Exploring the effect of COVID-19 on driver injury severities in freeway single-vehicle crashes accounting for unobserved heterogeneity

ABSTRACT

The COVID-19 pandemic, characterized by travel restrictions and reduced traffic volumes, heightened the risk of severe single-vehicle crashes on Florida's freeways. This study utilized random parameter multinomial logit models, accounting for heterogeneity in means and variances, to analyze driver injury severities in 2020 and compare variations in the magnitude of factors contributed to these injuries across different freeway systems. The estimated models identified 31 statistically significant variables across Florida's major freeways (I-4, I-10, I-75, and I-95). Among these variables, only two—normal driving and restraint usage—were statistically significant across all four freeway systems. Moreover, the model results indicated that factors contributing to severe driver injuries were most prominent on I-95 compared to the other freeways in 2020. These findings improve our understanding of freeway safety measures during a pandemic and provide valuable insights to enhance traffic management strategies for state highway agencies, prioritizing operational safety in potential future pandemics.

KEYWORDS:

• Single-vehicle crash
• driver injury severity
• COVID-19
• freeways
• random parameter multinomial logit models with heterogeneity in means and variances

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The focus on single-vehicle crashes is intended to specifically examine the driver errors without introducing the complexities associated with analyzing the responses of other drivers involved in multi-vehicle crashes.

2. The importance of accounting for unobserved heterogeneity in crash data modeling has been extensively discussed and justified in a study by Mannering, Shankar, and Bhat (2016). Ignoring unobserved heterogeneity and assuming that the effects of observable variables are the same for all observations can lead to a mis-specified model. As a result, the estimated parameters may be biased and inefficient, potentially leading to incorrect inferences and predictions. It is crucial to consider this paradigm shift from traditional models and acknowledge the need to account for unobserved heterogeneity.

3. This could be attributed to a lower level of deployment of law enforcement personnel and safety protocols for issuing traffic violation tickets. This observation is based on reviewed data from the Florida traffic violations of the Signal Four Analytics.

4. It is meaningful to present the mean and standard deviation of these indicator variables for each of the freeway systems since these variables were used in the statistically significant model estimations.

5. The modeling process involved the consideration of numerous variables from the combined dataset. However, it is beyond the scope of this paper to present descriptive statistics of all these variables. As such, only key variables that were determined to be statistically significant across these four freeway systems are included in this paper.

6. Possible Injury is any injury reported or claimed that is not a fatal injury, suspected serious injury or suspected minor injury. Examples include momentary loss of consciousness, claim of injury limping, complaint of pain or nausea. Possible injuries are those which are reported by the person or are indicated by his/her behavior, but no wounds or injuries are readily evident. Non-incapacitating injuries are non-disabling injuries, such as lacerations, scrapes, bruises, etc.

7. Incapacitating injuries are disabling injuries, such as broken bones, severed limbs, etc. These injuries usually require hospitalization and transport to medical facility.

Fatal Injury results in death within 30 days after the motor vehicle crash in which the injury occurred. If the person did not die at the scene but died within 30 days of the motor vehicle crash in which the injury occurred, the injury classification should be changed from the previously assigned to Fatal Injury.

8. The focus of this study is on injury-severity analysis considering the details of driver characteristics in Florida crash data accounting for unobserved heterogeneity. As such, discrete outcome models capturing the injury severity that account for unobserved heterogeneity were found to be more effective and promising in identifying the contributing factors leading to crashes compared to frequency models.