Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Objective The research objective, as outlined in the original request for proposal, was to âprovide a multidisciplinary analysis of the relative influence of the types of factors that contrib- uted to the recent national decline in the number of highway fatalities and rates in the United States.â Between 2005 and 2011, peak to trough, the number of traffic fatalities in the United States declined by 11,031, from 43,510 in 2005 to 32,479 in 2011 (see Figure S-1). This decline amounted to a reduction in traffic-related deaths by 25.4%, by far the greatest decline over a comparable period in the last 30 years. Historically, significant drops in traffic fatalities over a short period have coincided with economic recessions. Figure S-2 displays the number of traffic fatalities by year from 1966 through 2016, along with the periods of the seven recessions during that span. Traffic fatali- ties for 2016 were projected from an early estimate by NHTSA of the number of fatalities during the first half of 2016 (NHTSA 2016a). Longer recessions have coincided with deeper declines in the number of traffic fatalities. This project provides an analysis that identifies the specific factors in the economic decline that affected fatal crash risk, while taking into account the long-term factors that determine the level of traffic safety. The fundamental approach of this research project was based on the understanding that the number of fatalities in crashes is the product of risk times exposure: = ÃFatalities Risk Exposure Eq. S-1 Figure S-3 displays the relationship between vehicle miles traveled (VMT) (exposure), traffic fatalities (outcomes), and traffic fatality rates by VMT (risk), from 2001 to 2012. The figure shows the ratio of the values for each year to the base year of 2001. Exposure as mea- sured by VMT was relatively stable. However, a reduction in the risk of travel pulled down the number of traffic fatalities. Thus, the fatality risk of travel contributed significantly to the substantial decline in fatalities over the period. The decrease in exposure due to the recession and subsequent slowdown in economic activity contributed less. The goal of this project was to identify the sources of reduced risk. Analysis Approach Factors that affected the incidence and risk of fatal crashes and fatal crash injuries over the period were organized using the Haddon Matrix. The Haddon Matrix provides a framework that covers the factors comprehensively (Williams 1999). The utility of the S U M M A R Y Identification of Factors Contributing to the Decline of Traffic Fatalities in the United States from 2008 to 2012
2 Identification of Factors Contributing to the Decline of Traffic Fatalities in the United States from 2008 to 2012 Haddon Matrix was to ensure that all components of what might be called the crash systemâvehicle, drivers, and environmentâwere considered. Two fundamental processes were at work over the period. The first process set the baseline level of safety that influenced the number of traffic deaths each year. The baseline level was the product of long-term trends in factors known to affect traffic safety, such as safety-belt use, improvements in the crashworthiness of cars, highway infrastructure, traffic enforce- ment and safety campaigns, driver license laws, and other efforts to reduce the number of Figure S-1. Traffic fatalities, 2001â2012. Figure S-2. Motor vehicle traffic fatalities and periods of recession, 1966â2012.
Summary 3 fatalities on U.S. roads. Most of these factors operated incrementally and changed relatively slowly over time. Highway infrastructure cannot change dramatically over a short period. Safety-belt use is known to be a primary safety intervention, but belt use increased slowly and monotonically over the period. The second process consisted of the factors that precipitated the sharp decline in fatal crashes and deaths from 2008 to 2011. The major event in this period was the recession that started in December 2007 and ended in June 2009 (NBER 2010). Of the components of the crash systemâvehicle, drivers, and environmentârecessions have a short-term and substantial impact on drivers. Moreover, the recession affected some high-risk groups more than others, particularly younger drivers, so it may have taken some risky drivers off the road and reduced the amount of risky driving. At the same time, the long-term factors that influenced safety continued, such as incre- mental increases in safety-belt use, the penetration of more crashworthy passenger vehicles into the fleet, safety campaigns to improve driver behavior, infrastructure improvements, and other factors. Explaining the drop in traffic fatalities between 2008 and 2011 was a major goal of the project, but the explanation is undertaken within the context of overall trends in traffic safety over the period. Findings and Statistical Models A key insight into the analysis of factors producing the sharp drop in traffic fatalities was that the young contributed disproportionately to the drop-off in traffic fatalities. Of the reduction in traffic fatalities from 2007 to 2011, persons younger than 26 years of age accounted for almost 48% of the drop, though they were only about 28% of total traffic fatalities prior to the decline. Figure S-4 shows that traffic deaths among persons 25 years old or younger dropped substantially more than other groups. Young drivers are known to be a high-risk group and can be readily identified in the crash data. Other high-risk groups also likely contributed to the decline, but they cannot be identified as well as age can. Using the Haddon Matrix and the research teamâs broad expertise, a comprehensive set of parameters was identified, and their contribution to the drop in traffic fatalities after Figure S-3. Fatality rates by VMT and fatalities, normalized to 2001.
4 Identification of Factors Contributing to the Decline of Traffic Fatalities in the United States from 2008 to 2012 2007 was evaluated. Statistical models of the incidence of traffic fatalities over the period were developed using these parameters. Table S-1 provides the expected association of the various parameters in the models and describes the mechanisms through which they were hypothesized to affect the number of traffic fatalities. Most of the modeling results were consistent with these expectations, though not all. Two basic approaches were used to model the factors that were associated with the drop in traffic fatalities after 2007. The first approach was a set of count models, using negative binomial models to examine the associations between predictors (the variables in Table S-1) and raw fatality counts. Two count models were developed. One used a state fixed effect to remove the stable differences between states and focus on changes over time; this was labeled the model considering state or models controlling for state (MCS) model. The other model left out this fixed effect, allowing differences between states to be captured by the measured predictors; this was labeled the model not consid- ering state (MNCS). The other statistical modeling technique was a log-change regres- sion model, to model the association between the change in predictor variables in one year with the change in the outcome variable (traffic fatalities) in the following year. Table S-2 shows the results from the two count models. Table S-3 provides the results from the change model. The two modeling approaches were in broad agreement. The most significant con- tributors to the drop in traffic fatalities after 2007 were the substantial increase in teen and young adult unemployment, the reductions in median household income, and the reduction in gross domestic product (GDP) per capita income. The right-most col- umn in Table S-3 estimates the percentage decline contributed by each factor in the change model. The decline in rural VMT, increased strictness of DUI laws, and decreased beer consumption also contributed. State highway spending was not a significant con- tributor to the drop; the effect of changes in infrastructure was likely more cumulative and longer term. Changes in safety-belt use rates and fuel prices were not significant contributors to the decline in traffic fatalities after 2007 because they did not change much over the period. Failing to find that certain well-established safety interventions (safety-belt usage, high- way capital improvements) did not contribute significantly to the sharp drop in traffic fatalities during the recession does not mean that they are not essential tools to reduce Figure S-4. Ratio of traffic fatalities by age groups, normalized to 2001.
Summary 5 Variable Expected association with traffic fatalities Expected mechanism Total VMT Positive Increase in VMT increases exposure to traffic crashes and therefore fatalities. Proportion rural VMT Positive Increased proportion of rural VMT increases proportion of travel on riskier roads, leading to more fatalities. Pump price Negative Increased pump price raises cost of travel, reducing total travel and discretionary travel, reducing exposure to fatal crashes. Gross domestic product (GDP) per capita Positive GDP per capita reflects economic activity, which in turn leads to more travel, more exposure to crashes, and more fatalities. Median income Positive Increased median income increases discretionary and leisure travel, resulting in more exposure and more fatalities. 16â24 unemployment Negative Increased unemployment reduces total travel and discretionary leisure travel, resulting in fewer fatalities. Capital spending/mile (lag) Mixed Improved infrastructure would be expected to shift travel to higher quality roads. It may also induce more travel, thus more exposure to fatalities. Safety spending/mile (lag) Negative Increased traffic enforcement, education, and safety programs would reduce risky driving and reduce fatalities. Belt-use rate Negative Increased belt use provides more protection to vehicle occupants and reduces the probability of fatal injury, given a crash. Driving under the influence (DUI) law rating Negative Increased stringency of DUI laws reduces drunk (risky) driving and traffic fatalities. Motorcycle helmet law rating Negative Increased stringency of motorcycle helmet laws provides more protection to motorcycle riders and reduces the probability of fatality, given a crash. Beer consumption Positive Increased beer consumption may increase driving while under the influence of alcohol, increase risky driving, and increase traffic fatalities. Wine consumption Positive Increased wine consumption may increase driving while under the influence of alcohol, increase risky driving, and increase traffic fatalities. Penetration of model year >1991 Negative Increased penetration of vehicles that provide more occupant protection and more safety features reduces the probability of a crash, and reduces the probability of fatal injury, given a crash. Table S-1. Explanatory factors and expected mechanisms of activity.
6 Identification of Factors Contributing to the Decline of Traffic Fatalities in the United States from 2008 to 2012 Variable Change in parameter value from 2007 to 2011 MNCS model MCS model Predicted change in fatalities Statistically significant at 5% level? Predicted change in fatalities Statistically significant at 5% level? Rural VMT proportion as percent of total â0.8 % â103 Yes 95 Yes State gross domestic product per capita â$6,301 per person â617 Yes â1236 Yes Unemployment rate for 16- to 24-year-olds +6.39% â3305 Yes â3125 Yes Pump price +$0.55/gallon â877 No 127 No Per capita beer consumption â0.08 gallon/person â835 Yes â1312 Yes Median income â$3,760 2,677 Yes â466 Yes DUI laws rating â1.05 â120 No â261 Yes Safety-belt laws rating â0.16 5 No â28 No Motorcycle helmet law rating 0 0 Yes 0 No Table S-2. Effects of count-model factors. traffic fatalities. It means that their impact was not detectable given the magnitude of the short-term effects of other factors. The long-term factors that set the baseline of traffic safety continued to operate. Overlaid on them were the short-term shock of the reces- sion, which drove up unemployment, particularly among teens and young adults, and declining median income, which likely reduced driving and risky driving among high-risk populations. Implications and Further Research â¢ Teens and young adults contributed disproportionately to the reduction in traffic fatali- ties from 2008 through 2011. It is suggested here that the mechanism was economic con- straints that reduced total travel and risky (discretionary and leisure) travel. It has long been known that teens and young adults have disproportionately high crash risk, but the results from this study suggest that their behavior can be significantly modified over the short run, substantially reducing fatalities. â¢ The findings related to median household income are consistent with an income effect. This finding warrants further investigation, but interventions aimed at lower-income groups may have a disproportionately positive effect, similar to reducing crash risk among teens and young adults. â¢ DUI laws showed a significant positive effect in reducing traffic fatalities, even over the short term of this study and even within the substantial impact of the economic contrac- tion. Reduced beer consumption similarly showed a significant positive effect. It is clear that continuing to focus on reducing drunk driving can have a disproportionate effect on reducing traffic fatalities.
Summary 7 â¢ Rural VMT bears a higher risk of fatal crashes across all road types; reduction in the pro- portion of rural VMT was significant in all models. Programs aimed at reducing the risk of rural travel can substantially reduce traffic fatalities. â¢ It may be difficult to discern in any given year the effects of safety countermeasures, due to the significant influence of other factors on traffic fatalities. There is a need to more fully document and assess safety advances from countermeasures because these other factors may obscure them. â¢ The results here clearly illustrate that factors outside the authority of safety profes- sionals can have highly significant impacts on the level of highway safety. In the short term, shocks in the economy can overwhelm the effect of safety interventions that generally influence crash risk in the long term. It is clear that exogenous factors such as economic trends should be accounted for in setting realistic goals and evaluating traffic safety programs. Variable 2007 Mean 2011 Mean Percent change in predictor 2007â2011 Percent change in predicted fatalities 2007â2011 Total VMT 3,031,124 2,962,740 â2.3% â1.2% Proportion rural VMT 0.33 0.32 â1.6% â0.1% Pump price change 3.11 3.20 2.6% â0.1% Gross domestic product per capita change 59,687 54,519 â7.5% â1.2% Median income change 56,081 53,621 â4.3% â2.2% 16â24 unemployment change 10.59 16.69 55.7% â6.1% Capital spending/mile (lag) change 73.69 81.27 7.9% â0.1% Safety spending/mile (lag) change 13.61 14.68 9.3% 0.1% Belt-use rate change 85.77 88.10 2.4% â0.1% DUI law rating change 19.77 20.50 4.0% â0.7% Motorcycle helmet law rating change 2.91 2.91 0.0% 0.0% Beer consumption change 1.21 1.15 â3.5% â0.7% Wine consumption change 0.37 0.39 5.0% â0.1% MY>1991 change 95.80 97.11 1.4% 0.1% Table S-3. Effects of change-model predictors for 2007â2011.