Measuring the impact of the COVID-19 pandemic on road safety outcomes in Victoria, Australia (MUARC Report 366)

This project aimed to understand the impact that the COVID-19 pandemic has had on travel exposure and risk-taking behaviour and the subsequent impact of these factors on observed changes in serious road trauma in Victoria. This has been particularly important during 2020 and 2021 but is also relevant for understanding the impact of any future major events. Through the interrogation of available data, the project aimed to develop quantitative measures of the impact of both exposure and crash risk per unit exposure. A consideration of these two aspects in tandem allowed interpretation of the observed deviation in road trauma from projected long-term trends. Notably, it was found that reductions in exposure did not directly translate into reductions in road trauma. Rather it was important to take into account increases in risk whilst also accounting for the effect of changes to countermeasure delivery. 

One way to assess and monitor the effect of risk-taking behaviours on road trauma is through the definition and calculation of intermediate outcome measures. An intermediate outcome measure is one that is used in addition to crashes or injuries to measure changes in road safety outcomes. There should be a causal relationship between an intermediate outcome measure and final outcomes (crashes or injuries). The current project risk-weighted excessive speed data from the Fixed Digital Road Safety Camera (FDRSC) network to create an indicator of risk (intermediate outcome measure) linked to road trauma. Such a measure is useful for road safety strategy monitoring generally. It is also particularly useful when shocks such as the COVID-19 pandemic occur that can lead to unexpected changes in road trauma with very little lead time to consider their effect on road safety strategy targets. The calculation of an intermediate outcome measure based on FDRSC data provided a quantitative measure of the increase in risk on the road network at the same time as there was a drop in travel exposure. Consideration of both risk and exposure was necessary to allow proper interpretation of the observed deviation in road trauma rather than simply expecting a reduction in road trauma proportionate with a drop in travel. The same approach was taken to define intermediate outcome measures based on mobile speed camera data. Other areas that involve risk-taking behaviours, namely drink- and drug-driving, could also be used as the basis for defining and calculating intermediate outcome measures to measure risk for road safety strategy monitoring.

Overall, it was found that increases in excessive speeding, and drink- and drug-driving added to the evidence that suggests such risk-taking behaviours have contributed to higher levels of road trauma than might have been expected based on changes to exposure alone. In terms of the impact of countermeasure delivery, it was found that reduced roadside alcohol testing during the pandemic resulted in negative fatality savings, i.e. an increase in forecast fatalities. No differences were predicted in fatalities as a result of changes to roadside drug testing. At the same time, increases in the mobile speed camera program through increased enforcement hours has translated into crash savings or a decrease in forecast fatalities. Together with increased infrastructure investment, which also resulted in decreased forecast fatalities, net road trauma savings were predicted for 2020 and 2021 due to changes in countermeasure delivery. These findings emphasise the importance of maintaining enforcement activities and countermeasure delivery, especially during major shocks such as the COVID-19 pandemic, if road safety targets are to be met. The project also identified how enhancements to future Victorian road safety strategy baseline models through the incorporation of learnings from this project around both exposure and risk would allow future road trauma outcomes, both fatalities and serious injuries, to be better predicted.