A comparative analysis of accident modification functions for traffic law enforcement.

Traffic law enforcement is a road safety measure whose effects on accidents or injuries is best described by means of a function rather than a point estimate. An informative function should comprise both increases and decreases in enforcement. Currently available accident modification functions cannot serve this need. A fruitful approach to developing accident modification functions covering both increases and decreases in enforcement is differences-in-differences estimates based on multivariate accident prediction models. The paper explains how to develop such estimates and illustrates them. The interpretation of the results of empirical studies can be informed by a game-theoretic model of the effects of enforcement, previously published in Accident Analysis and Prevention (Bjørnskau and Elvik 1992, 507-520).

Risk of road traffic injury in Norway 1970-2022.

This paper describes changes in the risk of road traffic injury in Norway during the period from 1970 to 2022. During this period, the risk of fatal and personal injury declined by more than 70 % for most groups of road users. There are five main potential explanations of a decline in the risk of injury: (1) a reduced probability of accidents that have the potential for causing injury; (2) an improved protection against injury given that an accident has occurred; (3) improved medical care increasing the survival rate, given an injury (this would reduce the number of fatalities, but not the number of injuries); (4) a tendency for the reporting of injuries in official accident statistics to decline over time; (5) uncertain or erroneous estimates of the exposure to the risk of injury. The decline in the risk of road traffic injuries in Norway after 1970 can probably be attributed to a combination of reduced reporting of injuries in official statistics, improved protection against injury in accidents, and (for fatal injuries) improved medical care. Insurance data, available from 1992, do not indicate a reduction in the risk of accidents leading to insurance claims. Incomplete and possibly erroneous data for mopeds and motorcycles make it impossible to identify sources of changes in injury risk over time for these modes of transport.

The development of a road safety policy index and its application in evaluating the effects of road safety policy.

The paper presents an exploratory study of a road safety policy index developed for Norway. The index consists of ten road safety measures for which data on their use from 1980 to 2021 are available. The ten measures were combined into an index which had an initial value of 50 in 1980 and increased to a value of 185 in 2021. To assess the application of the index in evaluating the effects of road safety policy, negative binomial regression models and multivariate time series models were developed for traffic fatalities, fatalities and serious injuries, and all injuries. The coefficient for the policy index was negative, indicating the road safety policy has contributed to reducing the number of fatalities and injuries. The size of this contribution can be estimated by means of at least three estimators that do not always produce identical values. There is little doubt about the sign of the relationship: a stronger road safety policy (as indicated by index values) is associated with a larger decline in fatalities and injuries. A precise quantification is, however, not possible. Different estimators of effect, all of which can be regarded as plausible, yield different results.

Risk factors as causes of accidents: Criterion of causality, logical structure of relationship to accidents and completeness of explanations.

The causes of accidents are studied in the belief that by finding causes, accidents can be prevented by removing or controlling their causes. It follows that the risk factors that have traditionally been regarded as contributing to accidents can only be regarded as causes if it is possible to alter them by means of one or more road safety measures. Risk factors are causes if their relationship to accidents can be changed by implementing one or more road safety measures influencing the risk factors. Hence, road safety measures that could have been implemented to change risk factors identified as contributing to an accident, but have not, are also causes of accidents. Many of the human factors that have traditionally been identified as risk factors for accidents, like age, gender, driving experience, expectations or involuntary inattention are not causes of accidents, because they cannot be changed by means of any realistic road safety measure. What cannot be changed (could not have been different) is not a cause. It is possible, both in case studies and in statistical analyses, to determine when a set of factors precipitating or contributing to accidents is complete. A list of road safety measures that could have been implemented is only limited by our creativity and imagination and will therefore never be complete.

Safety effects of bypass lanes at unsignalized three-leg junctions.

Bypass lanes are a low-cost measure to increase capacity at unsignalized T-junctions without left-turn lanes that allow through-traffic to pass left-turning vehicles on the right. There is very limited knowledge about the safety effects of bypass lanes. We found six previous studies that could be summarized by means of meta-analysis, and the results show an average accident reduction of 10 percent. However, the results from previous studies are inconsistent and may be biased. Therefore, the present study has estimated safety effects of by-pass lanes in Norway, based on a sample of 2,227 T-junctions (incl. 94 with bypass-lanes) for which relevant data was available for a period of up to 10 years. We developed accident prediction models and conducted before-after analyses. The accident prediction models show that junctions with bypass lanes have 82 percent more accidents than junctions without bypass lanes, when controlling for endogeneity. Endogeneity occurs when the implementation of a measure is conditional on the frequency of crashes, as has been the case with bypass lanes. The before-after analysis shows that average accident numbers decrease after the installation of bypass lanes. However, when controlling for regression-to-the-mean (RTM), average accident numbers increase. RTM means that accident numbers would have been likely to decrease even without any measure because they had been exceptionally high in the before period. The control for potential biases in our study is likely to contribute to the discrepancy between results from our study and previous studies, most of which have not controlled for the same potential biases. We conclude therefore that bypass lanes, although favorable for capacity, are likely to be unfavorable for safety when compared to other unsignalized T-junctions without left-turn lanes. Unfavorable safety effects may partly be due to site specific conditions, such as road alignment and sight conditions, that contribute to rear-end collision risk or inappropriate driver behavior. However, this does not necessarily mean that bypass lanes never should be used. For example, at junctions where a bypass lane may solve capacity problems, and where site-specific conditions are favorable, bypass lanes may still be an acceptable solution.

Driver mileage and accident involvement: A synthesis of evidence.

The relationship between driver mileage and accident involvement has been a controversial topic for at least 20 years. The key issue is whether driver accident involvement rate increases in proportion to miles driven or has a non-linear relationship to miles driven. This paper presents a synthesis of evidence from studies of how the number of accidents per driver per unit of time relates to distance driven in the same period. Most studies of this relationship are methodologically weak and their results highly inconsistent and potentially misleading. Unreliable data and poor control for confounding factors characterise most studies. Only a few studies based on multivariate statistical models control for at least some of the confounding factors that may influence the relationship between distance driven and accident involvement. These studies consistently show that the number of accidents per driver per year increases less than in proportion to distance driven. A good approximation is that the number of accidents per driver per unit of time is proportional to the square root of distance driven. Potential methodological and substantive explanations of this finding are discussed.

Effects on accidents of technical inspections of heavy goods vehicles in Norway: A re-analysis and a replication.

INTRODUCTION: This paper presents a re-analysis of a previous study of the effects on accidents of technical inspections of heavy vehicles in Norway and a replication of the study using more recent data. METHOD: Increasing the number of technical inspections is associated with a reduction in the number of accidents. Reducing the number of inspections is associated with an increase in the number of accidents. The relationship between changes in the number of inspections and changes in the number of accidents is well described by means of logarithmic dose-response curves. RESULTS: These curves show that inspections had a larger effect on accidents in the recent period (2008-2020) than in the first period (1985-1997). Based on recent data, a 20% increase in the number of inspections is associated with a 4-6% reduction in the number of accidents. A 20% reduction of the number of inspections is associated with a 5-8% increase in the number of accidents.

What would a road safety policy fully consistent with safe system principles mean for road safety?

If fully implemented, the Safe System principles as formulated by the International Transport Forum would lead to a considerably safer road transport system. The aims of this paper are: (1) To define operationally what full consistency with Safe System principles means; (2) To estimate the potential effects on traffic fatalities of full compliance with the Safe System principles. Operational definitions of full consistency with Safe System principles are proposed for speed limits, road design, road maintenance, vehicle safety and road user compliance with road traffic law. Estimates for Norway indicate that by complying perfectly with Safe System principles in all these areas, the number of fatalities could be reduced by 50-70 %. This is a conservative estimate. This shows that the Safe System principles are well justified scientifically: adhering to them would greatly improve road safety. However, currently road safety policy in many countries, including Norway, fails to realise these improvements in safety by not complying with the Safe System principles.

Further compelling evidence for safety-in-numbers: It is more than meets the eye.

In the extant road safety literature, estimating safety-in-numbers is dominated by conventional cross-sectional methods in which active mode (pedestrian or cyclist) volume together with motorised traffic volume are present in regression models explaining active mode safety directly. There is "direct" evidence for safety-in-numbers when the coefficient associated with active mode volume is negative (safety improves as volume increases) or when it is smaller than one (safety decreases at a lower rate compared to the rate of increase in active mode volume). In this article we extend the concept of safety-in-numbers in the traffic safety field, introducing "indirect" safety-in-numbers, which constitutes a new form of evidence for this phenomenon. We provide empirical evidence to support this, discussing that using an approach based on heterogeneity in mean modelling-a form of random parameters (slopes) models-it is possible to reveal "indirect" safety-in-numbers effects. Therefore, such models can reveal further compelling evidence for safety-in-numbers. Accurate knowledge of safety-in-numbers effects (both direct and indirect) and their underlying mechanisms can help provide robust motives for promoting active travel and will have valuable implications for the design of road safety interventions.

Changes over time in the relationship between road accidents and factors influencing them: The case of Norway.

This paper compares the results found in successive accident prediction models developed at the national level for Norway. Over time, the models have become more comprehensive in terms of the roads and the variables included in them. It is found that traffic volume has consistently had the strongest association with the number of accidents. It explains nearly all the systematic variation in the number of accidents. The second most important variable has consistently been the speed limit of 50 km/h, which indicates an urban area (the default speed limit in urban areas in Norway is 50 km/h). This variable has become less important over time. Motorways (freeways) have consistently had a lower accident rate than other roads. The mean number of accidents per road section declined considerably from 1986 to 89 to 2010-15. Systematic variation in the number of accidents between road sections was greatly reduced. At present, the variation in the annual number of accidents between road sections is mostly random.

Does empirical evidence support the effectiveness of the Safe System approach to road safety management?

The objective of this paper is to evaluate the effectiveness of the Safe System approach to road safety management, as implemented in Norway. The paper proposes simple operational definitions of key elements of the Safe System approach to road safety management. The relationship between these elements and changes over time in the number of killed or seriously injured road users in Norway is studied by means of negative binomial regression models. These models do not support a causal interpretation of the findings, but predict systematic patterns in findings that, if replicated in other data sets, at least make a causal interpretation plausible, although not incontestable. The findings reported in this paper are broadly consistent with theoretical predictions and therefore support the effectiveness of the Safe System approach. It is highly likely that the adoption of the Safe System approach to road safety management in Norway has contributed to a larger improvement in road safety than would otherwise have occurred.

Speed limits, enforcement, and health consequences.

This review summarizes current knowledge regarding the effects of speed limit enforcement on public health. Speed limits are commonly used around the world to regulate the maximum speed at which motor vehicles can be operated on public roads. Speed limits are statutory, and violations of them are normally sanctioned by means of fixed penalties (traffic tickets) or, in the event of serious violations, suspension of the driver's license and imposition of prison sentences. Speed limit violations are widespread in all countries for which statistics can be found. Speeding contributes more to the risk of traffic injury than do other risk factors for which estimates of population-attributable risk are available. Traffic speed strongly influences impact speed in crashes and therefore has major implications for public health.

Why are there so few experimental road safety evaluation studies: Could their findings explain it?

Randomised controlled trials (also known as experiments) are widely regarded as the best design of studies that aim to estimate the effects of a treatment, like a road safety measures. However, as noted by Hauer (2016) there are few randomised controlled trials in the field of road safety. Hauer has suggested (2019) that not finding an effect of the measure being evaluated could be one reason for this. This paper provides an inventory of experimental evaluations of road safety measures. 24 different road safety measures have been evaluated experimentally. More than one experiment was reported for 8 road safety measures; for the other 16 only a single experiment was found. With few exceptions, experiments find no effect on accidents of the measures subjected to experiments. Most of the exceptions are due either to failure of randomisation or refer to measures whose effects tend to erode as they become more commonly used. Thus, the effects found in initial experiments with daytime running lights and high-mounted stop lamps have gradually eroded as more cars got these systems. Driver training is one of the few road safety measures for which more than one experiment has been carried out. Results consistently show no effect on accidents of driver training.

Democracy, governance, and road safety.

Countries scoring high on the Democracy Index developed by The Economist Intelligence Unit have fewer traffic fatalities per 100,000 inhabitants than countries scoring low on this index. The statistical relationship between democracy score and fatalities per 100,000 inhabitants is statistically highly significant and robust with respect to control for potentially confounding factors. A similar relationship exists between democracy score and the number of traffic fatalities per 100,000 motor vehicles. The statistical relationship between level of democracy and level of road safety is strong, although the analyses reported in this paper do not justify a causal interpretation of the relationship. Changes over time in government effectiveness (one of the indicators of the World Governance Index developed by the World Bank) are weakly associated with changes in road safety performance.

How can regulatory authorities improve safety in organizations by influencing safety culture? A conceptual model of the relationships and a discussion of implications.

Regulators have increasingly started to focus on safety culture. The causal link between regulatory initiatives to improve safety culture and a potential decline in accidents may, however, appear like a "black box", involving social processes that seem hard to foresee and influence. We need a better conceptual understanding of this. The aims of our study are to: 1) Map studies of regulatory efforts to influence safety culture in companies, 2) Use the identified studies to develop a conceptual model of the analytical relationships between regulatory initiatives to improve safety culture and accidents in these studies, including the factors influencing these analytical relationships and 3) discuss practical implications. The review is reported according to PRISMA-guidelines, and focuses on professional transport (aviation, sea, rail, road) and the Norwegian petroleum sector. Our review indicates at least six analytical relationships, mediating between regulatory attempts to influence organizational safety culture and accidents. These are between: 1) Rules and regulators, 2) Regulators and companies, 3) Managers and employees in the companies, 4) Organizational members' shared ways of thinking and acting, which are the two key elements of safety culture, 5) Safety culture and safety behaviour, and between 6) Safety behaviour and accidents. Regulatory attempts to influence safety culture may fail or succeed at each level, through factors involved in the different relationships.

First steps towards a holistic impact assessment methodology for connected and automated vehicles.

Connected and automated vehicles have become more common in recent years, increasing the need to assess their societal level impacts. In this paper a methodology is presented to explore and define relevant impact areas as a starting point for quantitative impact assessment. The many interrelations between impact areas increases the complexity of obtaining a complete overview. Therefore, a structured approach is used, which shows many similarities with the modelling of causal-loop-diagrams. Feedback loops between impact areas are taken into account at an early stage and methods of literature research, project team feedback, interrelation assessment and grouping are used to produce a holistic overview of impacts. The methodology was developed and applied in the European H2020 project LEVITATE. The impact taxonomy and interrelations between impact areas resulting from this project are presented and further steps needed to perform a quantitative evaluation of the impacts are discussed.

Can the use of road safety measures on national roads in Norway be interpreted as an informal application of the ALARP principle?

The ALARP principle, stating that risks should be reduced to a level "As Low As Reasonably Practicable", is widely known and discussed in risk management. The principle is flexible, as the interpretation of the key concepts of reasonable and practicable can be adapted to different contexts. This paper discusses whether the use of road safety measures on national roads in Norway can be interpreted as an informal application of the ALARP-principle. According to official guidelines, priority setting for major road investments should be based on cost-benefit analysis. Most road safety measures are low-cost projects that have traditionally not been subject to cost-benefit analysis. A use of these measures regarded as reasonable in the ALARP sense may include considerations of cost, efficiency and fair distribution. Data on 328 road safety measures implemented around 2000 is used to evaluate factors influencing their use. It is argued that the use of these measures is consistent with an informal application of the ALARP-principle.

Crash risk on entrance versus exit zones of road bridges in Norway.

Using data from the national register of police-reported crashes and from the bridge register of the Norwegian Public Roads Administration, we estimated rates of single-vehicle crashes on road sections adjacent to road bridges and on different sections of the bridges. Data included all single-vehicle personal injury crashes occurring on or close to road bridges in Norway between 2010 and 2016, a total of 219 crashes. All bridges on state and county roads were included. Crash rate was found to be highest in the approach zone of short bridges (last 50 m before bridge) and lowest in the middle of long bridges. On bridges shorter than about 100 m, crash rate was higher in the first than in the last bridge zone. Total crash rate on bridges was close to the figure for the total road network. However, for the approach to short bridges, crash rate was significantly higher than for the total road network, and for the middle part of long bridges it was significantly lower. A supplementary analysis of in-depth data from 31 fatal crashes including both single-vehicle and multiple-vehicle crashes supported the results from the main analysis. A higher proportion of fatal crashes occurred on approaching or entering a bridge than when leaving the bridge, as seen from the direction of travel of the at-fault vehicle. Concerning countermeasures against bridge accidents, particular attention should be payed to the approach zone and to the design of barriers.

How to trade safety against cost, time and other impacts of road safety measures.

Public policy, including road safety policy, involves balancing competing values against each other. Several techniques of policy analysis, most prominently cost-benefit analysis, have been developed to help policy makers prioritize between different values. Valuation studies have not produced credible monetary values of life and limb. Cost-benefit analysis therefore cannot tell when the "right" balance has been struck between road safety and other objectives of transport policy. All formal tools of policy analysis are likely to reflect analyst values to a major extent, not the values of policy makers only. It is argued that policy choices and tradeoffs can be informed simply by providing factual information about impacts and not attempting to impose any value judgements. A widely applicable metric is to state impacts as changes in human longevity and health state.

A comprehensive and unified framework for analysing the effects on injuries of measures influencing speed.

This paper proposes a comprehensive and unified framework for analysing the impacts on traffic injury of measures influencing speed. The key tool for analysis is a specification of the speed distribution, which in most cases closely approximates a standard normal distribution. The speed distribution can be represented, for example, by twelve intervals each comprising one half standard deviation. The exponential model of the relationship between speed and the number of injured road users is applied to estimate the expected injury rate for drivers travelling at the mean speed of any part of the distribution. The relationship between individual driver speed and accident involvement is then incorporated into the speed distribution. A speed distribution specified this way represents both the mean speed of traffic and the variation in speed-related risk between drivers. Impacts of changes in speed that can be modelled include: (1) Shifting the whole speed distribution, (2) Compressing the upper end of the speed distribution, (3) Enlarging or reducing the variance of the speed distribution, (4) Selective changes in specific regions of the speed distribution. Examples are given of how knowledge of the impacts of measures on speed can be translated into expected changes in the number of injured road users by relying on the analytic framework.