Peter T. Savolainen

The statistical analysis of highway crash-injury severities: A review and assessment of methodological alternatives

Reducing the severity of injuries resulting from motor-vehicle crashes has long been a primary emphasis of highway agencies and motor-vehicle manufacturers. While progress can be simply measured by the reduction in injury levels over time, insights into the effectiveness of injury-reduction technologies, policies, and regulations require a more detailed empirical assessment of the complex interactions that vehicle, roadway, and human factors have on resulting crash-injury severities. Over the years, researchers have used a wide range of methodological tools to assess the impact of such factors on disaggregate-level injury-severity data, and recent methodological advances have enabled the development of sophisticated models capable of more precisely determining the influence of these factors. This paper summarizes the evolution of research and current thinking as it relates to the statistical analysis of motor-vehicle injury severities, and provides a discussion of future methodological directions.

Mixed logit analysis of bicyclist injury severity resulting from motor vehicle crashes at intersection and non-intersection locations

Standard multinomial logit (MNL) and mixed logit (MXL) models are developed to estimate the degree of influence that bicyclist, driver, motor vehicle, geometric, environmental, and crash type characteristics have on bicyclist injury severity, classified as property damage only, possible, nonincapacitating or severe (i.e., incapacitating or fatal) injury. This study is based on 10,029 bicycleinvolved crashes that occurred in the State of Ohio from 2002 to 2008. Results of likelihood ratio tests reveal that some of the factors affecting bicyclist injury severity at intersection and non-intersection locations are substantively different and using a common model to jointly estimate impacts on severity at both types of locations may result in biased or inconsistent estimates. Consequently, separate models are developed to independently assess the impacts of various factors on the degree of bicyclist injury severity resulting from crashes at intersection and non-intersection locations. Several covariates are found to have similar impacts on injury severity at both intersection and non-intersection locations. Conversely, six variables were found to significantly influence injury severity at intersection locations but not non-intersection locations while four variables influenced bicyclist injury severity only at non-intersection locations. In crashes occurring at intersection locations, the likelihood of severe bicyclist injury increases by 14.8 percent if the bicyclist is not wearing a helmet, 82.2 percent if the motorist is under the influence of alcohol, 141.3 percent if the crash-involved motor vehicle is a van, 40.6 percent if the motor vehicle strikes the side of the bicycle, and 182.6 percent if the crash occurs on a horizontal curve with a grade. Results from non-intersection locations show the likelihood of severe injuries increases by 374.5 percent if the bicyclist is under the influence of drugs, 150.1 percent if the motorist is under the influence of alcohol, 53.5 percent if the motor vehicle strikes the side of the bicycle and 99.9 percent if the crash-involved motor vehicle is a heavy-duty truck.

Examination of factors determining fault in two-vehicle motorcycle crashes

Motorcycle crashes frequently involve a combination of high-risk behaviors by the motorcyclist or the other crash-involved driver. Such behaviors may include riding or driving without appropriate licensure or while under the influence of alcohol, as well as deciding not to use a safety device such as a helmet or safety belt. Given that these factors frequently occur in combination with one another, it is difficult to untangle the specific effects of individual factors leading up to the crash outcome. This study assesses how various rider-, driver-, and other crash-specific factors contribute to at-fault status in two-vehicle motorcycle crashes, as well as how these same factors affect the propensity for other high-risk behaviors. Furthermore, the interrelationships among fault status and these other behaviors are also examined using a multivariate probit model. This model is developed using police-reported crash data for the years 2006-2010 from the State of Ohio. The results show that younger motorcyclists are more likely to be at-fault in the event of a collision, as are riders who are under the influence of alcohol, riding without insurance, or not wearing a helmet. Similarly, motorcyclists were less likely to be at-fault when the other driver was of younger age or was driving under the influence of alcohol, without insurance, or not wearing their safety belt. Crash-involved parties who engaged in one high-risk behavior were more likely to engage in other such behaviors, as well, and this finding was consistent for both motorcyclists and drivers. The results of this study suggest that educational and enforcement strategies aimed at addressing any one of these behaviors are likely to have tangential impacts on the other behaviors, as well.

Safety impacts at intersections on curved segments

Indiana geometric design policy, consistent with national standards, allows for the design of intersections on superelevated curves if other solutions are prohibitively expensive. Consequently, the Indiana Department of Transportation (DOT) has built a number of such intersections. Following a series of fatal crashes at one of these intersections, Indiana DOT made a decision to avoid designing intersections on segments with steep superelevation. This design restriction calls for expensive alternatives, such as realigning roads or adding grade separations. This research was done to determine whether superelevated intersections were more hazardous than similar intersections located on tangents and, if so, to determine what combination of factors made this true. The research focused on two-way stop-controlled intersections where the mainline was a high-speed four-lane divided highway located on a superelevated curve. An attempt was made to analyze as many factors as possible by using appropriate comparison techniques. Negative binomial models were developed to determine the statistical relationship between crash occurrence and intersection geometric characteristics, including curvature of the main road. Crash severity and the joint impact of curvature with weather and lighting conditions were examined by using binomial comparisons of proportions. Research findings show significant increases in crash frequency and severity at intersections located on superelevated curves.

Driver Injury Severity Resulting from Single-Vehicle Crashes Along Horizontal Curves on Rural Two-Lane Highways

Horizontal curves have been identified as a target area for improving safety on rural two-lane highways in Texas. This study involved the development of multinomial logit models to assess driver injury severity resulting from single-vehicle crashes on such roads. Likelihood ratio tests warranted the development of separate injury severity models for curves of small, medium, and large radius. Various driver, vehicle, roadway, and environmental characteristics were found to affect injury severity among the 10,029 crashes analyzed. Run-off-the-road crashes, particularly those resulting in collisions with roadside objects, were found to increase injury severity significantly. Females were more likely to sustain injury and older drivers to be critically injured, particularly on curves of smaller radius. Various driver actions and behaviors were also significant determinants of injury severity. Unbelted drivers were up to 10 times more likely to suffer fatal injuries, and drivers who were uninsured, fatigued, or under the influence of drugs or alcohol were more likely to be seriously injured. Several of these behavioral factors were more pronounced on sharper curves.

Effects of Horizontal Curvature on Single-Vehicle Motorcycle Crashes Along Rural Two-Lane Highways

Motorcycle crashes and fatalities have been increasing rapidly during the past 12 years both in Ohio and across the rest of the United States. In response to these issues, various studies have examined aspects of motorcycle safety in recent years. However, there has been limited research on the effects of site-specific roadway geometry on the frequency of motorcycle crashes, particularly at nonintersection locations. Typically, researchers employ Poisson and negative binomial crash prediction modeling techniques in these types of studies. The research presented in this paper uses a negative binomial model, applying full Bayes methods to improve model performance and to assess the impacts of horizontal curvature and other geometric features on the frequency of single-vehicle motorcycle crashes along segments of rural two-lane highways. The data used in this study include crash records for the years 2002 through the spring of 2008, in combination with available geometric design information, for those curves maintained by the State of Ohio. The analysis data set includes 30,379 horizontal curves that experienced a total of 225 motorcycle crashes during the study period. The findings show that the radius and length of each horizontal curve significantly influence the frequency of motorcycle crashes, as do shoulder width, annual average daily traffic, and the location of the road segment in relation to the curve.

Effectiveness of Motorcycle Training and Motorcyclists' Risk-Taking Behavior

Persistent increases in motorcycle fatalities and injuries in recent years have heightened safety awareness and have focused attention on the role that motorcyclist training and education can play in reducing accident rates. In this study a 2005 sample of Indiana motorcyclists was used to estimate statistical models of the effectiveness of existing training programs in reducing accident probabilities. Statistical models relating to motorcyclist speed choice and helmet usage behavior were also estimated. The findings showed that those individuals who took beginning rider training courses were more likely to be involved in an accident than those who did not and that those who took the beginning course more than once were much more likely to be involved in an accident. Although explanations for these findings can range from the use of ineffective course material to changes in risk perception as a result of taking the course, another explanation is that riders who take the course are inherently less skilled than those who do not. The findings underscore the need for a careful and comprehensive study of rider skills and risk perceptions to maximize the effectiveness of motorcycle training courses.

Optical bioassay for measuring sublethal toxicity of insecticides in Daphnia pulex

Many emerging contaminants tend to be biologically active at very low concentrations, occur in water as part of complex mixtures, and impact biota in ways that are not detected using traditional toxicity tests (e.g., median lethal concentration). To evaluate emerging contaminants, the authors developed a method for detecting sublethal behavioral effects by quantifying the swimming behavior of Daphnia pulex, a model organism for studying aquatic toxicity. This optical tracking technique is capable of measuring many swimming parameters, 2 of which-cumulative distance and angular change-are presented. To validate this technique, 2 prototypical compounds that exhibit different modes of action as well as corresponding insecticides that are commonly found in surface waters were investigated. The acetylcholinesterase (AChE) inhibitor physostigmine was used as the prototypical compound for the large number of AChE inhibitor insecticides (e.g., chlorpyrifos). Nicotine was used as the prototypical compound for neonicotinoid insecticides (e.g., imidacloprid). Results demonstrate that this assay is capable of detecting sublethal behavioral effects that are concentration-dependent and that insecticides with the same mode of action yield similar results. The method can easily be scaled up to serve as a high-throughput screening tool to detect sublethal toxic effects of a variety of chemicals. This method is likely to aid in enhancing the current understanding of emerging contaminants and to serve as a novel water-quality screening tool.

Comparison of Severity of Motorcyclist Injury by Crash Types

The purpose of this study was to examine statewide motorcycle crash data from the state of Ohio to identify those factors most strongly associated with the level of injury sustained by motorcyclists involved in crashes. The year 2009 marked the first time since 1997 that Ohio experienced fewer motorcycle fatalities than in the previous year; this result was consistent with national trends over this same period. Since anecdotal evidence suggested that much of this result was due to decreases in motorcycle travel, an investigation of rider, crash, and environmental factors that influenced injury severity was an important step toward sustaining this reduction. The effects of such factors in various types of motorcycle crashes were compared. Multinomial logit models were developed for single-vehicle and multivehicle motorcycle crashes at both intersection and nonintersection locations. The results showed that the impacts of relevant crash factors varied by crash type and location and that severe injuries were more likely when high speed or alcohol was involved. Collisions with fixed objects and high-impact crashes, such as angle and head-on collisions, also produced dramatic increases in injury severity. Helmet use continued to provide the most promising means of reducing severe and fatal injuries in all types of crashes. Helmet use was found to reduce the likelihood of injury by 34.4% in single-vehicle crashes and between 12.4% and 14.1% in multivehicle crashes.

A comparison of contributing factors between alcohol related single vehicle motorcycle and car crashes

INTRODUCTION Alcohol related crashes have accounted for approximately 35% of fatal crashes per year since 1994 nationwide, with approximately 30% involving impairment over the legal blood alcohol content limit of 0.08%. Educational campaigns and law enforcement efforts are two components of multi-faceted programs aimed toward reducing impaired driving. It is crucial that further research be conducted to guide the implementation of enforcement and educational programs. METHOD This research attempts to provide such guidance by examining differences in alcohol-involved crashes involving motorcycles and passenger cars. Prior safety research has shown that motorcyclists follow a significantly different culture than the average passenger car operator. These cultural differences may be reflected by differences in the contributing factors affecting crashes and the severity of the resulting injuries sustained by the driver or motorcyclist. This research is focused on single-vehicle crashes only, in order to isolate modal effects from the contribution of additional vehicles. The crash data provided for this study are from the Ohio Department of Public Safety from 2009 through 2012. RESULTS The injury severity data are analysed through the development of two mixed logit models, one for motorcyclists and one for passenger car drivers. The models quantify the effects of various factors, including horizontal curves, speeds, seatbelt use, and helmet use, which indicate that the required motor skills and balance needed for proper motorcycle operation compounded with a lack of mechanical protection make motorcyclists more prone to severe injuries, particularly on curves and in collisions with roadside objects. PRACTICAL APPLICATIONS The findings of this study have been incorporated into combined motorcycle and sober driving educational safety campaigns. The results have shown to be favorable in supporting national campaign messages with local justification and backing.