In road safety, a commonly-used measure of treatment effectiveness is the crash modification factor, usually defined as a ratio of the expected crash frequency with the treatment to the expected frequency without. This paper explores the possibility of using surrogates to estimate crash modification factors. As in other situations where observational data are used to estimate causal effects, it is necessary to leverage background causal knowledge with the observational results. When the background knowledge is such that a crash-generating mechanism can be represented with a directed acyclic graph, the connectivity structure of the graph can be used to identify candidate surrogates. The modification factor associated with a safety-related improvement can then, in principle, be estimated from knowledge of how the improvement affects the surrogates, together with information on how the surrogates are distributed in the population of crashes. After developing this relationship between surrogates and crash modification factors, its potential usefulness is illustrated with two simulation studies, where estimates of CMFs using surrogates are compared to estimates computed from crash frequencies.
Bibliographical noteFunding Information:
This research was supported in part by the Roadway Safety Institute at the University of Minnesota .
Copyright 2021 Elsevier B.V., All rights reserved.
- Crash modification factor
- Graphical model
- Offset left turn lane
- Pedestrian hybrid beacon
- Surrogate outcome
PubMed: MeSH publication types
- Journal Article