The Estimated Benefit of Lights and Sirens During Ambulance Transport

The use of lights and sirens triples the odds of ambulance collisions, which are most common during patient transport. The time benefit of lights and sirens is reportedly small, though many studies were limited by small sample sizes and potential methodologic biases. The objective of this study was to estimate the effect of lights and sirens on trip duration and trip pace using a modern transportation model.

We conducted an observational study of actual trip durations versus predicted trip durations using one year of dispatch data for a mid-sized emergency medical system in the United States. We obtained predictions of non-emergent trip duration between each incident location and destination hospital using Google's transportation model. To validate the model, we fit a linear regression model with actual non-emergent trip durations versus the model's predictions. Next, we subtracted actual emergent trip durations from predictions of their non-emergent duration to estimate the effect of lights and sirens on trip duration (min) and trip pace (min/km). We used the Mann-Whitney U test to assess differences in trip duration and trip pace between emergent and non-emergent groups and reported the median differences with bootstrapped confidence intervals. We repeated our analysis using a spatially aggregated approach with the median trip duration and trip pace per 1 km² area to minimize bias from high incident densities.

We analyzed 25,902 incidents, of which 21.0% were transported using lights and sirens. The Google model fit non-emergent study data well (R² = 0.78, F-statistic 70,130). Lights and sirens were associated with a shorter trip duration and faster trip pace of 3.0 min (95%CI, 2.9-3.1) and 0.3 min/km (95%CI, 0.3-0.3), respectively. In the spatially-aggregated analysis, lights and sirens were associated with a shorter trip duration and faster trip pace of 3.3 min (95%CI, 2.9-3.6) and 0.3 min/km (95%CI, 0.2-0.3), respectively.

Using a modern transportation model, we estimated the median benefit of lights and sirens to be 18 s/km, or 3.0-3.3 min per trip in our system. Considering the well-documented risks of emergency driving, additional work is needed to identify the conditions when this benefit outweighs risk.