Assessing Active-Mode Fatality Disproportionality in U.S. Cities: A Comparison of Fatality and Commute Shares

The fatality risk of pedestrians and bicyclists is often evaluated using exposure-adjusted measures, such as fatalities per population, commuter, trip, or distance traveled. While these measures are essential for estimating risk, they do not directly describe how fatal outcomes are distributed across travel modes relative to a consistent city-level benchmark. This study develops and applies a commute-share-based Active-Mode Fatality Burden framework as a descriptive monitoring tool for pedestrian and bicyclist fatality-share disproportionality. Using a longitudinal city-level panel of 24 U.S. cities from 2010 to 2023, the study compares active fatality share, defined as the share of total traffic fatalities involving pedestrians and bicyclists, with active commute share, defined as the combined share of workers commuting by walking and bicycling. The resulting Active-Mode Fatality Burden Ratio is explicitly interpreted as disproportionality relative to a commute-share benchmark; it is not a per-trip, per-mile, or otherwise exposure-adjusted risk ratio. The results show that the observed active fatality share exceeded the commute-share benchmark throughout the study period. The city-year mean burden ratio increased from 6.05 in 2010 to 8.07 in 2023, while active fatality share rose from 33.89% to 39.83% and active commute share changed from 6.96% to 7.75%. During 2021–2023, all 24 observed city ratios were above 1, although these city-specific contrasts are descriptive rather than formal hypothesis tests against parity. Re-estimating the city fixed-effects trend after excluding the pandemic-disrupted years 2020–2021 retained a positive year coefficient (β = 0.128, p = 0.002; N = 288), and a city-year log-transformed fixed-effects specification also remained positive (β = 0.012, p = 0.010; N = 288). City-level typologies further identified contexts with low active commute shares but high fatality-share disproportionality. Because commute data omit non-work active travel and ratio estimates can be sensitive to small denominators, the framework should be used alongside trip-based exposure, serious-injury, and conventional risk measures. The proposed indicator is therefore intended for comparative monitoring and peer-learning within Safe System and equity-oriented safety planning, rather than for causal inference or individual-risk estimation.