Traffic Safety
Improved Reporting and Performance Measures Would Enhance Evaluation of High-Visibility Campaigns Gao ID: GAO-08-477 April 25, 2008Two primary risk behaviors related to fatal traffic crashes are failure to use safety belts and driving while impaired by alcohol. High-visibility enforcement (HVE) campaigns that combine enforcement of a traffic safety law with media to inform the public about the campaign are effective in reducing these behaviors. In 2005, the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users authorized funding of an HVE program, including safety belt and impaired-driving campaigns. The National Highway Traffic Safety Administration (NHTSA) within the Department of Transportation (DOT) provides media and coordinates with states to provide enforcement activities for the campaigns. This report addresses (1) the extent to which NHTSA has implemented the HVE program and (2) for selected states, the impact of the campaigns and challenges that exist in conducting the campaigns. To conduct this work, GAO analyzed fatality data, plans, and evaluations and interviewed officials from DOT and seven selected states.
NHTSA has fully implemented the high-visibility enforcement program by (1) developing and disseminating advertising, (2) coordinating with states on media and enforcement activities, and (3) annually evaluating the effectiveness of the two HVE campaigns; however, NHTSA's evaluations have shortcomings that limit the agency's ability to determine the effectiveness of the campaigns. Regarding advertising, NHTSA introduced an annual plan in 2005 that sets forth a strategy for the campaign advertisements, developed advertisements, and purchased national media time for the advertisements. To coordinate with states, NHTSA provides an overall strategy and guidance to assist states in conducting the campaigns, as well as technical assistance and collateral materials, such as posters and model press releases. Officials in selected states reported that NHTSA's coordination efforts provided the support and interaction needed to conduct HVE campaigns. Although NHTSA's annual evaluations of campaign effectiveness indicate that the campaigns are helping to improve safety belt use and reduce impaired driving, the evaluations have shortcomings that limit NHTSA's ability to assess the level of state and local activity--a key component of the campaigns--and the overall effectiveness of the campaigns. For example, the information that NHTSA has on states' activities is inconsistent and incomplete because reporting of such data is generally voluntary for local law enforcement agencies. As a result, NHTSA has reported that it cannot provide meaningful analyses and comparisons of state activities. NHTSA's ability to measure the campaigns' overall effectiveness is also hindered because the performance measures used to evaluate the campaigns are not comprehensive. For example, while NHTSA measures daytime safety belt use, it does not directly measure nighttime safety belt use, despite recent efforts to increase safety belt use at night. In addition, NHTSA's evaluations do not include measures of the effectiveness of the campaigns at reaching all target audiences. NHTSA is working to develop more comprehensive performance measures. According to officials in selected states GAO visited, the campaigns are contributing to increased safety belt use and reduced alcohol-involved fatalities, but these states face challenges in conducting the campaigns and achieving desired results. From 1997 to 2006, safety belt use increased in all seven of the selected states, and each state experienced a decrease in the alcohol fatality rate. Officials in the selected states said that the campaigns provide additional benefits, such as apprehending suspects involved in other crimes. However, officials in those selected states identified several challenges, such as increasing safety belt use and reducing impaired driving among resistant populations; insufficient staff to conduct the campaigns; and weak prosecution of impaired-driving arrests. NHTSA has initiatives under way to help states address some of these challenges. For example, NHTSA has sponsored a campaign to increase safety belt use in rural areas. In addition, NHTSA provides funds that can be used by states to purchase equipment for local law enforcement agencies, such as breath-testing units, to encourage the agencies to participate in campaigns.
RecommendationsOur recommendations from this work are listed below with a Contact for more information. Status will change from "In process" to "Open," "Closed - implemented," or "Closed - not implemented" based on our follow up work.
Director: Team: Phone:GAO-08-477, Traffic Safety: Improved Reporting and Performance Measures Would Enhance Evaluation of High-Visibility Campaigns
This is the accessible text file for GAO report number GAO-08-477
entitled 'Traffic Safety: Improved Reporting and Performance Measures
Would Enhance Evaluation of High-Visibility Campaigns' which was
released on April 25, 2008.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
United States Government Accountability Office:
GAO:
Report to the Chairman, Committee on Transportation and Infrastructure,
House of Representatives.
April 2008:
Traffic Safety:
Improved Reporting and Performance Measures Would Enhance Evaluation of
High-Visibility Campaigns:
GAO-08-477:
GAO Highlights:
Highlights of GAO-08-477, a report to the Chairman, Committee on
Transportation and Infrastructure, House of Representatives.
Why GAO Did This Study:
Two primary risk behaviors related to fatal traffic crashes are failure
to use safety belts and driving while impaired by alcohol. High-
visibility enforcement (HVE) campaigns that combine enforcement of a
traffic safety law with media to inform the public about the campaign
are effective in reducing these behaviors. In 2005, the Safe,
Accountable, Flexible, Efficient Transportation Equity Act: A Legacy
for Users authorized funding of an HVE program, including safety belt
and impaired-driving campaigns. The National Highway Traffic Safety
Administration (NHTSA) within the Department of Transportation (DOT)
provides media and coordinates with states to provide enforcement
activities for the campaigns. This report addresses (1) the extent to
which NHTSA has implemented the HVE program and (2) for selected
states, the impact of the campaigns and challenges that exist in
conducting the campaigns. To conduct this work, GAO analyzed fatality
data, plans, and evaluations and interviewed officials from DOT and
seven selected states.
What GAO Found:
NHTSA has fully implemented the high-visibility enforcement program by
(1) developing and disseminating advertising, (2) coordinating with
states on media and enforcement activities, and (3) annually evaluating
the effectiveness of the two HVE campaigns; however, NHTSA‘s
evaluations have shortcomings that limit the agency‘s ability to
determine the effectiveness of the campaigns. Regarding advertising,
NHTSA introduced an annual plan in 2005 that sets forth a strategy for
the campaign advertisements, developed advertisements, and purchased
national media time for the advertisements. To coordinate with states,
NHTSA provides an overall strategy and guidance to assist states in
conducting the campaigns, as well as technical assistance and
collateral materials, such as posters and model press releases.
Officials in selected states reported that NHTSA‘s coordination efforts
provided the support and interaction needed to conduct HVE campaigns.
Although NHTSA‘s annual evaluations of campaign effectiveness indicate
that the campaigns are helping to improve safety belt use and reduce
impaired driving, the evaluations have shortcomings that limit NHTSA‘s
ability to assess the level of state and local activity”a key component
of the campaigns”and the overall effectiveness of the campaigns. For
example, the information that NHTSA has on states‘ activities is
inconsistent and incomplete because reporting of such data is generally
voluntary for local law enforcement agencies. As a result, NHTSA has
reported that it cannot provide meaningful analyses and comparisons of
state activities. NHTSA‘s ability to measure the campaigns‘ overall
effectiveness is also hindered because the performance measures used to
evaluate the campaigns are not comprehensive. For example, while NHTSA
measures daytime safety belt use, it does not directly measure
nighttime safety belt use, despite recent efforts to increase safety
belt use at night. In addition, NHTSA‘s evaluations do not include
measures of the effectiveness of the campaigns at reaching all target
audiences. NHTSA is working to develop more comprehensive performance
measures.
According to officials in selected states GAO visited, the campaigns
are contributing to increased safety belt use and reduced alcohol-
involved fatalities, but these states face challenges in conducting the
campaigns and achieving desired results. From 1997 to 2006, safety belt
use increased in all seven of the selected states, and each state
experienced a decrease in the alcohol fatality rate. Officials in the
selected states said that the campaigns provide additional benefits,
such as apprehending suspects involved in other crimes. However,
officials in those selected states identified several challenges, such
as increasing safety belt use and reducing impaired driving among
resistant populations; insufficient staff to conduct the campaigns; and
weak prosecution of impaired-driving arrests. NHTSA has initiatives
under way to help states address some of these challenges. For example,
NHTSA has sponsored a campaign to increase safety belt use in rural
areas. In addition, NHTSA provides funds that can be used by states to
purchase equipment for local law enforcement agencies, such as breath-
testing units, to encourage the agencies to participate in campaigns.
What GAO Recommends:
GAO recommends that the Secretary of Transportation direct NHTSA to
establish a minimum set of reporting requirements for states to report
HVE activities that are federally funded and include additional
performance measures in campaign evaluations. DOT officials generally
agreed with the findings and recommendations of the report.
To view the full product, including the scope and methodology, click on
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-477]. For more
information, contact Katherine A. Siggerud at (202) 512-2834 or
siggerudk@gao.gov.
[End of section]
Contents:
Letter:
Results in Brief:
Background:
NHTSA Has Implemented the HVE Program but Could Improve the Annual
Evaluations of Campaign Effectiveness:
Selected State Officials Report That HVE Campaigns Are Contributing to
Increased Safety Belt Use and Reduced Fatalities, but Several
Challenges Hinder Further Progress:
Conclusions:
Recommendations for Executive Action:
Agency Comments:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: HVE Campaigns in Australia, Canada, and the Netherlands:
Appendix III: GAO Contact and Staff Acknowledgments:
Table:
Table 1: Funding for HVE Paid Media and Evaluations, Fiscal Years 2003-
2007:
Figures:
Figure 1: Total Fatalities and Total Fatalities per 100 Million Vehicle
Miles Traveled (1985-2006):
Figure 2: Unrestrained Vehicle Occupant Fatalities and Unrestrained
Fatalities per 100 Million Vehicle Miles Traveled (1985-2006):
Figure 3: Alcohol-Involved Fatalities and Alcohol-Involved Fatalities
per 100 Million Vehicle Miles Traveled (1985-2006):
Figure 4: HVE Campaign Activities and Timeline:
Figure 5: Increase in Safety Belt Use for Selected States Compared with
Overall U.S. Increase (1997-2006):
Figure 6: Safety Belt Use Compared with Federal Goal, Selected States
(2006):
Figure 7: Decrease in Alcohol-Involved Fatality Rate per 100 Million
Vehicle Miles Traveled for Selected States Compared with Average
Decrease in the United States (1997-2006):
Figure 8: Fatalities with a BAC of 0.08 or Greater per 100 Million
Vehicle Miles Traveled for Selected States Compared with Federal Goal
(2006):
Abbreviations:
BAC: blood alcohol content:
CIOT: Click It Or Ticket:
DOT: Department of Transportation:
DUI: driving under the influence:
FARS: Fatality Analysis Reporting System:
FHWA: Federal Highway Administration:
GHSA: Governors Highway Safety Association:
HPMS: Highway Performance Monitoring System:
HVE: high-visibility enforcement:
NASJE: National Association of State Judicial Educators:
NHTSA: National Highway Traffic Safety Administration:
NTSB: National Transportation Safety Board:
OTLUA: Drunk Driving. Over the Limit. Under Arrest:
RBT: random breath test:
SAFETEA-LU: The Safe, Accountable, Flexible, and Efficient
Transportation Equity Act: A Legacy for Users:
TEA-21: Transportation Equity Act for the 21st Century:
VMT: vehicle miles traveled:
[End of section]
United States Government Accountability Office: Washington, DC 20548:
April 25, 2008:
The Honorable James L. Oberstar:
Chairman:
Committee on Transportation and Infrastructure: House of
Representatives:
Dear Mr. Chairman:
More than 42,600 people died in traffic accidents during 2006. The
failure to use safety belts and driving while impaired by alcohol are
two primary risk behaviors related to these accidents. High-visibility
enforcement (HVE) campaigns that combine intensive enforcement of a
specific traffic safety law with extensive media communication to
inform the public about the campaign have been found effective in the
United States and other countries in helping reduce these behaviors.
The Safe, Accountable, Flexible, Efficient Transportation Equity Act: A
Legacy for Users (SAFETEA-LU), enacted in 2005, authorized funding for
an HVE program, including two primary HVE campaigns: Click It Or Ticket
(CIOT), to increase safety belt use, and Drunk Driving, Over the Limit,
Under Arrest (OTLUA), to decrease the number of impaired drivers.
[Footnote 1] SAFETEA-LU specified that the National Highway Traffic
Safety Administration (NHTSA) within the Department of Transportation
(DOT) should implement this program by developing and disseminating
national advertisements for the campaigns, coordinating with states to
conduct the campaigns, and evaluating the results of the campaigns; the
law authorized $29 million annually for NHTSA to implement the program.
State and local governments provide law enforcement resources for the
campaigns--such as officers, cars, and equipment for patrols or
checkpoints--and may supplement NHTSA's national advertisements; these
entities may use federal traffic safety grants for such activities.
You requested that we assess the HVE program and campaigns.
Accordingly, this report addresses (1) the extent to which NHTSA has
implemented the HVE program and (2) for selected states, the impact of
the HVE campaigns and challenges that exist in conducting the
campaigns. This report also includes additional information on the key
components of HVE campaigns used by Australia, Canada, and the
Netherlands (see app. II).
To determine the extent that NHTSA has implemented the HVE program, we
analyzed information and interviewed officials from NHTSA headquarters
and regions; the Federal Highway Administration (FHWA); and state
traffic safety offices, state police, local police, and police advocacy
organizations in seven states--Arkansas, Illinois, Iowa, North
Carolina, North Dakota, Rhode Island, and Washington. We judgmentally
selected the states by including: states that have enacted various laws
that may affect how states conduct enforcement campaigns; states with a
wide range of traffic safety performance levels, such as extent of
safety belt use and number of alcohol-involved fatalities in each
state; states with differences in average size of law enforcement
agencies; states that exhibited various degrees of participation by
state and local law enforcement agencies in campaigns; and states that
were geographically dispersed. Since we used a nongeneralizable
sampling approach, our findings cannot be used to make inferences about
all states that implemented the HVE program. We also interviewed
representatives of nongovernmental organizations, including the
American Association of State Highway and Transportation Officials,
Governors Highway Safety Association (GHSA), International Association
of Chiefs of Police, Mothers Against Drunk Driving, National Safety
Council, and the National Sheriffs Association. In addition, we
reviewed studies, reports, and laws relevant to the implementation of
the NHTSA HVE program. To determine, for selected states, what impact
the HVE campaigns have had and what challenges exist, we analyzed
safety belt use and alcohol-involved fatality data and interviewed
officials from state traffic safety offices, state police, local
police, and police advocacy organizations in the seven selected states.
We used data contained in NHTSA's Fatality Analysis Reporting System
(FARS) database and vehicle miles traveled data maintained by FHWA in
its Highway Performance Monitoring System database. We determined the
data to be sufficiently reliable for the purposes of this report. For
further details of our objectives, scope, and methodology, see appendix
I. We also provide a summary of high-visibility campaigns in Australia,
Canada, and the Netherlands, which can be found in appendix II. We
conducted this performance audit from March 2007 to April 2008 in
accordance with generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.
Results in Brief:
NHTSA has implemented the HVE program by developing and disseminating
advertisements, coordinating with all states on advertisement and
enforcement activities, and evaluating the effectiveness of the two HVE
campaigns; however, NHTSA's evaluations of these campaigns have
shortcomings that limit the extent to which NHTSA can determine the
effectiveness of the campaigns.
* Advertisements: To develop and disseminate advertisements, NHTSA
introduced an annual National Communications Plan in 2005 that sets
forth a strategy for the campaigns, including goals, dates, target
audiences, and messages for the campaigns. Through a contactor, NHTSA
also developed advertisements in multiple languages and media formats-
-such as broadcast television, cable television, and radio--and
purchased national media time for the advertisements. Purchasing media
time for these advertisements accounted for the majority of NHTSA's
annual $29 million appropriation for the HVE program. For example,
NHTSA allocated nearly $28 million to purchase media time for
advertisements in fiscal year 2006.
* Coordination: To coordinate with states on advertisement and
enforcement activities, NHTSA provides an overall strategy and guidance
to assist states in conducting the campaigns, as well as technical
assistance and collateral materials--such as posters and model press
releases--to help state officials with their advertisements. According
to officials in our seven selected states, NHTSA's coordination efforts
have provided the support and interaction needed to conduct HVE
campaigns. For example, officials from one state noted that NHTSA
assisted them in applying for traffic safety grants to conduct
campaigns and provided tool kits that were useful in developing the
campaigns. Officials from another state reported that NHTSA had
improved the quality and timeliness of advertising materials, allowing
them to devote more state resources to purchasing radio and television
ads rather than developing the ads. NHTSA's campaign coordination
efforts are included as part of the agency's day-to-day coordination
efforts with the states and are not funded by the $29 million
appropriation for the HVE program.
* Evaluation: NHTSA evaluates the effectiveness of the campaigns
annually, but the evaluations have shortcomings that limit NHTSA's
ability to assess the level of state activity and the overall
effectiveness of the campaigns. The evaluations accounted for $750,000
of NHTSA's $29 million appropriation in fiscal years 2006 and 2007.
NHTSA's evaluations include information on the level of enforcement
activity by states and the results of the campaigns based on
performance measures, such as message awareness, media activity, safety
belt use, and fatality and injury statistics. However, the information
that NHTSA has on states' activities is inconsistent and incomplete in
part because states are not required to report such data, although
NHTSA officials said that the agencies receiving federal traffic safety
grants for campaign activities generally voluntarily report on these
activities. As a result, NHTSA is not able to fully account for state
and local law enforcement campaign activity--a critical component of
HVE campaigns for which states may use federal traffic safety grants.
NHTSA's ability to measure the campaigns' overall effectiveness is also
hindered in part because the performance measures used to evaluate the
campaigns are not comprehensive or consistent. For example, while NHTSA
measures the change in daytime safety belt use, it does not directly
measure nighttime safety belt use, despite recent efforts to increase
the use of safety belts at night. Furthermore, measures of the
effectiveness of NHTSA's national advertising campaign in reaching all
target audiences were limited. For example, both the safety belt and
impaired-driving campaign evaluations contained information about the
effectiveness of the campaigns at reaching their primary target
audiences but no information on the effectiveness in reaching other
target audiences that were listed in the National Communications Plan.
NHTSA is working to develop more comprehensive measures of the
effectiveness of the campaigns.
According to officials in the selected states we visited, HVE campaigns
are contributing to increased safety belt use and reduced alcohol-
involved fatalities, but these states face challenges such as reaching
resistant populations, finding sufficient resources to conduct the
campaigns, and weak prosecution of impaired-driving offenders. From
1997 to 2006, safety belt use increased in all seven of the selected
states, and four of those states exceeded the 2006 NHTSA goal for
safety belt use (82 percent). HVE campaigns in the selected states are
also contributing to reduced alcohol-involved driving fatalities. From
1997 to 2006, each of the selected states experienced a decrease in the
alcohol fatality rate (per 100 million vehicle miles traveled). In
2006, five of the seven selected states met the NHTSA goal pertaining
to alcohol-involved fatality rates. Officials said that high-visibility
campaigns provide other benefits beyond those for which the campaigns
are designed. For example, officials in North Carolina and Iowa said
that stopping drivers for potential safety belt or impaired-driving
violations also allowed them to increase overall traffic safety by
writing citations for other traffic violations, as well as apprehend
fugitives and recover stolen vehicles. Despite the progress made so
far, states face several challenges in conducting the high-visibility
campaigns and achieving desired results, including improving safety
belt use and reducing impaired driving among resistant populations--
such as pickup truck drivers--and recruiting sufficient officers to
conduct the campaigns when other law enforcement needs compete for
resources. Officials from NHTSA and some of the selected states also
cited weak prosecution of existing driving under the influence (DUI)
laws as an obstacle. NHTSA has initiatives under way to help states
address some of these challenges. For example, NHTSA has sponsored a
campaign--Buckle Up In Your Truck--to increase safety belt use by
pickup truck drivers. In addition, NHTSA provides funds that can be
used by states to purchase equipment for local law enforcement
agencies, such as breath-testing units, to encourage the agencies to
participate in impaired-driving campaigns.
To improve NHTSA's evaluations of the HVE campaigns, we recommend that
the Secretary of Transportation direct NHTSA to develop a minimum core
set of reporting requirements for states to report their federally
funded HVE law enforcement and media activities. In addition, we
recommend that the Secretary of Transportation direct NHTSA to include
additional performance measures--such as a measure for nighttime safety
belt use and additional measures of media effectiveness--in the
agency's annual evaluations of the effectiveness of the two campaigns.
DOT officials generally agreed with the findings and recommendations.
Background:
During 2006, more than 42,600 drivers, occupants, cyclists, and
pedestrians died as a result of motor vehicle crashes. Over the 10-year
period from 1997 through 2006, the number of motor vehicle fatalities
per 100 million vehicle miles traveled (VMT) has decreased by 14.1
percent, from 1.65 to 1.41. However, the number of fatalities annually
has remained relatively constant, showing only a slight increase of 1.5
percent, from 42,013 in 1997 to 42,642 in 2006 (see fig. 1).
Figure 1: Total Fatalities and Total Fatalities per 100 Million Vehicle
Miles Traveled (1985 to 2006):
[See PDF for image]
This figure is a combination vertical bar and line graph depicting the
following data:
Year: 1985;
Total Fatalities: 43,825
Total fatalities per 100 million VMT: 2.47.
Year: 1986;
Total Fatalities: 46,087;
Total fatalities per 100 million VMT: 2.51.
Year: 1987;
Total Fatalities: 46,390;
Total fatalities per 100 million VMT: 2.41.
Year: 1988;
Total Fatalities: 47,087;
Total fatalities per 100 million VMT: 2.32.
Year: 1989;
Total Fatalities: 45,582;
Total fatalities per 100 million VMT: 2.17.
Year: 1990;
Total Fatalities: 44,599;
Total fatalities per 100 million VMT: 2.08.
Year: 1991;
Total Fatalities: 41,508;
Total fatalities per 100 million VMT: 1.91.
Year: 1992;
Total Fatalities: 39,250;
Total fatalities per 100 million VMT: 1.75.
Year: 1993;
Total Fatalities: 40,150;
Total fatalities per 100 million VMT: 1.75.
Year: 1994;
Total Fatalities: 40,716;
Total fatalities per 100 million VMT: 1.73.
Year: 1995;
Total Fatalities: 41,817;
Total fatalities per 100 million VMT: 1.73.
Year: 1996;
Total Fatalities: 42,065;
Total fatalities per 100 million VMT: 1.69.
Year: 1997;
Total Fatalities: 42,013;
Total fatalities per 100 million VMT: 1.65.
Year: 1998;
Total Fatalities: 41,501;
Total fatalities per 100 million VMT: 1.58.
Year: 1999;
Total Fatalities: 41,717;
Total fatalities per 100 million VMT: 1.55.
Year: 2000;
Total Fatalities: 41,945;
Total fatalities per 100 million VMT: 1.53.
Year: 2001;
Total Fatalities: 42,196;
Total fatalities per 100 million VMT: 1.51.
Year: 2002;
Total Fatalities: 43,005;
Total fatalities per 100 million VMT: 1.51.
Year: 2003;
Total Fatalities: 42,884;
Total fatalities per 100 million VMT: 1.48.
Year: 2004;
Total Fatalities: 42,836;
Total fatalities per 100 million VMT: 1.45.
Year: 2005;
Total Fatalities: 43,510;
Total fatalities per 100 million VMT: 1.46.
Year: 2006;
Total Fatalities: 42,642;
Total fatalities per 100 million VMT: 1.41.
Source: GAO analysis of NHTSA and FHWA data.
[End of figure]
Two primary behaviors related to fatal crashes are failure to use
safety belts and driving while impaired by alcohol. Research has found
that using lap and shoulder safety belts reduces the risk of fatal
injury to front-seat passenger car occupants by 45 percent and light-
truck occupants by 60 percent. Overall, unrestrained fatalities
[Footnote 2] have decreased over the last two decades. From 1985 to
2006, the number of unrestrained fatalities decreased from 23,236 in
1985 to 16,053 in 2006, while the unrestrained fatality rate decreased
by 0.78, from 1.31 to 0.53 fatalities per 100 million vehicle miles
traveled (see fig. 2). The greatest improvements were achieved from
1989 to 1993, a period when most states passed initial safety belt use
laws. From 1984 to 1992, 8 states passed primary safety belt laws that
allow law enforcement officers to stop a driver for not wearing a
safety belt and issue a citation, and 33 states passed secondary safety
belt laws that allow law enforcement officers to issue a citation for
not wearing a safety belt only after the driver has been stopped for a
separate offense.
Figure 2: Unrestrained Vehicle Occupant Fatalities and Unrestrained
Fatalities per 100 Million Vehicle Miles Traveled (1985 to 2006):
[See PDF for image]
This figure is a combination vertical bar and line graph depicting the
following data:
Year: 1985;
Number of fatalities, restraint not used: 23,236;
Unrestrained fatalities per 100 million VMT: 1.31.
Year: 1986;
Number of fatalities, restraint not used: 24,373;
Unrestrained fatalities per 100 million VMT: 1.33.
Year: 1987;
Number of fatalities, restraint not used: 24,707;
Unrestrained fatalities per 100 million VMT: 1.29.
Year: 1988;
Number of fatalities, restraint not used: 25,308;
Unrestrained fatalities per 100 million VMT: 1.25.
Year: 1989;
Number of fatalities, restraint not used: 24,449;
Unrestrained fatalities per 100 million VMT: 1.17.
Year: 1990;
Number of fatalities, restraint not used: 23,298;
Unrestrained fatalities per 100 million VMT: 1.09.
Year: 1991;
Number of fatalities, restraint not used: 21,165;
Unrestrained fatalities per 100 million VMT: 0.97.
Year: 1992;
Number of fatalities, restraint not used: 19,589;
Unrestrained fatalities per 100 million VMT: 0.87.
Year: 1993;
Number of fatalities, restraint not used: 19,138;
Unrestrained fatalities per 100 million VMT: 0.83.
Year: 1994;
Number of fatalities, restraint not used: 19,243;
Unrestrained fatalities per 100 million VMT: 0.82.
Year: 1995;
Number of fatalities, restraint not used: 19,683;
Unrestrained fatalities per 100 million VMT: 0.81.
Year: 1996;
Number of fatalities, restraint not used: 19,425;
Unrestrained fatalities per 100 million VMT: 0.78.
Year: 1997;
Number of fatalities, restraint not used: 19,254;
Unrestrained fatalities per 100 million VMT: 0.75.
Year: 1998;
Number of fatalities, restraint not used: 18,621;
Unrestrained fatalities per 100 million VMT: 0.71.
Year: 1999;
Number of fatalities, restraint not used: 18,890;
Unrestrained fatalities per 100 million VMT: 0.70.
Year: 2000;
Number of fatalities, restraint not used: 18,382;
Unrestrained fatalities per 100 million VMT: 0.67.
Year: 2001;
Number of fatalities, restraint not used: 18,051;
Unrestrained fatalities per 100 million VMT: 0.65.
Year: 2002;
Number of fatalities, restraint not used: 18,342;
Unrestrained fatalities per 100 million VMT: 0.64.
Year: 2003;
Number of fatalities, restraint not used: 17,323;
Unrestrained fatalities per 100 million VMT: 0.60.
Year: 2004;
Number of fatalities, restraint not used: 16,993;
Unrestrained fatalities per 100 million VMT: 0.57.
Year: 2005;
Number of fatalities, restraint not used: 16,869;
Unrestrained fatalities per 100 million VMT: 0.56.
Year: 2006;
Number of fatalities, restraint not used: 16,053;
Unrestrained fatalities per 100 million VMT: 0.53.
Source: GAO analysis of NHTSA and FHWA data.
Note: Data for 1985 through 1987 data included unknown restraint use in
fatalities ranging from 12.6 percent to 18.0 percent compared with
unknown restraint use from 1988 through 2006 of 7.2 percent to 10.6
percent.
[End of figure]
While alcohol-impaired driving showed similar improvements from 1986 to
1994, progress has slowed, with a fluctuating number of alcohol-
involved fatalities[Footnote 3] and generally a declining alcohol-
involved fatality rate from 1994 to 2006 (see fig. 3). From 1985 to
2006, the number of alcohol-involved fatalities decreased by 4,964
people per year, and the alcohol-involved fatality rate decreased by
0.63, from 1.13 to 0.50 fatalities per 100 million vehicle miles
traveled. According to NHTSA, the improvements during the 1980s and
early 1990s were influenced by the passage in 1984 of a law that
withheld a portion of a state's federal highway funds unless the state
established a 21-year-old minimum drinking age[Footnote 4], the efforts
of two groups--the Presidential Commission Against Drunk Driving and
Mothers Against Drunk Driving--to galvanize public opinion about the
damage caused by impaired drivers, and states' efforts to strengthen
their impaired-driving laws and increase enforcement of those laws.
States received incentives to strengthen their laws and enforcement of
the laws through the Transportation Equity Act for the 21st Century
(TEA-21), which was enacted in 1998. For example, TEA-21 authorized
incentive grants to states to enact a law to establish 0.08 blood
alcohol content (BAC) as the legal limit for drunken driving offenses.
The TEA-21 Restoration Act provided added incentives to encourage
states to adopt an open container law that prohibits the possession of
any open alcohol beverage container in a motor vehicle and enact a law
that provides for specific penalties for individuals convicted of a
second or subsequent drunken driving offense. For states that did not
enact the open alcoholic beverage container and repeat drunken driving
laws, the TEA-21 Restoration Act also included a provision to transfer
a portion of those states' highway construction and maintenance funds
to the state's highway safety program.
Figure 3: Alcohol-Involved Fatalities and Alcohol-Involved Fatalities
per 100 Million Vehicle Miles Traveled (1985 to 2006):
[See PDF for image]
This figure is a combination vertical bar and line graph depicting the
following data:
Year: 1985;
Number of alcohol-involved fatalities: 20,086;
Alcohol-involved fatalities per 100 million VMT: 1.13;
Year: 1986;
Number of alcohol-involved fatalities: 21,471;
Alcohol-involved fatalities per 100 million VMT: 1.17.
Year: 1987;
Number of alcohol-involved fatalities: 20,696;
Alcohol-involved fatalities per 100 million VMT: 1.08.
Year: 1988;
Number of alcohol-involved fatalities: 20,599;
Alcohol-involved fatalities per 100 million VMT: 1.02.
Year: 1989;
Number of alcohol-involved fatalities: 19,531;
Alcohol-involved fatalities per 100 million VMT: 0.93.
Year: 1990;
Number of alcohol-involved fatalities: 19.607;
Alcohol-involved fatalities per 100 million VMT: 0.91.
Year: 1991;
Number of alcohol-involved fatalities: 17,599;
Alcohol-involved fatalities per 100 million VMT: 0.81.
Year: 1992;
Number of alcohol-involved fatalities: 15,847;
Alcohol-involved fatalities per 100 million VMT: 0.71.
Year: 1993;
Number of alcohol-involved fatalities: 15,547;
Alcohol-involved fatalities per 100 million VMT: 0.68;
Year: 1994;
Number of alcohol-involved fatalities: 14,985;
Alcohol-involved fatalities per 100 million VMT: 0.64.
Year: 1995;
Number of alcohol-involved fatalities: 15,242;
Alcohol-involved fatalities per 100 million VMT: 0.63.
Year: 1996;
Number of alcohol-involved fatalities: 15,263;
Alcohol-involved fatalities per 100 million VMT: 0.61.
Year: 1997;
Number of alcohol-involved fatalities: 14,421;
Alcohol-involved fatalities per 100 million VMT: 0.57.
Year: 1998;
Number of alcohol-involved fatalities: 14,207;
Alcohol-involved fatalities per 100 million VMT: 0.54.
Year: 1999;
Number of alcohol-involved fatalities: 14,250
Alcohol-involved fatalities per 100 million VMT: 0.53.
Year: 2000;
Number of alcohol-involved fatalities: 14,870;
Alcohol-involved fatalities per 100 million VMT: 0.54;
Year: 2001;
Number of alcohol-involved fatalities: 14,858;
Alcohol-involved fatalities per 100 million VMT: 0.53.
Year: 2002;
Number of alcohol-involved fatalities: 15,093;
Alcohol-involved fatalities per 100 million VMT: 0.53.
Year: 2003;
Number of alcohol-involved fatalities: 14,678;
Alcohol-involved fatalities per 100 million VMT: 0.51.
Year: 2004;
Number of alcohol-involved fatalities: 14,593;
Alcohol-involved fatalities per 100 million VMT: 0.49.
Year: 2005;
Number of alcohol-involved fatalities: 15,102;
Alcohol-involved fatalities per 100 million VMT: 0.51.
Year: 2006;
Number of alcohol-involved fatalities: 15,121;
Alcohol-involved fatalities per 100 million VMT: 0.50.
Source: GAO analysis of NHTSA and FHWA data.
[End of figure]
High-Visibility Enforcement (HVE) campaigns[Footnote 5] have been found
effective in the United States and other countries in helping to reduce
these two primary risk behaviors associated with fatal crashes. An HVE
campaign combines intensive enforcement of a specific traffic safety
law with extensive communication, education, and outreach informing the
public about the enforcement activity. For example, a safety belt
campaign could include several weeks during which television and radio
commercials warn motorists to buckle their safety belt or risk
receiving a ticket from increased law enforcement patrols, coupled with
zero tolerance enforcement of safety belt laws highly visible to
motorists through law enforcement techniques such as checkpoints and
saturation patrols. Such a combination of activities is designed to
increase the public's perception that people who violate the law will
be ticketed, arrested, convicted, or punished, thereby persuading them
to adhere to the law. HVE campaigns have been used for several decades
in the United States and other countries to improve safety belt use and
reduce impaired driving. Canada initiated the first safety belt HVE
campaigns in North America in the 1980s, during which time a 1-month
program in Ottawa, Ontario, increased belt use from 58 percent to 80
percent. Based on the Canadian HVE campaigns, the community of Elmira,
New York, conducted the first safety belt HVE effort in the United
States in 1985 and raised its safety belt use rate from 49 percent to
77 percent in 3 weeks. Impaired-driving HVE campaigns have also been
shown to be effective at reducing alcohol-impaired driving since 1967
in Britain and since 1980 in New Zealand. In the 1980s, law enforcement
agencies around the United States began using sobriety checkpoints to
deter impaired driving. For example, a yearlong checkpoint program in
1984 in Charlottesville, Virginia, was associated with a 13 percent
reduction in alcohol-related crashes.[Footnote 6]
While HVE campaigns have proved effective in the United States and
other countries, selected other countries GAO reviewed generally have
higher safety belt use rates and lower impaired-driving fatality
percentages than the United States. For example, while the United
States had a 2007 safety belt use rate of 82 percent, Canada had a 2006-
2007 safety belt use rate of 93 percent, Australia has a safety belt
use rate of around 96 percent, and the Netherlands had a 2005 safety
belt use rate of 90 percent. An official from Canada noted that, while
HVE campaigns in the 1980s and early 1990s had been successful in
improving Canada's safety belt use rate, the rate has remained stagnant
over the last 10 years and that approximately 40 percent of Canada's
traffic fatalities still involved unbelted persons. The official
attributed the lack of further progress to the fact that most of the
last 10 percent of persons not wearing their safety belts are actively
choosing not to wear the belts. Officials from Australia and the
Netherlands noted that impaired-driving fatalities have been reduced in
their countries because law enforcement officials are allowed to stop
drivers at random to test the driver's breath for alcohol. However,
this deterrent may be difficult to implement in the United States
because other than at sobriety checkpoints (which are not allowed in
some states), drivers cannot be stopped unless there is suspicious
behavior or another traffic offense involved. Appendix II provides
further details on HVE campaigns in other countries.
The CIOT and OTLUA campaigns typically span about 7 weeks (see fig. 4).
The CIOT campaign is conducted during May to coincide with Memorial
Day, and the OTLUA campaign is conducted during August and September to
cover Labor Day and again in December to cover the holiday season and
New Year's Eve. The campaign activities conducted by NHTSA, state
traffic safety offices, state law enforcement agencies, and local law
enforcement agencies over the 7 weeks generally include the following:
* Precampaign evaluation. Data such as safety belt use and public
knowledge and attitudes about traffic enforcement programs are
collected prior to the campaign to provide a baseline. States generally
collect these data through safety belt use surveys, department of motor
vehicle driver surveys, and telephone surveys.
* Earned media. Earned media is unpaid coverage by broadcast and
published news services, such as a press conference or press release
provided by the state or local law enforcement officials. These media
events are used to announce the upcoming campaign, bring news coverage
to the ongoing enforcement effort, and update the public on the
progress and results of the campaign.
* Paid media. Paid media includes advertisements on television and
radio. NHTSA purchases these advertisements nationwide, which are
strategically placed at times and places intended to maximize exposure
to selected audiences. For example, advertisements targeted toward 21-
to 34-year-old men who are more likely to drive impaired might air on
sports programs during a time when the most people in the target
audience are likely to be watching and listening. States may augment
the national advertising with advertisements directed at state-level
high-risk populations such as pickup truck drivers or with taglines to
let the audience know that their local law enforcement agencies are
involved in the campaign.
* Enforcement. Enforcement techniques by state and local law
enforcement agencies may include aggressive enforcement by routine
patrols, "saturation" patrols that increase the number of officers on
patrol in a specific area, and stationary checkpoints along roadsides.
States and local agencies may use traffic safety grant funds
administered by NHTSA through their state highway safety office to
provide the increased level of enforcement.
* Postcampaign evaluation. Data are collected by states and local
agencies after the campaign in the same manner as the precampaign
evaluations and compared with precampaign data to identify changes in
awareness of the enforcement effort, measure progress toward campaign
goals, and measure the impact on traffic safety.
Figure 4: HVE Campaign Activities and Timeline:
[See PDF for image]
This figure is an illustration of HVE Campaign Activities and Timeline,
as follows:
Precampain evaluation: Week 1 and 2;
Earned Media: Week 3 through Week 7;
Paid Media: Week 4 through Week 6;
Enforcement: Middle of Week 4 through Week 6;
Postcampaign evaluation: Week 7.
Source: GAO and NHTSA.
Note: The timeline is a general representation of campaign activities.
The actual timing of the activities varies slightly between the CIOT
and the OTLUA campaigns.
[End of figure]
Campaigns that were held prior to 2003--often referred to as Selective
Traffic Enforcement Programs--typically relied on earned media such as
unpaid television and radio news stories to advertise the campaigns;
however, since 2003, Congress has funded nationwide paid advertising
for safety belt and impaired-driving campaigns. (See table 1.) The use
of paid media allows advertisements to be placed at optimal times with
high-quality messages so the campaign can better reach its target
audiences and maximize the probability that the audience will pay
attention to the advertisements, whereas earned media placement and
frequency are usually controlled by station managers and may not be
placed at optimal times. NHTSA used funds authorized under TEA-21 for
activities such as developing and producing broadcast and print
advertisements and providing media technical assistance to the states.
Although TEA-21, when enacted, did not authorize funding to purchase
national advertising for the campaigns, Congress appropriated funding
of $19 million in 2003 and $24 million in 2004 and 2005 for NHTSA to
provide paid national advertising for both campaigns. In 2005, SAFETEA-
LU authorized $29 million in each of fiscal years 2006 through 2009 for
NHTSA to conduct a nationwide HVE program. The program requirements
included developing and disseminating advertisements, coordinating with
states, and annually evaluating the effectiveness of the program. NHTSA
uses available funding to purchase national media time and conduct
evaluations for the campaigns (see table 1). NHTSA also uses funding
from other highway safety programs to develop the advertising and
includes campaign coordination efforts as a part of the agency's
ongoing coordination efforts with states for other highway safety
programs.[Footnote 7] NHTSA officials report that the increase in
funding authorized by SAFETEA-LU--$5 million above the annual funding
level immediately prior to SAFETEA-LU--has allowed them to increase the
visibility and frequency of advertising for the two enforcement
campaigns during Memorial Day and Labor Day, as well as allowed them to
provide additional national advertising for the impaired-driving
campaign during the December holiday season.
Table 1: Funding for HVE Paid Media and Evaluations, Fiscal Years 2003-
2007 (Dollars in millions):
Fiscal year: 2003;
CIOT: $8.00;
OTLUA: $11.00;
Evaluation: $0;
Available funding: $19.
Fiscal year: 2004;
CIOT: $10.00;
OTLUA: $14.00;
Evaluation: $0;
Available funding: $24.
Fiscal year: 2005;
CIOT: $9.92;
OTLUA: $13.89;
Evaluation: $0;
Available funding: $24.
Fiscal year: 2006;
CIOT: $10.00;
OTLUA: $17.96;
Evaluation: $0.75;
Available funding: $29.
Fiscal year: 2007;
CIOT: $10.00;
OTLUA: $18.25;
Evaluation: $0.75;
Available funding: $29.
Source: GAO analysis of NHTSA data.
[End of table]
States have used federal traffic safety grants authorized in TEA-21 and
SAFETEA-LU--such as State and Community Highway Safety grants, Safety
Belt Use grants, and Alcohol-Impaired Driving Countermeasures grants--
to fund state and local campaign activities.[Footnote 8] States and
local governments use grant funds for activities such as paying
overtime for law enforcement officials to conduct sobriety checkpoints
and saturation patrols, purchasing paid advertising, training,
conducting safety belt surveys, and buying enforcement equipment. The
states and local governments are also allowed to use highway safety
funding to provide local advertising and educational campaigns in
conjunction with the national media campaign. When included in the
state's annual highway safety plan, a state may also conduct campaigns
in addition to the national campaigns and provide sustained enforcement
utilizing federal funds. States also use grant funds to purchase law
enforcement equipment such as alcohol breath testers, radar units, and
in-car video cameras to provide incentives for local law enforcement
agencies to participate in the campaigns.
NHTSA Has Implemented the HVE Program but Could Improve the Annual
Evaluations of Campaign Effectiveness:
NHTSA has implemented the HVE program, including two high-visibility
traffic safety law enforcement campaigns to improve safety belt use and
reduce impaired driving. Specifically, to meet the requirements in
place since SAFETEA-LU, NHTSA has (1) developed and disseminated
advertisements, (2) coordinated with states to conduct the HVE
campaigns, and (3) evaluated the results of the campaigns. However, the
evaluations have shortcomings that limit NHTSA's ability to assess the
level of state activity and the overall effectiveness of the campaigns.
NHTSA Has Developed an Advertising Plan, Created Advertising, and
Purchased Media Time:
To develop and disseminate advertising for the CIOT and OTLUA
campaigns, NHTSA has developed an advertising plan and hired a
contractor to create advertising materials for national and state use
and purchase national media time. Since 2005, NHTSA has annually
developed a National Communications Plan that sets forth a national HVE
campaign advertising strategy. For example, the plan[Footnote 9]
specifies goals, dates, target audiences, and core campaign messages
for the campaigns. The plan also identifies how the campaign
advertising should be developed and purchased to cost-effectively reach
target audiences and includes links to Web sites that contain
additional guidance and advertising materials. State traffic safety
agencies can use these materials and develop supplemental advertising
materials following the guidance provided in the plan.
To create advertising materials, NHTSA contracts with a private
advertising firm to provide technical assistance and ad production,
including:
* producing national ads;
* modifying or updating national and state ads;
* developing a national plan to purchase media;
* reviewing states' plans to purchase media, and;
* negotiating and purchasing air time for national ads.
National ads are produced in several media formats and languages. Media
formats include television, radio, magazines, newspapers, and
alternative media. Broadcast television, cable television, and radio
are the three most used media formats to advertise HVE campaigns,
accounting for about 85 percent of the amount that NHTSA spends on
campaign advertising. NHTSA has also begun to use the Internet to reach
the target audience of young males by placing advertising messages into
online games, social sites such as Face Book, and sports sites such as
ESPN.com. These national ads are primarily produced in English and
Spanish. NHTSA officials reported that they considered developing
advertising for additional non-English-speaking populations and made
the decision that it was not cost effective. However, they encourage
states to develop materials for other non-English-speaking populations
that are prevalent in the state's population. NHTSA's contractor also
refreshes existing ads because, according to NHTSA officials, they can
reduce costs by updating ads with new taglines or messages rather than
creating new ads each year. For example, NHTSA darkened an existing
television ad that had been filmed in the daytime to make it appear as
though it were night to support an enforcement message for nighttime
safety belt use.
To purchase media time for the national ads, the NHTSA contractor
prepares a plan to purchase media for NHTSA's approval and release
before each national campaign. This plan identifies the advertising
period, the media budget, target audience profiles, a strategy for
purchasing the media, and the allocation of funds for different media
formats. The allocation is based on reaching the campaign target
audience as frequently and cost effectively as possible, the target
audiences' use of the various media types, and the cost of placing the
advertisements. For example, prime-time broadcast television reaches
many young men, but because it is expensive, NHTSA may build the
desired frequency of reaching the young men with cable television or
radio, which is less expensive. NHTSA is increasing funds allocated for
Hispanic media outlets, based on fatality and census data, and
alternative media outlets as young men spend more time on the Internet.
For example, the allocation for Hispanic media for the impaired-driving
campaign increased from 5 percent to 12 percent from 2004 to 2007, and
the allocation for alternative media in the safety belt campaign has
increased from 0 percent to 5 percent from 2004 to 2007.
Once NHTSA approves the plan to purchase media, the contractor
negotiates with media providers--such as television or radio networks-
-to purchase media. The contractor also negotiates for value-added
media, which is advertising time that a television or radio network may
provide to NHTSA at no additional cost because the network supports the
campaign message. After a campaign, the media contractor provides an
analysis of the effectiveness of the media formats, including the
extent to which the formats reached the target audience, the cost to
reach the audience, and the dollar value of the value-added media.
The National Communications Plan, the plans to purchase media, and
other resources that NHTSA uses to advertise its campaigns include
elements of the key practices we have previously identified through an
expert panel[Footnote 10] as important to planning a consumer education
campaign, motivating a target audience, and alleviating challenges in a
campaign.[Footnote 11] The key practices include the following:
* Define goals and objectives. NHTSA has established goals to reduce
deaths and injuries from crashes on our nation's highways by increasing
the number of people regularly using safety belts and decrease the
number of impaired drivers on the road.
* Analyze the situation. NHTSA applies research, lessons learned, and
other knowledge such as program evaluations to develop an integrated
year-round marketing campaign designed to modify behavior with a
calendar of events timed to coincide with national holidays and
celebrations, which have an increased number of traffic fatalities.
* Identify stakeholders. NHTSA engages national, state, and local
partners, such as coalitions, highway safety offices, and law
enforcement agencies across the country, to be involved in the calendar
of events. In addition, the communications plan identifies the need to
look for marketing alliances with sports organizations and other
corporations that already carry powerful brands important to NHTSA's
target audiences.
* Identify resources. Annual funding for the national campaigns is
authorized in SAFETEA-LU through fiscal year 2009. NHTSA identifies the
resources available for each campaign in the plan for purchasing media.
* Research target audiences. NHTSA reviews existing research and
surveys to help segment and target the key audiences, identifying
preferences, beliefs, competing behaviors, and motivators.
* Develop consistent, clear messages. The National Communications Plan
specifies the need to maintain a strong CIOT and OTLUA brand-name
status through consistency in presentation and broad geographical
coverage.
* Identify credible messengers. The credibility of the message lies in
the combination of the message with a high level of enforcement,
creating a general deterrence effect that increases the public
perception that drivers are likely to receive a citation. The
advertisements use law enforcement officers, who increase the
credibility of the message. Additionally, the messages are produced and
presented through media intended to appeal to the intended target
audience, such as teens, pickup truck drivers, and rural audiences.
* Design media mix. NHTSA identifies the target audiences and
appropriate media for target audiences to allocate media funding.
* Establish metrics to measure success. NHTSA's media contractor
analyzes the results of each campaign to evaluate whether the message
reached the intended target audience in the time period intended, and
NHTSA reports how the campaign reached the target market.
State Officials Report That NHTSA's Coordination Efforts Help Them
Conduct Successful Campaigns:
NHTSA coordinates with the states and provides resources to help states
carry out the campaigns through several means, including the National
Communications Plan, guidance on conducting HVE campaigns, technical
assistance on advertisements, and collateral advertising materials.
Officials in selected states reported that NHTSA's coordination efforts
provided the support and interaction needed to successfully conduct HVE
campaigns. For example, officials from one state noted that NHTSA
assisted them in applying for federal traffic safety grants to conduct
campaigns and provided tool kits that were useful in developing the
campaigns. Officials from another state reported that NHTSA had
improved the quality and timeliness of advertising materials, allowing
them to devote more state resources to purchasing radio and television
ads rather than developing the ads.
The annual National Communications Plan disseminates a strategy for
states to conduct occupant protection and impaired-driving events
throughout the year, including the CIOT and OTLUA campaigns.
Specifically, the National Communications Plan sets out the following:
* a primary purpose for each event, such as "to support enforcement
activities and to remind all partygoers of the dangers of impaired
driving";
* dates for the events;
* messages to be emphasized, such as "Drunk Driving, Over the Limit,
Under Arrest";
* primary and secondary target audiences, such as men, ages 21 to 34;
and;
* potential themes, such as "Buzzed Driving is Drunk Driving--Designate
a Sober Driver"
The National Communication Plan also provides links to Web sites
containing guidance for states in conducting campaigns, such as
[hyperlink, http://www.TrafficSafetyMarketing.gov]. NHTSA also provides
further guidance, such as the Uniform Guidelines for State Highway
Safety Programs, which includes guidance for both occupant protection
and impaired driving; and case studies of HVE campaigns, such as
NHTSA's "Creating Impaired Driving General Deterrence--Eight Case
Studies of Sustained, High-Visibility, Impaired-Driving
Enforcement."[Footnote 12]
NHTSA also furnishes technical advice and collateral materials to
assist states with advertising for the campaigns. To provide technical
support to states, NHTSA's contractor may, when requested by state
officials, evaluate states' proposed media purchases and make
suggestions for improvement. The evaluation includes reviewing states'
proposed target demographics, budget, and purchase of advertising time
to provide guidance on the appropriateness of the purchase. These
evaluations are intended to help states effectively reach target
audiences. NHTSA also provides collateral materials such as posters,
Web banners, talking points, and model press releases. States may
download these materials directly from NHTSA's Web site. These
materials are designed to support the various events set out in the
National Communications Plan.
NHTSA's Annual Evaluations Are Limited by Inconsistent and Incomplete
Data:
NHTSA's annual evaluations of the HVE campaigns include information on
the level of enforcement activity and the results of the campaigns
based on performance measures, such as message awareness, earned media
activity, safety belt use, and fatality and injury statistics.[Footnote
13] For example, the CIOT evaluation[Footnote 14] includes information
on the number of law enforcement agencies that reported enforcement
activities and the number of safety belt citations issued by these
agencies; this information showed that the number of citations issued
increased from 2004 to 2005, even though the number of reporting
agencies declined. Regarding message awareness, the OTLUA annual
evaluation[Footnote 15] includes information from NHTSA's annual
national telephone surveys, which found that the impaired-driving
message was reaching the general public--especially the 18-to 34-year-
old target audience--although the awareness did not carry over from
campaign to campaign.[Footnote 16] The CIOT report showed that safety
belt use rates generally increased following the 2005 campaign, and the
OTLUA report showed that the number of alcohol-impaired drivers
involved in fatal crashes decreased overall from 2001 to 2005.
However, the data on HVE campaign activity--such as the number of
agencies participating in the campaigns, hours worked by law
enforcement officers, citations issued, DUI enforcement actions, and
advertisements purchased by states--that states report to NHTSA are not
complete or consistent; this situation limits NHTSA's ability to
evaluate the overall level of state enforcement and advertising
activity and the extent to which states use federal funding--through
traffic safety grants--to support HVE campaigns. Because these
campaigns--other than the media developed and purchased by NHTSA--are
carried out by states, these data are the only way to determine whether
the level of activity is changing from year to year and whether NHTSA
is effectively leveraging state and local resources. According to NHTSA
officials, states are not required to report all HVE activity, although
in recent years states have voluntarily reported the level of activity
for selected law enforcement agencies--generally those agencies that
receive federal grants for HVE activities.[Footnote 17] However, such
voluntary reporting can cause substantial variances in data from
campaign to campaign and year to year. For the campaigns conducted from
2003 through 2006, an average of three states did not report on
campaign activity for each campaign, and between 22 percent to 52
percent[Footnote 18] of the law enforcement agencies that indicated
they would participate in the campaigns did not report on campaign
activity.
Of the agencies that do report, the data reported are not consistent
among law enforcement agencies or states. For example, some agencies
include all activities and others include only the federally funded
portion of their activities. In addition, some states only require a
portion of the activities to be reported by agencies and leave
reporting on other NHTSA requested activities as optional. As a result,
the types of activity data collected from state to state vary. For
example, in the May 2006 CIOT campaign, while 49 states reported having
participating agencies, only 37 states reported the hours worked and
only 36 states reported the number of earned media TV spots. Due to
these inconsistencies, NHTSA has reported that it is not possible to
provide meaningful analyses and comparisons of state activities to
conduct HVE campaigns.
NHTSA and Governors Highway Safety Association (GHSA) officials
stressed that law enforcement agencies are less likely to report if
they receive little or no federal highway safety grant funding for
enforcement activities. GHSA officials--who represent the state highway
safety offices--suggested that simplifying reports and limiting the
amount of data required may improve reporting. While it is important
for NHTSA to limit reporting requirements for states, state and local
law enforcement is a critical component of HVE campaigns that NHTSA
currently cannot measure. A minimum set of core reporting requirements-
-such as the number of agencies reporting data, number of law
enforcement labor hours applied to the mobilization, number of impaired-
driving arrests made during the mobilization, number of safety belt
violation citations issued, and amount spent on television, radio,
print, and other ads from agencies that receive federal funding for
these activities--would minimize the reporting burden while allowing
NHTSA to more thoroughly and consistently measure the level of state
activity over time and provide accountability for federal funding.
To learn more about the extent of participation by local law
enforcement agencies, NHTSA collected information from a sample of law
enforcement agencies that conducted campaigns independent of the
national mobilization evaluations or that had received federal traffic
safety grants. NHTSA sought a representative sample of law enforcement
agencies that had at least 10 years of citation and arrest data on a
monthly basis in order to track enforcement activities over time and
compare enforcement activities before, during, and after the CIOT and
OTLUA campaigns. The results of this data collection effort will be
reported in the Evaluation of the 2006 CIOT Campaign (to be released
later in 2008). This attempt to obtain more reliable and representative
information has faced several obstacles, including difficulties in
defining a representative sample of law enforcement agencies, locating
and securing the cooperation of agencies that had 10 years of citation
and arrest data, and finding agencies that would provide the data on a
monthly basis.
NHTSA's effort to evaluate the effectiveness of the HVE campaigns is
also hindered, in part, because NHTSA's performance measures are not
comprehensive. For example, while NHTSA measures the change in daytime
safety belt use for the driver and right front passenger in passenger
cars, vans, sports utility vehicles, and pickup trucks, it does not
directly measure nighttime safety belt use, despite recent efforts to
increase the use of safety belts at night. Specifically, NHTSA is
working with the states of Washington, North Carolina, and West
Virginia in pilot programs to evaluate the use of different enforcement
strategies for increasing nighttime safety belt use. NHTSA and the
states expect to report the results of these pilot programs by the end
of 2008. In addition, NHTSA is preparing an enforcement guide on the
different approaches states may use for nighttime enforcement during
the 2008 CIOT campaign.
The annual evaluations also include limited information on performance
measures for the effectiveness of NHTSA's advertisements. For example,
while the evaluations include information on the extent to which the
advertisements are reaching the primary target audience, the
evaluations did not measure the extent to which the advertisements
reached special-emphasis audiences identified in the 2005 National
Communications Plan,[Footnote 19] such as pickup truck drivers and
Hispanics for the CIOT campaign. The impaired-driving report also did
not evaluate the extent to which the advertisements had reached other
targeted audiences, such as college students, men ages 35 to 59, and
young women 21 to 25 who were also identified in the 2005 National
Communications Plan,[Footnote 20] and did not include the media dollar
allocation to show how NHTSA had advertised to non-English-speaking
populations and used nontraditional media.[Footnote 21] Without this
information, NHTSA cannot evaluate the extent to which the campaigns
are meeting the goals set out in the National Communications Plans.
NHTSA officials recognize the need for more comprehensive performance
measures, and--through a contractor--the agency is developing
additional performance measures to address these issues. The statement
of work for the contractor specifies that the purpose of the project is
to develop a minimum set of performance measures that could be used by
federal, state, and local governments for traffic safety areas,
including high-visibility enforcement campaigns. The deadline for this
work is August 2008. However, NHTSA officials stated that the key
requirement for developing effective performance measures is accurate
and comprehensive data and that existing data available to states are
not sufficient to mandate more specific performance measures. As a
result, NHTSA plans to recommend--not require--the new performance
measures to states. It is unlikely that all states will voluntarily
report the same performance indicators in a consistent and
comprehensive manner sufficient to allow national comparisons without
specific required measures.
Selected State Officials Report That HVE Campaigns Are Contributing to
Increased Safety Belt Use and Reduced Fatalities, but Several
Challenges Hinder Further Progress:
According to officials in selected states we visited, HVE campaigns are
contributing to increased safety belt use and reduced alcohol-involved
fatalities. From 1997 to 2006, safety belt use increased in all seven
of the selected states, and four of those states exceeded the 2006
NHTSA goal for safety belt use (82 percent). In addition to increases
in safety belt use, from 1997 to 2006, each selected state's alcohol
fatality rate decreased, and, in 2006, five of the seven states met the
NHTSA goal pertaining to alcohol-involved fatality rates. Despite the
gains made so far, officials from these states reported facing several
challenges: increasing safety belt use and reducing impaired driving
among resistant populations; insufficient staff to conduct the
campaigns; and weak prosecution of DUI arrests. NHTSA and the states
are taking steps to help address these challenges.
Officials in Selected States Report That HVE Campaigns Are Contributing
to Increased Use of Safety Belts and Reduced Fatalities from Impaired
Drivers:
HVE campaigns are contributing to increased safety belt use and reduced
alcohol-involved fatalities, according to officials in selected states
we visited. Specifically, all of the selected states experienced
increased safety belt use and reduced alcohol-involved fatality rates
in the last 10 years, and state officials attributed these
improvements, in part, to participation in HVE campaigns.
According to NHTSA data, between 1997 and 2006, safety belt use
increased in all of the selected states, although some states
experienced larger increases than other states. According to NHTSA
survey data on safety belt use, the increase in safety belt use from
1997 to 2006 ranged from a 6.5 percentage point increase in North
Carolina to a 29.6 percentage point increase in North Dakota. The
overall increase in safety belt use nationwide from 1997 to 2006 was 12
percentage points (see fig. 5).
Figure 5: Increase in Safety Belt Use for Selected States Compared with
Overall U.S. Increase, 1997 through 2006:
[See PDF for image]
This figure is a horizontal bar graph depicting the following data:
Increase in Safety Belt Use for Selected States Compared with Overall
U.S. Increase, 1997 through 2006:
State: North Dakota;
Percent increase: 29.6%.
State: Illinois;
Percent increase: 25.7%.
State: Washington;
Percent increase: 19%.
State: Arkansas;
Percent increase: 18.8%.
State: Rhode Island;
Percent increase: 15%.
State: Iowa;
Percent increase: 14.7%.
State: North Carolina;
Percent increase: 6.5%.
State: Overall U.S.;
Percent increase: 12%.
Source: GAO analysis of NHTSA data.
[End of figure]
The range in improvements in safety belt use rates from 1997 to 2006
can be attributed in part to the safety belt use rate each state had
achieved by 1997. For example, in 1997, North Carolina had achieved an
82 percent safety belt use rate, which exceeded the U.S. average safety
belt use rate of 69 percent at that time. In contrast, North Dakota's
safety belt use rate in 1997 was only 49 percent.
In 2006, safety belt use in the seven selected states ranged from 69
percent in Arkansas to 96 percent in Washington. Nationwide, safety
belt use in 2006 ranged from a low of 64 percent in New Hampshire and
Wyoming to Washington's 96 percent. Safety belt use in four of the
selected states we visited--Illinois, Iowa, North Carolina, and
Washington--exceeded the 2006 NHTSA safety belt use goal of 82 percent.
All four of these states had a primary safety belt law in place by
2006. Safety belt use rates for 2006 in Arkansas, Rhode Island, and
North Dakota--states without a primary safety belt law--fell short of
the 2006 federal goal (see fig. 6).[Footnote 22]
Figure 6: Safety Belt Use Compared with Federal Goal, Selected States
(2006):
[See PDF for image]
This figure is a horizontal bar graph depicting the following data:
Safety Belt Use Compared with Federal Goal, Selected States (2006):
State: Arkansas;
Belt use (percent): 69.3%.
State: Rhode Island;
Belt use (percent): 74%.
State: North Dakota;
Belt use (percent): 79%.
State: Illinois;
Belt use (percent): 87.8%.
State: North Carolina;
Belt use (percent): 88.5%.
State: Iowa;
Belt use (percent): 89.6%.
State: Washington;
Belt use (percent): 96.3%.
State: NHTSA goal;
Belt use (percent): 82%.
Source: GAO analysis of NHTSA data.
[End of figure]
According to officials, selected states' HVE campaigns are also
contributing to reducing alcohol-involved fatality rates. From 1997 to
2006, all of the selected states experienced a decrease in alcohol-
involved fatality rates. The decrease in the alcohol-involved fatality
rate during this period ranged from 22 percent in Rhode Island and
North Dakota to 3 percent in Arkansas. Five of the seven selected
states experienced declines in alcohol-involved fatality rates that
were greater than the overall U.S. decrease of 12 percent (see fig. 7).
Figure 7: Decrease in Alcohol-Involved Fatality Rate per 100 Million
Vehicle Miles Traveled for Selected States Compared with Average
Decrease in the United States (1997 to 2006):
[See PDF for image]
This figure is a horizontal bar graph depicting the following data:
Decrease in Alcohol-Involved Fatality Rate per 100 Million Vehicle
Miles Traveled:
State: Rhode Island;
Percent decrease: 22%.
State: North Dakota;
Percent decrease: 22%.
State: Iowa;
Percent decrease: 21%.
State: North Carolina;
Percent decrease: 18%.
State: Washington;
Percent decrease: 13%.
State: Illinois;
Percent decrease: 9%.
State: Arkansas;
Percent decrease: 3%.
State: U.S. average;
Percent decrease: 12%.
Source: GAO analysis of NHTSA and FHWA data.
[End of figure]
In 2006, alcohol-involved fatality rates in the seven selected states
ranged from 0.62 fatalities per 100 million vehicle miles traveled in
Arkansas to 0.4 in Rhode Island. Nationwide, the average alcohol-
involved fatality rate in 2006 was 0.50 fatalities per 100 million
vehicle miles traveled. In 2006, five of the seven selected states
experienced fatality rates that were lower than the NHTSA goal of 0.51
fatalities per 100 million vehicle miles traveled. North Dakota and
Arkansas experienced alcohol-involved fatality rates in 2006 that fell
short of the NHTSA goal (see fig. 8).
Figure 8: Fatalities with a BAC of 0.08 or Greater per 100 Million
Vehicle Miles Traveled for Selected States Compared with Federal Goal
(2006):
[See PDF for image]
This figure is a horizontal bar graph depicting the following data:
Fatalities with a BAC of 0.08 or Greater per 100 Million Vehicle Miles
Traveled:
State: Arkansas;
Fatalities per 100 million VMT: 0.62.
State: North Dakota;
Fatalities per 100 million VMT: 0.56.
State: North Carolina;
Fatalities per 100 million VMT: 0.47.
State: Illinois;
Fatalities per 100 million VMT: 0.46.
State: Washington;
Fatalities per 100 million VMT: 0.44.
State: Iowa;
Fatalities per 100 million VMT: 0.41.
State: Rhode Island;
Fatalities per 100 million VMT: 0.40.
State: Federal Goal;
Fatalities per 100 million VMT: 0.51.
Source: GAO analysis of NHTSA and FHWA data.
[End of figure]
Officials from the selected states we visited also identified other
benefits that result from participating in HVE campaigns. For example,
North Carolina and Iowa officials said that, during HVE campaigns, the
additional officers staffing checkpoints or on patrol are able to
apprehend suspects in other crimes and write citations for traffic
violations such as speeding; this increased level of enforcement
activity contributes to improvements in overall traffic safety. For
example, during a 2006 CIOT campaign in North Carolina, the state
Department of Transportation reported that it issued over 23,000
speeding tickets but also arrested 699 fugitives and recovered 141
stolen vehicles. Iowa officials also said that HVE campaigns yield
benefits in that they will often find drugs or stolen property when
stopping vehicles, and the campaigns serve to improve relationships
between law enforcement personnel and the community. In addition,
another benefit is that law enforcement agencies can use the equipment
purchased for HVE campaigns--such as patrol cars, vehicle cameras, and
BAC testing equipment--to enhance traffic safety enforcement efforts
throughout the year.
Selected States Face Several Challenges in Conducting HVE Campaigns:
Despite the gains made so far, several challenges hinder further
progress in carrying out the HVE campaigns. The challenges cited by
officials from the states we visited include: increasing safety belt
use and reducing impaired driving among resistant populations;
insufficient staff to conduct the campaigns; and weak prosecution of
DUI arrests. Reviews of NHTSA's HVE campaigns from DOT's Office of the
Inspector General and the National Transportation Safety Board (NTSB)
have found similar challenges.[Footnote 23]
Challenge in Increasing Safety Belt Use and Reducing Impaired Driving
among Resistant Populations:
Officials in selected states face the challenge of increasing safety
belt use and reducing impaired driving among resistant populations,
such as drivers in rural areas, pickup truck drivers, and hardcore
drinking drivers. For example, statistics show that more drivers in
rural areas resist wearing safety belts. Though recent progress has
been made, in general, rural areas have a higher proportion of fatal
crashes and traffic fatalities than in urban areas. In a recent NHTSA
analysis of urban and rural fatalities, NHTSA reported that rural
fatalities accounted for 55 percent of fatal crashes and 57 percent of
traffic fatalities in 2006 even though only 23 percent of the U.S.
population lived in rural areas, according to 2006 Census estimates.
[Footnote 24] Several factors in addition to lower safety belt use
contribute to this disparity, including higher alcohol-involved crash
rates, higher speed, rural roads that are narrow or have sharp curves,
and less access to emergency services in rural areas. Crashes in rural
areas are also more likely to involve occupants who are ejected from
vehicles because they are not wearing safety belts. According to one
study, of the 5,959 people who died in rural crashes where a vehicle
occupant was ejected or partially ejected from the vehicle, 92 percent
were unbelted or were not properly restrained in a child safety seat.
Statistics show that pickup truck drivers are also resistant to
changing their safety belt habits. Compared with other drivers, pickup
truck drivers are more resistant to wearing safety belts. In 2007,
according to NHTSA, 72 percent of pickup truck drivers were belted
compared with 84 percent of passenger car drivers. This lower safety
belt use by pickup truck drivers and their passengers can lead to more
vehicle occupant ejections and fatalities.
NHTSA and the states are taking steps to increase rural safety belt use
and have developed programs targeting pickup truck drivers. In 2006,
for example, a NHTSA report outlined several strategies that states and
local communities could use to improve their rural safety belt programs
and provided examples of leading enforcement and communication programs
in many states.[Footnote 25] Illinois, one of selected states we
visited, targeted rural safety belt use in 2005 by participating in a
NHTSA Great Lakes Region Rural Demonstration program. The program
involved intensified enforcement and paid media to alert residents in
targeted rural areas that safety belt laws would be enforced. During
the program, Illinois aired television and radio ads in five media
markets that include rural areas and conducted safety belt enforcement
zones by stopping vehicles if an unbuckled occupant was observed.
Observational surveys of safety belt use in the targeted rural areas in
Illinois showed belt use increased from a baseline of 78.5 percent
before the Rural Demonstration program to 81.5 percent after the
program. Other states are also taking steps to increase safety belt use
in rural areas. For example, Washington state developed a corridor
program to improve traffic safety and safety belt use for one of its
rural roads by increasing enforcement and installing signs to remind
drivers to buckle up. Another example is in Iowa, where a rural youth
organization expanded a program called "Farm Safety Just 4 Kids" to
improve safety belt use among younger drivers in rural areas.
NHTSA and the states have also taken actions to address the challenge
of increasing belt use by drivers and occupants of pickup trucks. For
example, in 2000, NHTSA initiated a study to identify safety belt use
rates and important characteristics of pickup truck drivers and
passengers, review public information campaigns intended for pickup
truck drivers, obtain qualitative information about pickup truck
drivers' knowledge and attitudes about safety belt use, and make
suggestions for the development of future campaigns targeting pickup
truck drivers. Some of the selected states, in concert with NHTSA, have
participated in special campaigns that are designed to increase safety
belt use in pickup trucks. Both Arkansas and North Carolina, working
with their NHTSA regional offices, participated in safety belt
campaigns in 2006 called "Buckle Up in Your Truck." In Arkansas, the
campaign ran during 2 weeks in May, and the ads for the campaign aired
in five media markets. The ads ran at times and on programs that are
popular with young males who are more likely to be driving pickup
trucks. An analysis of a 2005 "Buckle Up in Your Truck" campaign in
Alabama showed a greater than 4 percent increase in safety belt use in
pickup trucks.
Officials in selected states also face the challenge of reducing
impaired driving among hardcore drinking drivers. Hardcore drinking
drivers are those who drive with a BAC of 0.15 or greater. According to
the NTSB, hardcore drinking drivers are involved in 54 percent of
alcohol-involved fatalities, and these drivers are likely to be repeat
drinking drivers. NHTSA and the states have taken steps to address this
challenge. In August 2007, NHTSA recommended increased use of ignition
interlock devices--an in-car breath tester connected to the ignition
that prevents the vehicle from starting if the device registers a BAC
over a specified limit--as part of a penalty against repeat drunken
driving offenders. In addition, NHTSA has published uniform guidelines
for state highway safety programs that call for states to enact laws
such as high BAC and repeat offender laws with increased sanctions for
each offense. Many states have enacted these laws. For example, 39
states and the District of Columbia impose higher penalties for drivers
with BAC levels of 0.15 and above. In addition, 43 states and the
District of Columbia have repeat offender laws to discourage multiple
alcohol offenses. All of the seven selected states have enacted these
laws.
Challenge in Staffing:
Another challenge that hinders further progress in increasing safety
belt use and reducing impaired driving is that law enforcement agencies
report that they do not always have a sufficient number of officers to
conduct HVE campaigns, even though traffic safety grants can be used
for law enforcement officers to staff checkpoints or saturation
patrols. This shortage of officers may affect HVE activities that occur
during normal work hours and when HVE activities are staffed by
officers working overtime.
The challenge of finding a sufficient number of officers to conduct HVE
campaign work during normal work hours can occur for various reasons.
For example, some law enforcement agencies said they do not have
sufficient staffing levels to conduct both regular police work and
frequent HVE campaign enforcement activities. In one city we visited,
crime enforcement needs took priority over traffic enforcement,
officials said. For example, until recently, Providence, Rhode Island,
devoted most of its policing resources to reducing violent crime.
However, with a reduction in violent crime, Providence has been able to
increase the number of officers working on traffic safety and HVE
campaign activities. Officials also stated that regular staffing levels
were being depleted because many of their officers have been called up
for duty in the armed forces. Other factors that impact police
department staffing levels, according to one 2005 study, include
additional homeland security duties in many jurisdictions that lead to
an increased workload for local police, as well as a potentially
smaller pool of qualified applicants because of previous drug use and
lack of physical condition. In addition, some law enforcement agencies
may have too few personnel to conduct HVE activities--staffing a full-
scale sobriety checkpoint, for example, can require 10 or more
officers.
In addition to the challenge of finding sufficient staff for HVE
activities during regular hours, officials in some of the law
enforcement agencies we visited reported that they are also having a
difficult time getting enough officers to sign up for overtime to work
on HVE campaigns. For example, officials in North Dakota, Rhode Island,
and Washington said that one of the challenges in getting a sufficient
number of officers to work overtime is that there are often other
opportunities for overtime work, such as working in a work-zone patrol
car at a highway construction site or as a security guard at a mall.
Some officers prefer these opportunities over DUI enforcement because
DUI enforcement involves a greater amount of effort and paperwork
compared with other duties.
Though NHTSA continues to provide funding through traffic safety
grants, NHTSA has taken other steps to assist state and local law
enforcement agencies in providing adequate staff for HVE campaigns. For
example, NHTSA provides funds that states can use to provide equipment,
such as breath-testing units, that are used as incentives to improve
participation. NHTSA also provides guidance on how to make better use
of existing resources. For example, NHTSA created a set of guidelines
on sobriety checkpoints that outlines ways that small law enforcement
agencies with limited staff can conduct effective sobriety checkpoints
with fewer officers.
States are also taking steps to address the staffing challenge. For
example, to encourage officers to participate in enforcement campaigns,
some states have developed programs to recognize officers for their
contributions. In 2006, the Rhode Island Office of Highway Safety
increased recognition for HVE work by giving awards to officers who
worked overtime in enforcement campaigns targeting drunken driving and
drivers not wearing safety belts. In addition, Arkansas has initiated a
recognition program for exemplary performance in DUI work. One way of
addressing staffing limitations in states with relatively small,
neighboring law enforcement agencies is to conduct multijurisdictional
enforcement activities. By pooling operations, according to NHTSA
officials, 5 to10 small agencies are able to concentrate appreciable
resources during an HVE campaign mobilization period. For example,
North Dakota is developing a multiagency approach to address the
challenge of bringing together sufficient resources to staff DUI
checkpoints.
Weak Prosecution of DUI Arrests:
Another challenge for implementing HVE campaigns is weak prosecution of
DUI arrests. Based on our interviews with selected state and NHTSA
officials and our review of reports and studies, the main factors that
contribute to weak DUI prosecution are as follows:
* Court systems have heavy caseloads and limited resources. Therefore,
DUI cases may be given a lower priority compared with more violent
crimes. According to a 2002 panel convened by the National Association
of State Judicial Educators, this results in prolonged adjudication of
DUI cases and increased likelihood of dismissals and acquittals.
[Footnote 26] States often lack sufficient funds to establish special
courts to more effectively process DUI cases and to provide the
supervised probation and treatment that DUI offenders often require.
* Some law enforcement officials and prosecutors lack the necessary
knowledge and training to consistently prosecute DUI cases. In some
cases, the DUI charge may be dismissed because an officer lacks proper
training. For example, an arresting officer may not have taken enough
notes on the DUI arrest to testify in sufficient detail during the
trial. In other cases, the DUI charge may be dismissed because the
prosecutor did not have sufficient training to effectively prosecute
DUI cases. However, many prosecutors view themselves as often not
sufficiently prepared for their first DUI cases. A 2002 survey of
prosecutors found that 48 percent believed they did not have adequate
training or preparation before they began handling DUI cases. In
addition, prosecutors and judges have to be knowledgeable in the
complex aspects of DUI cases, including relevant legal rulings and
admissibility of evidence as well as more scientific and technical
issues such as blood alcohol testing procedures. The National
Association of State Judicial Educators (NASJE) panel cited several
challenges that judges handling DUI cases face, including that DUI
cases are frequently plea-bargained, which may undermine the deterrent
value of the arrest.
Though states have a primary role in improving prosecution, NHTSA has
provided guidance, funded training programs, and provided grants to
states that can be used for more effective prosecution of DUI
offenders.[Footnote 27] In February 2007, in cooperation with the
National District Attorneys Association, NHTSA issued guidelines for
improving testimony, note taking, and evidence gathering procedures by
officers. NHTSA has also developed courses designed to improve
prosecutorial skills, including one course on prosecuting DUI cases and
another course that examines complex cases involving alcohol-involved
crashes and provides training on how to prosecute these cases
effectively and respond to challenges presented by the defense.
[Footnote 28] NHTSA has also provided funding for systems that allow
states to share information on effective DUI prosecution. For example,
with NHTSA funding, NASJE is developing a clearinghouse for the
exchange of materials, techniques, and information on DUI prosecution.
Selected states have also developed initiatives to address the
challenge of more effectively prosecuting DUI cases. For example,
Arkansas has developed a judicial training project initiative that
includes training for judges and officers on DUI prosecution and
additional training for about 400 officers on conducting field sobriety
testing. In June 2006, Illinois used HVE 410 grant funding to conduct a
2-day seminar for 23 judges, which included training on sentencing and
evidentiary issues.
Conclusions:
HVE enforcement campaigns have been shown to be effective tools in
raising public awareness of and encouraging compliance with safety belt
and impaired-driving laws. NHTSA has fulfilled the requirements for
implementing an HVE program by developing advertising, coordinating
with states, and evaluating the effectiveness of the campaigns.
Although state officials we spoke with reported that NHTSA's
coordination efforts helped them implement HVE campaigns, they also
cited several challenges to conducting campaigns and achieving desired
results--resistant populations, insufficient staffing, and inconsistent
DUI prosecution--that may be limiting the effectiveness of the
campaigns. NHTSA is implementing some initiatives--such as the "Buckle
Up In Your Truck" campaign to promote safety belt use among rural
pickup truck drivers--to help states address these challenges. However,
the challenges of insufficient staffing and inconsistent DUI
prosecution are primarily state issues and largely out of NHTSA's
control to influence under the current program. While NHTSA's
performance measures indicate that the campaign messages are reaching
the primary target audiences and positively affecting behaviors,
NHTSA's assessments of campaign effectiveness do not provide a complete
picture of the impact of HVE campaigns. For example, NHTSA lacks the
data to consistently measure a key component of the campaigns--the
level of state and local activities--and evaluate the use of federal
funds used by states for campaign activities. Furthermore, the
performance measures reported in the assessments are limited and do not
provide information on the impact of the campaigns in areas such as
nighttime safety belt use and advertisements for all target audiences.
The assessments would be improved by more complete and consistent
information on federally funded state campaign activities and a more
comprehensive set of performance measures.
Recommendations for Executive Action:
To improve NHTSA's evaluations of the HVE campaigns, we recommend that
the Secretary of Transportation direct NHTSA to take the following two
actions:
* Develop a minimum core set of reporting requirements for states to
consistently report HVE law enforcement and media activities funded
with federal dollars. These requirements should be designed to achieve
a more consistent measure of state activity and accountability for
federal funding without presenting an undue burden to states.
* Develop and include additional performance measures--such as a
measure for nighttime safety belt use and additional measures of media
effectiveness--in the agency's annual evaluations of the effectiveness
of the two campaigns.
Agency Comments:
We provided a draft of this report to DOT for its review and comment.
DOT officials--including the Senior Associate Administrator of Traffic
Injury Control--generally agreed with the findings and recommendations
of the report.
We are sending copies of this report to interested congressional
committees and the Secretary of Transportation. We will also make
copies available to others upon request. In addition, the report will
be available at no cost on GAO's Web site at [hyperlink,
http://www.gao.gov].
If you or your staffs have any questions about this report, please
contact me at (202) 512-2834 or siggerudk@gao.gov. Contact points for
our Offices of Congressional Relations and Public Affairs may be found
on the last page of this report. GAO staff who made key contributions
to this report are listed in appendix III.
Signed by:
Katherine A. Siggerud:
Director, Physical Infrastructure Issues:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
To determine the extent that the National Highway Traffic Safety
Administration (NHTSA) has implemented the high-visibility enforcement
program, and evaluated results, we reviewed information and interviewed
officials from NHTSA and representatives of nongovernmental
organizations, including the American Association of State Highway and
Transportation Officials, Governors Highway Safety Association,
International Association of Chiefs of Police, Mothers Against Drunk
Driving, National Safety Council, and the National Sheriffs
Association. In addition, we reviewed studies, reports, and laws
relevant to the implementation and evaluation of NHTSA's and other high-
visibility enforcement programs. We also interviewed state officials to
obtain their views on NHTSA's assistance with campaign activities in
seven states--Arkansas, Illinois, Iowa, North Carolina, North Dakota,
Rhode Island, and Washington. We judgmentally selected the states to
include those with laws that may affect how states conduct enforcement
campaigns; states with a wide range of traffic safety performance
levels, such as safety belt use and number of alcohol-involved
fatalities, and states with different sizes of law enforcement agencies
and with various degrees of participation in campaigns. In selecting
states based on differences in laws that affect campaign enforcement,
we included states that adopted primary safety belt laws before or
after 1997 and states that have not enacted a primary safety belt law.
In selecting states based on levels of traffic safety performance, we
included states that, when ranked nationally, fell into the upper,
lower, and middle third in safety belt use, unbuckled fatalities, and
alcohol-related fatal crash ratio.[Footnote 29] To select states based
on size of state law enforcement agencies, we looked at the average
number of sworn officers reported by those agencies. To select states
based on participation by law enforcement agencies, we chose states
that had varying reported percentages of participation in HVE
campaigns, the extent that agencies reported campaign hours worked, and
whether law enforcement agencies used checkpoints while conducting the
HVE campaigns from 2003 through 2006. In selecting the states, we used
a nongeneralizable sampling approach, and, consequently, the results
cannot be used to make inferences about all of the states.
We used data contained in NHTSA's Fatality Analysis Reporting System
(FARS) database to analyze information on all traffic-related
fatalities. Each state provides NHTSA fatality data in a standardized
format. To be included in the database, a crash must result in the
death of an occupant or nonmotorist within 30 days of the incident. The
states obtain this information from such sources as police reports,
vehicle registration files, state driver licensing files, death
certificates, coroner or medical examiner reports, and hospital
records. It should be noted that while fatality data are useful in
understanding crashes, other factors in addition to those involved in
causing the crash might have contributed to the fatality. This would
include whether safety belt or other occupant protection measures were
used and functioned properly. Further, in providing information on
state fatality rate trends, we identified fatalities per million miles
traveled. To do so, we used vehicle miles traveled data maintained by
Federal Highway Administration in its Highway Performance Monitoring
System (HPMS). HPMS is a national-level highway information system that
includes data on the extent, condition, performance, use, and operating
characteristics of the nation's highways. HPMS obtains vehicle-miles-
traveled data from each state, and states have different methods for
collecting certain travel information. There are certain limitations
associated with using these data. For example, the quality of the data
in the system relies on state data collection techniques. HPMS guidance
is flexible, so that each state has its own approach, and some
approaches do not require annual revisions. In addition, vehicle-miles-
traveled data may not be comparable from state to state. We have
previously assessed the reliability of the FARS and HPMS data by
reviewing it for obvious errors in accuracy and completeness, reviewing
existing information about the data, and interviewing agency officials
knowledgeable about the data and determined that the data is
sufficiently reliable for the purposes of this report.
To determine the impact of the high-visibility enforcement campaigns
and what challenges exist, we interviewed officials from state highway
traffic safety offices, state police, and local police in the seven
selected states. For each state we visited, we also interviewed
officials in the applicable NHTSA regional office and, when available,
representatives of state associations of chiefs of police and sheriffs'
associations about the impact of the HVE campaigns and challenges they
faced. In addition, we reviewed the state highway safety plans and
annual reports and other relevant reports for information on HVE
campaign activities and challenges. We also provide a summary of high-
visibility campaigns in Australia, Canada, and the Netherlands, which
can be found in appendix II.
We conducted this performance audit from March 2007 to April 2008 in
accordance with generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.
[End of section]
Appendix II: HVE Campaigns in Australia, Canada, and the Netherlands:
Similar to the United States, Australia, Canada, and the Netherlands,
among other countries, combine high-visibility enforcement (HVE) with
advertising to improve safety belt use and reduce impaired
driving.[Footnote 30] In fact, Canada was the first country in North
America to demonstrate that highly publicized occupant protection
enforcement increases compliance with occupant protection laws. The
United States based its model for high-visibility campaigns on Canada's
occupant protection enforcement program.
A key difference among the U.S., Australia, and Netherlands HVE
programs is the use of random breath testing (RBT), which allows police
to pull over any driver at random to undergo a breath test for alcohol.
Officials from Australia and the Netherlands noted that impaired-
driving fatalities have been reduced in their countries because law
enforcement officials are allowed to stop drivers at random to test the
driver's breath for alcohol. Though other countries report that the RBT
technique is effective in combating impaired driving, it may be
difficult to replicate in the United States due to privacy concerns. In
the United States, drivers must be stopped for suspicious behavior or
another offense before being given a breathalyzer. However, states may
be able to use equipment, such as a passive alcohol sensor embedded in
a flashlight, to overcome potential privacy concerns.
* Canada bases its current high-visibility enforcement campaigns on its
national road safety plan, Road Safety Vision 2010, which commenced in
2002. The plan has ambitious goals--a 30 percent reduction in
fatalities and injuries based on average fatalities and serious
injuries from 1996 to 2001--and runs through 2010. Canadian HVE
campaigns concentrate on speed, impaired driving, and safety belts
among other things. Despite previous success with safety belt campaigns
in the 1980s to 1990s, Canada's safety belt use rates have remained
stagnant at approximately 90 percent to 93 percent during the last 10
years. Canada is currently trying to develop strategies to target the
"last 10 percent" since approximately 40 percent of their fatalities
involve unbuckled persons, many of whom are thrown from the vehicle. To
target impaired drivers, Canada conducts a Reduce Impaired Driving
Everywhere program, where police stop drivers to check for signs of
alcohol consumption. If there is suspicion of drinking, a roadside
screening test can be administered; if the driver fails this test, an
evidentiary breath test is given, usually at the police station.
Canadian law also stipulates that anyone with a blood alcohol content
above 0.08 is criminally liable. Finally, Canada has a Web site to
allow officers to share best practices in high-visibility campaigns.
* Australia has primarily relied on RBT as an effective strategy for
the country's high-visibility law enforcement campaigns against
impaired driving. Each Australian state enacted impaired-driving laws
between 1976 and 1992, setting a blood alcohol content of 0.05 as the
legal limit for drunken driving offenses. Each state has also enacted
RBT laws, beginning with the state of Victoria in 1976. It was not
until researchers determined in the early 1980s that one in two drivers
killed in traffic accidents had a blood alcohol content over the legal
limit, however, that Australia began widespread implementation of RBTs.
For example, the state of Victoria went from conducting nearly 200,000
RBTs per year in the mid-1980s to approximately 1.2 million RBTs per
year in the early 1990s. Since there were approximately 3 million
licensed drivers in the state of Victoria at that time, this means more
than one in three drivers were subject to an RBT. The proportion of
drivers killed in Australia traffic accidents that had a blood alcohol
content over the legal limit fell to one in five in 1992. Furthermore,
the number of intoxicated drivers in the state of Victoria who died in
accidents decreased from 350 per year in the 1980s to approximately 100
in 2006. In addition, Australia implemented a campaign to educate the
public about why police officers were conducting RBTs and ensured the
RBTs were conducted efficiently to ensure minimal delays to drivers.
Australia emphasized the RBTs have been successful because of the
general deterrence theory--that is, Australian drivers are afraid of
being pulled over and tested, and, thus, they are less likely to drink
and drive. Finally, anyone who refuses a breathalyzer test is presumed
to have a blood alcohol content above the legal limit.
* The Netherlands' HVE campaigns focus on a variety of areas set by the
Minister of Transport and employs a number of different strategies. The
five campaigns for 2007 emphasized helmet use for motorcycles and
mopeds, safety belt use, red light compliance, reducing impaired
driving, and lowering driving speeds. Each campaign lasts several weeks
and involves significant police activity using the country's 750
traffic safety officers. Approximately 1.5 million RBTs are done per
year in the Netherlands, and drivers are afraid of being pulled over
and tested, based on the general deterrence theory. The Netherlands
also has an aggressive speed enforcement program to ticket 100 percent
of offenders on the country's motorways, through an intricate speed
camera system. For example, a motorway will have a camera on mile 1 and
another camera on mile 10. A citation is then issued when drivers
exceed the speed limit between those two locations; the fine depends on
how far over the limit the driver was traveling. In the past, the
Netherlands has run a campaign featuring a cartoon character known as
"Goochem the Armadillo" to encourage children ages 4 to 12 to wear
safety belts. The campaign was intended to emphasize knowledge of
safety belt laws, increase use of safety belts, and encourage positive
attitudes about safety belts. The campaign was successful in raising
safety belt use rates by back seat passengers. Safety belt use rates by
back seat passengers increased from 40 percent in urban areas and 43
percent in rural areas in 1998 to 73 percent for both in 2006. Further,
although the campaign was aimed at children, a survey demonstrated that
the campaign reached approximately 90 percent of adults, as well. In
addition to the high-visibility campaigns, the Netherlands also
promotes sustained activity targeting each of the campaign areas
throughout the year.
[End of section]
Appendix III:
GAO Contact and Staff Acknowledgments:
GAO Contact:
Katherine A. Siggerud, (202) 512-2834 or siggerudk@gao.gov:
Staff Acknowledgements:
In addition to the contact named above, other key contributors to this
report were Sara Vermillion (Assistant Director), Elizabeth Curda,
Sandra DePaulis, Brad Dubbs, Colin Fallon, Bert Japikse, Tom James,
Leslie Locke, Terry Richardson, Beverly Ross, and Aron Szapiro.
[End of section]
Footnotes:
[1] The OTLUA campaign was termed "You Drink & Drive, You Lose" from
2003 to 2005.
[2] Unrestrained fatalities are those in which the deceased was not
wearing a shoulder belt, lap belt, lap and shoulder belt, child safety
seat, or other restraint and were occupants (except bus passengers) of
motor vehicles (except motorcycles, all terrain vehicles, or
snowmobiles).
[3] Alcohol-involved fatalities include all fatalities in a motor
vehicle crash where one or more involved drivers, pedestrians, or pedal-
cyclists in the crash had a blood alcohol content of 0.08 g/dL or
greater.
[4] Pub. L. 98-363 specified that 10 percent of a state's apportioned
funds for the National Highway System, the Surface Transportation
Program, and interstate reconstruction and maintenance would be
withheld if a state did not establish a 21-year-old minimum drinking
age.
[5] HVE campaigns are also known as Selective Traffic Enforcement
Programs.
[6] R.B. Voas, A.E. Rhodenizer, and C. Lynn, Evaluation of
Charlottesville Checkpoint Operation (final report), special report
prepared at the request of the U.S. Department of Transportation,
National Highway Traffic Safety Administration, May 1985.
[7] NHTSA has also continued to fund media technical assistance, the
development of advertisements and the production of advertisements
through Highway Safety Research and Development Section 403 funds
(i.e., for funds appropriated to carry out 23 U.S.C. § 403) and in
fiscal year 2007 reported expending $3,454,458.
[8] Information on the amount states spend on these activities is not
available because NHTSA does not require states to report such
information, and their grant tracking system reports on program areas
such as occupant protection and not on components within the program
such as HVE media.
[9] U.S. Department of Transportation, National Highway Traffic Safety
Administration, NHTSA National Communications Plan 2007 (Washington,
D.C., January 2007).
[10] The expert panel included strategic communications and social
marketing experts. We identified these elements in NHTSA's plans and
resources but did not evaluate each element.
[11] GAO, Digital Television Transition: Increased Federal Planning and
Risk Management Could Further Facilitate the DTV Transition,
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-43] (Washington,
D.C.: Nov. 19, 2007).
[12] U.S. Department of Transportation, National Highway Traffic Safety
Administration, Creating Impaired Driving General Deterrence: Eight
Case Studies of Sustained, High-Visibility, Impaired-Driving
Enforcement (Washington, D.C., March 2006).
[13] NHTSA issued its evaluation of the May 2005 CIOT Campaign in May
2007, and the evaluation of the Impaired Driving Campaigns for 2003-
2005 in July 2007. NHTSA plans to release evaluations of the 2006
campaigns--the first year of the national advertising campaigns
authorized by SAFETEA-LU--in the spring of 2008.
[14] U.S. Department of Transportation, National Highway Traffic Safety
Administration, Evaluation of the May 2005 Click It or Ticket
Mobilization to Increase Seat Belt Use, Report No. DOT HS 810 778
(Washington, D.C., May 2007).
[15] U.S. Department of Transportation, National Highway Traffic Safety
Administration, Evaluation of the National Impaired Driving High-
Visibility Enforcement Campaign: 2003-2005, Report No. DOT HS 810 789
(Washington, D.C., July 2007).
[16] This information is based on annual telephone surveys on the
effectiveness of impaired-driving campaigns from 2003 through 2005.
NHTSA officials also conducted regional telephone surveys for the CIOT
campaign in 2006 and a national telephone survey for the CIOT campaign
in 2007; data from these surveys will be included in NHTSA's 2006 and
2007 evaluations. NHTSA reported the 2006 evaluation should be released
in March 2008 and the 2007 evaluation in the spring of 2009.
[17] States that receive Alcohol-Impaired Driving Countermeasures
Incentive Grant funds are required to report on HVE activities those
funds support.
[18] Percentage is based on NHTSA data collected from states, the
District of Columbia, and Puerto Rico.
[19] U.S. Department of Transportation, National Highway Traffic Safety
Administration, 2005 National Communications Plan (Washington, D.C.,
November 2004).
[20] While the 2005 Integrated National Communications Plan listed
these other target audiences, the summary of the 2005 Impaired Driving
Campaign shown in the appendix of the 2006 National Communications plan
shows these additional targeted audiences were replaced with a "Newly
Arrived Latino Immigrants" secondary target audience.
[21] Media allocation was reported in an appendix of the 2006 National
Communications Plan.
[22] In general, throughout the U.S., safety belt use rates are higher
in primary law states, but some states without a primary law have
safety belt use rates that are higher than the nationwide average.
Minnesota, Nevada, Ohio, Pennsylvania, Utah, Vermont, and West
Virginias--states without a primary safety belt law according to NHTSA-
-experienced 2006 safety belt use rates that were higher than the
nationwide average of 81 percent.
[23] Since we used a nongeneralizable sample of seven states, these
challenges should not be used to make inferences about challenges in
all states that implement HVE campaigns. For similar challenges, see:
Statement of The Honorable Calvin L. Scovel III, Inspector General,
U.S. Department of Transportation on the Effectiveness of Federal Drunk
Driving Programs, Senate Committee on Environment and Public Works,
Subcommittee on Transportation Safety, Infrastructure Security, and
Water Quality, Oversight Hearing on Effectiveness of Federal Drunk
Driving Programs, 110th Cong., 1st sess., October 25, 2007; and
Testimony of Mark V. Rosenker, Chairman, National Transportation Safety
Board, Senate Committee on Environment and Public Works, Subcommittee
on Transportation Safety, Infrastructure Security, and Water Quality,
Oversight Hearing on Effectiveness of Federal Drunk Driving Programs,
110th Cong., 1st sess., October 25, 2007.
[24] NHTSA, Traffic Safety Facts: Rural/Urban Comparison, DOT HS 810-
812 (2006).
[25] NHTSA, Traffic Crashes Take Their Toll on America's Rural Roads:
The Need to Establish Rural Seat Belt Programs, DOT HS 810-658
(December 2006).
[26] National Association of State Judicial Educators and The Century
Council, Hardcore Drunk Driving Judicial Guide (Washington, D.C.,
2002).
[27] States may use Alcohol-impaired Driving Countermeasures Incentive
Grant funds to implement an outreach program to educate prosecutors and
judges.
[28] These courses are offered by the National Association of
Prosecutor Coordinators and the American Prosecutors Training
Institute.
[29] The alcohol-related fatal crash ratio is the number of alcohol-
related vehicle crashes with one or more fatalities divided by the
total number of vehicle crashes with one or more fatalities. A crash
was considered to be alcohol related if a vehicle operator, pedestrian,
or bicyclist involved in the crash had blood alcohol content at or
above 0.01.
[30] GAO selected these three countries based on expert recommendation
and the countries' performance in reducing road fatalities.
[End of section]
GAO's Mission:
The Government Accountability Office, the audit, evaluation and
investigative arm of Congress, exists to support Congress in meeting
its constitutional responsibilities and to help improve the performance
and accountability of the federal government for the American people.
GAO examines the use of public funds; evaluates federal programs and
policies; and provides analyses, recommendations, and other assistance
to help Congress make informed oversight, policy, and funding
decisions. GAO's commitment to good government is reflected in its core
values of accountability, integrity, and reliability.
Obtaining Copies of GAO Reports and Testimony:
The fastest and easiest way to obtain copies of GAO documents at no
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each
weekday, GAO posts newly released reports, testimony, and
correspondence on its Web site. To have GAO e-mail you a list of newly
posted products every afternoon, go to [hyperlink, http://www.gao.gov]
and select "E-mail Updates."
Order by Mail or Phone:
The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent
of Documents. GAO also accepts VISA and Mastercard. Orders for 100 or
more copies mailed to a single address are discounted 25 percent.
Orders should be sent to:
U.S. Government Accountability Office:
441 G Street NW, Room LM:
Washington, D.C. 20548:
To order by Phone:
Voice: (202) 512-6000:
TDD: (202) 512-2537:
Fax: (202) 512-6061:
To Report Fraud, Waste, and Abuse in Federal Programs:
Contact:
Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]:
E-mail: fraudnet@gao.gov:
Automated answering system: (800) 424-5454 or (202) 512-7470:
Congressional Relations:
Ralph Dawn, Managing Director, dawnr@gao.gov:
(202) 512-4400:
U.S. Government Accountability Office:
441 G Street NW, Room 7125:
Washington, D.C. 20548:
Public Affairs:
Chuck Young, Managing Director, youngc1@gao.gov:
(202) 512-4800:
U.S. Government Accountability Office:
441 G Street NW, Room 7149:
Washington, D.C. 20548:
