D P Bui1, C Hu1, A M Jung1, K M Pollack Porter2, S C Griffin3, D D French4,5,6, S Crothers7, J L Burgess3. 1. a The Mel and Enid Zuckerman College of Public Health, Department of Epidemiology and Biostatistics , The University of Arizona , Tucson , Arizona. 2. b Johns Hopkins Bloomberg School of Public Health , Johns Hopkins Center for Injury Research and Policy , Baltimore , Maryland. 3. c The Mel and Enid Zuckerman College of Public Health, Department of Community Environment and Policy , The University of Arizona , Tucson , Arizona. 4. d Department of Ophthalmology, Feinberg School of Medicine , Northwestern University , Chicago , Illinois. 5. e Feinberg School of Medicine, Center for Healthcare Studies, Northwestern University , Chicago , Illinois. 6. f Veterans Affairs Health Services Research and Development Service , Chicago , Illinois. 7. g Seattle Fire Department , Seattle , Washington.
Abstract
OBJECTIVE: Emergency service vehicle incidents are a leading cause of firefighter fatalities and are also hazardous to civilian road users. Modifiable driving behaviors may be associated with emergency service vehicle incidents. The goal of this study was to use telematics to identify driving behaviors associated with crashes in the fire service. METHODS: Forty-three emergency service vehicles in 2 fire departments were equipped with telematics devices (12 in Department A and 31 in Department B). The devices collected vehicle coordinates, speed, and g forces, which were monitored for exceptions to driving rules established by the fire departments regarding speeding, harsh braking, and hard cornering. Fire department administrative reports were used to identify vehicles involved in crashes and merged with daily telematics data. Penalized logistic regression was used to identify driving rules associated with crashes. Least absolute shrinkage and selection operator (LASSO) regression was used to generate a telematics-based risk index for emergency service vehicle incidents. RESULTS: Nearly 1.1 million km of driving data and 44 crashes were recorded among the 2 departments during the study. Harsh braking was associated with increased odds of crash in Department A (odds ratio [OR] = 2.22; 95% confidence interval [CI], 1.09-4.51) and Department B (OR = 1.55; 95% CI, 1.12-2.15). For every kilometer of nonemergency speeding, the odds of crash increased by 35% in Department A (OR = 1.35; 95% CI, 1.03-1.77) and by over 2-fold in Department B (OR = 2.09; 95% CI, 1.19-3.66). In Department B, hard cornering (OR = 1.14; 95% CI, 1.03-1.26) and emergency speeding (OR = 1.65; 95% CI, 1.06-2.57) were also associated with increased odds of crash. The final LASSO risk index model had a sensitivity of 73% and specificity of 57%. CONCLUSIONS: Harsh braking and excessive speeding were driving behaviors most associated with crash in the fire service. Telematics may be a useful tool for monitoring driver safety in the fire service.
OBJECTIVE: Emergency service vehicle incidents are a leading cause of firefighter fatalities and are also hazardous to civilian road users. Modifiable driving behaviors may be associated with emergency service vehicle incidents. The goal of this study was to use telematics to identify driving behaviors associated with crashes in the fire service. METHODS: Forty-three emergency service vehicles in 2 fire departments were equipped with telematics devices (12 in Department A and 31 in Department B). The devices collected vehicle coordinates, speed, and g forces, which were monitored for exceptions to driving rules established by the fire departments regarding speeding, harsh braking, and hard cornering. Fire department administrative reports were used to identify vehicles involved in crashes and merged with daily telematics data. Penalized logistic regression was used to identify driving rules associated with crashes. Least absolute shrinkage and selection operator (LASSO) regression was used to generate a telematics-based risk index for emergency service vehicle incidents. RESULTS: Nearly 1.1 million km of driving data and 44 crashes were recorded among the 2 departments during the study. Harsh braking was associated with increased odds of crash in Department A (odds ratio [OR] = 2.22; 95% confidence interval [CI], 1.09-4.51) and Department B (OR = 1.55; 95% CI, 1.12-2.15). For every kilometer of nonemergency speeding, the odds of crash increased by 35% in Department A (OR = 1.35; 95% CI, 1.03-1.77) and by over 2-fold in Department B (OR = 2.09; 95% CI, 1.19-3.66). In Department B, hard cornering (OR = 1.14; 95% CI, 1.03-1.26) and emergency speeding (OR = 1.65; 95% CI, 1.06-2.57) were also associated with increased odds of crash. The final LASSO risk index model had a sensitivity of 73% and specificity of 57%. CONCLUSIONS: Harsh braking and excessive speeding were driving behaviors most associated with crash in the fire service. Telematics may be a useful tool for monitoring driver safety in the fire service.