Motor Vehicle Injury Alcohol-Impaired Driving: Mass Media Campaigns

The CPSTF finding is based on evidence from a systematic review of 8 studies (search period through June 2000). The review was conducted on behalf of the CPSTF by scientists from CDC’s Division of Unintentional Injury Prevention with input from a team of specialists in systematic review methods and experts in research, practice and policy related to motor vehicle injury prevention.

Eight studies qualified for the systematic review.

• Total alcohol-related crashes: median decrease of 13% (interquartile interval: 6% to 14% decrease; 7 studies)
• Injury-producing alcohol-related crashes: median decrease of 10% (interquartile range: 6% to 14% decrease; 6 studies)
• Proportion of drivers who had consumed alcohol: net decreases of 30% and 158% (2 studies)
• Evaluated mass media campaigns had several components in common:
  • A theoretical framework in communications research
  • Pretested messages
  • High levels of audience exposure to the message, mostly through paid advertising
• Results did not differ according to the message appeals used.

• Cost benefit analyses were conducted for two of the campaigns evaluated in this review. One campaign was conducted in Australia, and the other was implemented in two cities in Kansas.
  • In all three sites evaluated, the estimated societal benefits substantially exceeded the costs of developing and airing the campaign messages.
  • Monetary values are presented in 1997 U.S. dollars.
• One analysis reported on the first 23 months of a campaign in Victoria, Australia.
  • The cost was $403,174 per month for advertisement development, supporting media, media placement, and concept research.
  • Estimated savings from medical costs, productivity losses, pain and suffering, and property damage were $8,324,532 per month, with $3,214,096 of these savings coming from averted medical costs.
• Analyses of six-month campaigns in Wichita (using paid media) and Kansas City, Kansas (using public service announcements) reported total costs of $454,060 and $322,660, respectively. These costs included planning and evaluation research, message production, and media scheduling.
  • Total savings from averted costs of insurance administration, premature funeral, legal and court, medical payments, property damage, rehabilitation, and employers’ losses were estimated at $3,431,305 for the Wichita campaign and $3,676,399 in Kansas City.

Results should be applicable to carefully planned and pretested mass media campaigns, with ads that reach the intended audience often enough, implemented in an environment with other ongoing prevention activities (e.g., grassroots activities, enhanced law enforcement efforts), and targeted to any audience of driving age.

CPSTF identified several areas that have limited information. Additional research and evaluation could help answer the following questions and fill remaining gaps in the evidence base. (What are evidence gaps?)

Results from the Community Guide review indicate that under some conditions, well-executed mass media campaigns can contribute to a reduction in alcohol-impaired driving (AID) and alcohol-related crashes. They also suggest that such campaigns are cost saving.

The characteristics of the campaigns evaluated in this review may serve as a preliminary guide to evaluating the potential for success of a proposed mass media campaign, but several research questions will need to be addressed to maximize the effectiveness and efficiency of future programs. A list of such questions is provided in the list below. Foremost among these is the question of the relative effectiveness of specific campaign themes and messages. It is unlikely that all potential messages are equally effective for changing drinking and driving behavior, and some may prove not to be effective at all or even to be counterproductive. Another important question relates to the potential impact of the changing media market, with increasing market segmentation, emerging technology to allow consumers to avoid exposure to broadcast messages, and opportunities for individually tailored message delivery via the Internet. The impact of these changes should be evaluated and future campaigns adapted to the changing media environment.

The campaigns reviewed tended to take place in areas with relatively high levels of law enforcement and other activities to prevent AID. These activities may have helped provide a context in which the audience was predisposed to react positively to the campaign messages. It is not clear whether these campaigns might have had similar effects in a setting where strong AID-prevention activities were not in place.

Evaluating Message Content Effects

• What is the relative effectiveness and cost-effectiveness of various campaign themes (e.g., law enforcement, legal penalties, social stigma, guilt, injury to self and others) for reducing AID and alcohol-related crashes? For influencing public support for stronger prevention activities?

Evaluating Message Delivery Effects

• What is the dose response curve for varying levels of advertising exposure (e.g., none, light, moderate, and heavy)? Does the shape of this curve vary according to message content and the outcome evaluated?
• What is the relative effectiveness and cost-effectiveness of different media types (TV, radio, etc.)? Paid advertising and public service announcements?
• What is the optimal exposure schedule for AID mass media campaigns (e.g., intermittent waves of messages vs. a steady flow)?
• How should mass media campaigns be adapted to the changing media environment (e.g., market segmentation, Internet, message filtering devices)?

Evaluating Message/Recipient Interactions

• To what extent are certain population groups more or less likely to be influenced by mass media campaigns?
• Are some themes more likely than others to influence “hard-to-reach” target groups (e.g., enforcement themes for “hard-core” drinking drivers)?

Improving Research Design

• What measurement issues need to be addressed to improve assessment of media and message exposure? What research designs can best address problems in measuring exposure?

• Evaluated mass media campaigns had several components in common:
  • A theoretical framework in communications research
  • Pretested messages
  • High levels of audience exposure to the message, mostly through paid advertising
• Campaigns were implemented in settings that had other prevention efforts in place, such as high-visibility enforcement of impaired driving laws.
• Campaign messages ranged from those focused on law enforcement activities and the legal consequences of drinking and driving to the social and health consequences of alcohol-impaired driving. Results did not differ according to the message appeals used.
• Included studies assessed intervention effectiveness on fatal crashes, fatal and nonfatal injury crashes combined, crashes that damage property, and drivers’ BACs.

Effectiveness Review

Analytic Framework see Figure 1 on page 59

When starting an effectiveness review, the systematic review team develops an analytic framework. The analytic framework illustrates how the intervention approach is thought to affect public health. It guides the search for evidence and may be used to summarize the evidence collected. The analytic framework often includes intermediate outcomes, potential effect modifiers, potential harms, and potential additional benefits.

Effectiveness Review

Summary Evidence Table

The number of studies and publications do not always correspond (e.g., a publication may include several studies or one study may be explained in several publications).

Effectiveness Review

Cameron M, Vulcan P. Evaluation review of the supplementary road safety package and its outcomes during the first two years. Auckland, New Zealand: Land Transport Safety Authority, 1998.

Epperlein T. Initial deterrent effects of the crackdown on drinking drivers in the state of Arizona. Accid Anal Prev 1987;19:285 303.

Lastovicka JL. Highway safety mass media youth project. Washington DC: U.S. Department of Transportation 1987 (contract DTNH22-85-C-15404).

McLean AJ, Kloeden CN, McCaul KA. Drink-driving in the general nighttime driving population, Adelaide 1989. Aust J Public Health 1991;15:190 3.

Newstead S, Cameron M, Gantzer S, Vulcan P. Modelling of some major factors influencing road trauma trends in Victoria 1989 1993. Melbourne: Monash University Accident Research Centre, 1995 (report 74).

Worden JK, Waller JA, Riley TJ. The Vermont public education campaign in alcohol and highway safety: a final review and evaluation. Waterbury: Vermont Department of Mental Health, 1975 (CRASH report I-5).

Economic Review

Cameron MH, Haworth N, Oxley J, Newstead SV, Le T. Evaluation of Transport Accident Commission road safety advertising. Melbourne: Monash University Accident Research Centre, 1993 (report RN52).

Murry JP, Stam A, Lastovicka JL. Paid- versus donated-media strategies for public service announcement campaigns. Public Opinion Q 1996;60:1 29.

The following outlines the search strategy used for reviews of these interventions to reduce alcohol-impaired driving: 0.08% Blood Alcohol Concentration (BAC) Laws; Lower BAC Laws for Young or Inexperienced Drivers; Maintaining Current Minimum Legal Drinking Age (MLDA) Laws; Mass Media Campaigns; Multicomponent Interventions with Community Mobilization; Ignition Interlocks; School-Based Programs; Designated Driver Promotion Programs; Sobriety Checkpoints (archived); Intervention Training Programs for Servers of Alcoholic Beverages (archived).

The reviews of interventions to reduce motor vehicle-related injury reflect systematic searches of multiple databases as well as reviews of reference lists and consultations with experts in the field. The team searched six computerized databases (MEDLINE, Embase, Psychlit, Sociological Abstracts, EI Compendex, and Transportation Research Information Services [TRIS]), which yielded 10,958 titles and abstracts for articles, book chapters, reports, and published papers from the Association for the Advancement of Automotive Medicine proceedings about safety belts, alcohol-impaired driving or child passenger safety. Studies were eligible for inclusion if:

• They were published from the originating date of the database through June 2000 (March 1998 for child safety seat interventions)
• They involved primary studies, not guidelines or reviews
• They were published in English
• They were relevant to the interventions selected for review
• The evaluation included a comparison to an unexposed or less-exposed population
• The evaluation measured outcomes defined by the analytic framework for the intervention

For alcohol-impaired driving reviews, supplementary searches were conducted to address specialized questions and to update searches for reviews published after 2001. The final search using the primary alcohol-impaired driving search strategy was conducted through December 2004. For the most recent review in this series, “Effectiveness of Multicomponent Programs with Community Mobilization for Reducing Alcohol-Impaired Driving,” this database was supplemented by a hand search of the “Alcohol and Other Drugs” and “Transportation” sections of the SafetlyLit injury literature update service for the period from January through June 2005.

Effectiveness Review

Primary Search Strategy

1. S MOTOR(W)VEHICLE? OR CAR OR CARS OR AUTOMOBILE? OR MOTORCYCLE? OR TRUCK? OR TRAFFIC(2N)ACCIDENT? OR DRIVING OR DRIVER?
2. S ALCOHOL OR ALCOHOLIC(W)BEVERAGE? OR ALCOHOL(3N)DRINKING OR ETHANOL OR ALCOHOLISM OR DWI OR DUI OR (DRIVING(3N)(INTOXICATED OR INFLUENCE OR DRUNK OR DRINKING OR IMPAIRED))
3. S INTERVENTION? OR OUTREACH? OR PREVENTION OR (COMMUNITY(3N)(RELATION? OR PROGRAM? OR ACTION)) OR DETERRENT? OR PROGRAM? OR LEGISLATION OR LAW? OR EDUCATION OR DETERENCE OR COUNSELING OR CLASS OR CLASSES OR HEALTH(W)PROMOTION
4. S FOOD(W)INDUSTRY OR AIRPLANE? OR AIRCRAFT? OR PILOT? OR SOLVENT? OR SLEEP(W)APNEA OR EMISSION? OR AIR(W)QUALITY OR POLLUTION
5. S (S1 AND S2 AND S3 ) NOT S4

Higher Education-based Interventions

S1 MOTOR(W)VEHICLE? OR CAR OR CARS OR AUTOMOBILE? OR MOTORCYCLE? OR TRUCK? OR TRAFFIC(2N)ACCIDENT? OR DRIVING OR DRIVER?
S2 ALCOHOL OR ALCOHOLIC(W)BEVERAGE? OR ALCOHOL(3N)DRINKING OR ETHANOL OR ALCOHOLISM OR DWI OR DUI OR (DRIVING(3N)(INTOXICATED OR INFLUENCE OR DRUNK OR DRINKING OR IMPAIRED))
S3 UNIVERSIT? OR COLLEGE? OR CAMPUS? OR (EDUCATION?(2N)(HIGER OR INSTITUTION? OR FACILIT? OR PROGRAM? OR SURVEY?))
S4 S1 AND S2 AND S3
S5 CURRICULUM OR INSTRUCTION OR EDUCATION OR TRAINING OR WORKSHOPS OR PROGRAMS OR COURSE? OR TEACH? OR (SOCIAL(W)NORM?)
S6 STUDENT? OR YOUTH? OR TEEN? OR (YOUNG(W)ADULT?)
S7 S4 AND S5 AND S6

School-based Interventions

S1 MOTOR(W)VEHICLE? OR CAR OR CARS OR AUTOMOBILE? OR MOTORCYCLE? OR TRUCK? OR TRAFFIC(2N)ACCIDENT? OR DRIVING OR DRIVER?
S2 ALCOHOL OR ALCOHOLIC(W)BEVERAGE? OR ALCOHOL(3N)DRINKING OR ETHANOL OR ALCOHOLISM OR DWI OR DUI OR (DRIVING(3N)(INTOXICATED OR INFLUENCE OR DRUNK OR DRINKING OR IMPAIRED))
S3 SCHOOL?(5N)(BASED OR SETTING OR PROGRAM? OR PRIMARY OR ELEMENTARY OR SECONDARY OR ((JUNIOR OR SENIOR)(W)HIGH) OR MIDDLE) OR (EDUCATION?(2N)(INSTITUTION? OR FACILIT? OR PROGRAM? OR SURVEY?))
S4 S1 AND S2 AND S3
S5 CURRICULUM OR INSTRUCTION OR EDUCATION OR TRAINING OR WORKSHOPS OR PROGRAMS OR COURSE? OR TEACH?
S6 STUDENT? OR ADOLESCENT? OR YOUTH? OR TEEN? OR CHILD? OR TEACHER?
S7 525 S4 AND S5 AND S6

Cost Analyses

1. S MOTOR(W)VEHICLE? OR CAR OR CARS OR AUTOMOBILE? OR MOTORCYCLE? OR TRUCK? OR TRAFFIC(2N)ACCIDENT? OR DRIVING OR DRIVER?
2. S ALCOHOL OR ALCOHOLIC(W)BEVERAGE? OR ALCOHOL(3N)DRINKING OR ETHANOL OR ALCOHOLISM OR DWI OR DUI OR (DRIVING(3N)(INTOXICATED OR INFLUENCE OR DRUNK OR DRINKING OR IMPAIRED))
3. S INTERVENTION? OR OUTREACH? OR PREVENTION OR COMMUNITY(3N)(RELATION? OR PROGRAM? OR ACTION)) OR DETERRENT? OR PROGRAM? OR LEGISLATION OR LAW? OR EDUCATION OR DETERENCE OR COUNSELING OR CLASS OR CLASSES OR HEALTH(W)PROMOTION
4. S FOOD(W)INDUSTRY OR AIRPLANE? OR AIRCRAFT? OR PILOT? OR SOLVENT? OR SLEEP(W)APNEA OR EMISSION? OR AIR(W)QUALITY OR POLLUTION
5. S COST? OR ECONOMIC? OR ECONOMETRIC?
6. S (S1 AND S2 AND S3 AND S5) NOT S4

The following considerations are drawn from studies included in the evidence review, the broader literature, and expert opinion.

• Mass media campaigns may raise awareness of the dangers that drinking and driving pose for a community, thereby helping to generate an interest in strengthening legislation.
• Those thinking about implementing a mass media campaign to reduce alcohol-related crashes are cautioned to do so only if the necessary resources and supports are in place. Campaigns implemented without adequate planning, pretesting of messages, ad placement, and support activities cannot be expected to reduce alcohol-related crashes.
• Key factors in the design of mass media campaigns are related to both the content and the delivery of the messages used.
  • Content:
    • Several themes are commonly used to motivate people, such as fear of arrest; fear of harming oneself, others, or property; a positive social norm in which drinking and driving don’t mix; and portraying people who drink and drive as irresponsible and dangerous.
    • Content varies with regard to suggested actions, including abstaining from or drinking in moderation when driving, using a designated driver, or taking the keys from someone who has had too much to drink.
  • Delivery:
    • Optimal delivery of messages to the intended audience requires both control over when and where ads appear and control over the quality of the ads themselves.
    • To maximize exposure of target audiences to the message of the campaign, placement of ads is vital, and purchasing advertising space or time ensures control over placement; relying on free public service announcements leaves the scheduling of ads to media personnel who may not be concerned with the goals of the campaign.
    • Messages should be tested before the campaign is launched (pretested) to see which themes or concepts are most appropriate for target audiences. Conducting a campaign with untested messages can result in a diluted message, or the wrong message, being received.

Healthy People 2030

Healthy People 2030 includes the following objectives related to this CPSTF recommendation.

• Reduce the proportion of motor vehicle crash deaths that involve a drunk driver — SU‑11