Forest plot showing the relative risk of pedestrian fatality between 5 pm and 11:59 pm on Halloween compared with the same time interval on control evenings exactly 1 week earlier and 1 week later. Solid squares indicate point estimate; relative dimensions, sample size; horizontal lines, 95% CIs.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Staples JA, Yip C, Redelmeier DA. Pedestrian Fatalities Associated With Halloween in the United States. JAMA Pediatr. 2019;173(1):101–103. doi:10.1001/jamapediatrics.2018.4052
On October 31 each year, millions of children in the United States celebrate Halloween by walking door to door to collect candy from neighbors, while adults and adolescents engage in Halloween festivities. The holiday may heighten pedestrian traffic risk, because celebrations occur at dusk, masks restrict peripheral vision, costumes limit visibility, street-crossing safety is neglected, and some partygoers are impaired by alcohol.1 Mitigating factors include broad public awareness of Halloween, widespread parental supervision of younger children, and the potential for improved safety as pedestrian numbers increase. Prior studies of Halloween traffic risks have been limited to brief observations, failed to test for statistical significance, or lacked appropriate control groups.2,3 We therefore examined 4 decades of national data to systematically evaluate pedestrian fatality risks on Halloween and highlight opportunities for year-round injury prevention.
Data were obtained from the National Highway Traffic Safety Administration's Fatality Analysis Reporting System. The 42-year study interval extended from the first year that data were collected to the most recent year for which data were available (1975 to 2016). The primary analysis compared the number of pedestrian fatalities that occurred between 5 pm and 11:59 pm on October 31 each year with the number that occurred during the same hours on control evenings 1 week earlier (October 24) and 1 week later (November 7). As in prior research, exact binomial tests were used to examine the significance of deviations from the expected ratio of 1:2.4 Hourly variation in risk was evaluated by stratifying pedestrian fatalities by clock time.
This study received a waiver of approval from the University of British Columbia research ethics board. This included a waiver of the requirement for informed consent.
Statistical analyses used 2-sided tests and statistical significance was inferred from P values less than 0.05. Analyses were performed using R version 3.4 (R Foundation for Statistical Computing). Data analysis occurred from February to September 2018.
The entire study interval included 1 580 608 fatal traffic crashes involving 2 333 302 drivers and 268 468 pedestrians. A total of 608 pedestrian fatalities occurred on the 42 Halloween evenings, whereas 851 pedestrian fatalities occurred on the 84 control evenings. Absolute mortality rates averaged 2.07 and 1.45 pedestrian fatalities per hour, respectively. The relative risk of a pedestrian fatality was 43% higher on Halloween compared with control evenings (odds ratio, 1.43 [95% CI, 1.29-1.59]; P < .001). The average Halloween resulted in 4 additional pedestrian deaths.
Subgroup analysis revealed the highest relative risk increase was among children, with pedestrians aged 4 to 8 years exhibiting a 10-fold increase in pedestrian fatality risk on Halloween (odds ratio, 10.00 [95% CI, 5.23-19.11]; P < .001; Figure 1). Relative risks remained stable throughout the study interval, but the absolute risk per 100 million Americans was small and declined from 4.9 to 2.5 between the first and final study decades. Risks were highest around 6 pm (Figure 2). Sensitivity analyses adjusting for annual variability in the end date of Daylight Saving Time yielded similar results. Driver demographics, crash location, vehicle size, precrash maneuver, and police-reported alcohol involvement or excess speed did not differ between Halloween and control days.
Halloween traffic fatalities are a tragic annual reminder of routine gaps in traffic safety. On Halloween and throughout the year, most childhood pedestrian deaths occur within residential neighborhoods.5 Such events highlight deficiencies of the built environment (eg, lack of sidewalks, unsafe street crossings), shortcomings in public policy (eg, insufficient space for play), and failures in traffic control (eg, excessive speed).
Event-specific interventions that may prevent Halloween child pedestrian fatalities include traffic calming and automated speed enforcement in residential neighborhoods. Pedestrian visibility could be improved by limiting on-street parking and incorporating reflective patches into clothing. But restricting these interventions to 1 night per year misses the point, since year-round application of effective traffic safety interventions will foster much greater progress toward eliminating pedestrian fatalities altogether.6
Halloween trick-or-treating encourages creativity, physical activity, and neighborhood engagement. Trick-or-treating should not be abolished in a misguided effort to eliminate Halloween-associated risk. Instead, policymakers, physicians, and parents should act to make residential streets safer for pedestrians on Halloween and throughout the year.
Corresponding Author: John A. Staples, MD, MPH, St Paul’s Hospital, 1081 Burrard St, Burrard Bldg, Rm 5910, Vancouver, BC V6Z 1Y6, Canada (email@example.com).
Published Online: October 30, 2018. doi:10.1001/jamapediatrics.2018.4052
Author Contributions: Dr Staples had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Staples, Redelmeier.
Obtained funding: Staples, Redelmeier.
Administrative, technical, or material support: Staples.
Supervision: Staples, Redelmeier.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was supported by the Vancouver Coastal Health Research Institute, the Canadian Institutes of Health Research, and the Canada Research Chair in Medical Decision Science.
Role of the Funder/Sponsor: The study sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: