A 2018 JAMA Pediatrics study led by UBC found pedestrian deaths were 43% higher on Halloween evenings, and 10 times higher for children aged 4 to 8.
In 42 years of US crash data, pedestrians aged 4 to 8 were 10 times more likely to be killed on Halloween evening than on the same evening a week earlier or later.
That is from a research letter in JAMA Pediatrics, published in 2018 and led by John Staples of the University of British Columbia in Vancouver, with Candace Yip and Donald Redelmeier.
What the team measured
The researchers used the US National Highway Traffic Safety Administration's Fatality Analysis Reporting System for 1975 to 2016. They counted pedestrian deaths between 5 p.m. and 11:59 p.m. on October 31 each year, and compared them with the same hours one week earlier (October 24) and one week later (November 7).
The numbers
- 608 pedestrians died on the 42 Halloween evenings, against 851 on the 84 control evenings.
- That is about 2.07 deaths an hour on Halloween evening, against 1.45 on a control evening.
- The risk was 43% higher on Halloween. The average Halloween brought 4 extra pedestrian deaths.
- For pedestrians aged 4 to 8, the risk was 10 times higher.
- Risk peaked around 6 p.m.
What did not change
Driver demographics, where the crash happened, vehicle size, what the driver was doing just before, and police-reported alcohol or speeding did not differ between Halloween and the control evenings, the authors report. Adjusting for the end date of daylight saving time gave similar results.
What the study can't tell you
This is US data, not Canadian. The absolute risk is small: it fell from 4.9 to 2.5 per 100 million Americans between the first and last decades of the study, even as the relative jump on Halloween stayed about the same.
The authors do not call for an end to trick-or-treating. They point to slower, calmer residential streets, fewer parked cars blocking sight lines, and reflective patches on clothing, and say those fixes matter all year, not just one night.
