
Page 2
OverviewOverview
Michigan’s GDL evaluation
Other jurisdictions’ GDL evaluations
National GDL evaluations
Program to enhance parental involvement
Page 3
BackgroundBackground
US: MVC injury leading cause of teen deaths
Crash risk highest first few months driving solo
1996-2007: GDL adopted by nearly all states
Under 18 years old
3-stage license process
Extended learner phase (practice requirements)
Restrictions in intermediate phase (night , passenger)
Page 4
Segment 1Segment 1DriverDriver
EducationEducation
Level 1Level 1LicenseLicense
14 yr 9 mo14 yr 9 mo
Segment 2Segment 2DriverDriver
EducationEducation
Level 2Level 2LicenseLicense
16 yr16 yr
Level 3Level 3LicenseLicense
17 yr17 yr
24 hr class24 hr class
6 hr driving6 hr driving
Written examWritten exam
Drive only with Drive only with parent or adultparent or adult
6 hr class6 hr class
Drive alone Drive alone Night restrictionNight restriction
Michigan GDL/Driver EducationMichigan GDL/Driver EducationApril 1, 1997April 1, 1997
No restrictionsNo restrictions
HealthyHealthy
Parent signParent sign
6 mo Level 16 mo Level 1
Driven 50 hrDriven 50 hr
Road testRoad test
90 day clean90 day clean
Parent signParent sign
6 mo Level 26 mo Level 2
12 mo clean12 mo clean
Page 5
Parental Experience with Michigan’s Parental Experience with Michigan’s GDL Program (July 1998 Survey)GDL Program (July 1998 Survey)
Hours of practice: 9% less than required 23% required 50 hours 68% more (mean = 75.3 hours)
Quality of GDL experience: 97% good/very good
Waller, Olk, Shope. J Safety Research (2000) 31:9-15
Page 6
Michigan’s GDL: Early Impact on MVCs Michigan’s GDL: Early Impact on MVCs Among 16-Year-OldsAmong 16-Year-Olds
1996 vs 1999 crash data, adjusted1996 vs 1999 crash data, adjusted
All crashes: down 25%
Fatal plus nonfatal injury crashes: down 24%
Night crashes: down 53%
Shope, Molnar, Elliott, Waller. JAMA (2001) 286:1593-1598
Page 7
All Crashes: Counts, All Crashes: Counts, Pop Rates, Licensee RatesPop Rates, Licensee Rates
0
5,000
10,000
15,000
20,000
25,000
1994 1995 1996 1998 1999 2000 2001
Count
0
50
100
150
200
250
300Rate
Crash Involvement Crash/1,000 Population Crash/1,000 Drivers
Shope, Molnar. Journal of Safety Research 35 (2004) 337-344.
Page 8
Casualty Crashes: Counts, Casualty Crashes: Counts, Pop Rates, Licensee RatesPop Rates, Licensee Rates
0
1,000
2,000
3,000
4,000
5,000
6,000
7,000
1994 1995 1996 1998 1999 2000 2001
Count
01020
30405060
708090
Rate
Crash Involvement Crash/1,000 Population Crash/1,000 Drivers
Shope, Molnar. Journal of Safety Research 3 (2004) 337-344.
Page 9
Michigan’s GDL: First Four Years: Michigan’s GDL: First Four Years: 16 yo MVCs 1996 vs 1998-200116 yo MVCs 1996 vs 1998-2001
Significant crash reductions maintained (2001 all crashes down 19%, adjusted)
Reductions in #, crashes/population, crashes/driver
Reductions in both sexes, but men still higher
Evening crash reduction not significant after adjusting (3 X that of 25+ yo)
Crashes with passenger reduced (3 X that of 25+ yo)
Lower proportion of 16-year-olds licensed Shope, Molnar. J Safety Research (2004) 35:337-344
Page 10
Updated Michigan GDL ResultsUpdated Michigan GDL Results
Age of licensure increased somewhat
Time in each GDL level exceeds minimum
Number of crashes per driver less each GDL year
Time until first crash/offense longer each GDL year
Time until first injury crash: Each cohort longer than pre-GDL Each cohort longer than previous cohort
Page 11
Novice Teen Driving/GDLNovice Teen Driving/GDLInvited SymposiumInvited Symposium
February 5-7, 2007 in Tucson Shope: Review of GDL evaluations Williams: Components of GDL Papers to be published in April Journal of Safety Research
Page 12
GDL Evaluation Results (20): GDL Evaluation Results (20): Individual Jurisdiction StudiesIndividual Jurisdiction Studies
Can’t compare - different pre/post programs & evals
Consistent positive findings
Substantial crash reductions from 19 of 20 studies (20%-40%)
Convictions down in Iowa
Hospitalization and charges down in NC
California studies: differing methods and results
Page 13
GDL Evaluation Results (6): GDL Evaluation Results (6): Nationwide StudiesNationwide Studies
Consistent, positive findings
Reductions 6% (15-17 yo traffic fatalities) to 40% (16 yo driver involvement in injury crashes)
Greater reductions found with stronger GDL programs
Greater reductions among teen vs. older drivers
No increase in crash risk for 17 or 18 yo
No male/female differences in reductions
Page 14
What in GDL works?What in GDL works?
Whole program works as a package
GDL programs with recommended components more effective
Learner: 16 yrs, minimum 6 months, 30+ hours practice
Intermediate:
• Night restriction start 10 pm
• Passenger restriction - no more than one teen except family
Effectiveness of each component?
Allan Williams’ paper (April J Safety Research)
Page 15
How do we enhance GDL?How do we enhance GDL?
Even with GDL, teen drivers still crash
Based on research, enact the best GDL program
Implement the program well
Enhance parent involvement
Page 16
Checkpoints ProgramCheckpoints Program
Developed by:
National Institute of Child Health & Human Development (Bruce Simons-Morton & colleagues)
Purpose:
To facilitate parental management of teen driving and reduce adolescent driving risk
Page 17
Checkpoints ProgramCheckpoints Program
Parent-Teen Written Driving Agreement
Initially: low-risk conditions, can drive alone high-risk conditions, with adult
Later, increase privileges with experience and responsible behavior
Page 18
Checkpoints ProgramCheckpoints Program Persuasive Communications (mailed/DMV)
Video Newsletters
Agreement
3 studies completed (CT, MD)
1 study underway (RI) 2 studies underway (MI)
Page 19
Checkpoints Study ResultsCheckpoints Study Results(Simons-Morton & Colleagues)(Simons-Morton & Colleagues)
Parents set limits on teen drivers
Nearly all set limits; not strict; rapid decline (Prev Sci 2001, Inj Prev 2004, Am J Pub Hlth 2005)
More limits set in GDL vs non-GDL state (Acc Anal Prev 2005)
Greater parent limits associated with less risky driving, fewer violations and crashes (J Adol Res 2000, Prev Sci 2001, Hlth Ed Behav 2002, Traffic Inj Prev 2006)
Page 20
Segment 1Segment 1DriverDriver
EducationEducation
Level 1Level 1LicenseLicense
14 yr 9 mo14 yr 9 mo
Segment 2Segment 2DriverDriver
EducationEducation
Level 2Level 2LicenseLicense
16 yr16 yr
Level 3Level 3LicenseLicense
17 yr17 yr
24 hr class24 hr class
6 hr driving6 hr driving
Written examWritten exam
Drive only with Drive only with parent or adultparent or adult
6 hr class6 hr class
Drive alone Drive alone Night restrictionNight restriction
Michigan GDL/Driver EducationMichigan GDL/Driver Education
No restrictionsNo restrictions
HealthyHealthy
Parent signParent sign
6 mo Level 16 mo Level 1
Driven 50 hrDriven 50 hr
Road testRoad test
90 day clean90 day clean
Parent signParent sign
6 mo Level 26 mo Level 2
12 mo clean12 mo clean
CHECKPOINTSCHECKPOINTS
Page 21
Michigan “Checkpoints One”Michigan “Checkpoints One”Driver EducationDriver Education
(NICHD-funded)(NICHD-funded)
Randomized controlled trial
Driver education setting
Timing just prior to independent driving
Ensure parent/teen complete agreement
Conditions/privileges (night, passengers, weather, roads)
Rules: check in, risks, traffic laws (alcohol, safety belts)
Page 22
““Checkpoints One” InterventionCheckpoints One” Intervention
Recruited from Segment 2 classes (Sears)
Parent/teen session (30 minutes) taught by health educator (research staff) at end of Segment 2
Baseline survey, video, agreement discussed/completed
Mailing 1 week prior to expected Level 2 license date
Newsletter, agreement
Page 23
““Checkpoints One” StatusCheckpoints One” Status
Enrollment: 326 parent-teen dyads
Telephone surveys: licensure, 3 & 6 mo after
Teens: most at Level 2 licensure; in follow-up
Results soon from baseline and licensure surveys
Page 24
Michigan “Checkpoints Two” Michigan “Checkpoints Two” (CDC/NCIPC-funded)(CDC/NCIPC-funded)
Randomized controlled design
Baseline survey by mail; ask expected license date
Driver educators trained to teach parent/teen session (30 minutes) in Segment 2
Video, persuasion, agreement discussed/completed
Booklet / agreement
Telephone surveys: licensure, 3 & 6 months after
Page 25
““Checkpoints Two” StatusCheckpoints Two” Status
Permission granted for Checkpoints in Segment 2
Driving schools recruited and randomized (8)
Training video, materials developed
Driver educators trained
First classes scheduled
Recruitment starting soon
Page 26
SummarySummary
GDL reduces teen driver crashes and consequences
More needed, especially in first six months driving solo
Parental involvement, limit-setting important
Checkpoints approach effective
Checkpoints in Michigan’s driver education Segment 2 being evaluated with researchers and driver educators
If effective, could be implemented widely
Page 27
Thank you!
[email protected]