- Email: [email protected]
The Illinois .08 law
Journal of Safety Research 33 (2002) 73 – 80 www.elsevier.com/locate/jsr
The Illinois .08 law An evaluation Robert B. Voas*, A. Scott Tippetts, Eileen P. Taylor Pacific Institute for Research and Evaluation, Calverton Office Park, Suite 300, 11710 Beltsville Drive, Calverton, MD 20705, USA Received 23 May 2001; received in revised form 1 August 2001; accepted 10 October 2001
Abstract Lowering state blood alcohol concentration (BAC) limits to 0.08, though controversial, has been supported by most evaluation studies to date. The Illinois .08 BAC law implemented in 1997 provided a unique opportunity to evaluate the effect of the law without the simultaneous passage of an administrative license revocation (ALR) law, which has clouded some previous evaluations of the .08 laws. The proportion of all drinking drivers in fatal crashes was compared before versus after implementing the .08 law using time-series analysis to evaluate 12 years of fatal crash data for Illinois and five bordering states. The results showed that the proportion of drinking drivers in fatal crashes decreased by 14% in Illinois and increased by 3% in bordering states. The proportion of drinking drivers in fatal crashes in Illinois, though increasing since 1995, was sharply reduced after passage of the .08 law in 1997, saving more than 100 lives in 1998 and 1999 than it would have without the .08 law. D 2002 National Safety Council and Elsevier Science Ltd. All rights reserved. Keywords: Drinking and driving; Driving under the influence; .08 laws; Administrative license revocation laws; Alcohol-related crashes
1. Introduction Over the last several years, there has been a strong trend for state legislatures to lower the legal blood alcohol concentration (BAC) from 0.10 to 0.08. As of * Corresponding author. Tel.: +1-301-755-2700; fax: +1-301-755-2799. E-mail address: [email protected] (R.B. Voas). 0022-4375/02/$ – see front matter D 2002 National Safety Council and Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 4 3 7 5 ( 0 2 ) 0 0 0 0 4 - X
74
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
May 2001, 24 states have taken that action under pressure from safety advocates supported by several studies on the effect of the law (Apsler, Char, Harding, & Klein, 1999; Hingson, Heeren, & Winter, 1996, 2000; Johnson & Fell, 1995; Research and Evaluation Associates [REA], 1991; Rogers, 1995; Voas, Tippetts, & Fell, 2000). This process has gone forward despite opposition by the hospitality industry, some negative reports (e.g., Foss, Stewart, & Reinfurt, 1998), and a negative evaluation of .08 laws by the General Accounting Office (GAO, 1999). The preponderance of evidence from these studies demonstrates that the .08 legislation produces a modest (about 8%) reduction in alcohol-related fatal crashes. Nonetheless, it has been difficult in some states to evaluate the effectiveness of .08 laws because they have been enacted in conjunction with other drunk-driving legislation, particularly administrative license revocation (ALR) laws that are known to be effective in reducing crashes. Consequently, it has been hard to find an appropriate comparison for states that adopt .08 legislation. Recently, Hingson et al. (2000) provided evidence that states implementing .08 laws, independent of ALR laws, appeared to experience reductions in alcohol-related crashes. Illinois provided another opportunity to evaluate the .08 law because that state’s ALR law had been enacted on January 1, 1986, more than 11 years before implementation of the .08 law in July 1997. Further, none of the five states that border Illinois, all of which also have ALR laws, had enacted a .08 law. Thus, these states provided a good comparison for analyzing the effect of the Illinois .08 law. To take advantage of that opportunity, the National Highway Traffic Safety Administration (NHTSA) funded a recently published study entitled Effectiveness of the Illinois .08 Law (Voas, Taylor, Kelley Baker, & Tippetts, 2000). The objective of that study was to determine the effect of the passage of a .08 law on the criminal justice system. The results indicated that all driving-under-the-influence (DUI) arrests statewide increased by 10% and DUI arrests in the 0.08 –0.09 range increased from 1% before to 9% after enactment of the law. Although a high percentage of DUI suspects in Illinois refused to take the breath test (39%), there was no evidence in state records that the proportion of refusals increased following enactment of the law. Nor was there any evidence that offenders or their lawyers were able to reduce the penalties normally assessed by the courts or the motor vehicle department for the DUI offense. Conviction rates increased slightly, and the proportion of offenders who were suspended remained higher than 90%. This report uses 12 years of data (1988 –1999) from the Fatality Analysis Reporting System (FARS; NHTSA, 1999) for Illinois and five bordering states (Indiana, Iowa, Kentucky, Missouri, and Wisconsin) to evaluate the effect of the .08 law on the number of drinking drivers compared to nondrinking drivers in fatal crashes.
2. Methods The FARS provides a file for every crash occurring on a U.S. highway that produces a fatality within 24 hours. A subfile is created for every driver involved
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
75
in such a crash containing the BAC measured at the time of the crash or an estimated BAC based on the characteristics of the crash and the driver (Klein, 1986). These BAC data permit the division of the files into two groups: drinking drivers (those with BACs greater than zero) and nondrinking drivers (those with zero BACs). Table 1 summarizes the data from FARS used in this study. These data (shown in Table 1) cover four time series from 1988 to 1999, categorized by drinking and nondrinking drivers in Illinois and drinking and nondrinking drivers in the five comparison states. In Illinois, the number of drinking drivers in crashes reflects any reduction in the role of alcohol in fatal crashes resulting from the .08 law. Any changes occurring when the law went into effect in June 1997 can be compared with the other three series. The nondrinking drivers series in Illinois reflects factors influencing the number of fatal crashes within the state such as the quality of roadways and vehicles, the weather, and the economy. The drinking driver series for the five comparison states reflects the effect of factors outside Illinois that might influence national drinking – driving trends such as the national economy or alcohol prices. Finally, the nondrinking driver series for the comparison states reflects national trends that affect fatal crashes such as gasoline prices. For this study, separate ARIMA analyses (Box & Tiao, 1975) were applied to two time series: the number of drinking drivers in Illinois, and a similar series of drinking drivers in the five comparison states as a group. In each ARIMA analysis, the number of nondrinking drivers was used as a covariate. The intercept date was July 1, 1997. The results from the two analyses were then compared using a pooled error term. This provided a before – after experimental comparison design with a prelaw period of 9 1/2 years and a postlaw period of 30 months. Quantifying any injury savings attributable to the .08 law in Illinois involved two steps. First, we estimated the number of fatalities associated with each drinking driver involved in a fatal crash. Pedestrian fatalities were eliminated when the pedestrian had a positive BAC because the drinking pedestrian, rather than a driver, may have contributed to the cause of the crash. Second, we estimated the number of fatalities attributable to each driver in the crash. If two drivers were involved, then half the fatalities in that crash were attributed to each driver. The fatalities attributable to sober and drinking drivers were aggregated separately, and only the fatalities attributed to drinking drivers (BAC > .00) were
Table 1 Drinking and nondrinking drivers in fatal crashes from the FARS, 1988 – 1999 Illinois N Drinking drivers Nondrinking drivers Totals
7,159.2 17,830.8 24,990.0
Comparisons % Known BAC 79 33
N 18,908.9 47,857.1 66,766.0
% Known BAC 76 42
76
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
used in the calculation of the number of fatalities that could be attributed to the .08 law. These two steps likely produce a smaller number of fatalities identified as being ‘‘alcohol related’’ because drinking pedestrians have been eliminated and responsibility for the crash has been attributed equally to drinking and nondrinking drivers, even though it has been shown that drinking drivers are significantly more likely to cause fatal crashes than sober drivers. Thus, this approach produces a somewhat conservative estimate of the lives saved by the .08 law. For the nation as a whole, the FARS for the 2 years from 1998 to 1999 showed 26,228 alcohol-positive drivers who, using this attribution method, accounted for 22,622 fatalities. This produced a rate of 0.863 fatalities per drinking driver. Considering the state of Illinois alone, there were 1,016 alcohol-positive drivers to whom this procedure attributed 876 fatalities, yielding a rate of 0.862 fatalities per alcohol-positive driver, which is almost identical to the national rate (0.863) during the same 2-year period.
3. Results The results from the time-series analysis of drinking drivers in fatal crashes in Illinois are contrasted with the results from the analysis of drinking drivers in the five comparison states in Table 2. The change from the prelaw to the postlaw period in Illinois produced an estimated 13.65% reduction in the number of drinking drivers in fatal crashes in Illinois. This was statistically significant ( P =.038). The change from the same pre-July 1997 period to the post-July 1997 period for the neighboring states (combined) was a nonsignificant increase of 3.38%. If we contrast the ARIMA results for the five comparison states with that
Table 2 Illinois .08 —ARIMA intervention analysis of drinking drivers in fatal crashes in Illinois compared to five bordering states: Indiana, Iowa, Kentucky, Missouri, and Wisconsin combined, employing nondrinking drivers as a covariate Series Illinois Comparisons
b (coef ) 0.14681 0.03320
Contrasted models Diff (b) Illinois versus Comparison
0.16645
Effect 13.65% + 3.38% Cumulative effect 15.33%
S.E. (b) 0.08232 0.03383 Pooled S.E. (diff ) 0.09123
t ratio 1.783 0.981 t ratio 1.825
One-tail probability .038 .164 One-tail probability .035
Model specifications for the two ARIMA models: Illinois Alcohol Series (natural log transform). MA = 1; SAR = 1 (seasonal periodicity = 12); first-order (nonseasonal) differencing covariate = sober Illinois driver series; no constant. Comparison States’ Alcohol Series (natural log transform). SAR = 2 only (seasonal periodicity = 12); SMA = 1 (seasonal periodicity = 12); first-order seasonal differencing covariate = sober driver series for pooled comparison states, constant included.
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
77
Table 3 Actual and predicted number of drinking drivers in fatal crashes and fatalities in Illinois in 1998 and 1999 Drinking drivers in fatal crashes
Fatalities attributable to drinking drivers
Year
Actual
Predicted
Actual
Predicted
Lives saved
1998 1999 1998/1999
487 529 1016
540 598 1138
418 458 876
463 518 981
45 60 105
for Illinois, we obtain an overall net reduction of 15.33%, which is statistically significant at the P =.035 level. The time-series analyses results were used to estimate the number of alcoholinvolved drivers who would have been involved in fatal crashes had the .08 law not been implemented in Illinois. The predictions for 1998 and 1999 are shown in the second column of Table 3. Based upon the attribution rate of 0.862 fatalities per drinking driver in Illinois, the predicted number of deaths in 1998 and 1999, without the .08 law, would have been 981 (fourth column) as opposed to the 876 (third column) that actually occurred. Thus, the estimated number of lives saved during the 2 years was 105 (shown in the last column). Although the results of the ARIMA analyses clearly demonstrate that there was a significant reduction in drinking drivers in fatal crashes in Illinois associated with the .08 law, there was in fact an increase in the percentage of drinking drivers during the period following the implementation of the .08 law. Fig. 1 provides a time series of the monthly percentage of drinking drivers in fatal crashes in Illinois from 1988 through 1999, which helps to explain this apparent anomaly. This graph was produced with an ARIMA modeling procedure using time-series analysis (Box & Tiao, 1975). Between 1989 and 1993, the percentage of drivers with positive BACs in fatal crashes in Illinois varied from a low of 28% to a high of 34%. It then fell to between 23% and 26% in 1995, after
Fig. 1. Percentage of all drivers in fatal crashes in Illinois who had positive BACs using monthly percentages (ARIMA Model Fit).
78
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
which it rose to between 28% and 29% in 1996, the year before the .08 law became effective. A decrease in the percentage of drivers in fatal crashes just before new DUI laws are passed is a common occurrence. One reason is that the high level of publicity about impending changes in the law often creates a deterrent effect. The time-series analysis (ARIMA) model predicted that, had there not been an abrupt reduction associated with the July 1997 implementation of the .08 law, the 1999 percentage of drinking drivers in fatal crashes would have been 28 –29%. The actual percentage was 25– 27%. Fig. 1 suggests that the .08 law was associated with an immediate reduction (shift in level) in the percentage of drinking drivers involved in fatal crashes, but not a change in the slope or trend. Consequently, the percentage of drinking drivers involved in fatal crashes continued to rise at about the same rate as it had before the .08 law, but the reduction in level that occurred in 1997 was sustained in 1998 and 1999.
4. Discussion This study has several strengths. It used only crash data resulting in fatalities from the FARS, a highly accurate crash record system. It considered only drivers because drinking pedestrians who may have contributed to alcohol-related crashes are not likely to have been influenced by the .08 law. It used nondrinking drivers in crashes to isolate the influence of alcohol from other factors that can affect fatal crashes. It contrasted the trend in drinking drivers in Illinois with five bordering states that, like Illinois, already had ALR laws, so differences related to that important legislation did not play a role in the results. Finally, a conservative procedure was used to estimate the probable savings in alcoholrelated fatalities. In addition to our findings, an earlier study found that following enactment of the .08 law (Voas, Taylor, Kelley Baker, Tippetts, 2000), the total number of DUI arrests and arrests at the target BACs of 0.08 and 0.09 increased in Illinois. This earlier research also demonstrated that the court system did not reduce the effect of the .08 law by reducing the conviction rate or the severity of sentencing DUI offenders. Despite these strengths, this study shares with others of its type certain weaknesses. First, the study was limited to the lowering of the BAC limit to 0.08 in one state. Further, even though 9 1/2 years of predata were used for analysis, only 30 months of postdata were available. The separation of drivers into two groups, drinking and nondrinking, was a critical step in the analysis, but only 50% of the drivers’ records contained a measured BAC. The rest had to be imputed. Finally, alcohol-related crashes are complex events resulting from many factors. It is generally not possible to consider every extraneous nonlaw factor that could influence alcohol-related crashes; nonetheless, a valid result depends on using a method that reflects the major factors influencing such crashes. For this study, we used nondrinking drivers in crashes and drinking drivers in
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
79
adjacent states to control for these factors. The validity of the results is clearly dependent on the adequacy of these control measures.
5. Summary These results, building on an earlier study of DUI arrests and convictions following the .08 law in Illinois, suggest that the .08 law, introduced in 1997, reduced the predicted percentage of drinking drivers involved in fatal crashes by 13.65%. The effect of that downward shift in the number of drinking drivers to nondrinking drivers in crashes remains, even though the trend of drinking drivers in crashes continues upward as it did before implementing the .08 law. For the 2-year period (1998 and 1999), the .08 law is associated with a reduction of approximately 105 deaths.
Acknowledgments This study was supported by the National Highway Safety Administration (Contract #DTNH22-98-D-35079).
References Apsler, R., Char, A. R., Harding, W. M., & Klein, T. M. (1999, March). The effects of .08 BAC laws. Washington, DC: National Highway Traffic Safety Administration. Box, G. E. P., & Tiao, G. C. (1975). Intervention analysis with applications to economic and environmental problems. Journal of the American Statistical Association, 70(3), 70 – 79. Foss, R. D., Stewart, J. R., & Reinfurt, D. W. (1998, November 12). Evaluation of North Carolina’s 0.08% BAC law. (DOT HS 808 893). Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration. General Accounting Office (GAO) (1999). Report to Congressional Committees on highway safety: effectiveness of state .08 blood alcohol laws. (GAO/RCED-99-179). Washington, DC: U.S. General Accounting Office. Hingson, R., Heeren, T., & Winter, M. (1996). Lowering state legal blood alcohol limits to 0.08 percent: the effect on fatal motor vehicle crashes. American Journal of Public Health, 86, 1297 – 1299. Hingson, R., Heeren, T., & Winter, M. (2000). Effects of recent 0.08% legal blood alcohol limits on fatal crash involvement. Injury Prevention, 6, 109 – 114. Johnson, D., & Fell, J. (1995). The impact of lowering the illegal BAC limit to .08 in five states in the U.S. In 39th Annual Proceedings for the Association for the Advancement of Automotive Medicine, October 16 – 18, 1995, Chicago, IL ( pp. 45 – 64). Des Plaines, IL: AAAM. Klein, T. (1986, July). A method for estimating posterior BAC distributions for persons involved in fatal traffic accidents. (DOT HS 807 094). Washington, DC: National Highway Traffic Safety Administration. National Highway Traffic Safety Administration (NHTSA). (1999). Fatality Analysis Reporting System data files, 1982 – 1998. Washington, DC: National Centre for Statistics Analyses. Research and Evaluation Associates (REA). (1991). The effects following the implementation of an 0.08 BAC limit and an administrative per se law in California. (Report no. DOT HS 807 777). Washington, DC: National Highway Traffic Safety Administration.
80
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
Rogers, P. (1995). The general deterrent impact of California’s 0.08% blood alcohol concentration limit and administrative per se license suspension laws. Sacramento, CA: California Department of Motor Vehicles. Voas, R. B., Taylor, E., Kelley Baker, T., & Tippetts, A. S. (2000, December). Effectiveness of the Illinois .08 Law. (DOT HS 809 186). Springfield, VA: National Technical Information Service. Available at: www.nhtsa.dot.gov. Voas, R. B., Tippetts, A. S., & Fell, J. C. (2000). The relationship of alcohol safety laws to drinking drivers in fatal crashes. Accident Analysis and Prevention, 32, 483 – 492. Dr. Robert B. Voas, PhD, is a Senior Research Scientist with Pacific Institute in Calverton, Maryland. He has been involved in research on alcohol and highway safety for 30 years, initially as director of the NHTSA’s Office of Program Evaluation and more recently as Principal Investigator for government research programs in drinking – driving and community alcohol problem prevention. Dr. Voas is a Fellow of the American Psychological Association and a Past President of the International Council on Alcohol, Drugs and Traffic Safety. He is also a member of the Committee on Alcohol and Drugs, the National Safety Council, and the Committee on Alcohol and Other Drugs of the National Transportation Research Board and has served on the National Board of Mothers Against Drunk Driving. His recent research projects have included an evaluation of the zero tolerance law in California and studies of underage binge drinkers crossing the border at San Diego to drink in Tijuana, where the drinking age is 18. He has just completed a national study of alcohol safety laws and differences between ethnic groups in their involvement in alcohol-related fatal crashes, using the FARS. During his 13 years at Pacific Institute, Mr. A. Scott Tippetts, BA, has served as primary statistician for more than 50 federally funded research projects in the fields of Alcohol and Drug Prevention/Treatment, Traffic Safety Programs and Legislation, and Human Factors/Equipment Design. He has served as a member of the Scientific Advisory Board for the Center for Criminal Justice Studies, and has been on MADD’s ‘‘Rating the States’’ Task Force since its inception in 1990, directing the methodology and analysis for the triannual survey and comprehensive evaluation of state anti-DUI efforts. Eileen P. Taylor, MS, is a Senior Project Manager with the Pacific Institute. She is a former probation/parole officer who has worked on DUI-related research at Pacific Institute for 13 years. Ms. Taylor has managed numerous federally sponsored evaluation projects including ones on vehicle interlock, vehicle immobilization, and the current study of the effectiveness of the .08 BAC law.
The Illinois .08 law An evaluation Robert B. Voas*, A. Scott Tippetts, Eileen P. Taylor Pacific Institute for Research and Evaluation, Calverton Office Park, Suite 300, 11710 Beltsville Drive, Calverton, MD 20705, USA Received 23 May 2001; received in revised form 1 August 2001; accepted 10 October 2001
Abstract Lowering state blood alcohol concentration (BAC) limits to 0.08, though controversial, has been supported by most evaluation studies to date. The Illinois .08 BAC law implemented in 1997 provided a unique opportunity to evaluate the effect of the law without the simultaneous passage of an administrative license revocation (ALR) law, which has clouded some previous evaluations of the .08 laws. The proportion of all drinking drivers in fatal crashes was compared before versus after implementing the .08 law using time-series analysis to evaluate 12 years of fatal crash data for Illinois and five bordering states. The results showed that the proportion of drinking drivers in fatal crashes decreased by 14% in Illinois and increased by 3% in bordering states. The proportion of drinking drivers in fatal crashes in Illinois, though increasing since 1995, was sharply reduced after passage of the .08 law in 1997, saving more than 100 lives in 1998 and 1999 than it would have without the .08 law. D 2002 National Safety Council and Elsevier Science Ltd. All rights reserved. Keywords: Drinking and driving; Driving under the influence; .08 laws; Administrative license revocation laws; Alcohol-related crashes
1. Introduction Over the last several years, there has been a strong trend for state legislatures to lower the legal blood alcohol concentration (BAC) from 0.10 to 0.08. As of * Corresponding author. Tel.: +1-301-755-2700; fax: +1-301-755-2799. E-mail address: [email protected] (R.B. Voas). 0022-4375/02/$ – see front matter D 2002 National Safety Council and Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 2 - 4 3 7 5 ( 0 2 ) 0 0 0 0 4 - X
74
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
May 2001, 24 states have taken that action under pressure from safety advocates supported by several studies on the effect of the law (Apsler, Char, Harding, & Klein, 1999; Hingson, Heeren, & Winter, 1996, 2000; Johnson & Fell, 1995; Research and Evaluation Associates [REA], 1991; Rogers, 1995; Voas, Tippetts, & Fell, 2000). This process has gone forward despite opposition by the hospitality industry, some negative reports (e.g., Foss, Stewart, & Reinfurt, 1998), and a negative evaluation of .08 laws by the General Accounting Office (GAO, 1999). The preponderance of evidence from these studies demonstrates that the .08 legislation produces a modest (about 8%) reduction in alcohol-related fatal crashes. Nonetheless, it has been difficult in some states to evaluate the effectiveness of .08 laws because they have been enacted in conjunction with other drunk-driving legislation, particularly administrative license revocation (ALR) laws that are known to be effective in reducing crashes. Consequently, it has been hard to find an appropriate comparison for states that adopt .08 legislation. Recently, Hingson et al. (2000) provided evidence that states implementing .08 laws, independent of ALR laws, appeared to experience reductions in alcohol-related crashes. Illinois provided another opportunity to evaluate the .08 law because that state’s ALR law had been enacted on January 1, 1986, more than 11 years before implementation of the .08 law in July 1997. Further, none of the five states that border Illinois, all of which also have ALR laws, had enacted a .08 law. Thus, these states provided a good comparison for analyzing the effect of the Illinois .08 law. To take advantage of that opportunity, the National Highway Traffic Safety Administration (NHTSA) funded a recently published study entitled Effectiveness of the Illinois .08 Law (Voas, Taylor, Kelley Baker, & Tippetts, 2000). The objective of that study was to determine the effect of the passage of a .08 law on the criminal justice system. The results indicated that all driving-under-the-influence (DUI) arrests statewide increased by 10% and DUI arrests in the 0.08 –0.09 range increased from 1% before to 9% after enactment of the law. Although a high percentage of DUI suspects in Illinois refused to take the breath test (39%), there was no evidence in state records that the proportion of refusals increased following enactment of the law. Nor was there any evidence that offenders or their lawyers were able to reduce the penalties normally assessed by the courts or the motor vehicle department for the DUI offense. Conviction rates increased slightly, and the proportion of offenders who were suspended remained higher than 90%. This report uses 12 years of data (1988 –1999) from the Fatality Analysis Reporting System (FARS; NHTSA, 1999) for Illinois and five bordering states (Indiana, Iowa, Kentucky, Missouri, and Wisconsin) to evaluate the effect of the .08 law on the number of drinking drivers compared to nondrinking drivers in fatal crashes.
2. Methods The FARS provides a file for every crash occurring on a U.S. highway that produces a fatality within 24 hours. A subfile is created for every driver involved
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
75
in such a crash containing the BAC measured at the time of the crash or an estimated BAC based on the characteristics of the crash and the driver (Klein, 1986). These BAC data permit the division of the files into two groups: drinking drivers (those with BACs greater than zero) and nondrinking drivers (those with zero BACs). Table 1 summarizes the data from FARS used in this study. These data (shown in Table 1) cover four time series from 1988 to 1999, categorized by drinking and nondrinking drivers in Illinois and drinking and nondrinking drivers in the five comparison states. In Illinois, the number of drinking drivers in crashes reflects any reduction in the role of alcohol in fatal crashes resulting from the .08 law. Any changes occurring when the law went into effect in June 1997 can be compared with the other three series. The nondrinking drivers series in Illinois reflects factors influencing the number of fatal crashes within the state such as the quality of roadways and vehicles, the weather, and the economy. The drinking driver series for the five comparison states reflects the effect of factors outside Illinois that might influence national drinking – driving trends such as the national economy or alcohol prices. Finally, the nondrinking driver series for the comparison states reflects national trends that affect fatal crashes such as gasoline prices. For this study, separate ARIMA analyses (Box & Tiao, 1975) were applied to two time series: the number of drinking drivers in Illinois, and a similar series of drinking drivers in the five comparison states as a group. In each ARIMA analysis, the number of nondrinking drivers was used as a covariate. The intercept date was July 1, 1997. The results from the two analyses were then compared using a pooled error term. This provided a before – after experimental comparison design with a prelaw period of 9 1/2 years and a postlaw period of 30 months. Quantifying any injury savings attributable to the .08 law in Illinois involved two steps. First, we estimated the number of fatalities associated with each drinking driver involved in a fatal crash. Pedestrian fatalities were eliminated when the pedestrian had a positive BAC because the drinking pedestrian, rather than a driver, may have contributed to the cause of the crash. Second, we estimated the number of fatalities attributable to each driver in the crash. If two drivers were involved, then half the fatalities in that crash were attributed to each driver. The fatalities attributable to sober and drinking drivers were aggregated separately, and only the fatalities attributed to drinking drivers (BAC > .00) were
Table 1 Drinking and nondrinking drivers in fatal crashes from the FARS, 1988 – 1999 Illinois N Drinking drivers Nondrinking drivers Totals
7,159.2 17,830.8 24,990.0
Comparisons % Known BAC 79 33
N 18,908.9 47,857.1 66,766.0
% Known BAC 76 42
76
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
used in the calculation of the number of fatalities that could be attributed to the .08 law. These two steps likely produce a smaller number of fatalities identified as being ‘‘alcohol related’’ because drinking pedestrians have been eliminated and responsibility for the crash has been attributed equally to drinking and nondrinking drivers, even though it has been shown that drinking drivers are significantly more likely to cause fatal crashes than sober drivers. Thus, this approach produces a somewhat conservative estimate of the lives saved by the .08 law. For the nation as a whole, the FARS for the 2 years from 1998 to 1999 showed 26,228 alcohol-positive drivers who, using this attribution method, accounted for 22,622 fatalities. This produced a rate of 0.863 fatalities per drinking driver. Considering the state of Illinois alone, there were 1,016 alcohol-positive drivers to whom this procedure attributed 876 fatalities, yielding a rate of 0.862 fatalities per alcohol-positive driver, which is almost identical to the national rate (0.863) during the same 2-year period.
3. Results The results from the time-series analysis of drinking drivers in fatal crashes in Illinois are contrasted with the results from the analysis of drinking drivers in the five comparison states in Table 2. The change from the prelaw to the postlaw period in Illinois produced an estimated 13.65% reduction in the number of drinking drivers in fatal crashes in Illinois. This was statistically significant ( P =.038). The change from the same pre-July 1997 period to the post-July 1997 period for the neighboring states (combined) was a nonsignificant increase of 3.38%. If we contrast the ARIMA results for the five comparison states with that
Table 2 Illinois .08 —ARIMA intervention analysis of drinking drivers in fatal crashes in Illinois compared to five bordering states: Indiana, Iowa, Kentucky, Missouri, and Wisconsin combined, employing nondrinking drivers as a covariate Series Illinois Comparisons
b (coef ) 0.14681 0.03320
Contrasted models Diff (b) Illinois versus Comparison
0.16645
Effect 13.65% + 3.38% Cumulative effect 15.33%
S.E. (b) 0.08232 0.03383 Pooled S.E. (diff ) 0.09123
t ratio 1.783 0.981 t ratio 1.825
One-tail probability .038 .164 One-tail probability .035
Model specifications for the two ARIMA models: Illinois Alcohol Series (natural log transform). MA = 1; SAR = 1 (seasonal periodicity = 12); first-order (nonseasonal) differencing covariate = sober Illinois driver series; no constant. Comparison States’ Alcohol Series (natural log transform). SAR = 2 only (seasonal periodicity = 12); SMA = 1 (seasonal periodicity = 12); first-order seasonal differencing covariate = sober driver series for pooled comparison states, constant included.
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
77
Table 3 Actual and predicted number of drinking drivers in fatal crashes and fatalities in Illinois in 1998 and 1999 Drinking drivers in fatal crashes
Fatalities attributable to drinking drivers
Year
Actual
Predicted
Actual
Predicted
Lives saved
1998 1999 1998/1999
487 529 1016
540 598 1138
418 458 876
463 518 981
45 60 105
for Illinois, we obtain an overall net reduction of 15.33%, which is statistically significant at the P =.035 level. The time-series analyses results were used to estimate the number of alcoholinvolved drivers who would have been involved in fatal crashes had the .08 law not been implemented in Illinois. The predictions for 1998 and 1999 are shown in the second column of Table 3. Based upon the attribution rate of 0.862 fatalities per drinking driver in Illinois, the predicted number of deaths in 1998 and 1999, without the .08 law, would have been 981 (fourth column) as opposed to the 876 (third column) that actually occurred. Thus, the estimated number of lives saved during the 2 years was 105 (shown in the last column). Although the results of the ARIMA analyses clearly demonstrate that there was a significant reduction in drinking drivers in fatal crashes in Illinois associated with the .08 law, there was in fact an increase in the percentage of drinking drivers during the period following the implementation of the .08 law. Fig. 1 provides a time series of the monthly percentage of drinking drivers in fatal crashes in Illinois from 1988 through 1999, which helps to explain this apparent anomaly. This graph was produced with an ARIMA modeling procedure using time-series analysis (Box & Tiao, 1975). Between 1989 and 1993, the percentage of drivers with positive BACs in fatal crashes in Illinois varied from a low of 28% to a high of 34%. It then fell to between 23% and 26% in 1995, after
Fig. 1. Percentage of all drivers in fatal crashes in Illinois who had positive BACs using monthly percentages (ARIMA Model Fit).
78
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
which it rose to between 28% and 29% in 1996, the year before the .08 law became effective. A decrease in the percentage of drivers in fatal crashes just before new DUI laws are passed is a common occurrence. One reason is that the high level of publicity about impending changes in the law often creates a deterrent effect. The time-series analysis (ARIMA) model predicted that, had there not been an abrupt reduction associated with the July 1997 implementation of the .08 law, the 1999 percentage of drinking drivers in fatal crashes would have been 28 –29%. The actual percentage was 25– 27%. Fig. 1 suggests that the .08 law was associated with an immediate reduction (shift in level) in the percentage of drinking drivers involved in fatal crashes, but not a change in the slope or trend. Consequently, the percentage of drinking drivers involved in fatal crashes continued to rise at about the same rate as it had before the .08 law, but the reduction in level that occurred in 1997 was sustained in 1998 and 1999.
4. Discussion This study has several strengths. It used only crash data resulting in fatalities from the FARS, a highly accurate crash record system. It considered only drivers because drinking pedestrians who may have contributed to alcohol-related crashes are not likely to have been influenced by the .08 law. It used nondrinking drivers in crashes to isolate the influence of alcohol from other factors that can affect fatal crashes. It contrasted the trend in drinking drivers in Illinois with five bordering states that, like Illinois, already had ALR laws, so differences related to that important legislation did not play a role in the results. Finally, a conservative procedure was used to estimate the probable savings in alcoholrelated fatalities. In addition to our findings, an earlier study found that following enactment of the .08 law (Voas, Taylor, Kelley Baker, Tippetts, 2000), the total number of DUI arrests and arrests at the target BACs of 0.08 and 0.09 increased in Illinois. This earlier research also demonstrated that the court system did not reduce the effect of the .08 law by reducing the conviction rate or the severity of sentencing DUI offenders. Despite these strengths, this study shares with others of its type certain weaknesses. First, the study was limited to the lowering of the BAC limit to 0.08 in one state. Further, even though 9 1/2 years of predata were used for analysis, only 30 months of postdata were available. The separation of drivers into two groups, drinking and nondrinking, was a critical step in the analysis, but only 50% of the drivers’ records contained a measured BAC. The rest had to be imputed. Finally, alcohol-related crashes are complex events resulting from many factors. It is generally not possible to consider every extraneous nonlaw factor that could influence alcohol-related crashes; nonetheless, a valid result depends on using a method that reflects the major factors influencing such crashes. For this study, we used nondrinking drivers in crashes and drinking drivers in
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
79
adjacent states to control for these factors. The validity of the results is clearly dependent on the adequacy of these control measures.
5. Summary These results, building on an earlier study of DUI arrests and convictions following the .08 law in Illinois, suggest that the .08 law, introduced in 1997, reduced the predicted percentage of drinking drivers involved in fatal crashes by 13.65%. The effect of that downward shift in the number of drinking drivers to nondrinking drivers in crashes remains, even though the trend of drinking drivers in crashes continues upward as it did before implementing the .08 law. For the 2-year period (1998 and 1999), the .08 law is associated with a reduction of approximately 105 deaths.
Acknowledgments This study was supported by the National Highway Safety Administration (Contract #DTNH22-98-D-35079).
References Apsler, R., Char, A. R., Harding, W. M., & Klein, T. M. (1999, March). The effects of .08 BAC laws. Washington, DC: National Highway Traffic Safety Administration. Box, G. E. P., & Tiao, G. C. (1975). Intervention analysis with applications to economic and environmental problems. Journal of the American Statistical Association, 70(3), 70 – 79. Foss, R. D., Stewart, J. R., & Reinfurt, D. W. (1998, November 12). Evaluation of North Carolina’s 0.08% BAC law. (DOT HS 808 893). Washington, DC: U.S. Department of Transportation, National Highway Traffic Safety Administration. General Accounting Office (GAO) (1999). Report to Congressional Committees on highway safety: effectiveness of state .08 blood alcohol laws. (GAO/RCED-99-179). Washington, DC: U.S. General Accounting Office. Hingson, R., Heeren, T., & Winter, M. (1996). Lowering state legal blood alcohol limits to 0.08 percent: the effect on fatal motor vehicle crashes. American Journal of Public Health, 86, 1297 – 1299. Hingson, R., Heeren, T., & Winter, M. (2000). Effects of recent 0.08% legal blood alcohol limits on fatal crash involvement. Injury Prevention, 6, 109 – 114. Johnson, D., & Fell, J. (1995). The impact of lowering the illegal BAC limit to .08 in five states in the U.S. In 39th Annual Proceedings for the Association for the Advancement of Automotive Medicine, October 16 – 18, 1995, Chicago, IL ( pp. 45 – 64). Des Plaines, IL: AAAM. Klein, T. (1986, July). A method for estimating posterior BAC distributions for persons involved in fatal traffic accidents. (DOT HS 807 094). Washington, DC: National Highway Traffic Safety Administration. National Highway Traffic Safety Administration (NHTSA). (1999). Fatality Analysis Reporting System data files, 1982 – 1998. Washington, DC: National Centre for Statistics Analyses. Research and Evaluation Associates (REA). (1991). The effects following the implementation of an 0.08 BAC limit and an administrative per se law in California. (Report no. DOT HS 807 777). Washington, DC: National Highway Traffic Safety Administration.
80
R.B. Voas et al. / Journal of Safety Research 33 (2002) 73–80
Rogers, P. (1995). The general deterrent impact of California’s 0.08% blood alcohol concentration limit and administrative per se license suspension laws. Sacramento, CA: California Department of Motor Vehicles. Voas, R. B., Taylor, E., Kelley Baker, T., & Tippetts, A. S. (2000, December). Effectiveness of the Illinois .08 Law. (DOT HS 809 186). Springfield, VA: National Technical Information Service. Available at: www.nhtsa.dot.gov. Voas, R. B., Tippetts, A. S., & Fell, J. C. (2000). The relationship of alcohol safety laws to drinking drivers in fatal crashes. Accident Analysis and Prevention, 32, 483 – 492. Dr. Robert B. Voas, PhD, is a Senior Research Scientist with Pacific Institute in Calverton, Maryland. He has been involved in research on alcohol and highway safety for 30 years, initially as director of the NHTSA’s Office of Program Evaluation and more recently as Principal Investigator for government research programs in drinking – driving and community alcohol problem prevention. Dr. Voas is a Fellow of the American Psychological Association and a Past President of the International Council on Alcohol, Drugs and Traffic Safety. He is also a member of the Committee on Alcohol and Drugs, the National Safety Council, and the Committee on Alcohol and Other Drugs of the National Transportation Research Board and has served on the National Board of Mothers Against Drunk Driving. His recent research projects have included an evaluation of the zero tolerance law in California and studies of underage binge drinkers crossing the border at San Diego to drink in Tijuana, where the drinking age is 18. He has just completed a national study of alcohol safety laws and differences between ethnic groups in their involvement in alcohol-related fatal crashes, using the FARS. During his 13 years at Pacific Institute, Mr. A. Scott Tippetts, BA, has served as primary statistician for more than 50 federally funded research projects in the fields of Alcohol and Drug Prevention/Treatment, Traffic Safety Programs and Legislation, and Human Factors/Equipment Design. He has served as a member of the Scientific Advisory Board for the Center for Criminal Justice Studies, and has been on MADD’s ‘‘Rating the States’’ Task Force since its inception in 1990, directing the methodology and analysis for the triannual survey and comprehensive evaluation of state anti-DUI efforts. Eileen P. Taylor, MS, is a Senior Project Manager with the Pacific Institute. She is a former probation/parole officer who has worked on DUI-related research at Pacific Institute for 13 years. Ms. Taylor has managed numerous federally sponsored evaluation projects including ones on vehicle interlock, vehicle immobilization, and the current study of the effectiveness of the .08 BAC law.