Traffic fatalities in West Virginia and the remaining United States, 2008-2012.
Purpose: Research has shown that Appalachia has a higher traffic fatality rate than the non-Appalachian United States. This study compared traffic fatality rates in West Virginia to the rest of the United States.
Methods: Fatality Analysis Reporting System and Census data from 2008-2012 were used to calculate traffic fatality rates. Poisson regression was used to model rate ratios stratified by age, sex, rurality, and transportation type, as well as rate ratios per licensed driver, vehicles registered and miles travelled.
Results: The West Virginia traffic fatality rate was 71% higher than the United States. Fatality rates per-capita were elevated for vehicle passengers and drivers of passenger vehicles, motorcycles, all-terrain vehicles, and large trucks. In contrast, rates for pedestrians and bicyclists were lower than the rest of the country.
Conclusion: Public health intervention and further research are likely needed to discern reasons as to why this disparity exists in West Virginia.
West Virginia is home to more than 1.8 million residents and is the only state to lie entirely within the Appalachian region. (1) Due to its mountainous terrain, many communities within the state are geographically isolated with slightly more than 51% of the state's population living in a rural setting, as opposed to 21% for the rest of the nation. (2) Previous research has suggested that disparities in traffic safety may be extant in this area, as traffic fatality rates in Appalachia are considerably higher than the rest of the United States (U.S.). (3) Numerous studies have also shown that traffic fatalities are higher in rural areas compared to more urbanized locations and that rural drivers are more likely to engage in high risk driving behaviors compared to urban drivers. (4,6) Previous research has also shown that all-terrain vehicle (ATV) fatality rates in West Virginia are some of the highest in the nation, (7,8) and nationally more ATV collisions occur on roadways as opposed to off-road. (9,10) Therefore, the purpose of this study is to compare traffic fatality rates specifically in West Virginia with those of the rest of the U.S. Given the extant literature, it is hypothesized that traffic fatality rates are likely higher in West Virginia than the rest of the U.S. regardless of driver age, sex, rural or urban location (i.e. rurality) of the collision, and transportation type.
Traffic fatality data were obtained from the Fatality Analysis Reporting System (FARS) for 2008 through 2012. FARS is a publicly available database, maintained by the National Highway Traffic Safety Administration, which records all U.S. motor vehicle crashes that result in at least one fatality within 30 days of the crash. (11) To be included in the FARS database, the collision had to occur on a roadway customarily open to the public. (11) Data for each incident included date of crash, county location of crash, age and sex of the decedent, vehicle and person type (i.e. if the decedent was a passenger vehicle driver, passenger vehicle passenger, bus rider, large truck driver, motorcycle driver, ATV driver, pedestrian or bicyclist). The U.S. Department of Agriculture's 2013 urban influence codes (UIC) were used to classify the counties where the collisions occurred as rural or urban. (12) UlCs divide counties, county equivalents and independent cities into 12 groups based on population size. Counties in groups one and two were considered urban, while all other counties were considered rural. (12) The resident population of each county by age group and sex were obtained from the U.S. Census Bureau for 2008 through 2012. (13) Data for the number of licensed drivers, number of registered vehicles and number of miles traveled for West Virginia and the rest of the U.S. were obtained from the Federal Highway Administration highway statistics series. (14)
The FARS data were stratified by sex, age, rurality of the collision, and person/transportation type. Resident population data were used as the denominator to calculate traffic fatality rates per-capita for each region. The number of licensed drivers, number of registered vehicles and number of miles traveled were also used separately as denominators to calculate overall crude traffic fatality rates for each region. All rate ratios and 95% confidence intervals (CI) were then calculated via Poisson regression which modeled fatality counts for each region, using the log of the denominator as the offset. Rate ratios for pedestrians and bicyclists models were also adjusted for the rurality of the collision (binary coded), while all other person/transportation type models were adjusted for rurality of the collision and for instate residence (binary coded) of the decedent. A person was considered an instate resident if the state where the collision occurred was the same as their vehicle registration. Models were adjusted for these variables because traffic fatality rates are typically higher in rural areas and out-of-state residents may be unfamiliar with the topography of an area and be more prone to collision. For all rate ratios, the non-West Virginia United States served as the referent. All statistical analyses were performed using SAS version 9.4 with [alpha]=0.05.
There were 168,618 traffic fatalities occurring in the non-West Virginia United States from 2008 through 2012. Roughly one percent (N=1,723) of the national fatality total occurred in West Virginia (Table 1 ). The total traffic fatality rate per 100,000 residents was 18.7 for West Virginia and 11.0 for the rest of the United States. Similar elevations in fatality rate were seen when using licensed drivers, vehicle registrations, and vehicle miles travelled as the denominator. The crude traffic fatality rate was increased for West Virginia as compared with the rest of the U.S. for each year in the study period, as shown in Figure 1.
Males in West Virginia were killed in traffic crashes 1.74 times higher than in the rest of the United States, while females fared only slightly better at 1.64 (rate ratios (RR), 1.74 and 1.64, respectively) (Table 2). The incidence of fatality was highest in the 16-24 year old age group, both nationally and within the state, but the rate ratios across all age groups reflected statistically significant increases in fatality rate. Rural areas had increased incidences as well, with West Virginia's rate approaching that of the rest of the country. There was still a statistically significant increase in the rate ratio, however, as there was for all of the demographic variables examined. Crashes in rural counties accounted for roughly half of the fatalities in West Virginia, as opposed to less than one fourth in rural counties nationally.
Automobile drivers represented the bulk of traffic fatalities both within the state and nationwide. The adjusted rate of fatality among automobile drivers was two times higher (RR=2.02; 95% CI, 1.90-2.15) than the rest of the country (Table 3). For automobile passengers, there was a 74% increase in traffic fatalities when compared to the rest of the country (adjusted RR=1.74; 95% CI 1.56-1.94). The adjusted traffic fatality rate for motorcycle drivers was 19% higher (RR=1.19; 95% CI 1.01-1.39) than the U.S. motorcycle fatality rate. Large truck drivers sustained fatal injuries in West Virginia 1.72 times higher than the rest of the country (RR=1.72; 95% CI 1.18-2.52). ATV drivers were particularly susceptible to fatalities in West Virginia, as the adjusted rate was over 10 times higher than the rest of the U.S. rate when controlling for rurality of the collision and in-state residence (RR=10.08; 95% CI 8.06-12.63).
Conversely, pedestrians were 26% less likely (adjusted RR=0.74; 95% CI 0.61-0.91) to suffer a fatal injury in West Virginia as opposed to the rest of the United States, and bicyclists had a 70% (adjusted RR=0.30; 95% CI 0.13-0.66) reduction as well. No fatal bus crashes were recorded in West Virginia in the time period examined.
Studies comparing West Virginia and the rest of the United States are relatively scarce. To the authors' knowledge, this is the first study to compare traffic fatality rates in the state with those of the U.S. The main finding of this analysis was that the traffic fatality rates in West Virginia were significantly higher than the rest of the nation regardless of sex, age, rurality, and for all forms of transportation except walking and bicycling.
These findings were consistent with the current literature. Zhu et al examined the rates of fatal crashes in Appalachia compared to non-Appalachian counties, and found that there was a 45% increase in fatal crashes per capita in Appalachian counties. (3) This finding was consistent with the results from this study, which found rates that were even higher in West Virginia than those reported in the Appalachian study. Adekoya and Majumder found that West Virginia had a significantly higher rate of fatal traumatic brain injury than the nation as a whole, with motor vehicles being the largest non-suicide contributor to the rate. (15) The findings concerning ATV fatalities was also consistent with previous literature, though most studies did not differentiate between on or off-road collisions and no previous studies analyzed FARS data. (7,8,16-19)
There are several potential reasons to explain why the traffic fatality rates were significantly higher in West Virginia for various forms of transportation. Even though the models were adjusted for rurality of the collision, there may be other factors, such as driver behavior or environmental factors, which may also be contributing to these disparate rates.
It is possible that West Virginia has a lower safety culture compared to the rest of the nation. Previous research has shown that high risk safety behaviors, such as speeding, driver intoxication, disobeying traffic control devices, and lower safety restraint use, are much higher in rural motor vehicle collisions. (4,6) Research has shown that drug and alcohol use contribute greatly to driver intoxication in West Virginia, as ~25% of all fatally injured drivers from 2004-2005 were found drug and/or alcohol positive at time of collision. (20) In an ATV fatality study conducted in West Virginia using data from 1991-2000, alcohol and drug use were found in ~50% of ATV drivers. (16) A later ATV study using data from 2004-2006 showed ~50% of drivers tested positive for alcohol, while 22% tested positive for drugs of abuse. (8)
Besides alcohol and drugs, use of protective equipment may also be lower in West Virginia. A national study using data from 1993 showed that seat belt usage is lower in Appalachia compared to the rest of the U.S. (21) Helmet usage is also low in West Virginia, particularly in ATV collisions. Helmet usage among fatal ATV crashes in West Virginia was estimated at 15% and 22%. (8,16)
It is also possible that the vehicle fleet in West Virginia is older. Newer vehicles typically have more advanced safety features to protect the driver and passenger(s) in the event of a collision. These vehicle safety improvements have been associated with decreased injuries and fatalities in numerous studies. (22,23)
Environmental factors, such as the state's mountainous topography, may also affect traffic fatality rates. First, the topography can complicate access to emergency care post-collision. For example, only 58.7% of West Virginia residents have access to an emergency department within 30 minutes of their residence. (24)
Previous research has also shown that emergency medical service (EMS) response time is significantly longer in rural areas. (4) In one study of U.S. motor vehicle collisions, only half of rural crashes received an EMS response within 10 minutes compared to almost 90% of all urban crashes studied. (4) Second, the Appalachian Mountains are known to have microclimates, (25) which means that weather conditions can be highly varied or change quickly between locations that are geographically close. Changing weather conditions may pose additional challenges to drivers as there is a strong relationship between precipitation rate, visibility, and road geometry (i.e. steep slopes and curved road segments) in analyses of highway crashes in mountainous areas. (26)
Other environmental factors, such as wildlife collisions or road conditions, can also be problematic for drivers in West Virginia. West Virginia ranks as one of the top states for wildlife collisions by car (27) and even aircraft at landing and take-off. (28) The physical condition of the road may also affect collision risk. Many of the state's roads are rough and in poor repair. While literature in this area is sparse, one Australian study showed a distinct linear relationship between road roughness and collision rate. (29)
While the traffic fatality rates were higher for most forms of transportation, the pedestrian and bicyclist fatality rates in West Virginia were considerably lower than the rest of the nation. This may also be due to several factors. First, the denominator in this analysis was resident population (i.e. per capita). While a more appropriate denominator may be the number of people who walk or ride bicycles, this figure is unknown and nearly impossible to determine. It is possible that a greater proportion of West Virginia residents simply do not walk or do not ride bicycles compared to the rest of the nation. This effect would be masked when using resident population as the denominator. A second reason for the lower bicycle and pedestrian fatality rates could be the built environment. A substantial body of literature exists regarding how the environment affects the activity level of a population; it is well-documented that when individuals have less perceived environmental barriers, they are more likely to exercise. (30,31) Commonly cited barriers to exercise are hilly terrain, lack of sidewalks, unattended dogs, crime, no walking trails, poor scenery, not seeing others exercising, and poor lighting. (32,33) Rural roads, which are typical in West Virginia, generally lack sidewalks and paved road shoulders and are poorly lit. (34) Also, due to the topography of the state, many West Virginia roads are also steep, narrow, and windy which may not be conducive to walking or cycling (i.e. a perceived barrier). However, it is also possible that those who do walk or cycle in West Virginia use bike or mountain trails, which are generally free from most forms of traffic (except ATVs). Therefore, those who walk and bike may be doing these activities"
There are several limitations to this study. First, the classification of rural or urban was based on county of crash, which may not truly indicate the nature of the environment where the crash occurred. Additionally, the 2013 data was used from the Department of Agriculture UIC codes, which may not precisely reflect the status at the time of the crashes. Third, due to the nature of the FARS data, the fatality rates for ATVs only included those which happened on a public road; differences in rates likely exist for off-road collisions. Finally, while this study stratified traffic fatality rates by age, sex and rurality and adjusted person/transportation models by rurality and in-state residence, fatal crashes involve a multitude of factors, such as the use of safety equipment, alcohol use and road and weather conditions. These factors were not evaluated in this study as its primary objective was simply to discern if West Virginia traffic fatality rates were higher than the nation's rates.
The findings of this analysis suggest that there is a disparity in traffic safety in West Virginia. Because the traffic fatality rates were starkly higher in this state compared to the rest of the nation over the study period, this study has public health implications. Further research is necessary to discern how and why fatal crashes in West Virginia differ from others in the nation. Answers to these questions are necessary to guide future public health interventions, policies, or possibly law enforcement to mitigate death and disability in this population.
(1.) Appalachian Regional Commission. The Appalachia Regional Commission. http://www.arc.gov/. Accessed August 12, 2014.
(2.) Baker SP, Whitfield RA, O'Neill B. Geographic variations in mortality from motor vehicle crashes. N Engl J Med. May 1987:316(22):1384-1387.
(3.) Zhu M, Zhao S, Gurka KK, Kandati S, Coben JH. Appalachian versus non-Appalachian U.S. traffic fatalities, 2008-2010. Ann Epidemiol. Jun 2013:23(6):377-380.
(4.) National Highway Traffic Safety Administration. Contrasting rural and urban fatal crashes 1994-2003. www-nrd.nhtsa.dot.gov/Pubs/809896.pdf. Published December 2005. Accessed November 1, 2015.
(5.) Rakauskas ME, Ward NJ, Gerberich SG. Identification of differences between rural and urban safety cultures. Accid Anal Prev. Sep 2009:41 (5):931-937.
(6.) Thompson JP, Baldock MR, Mathias JL, Wundersitz LN. An examination of the environmental, driver and vehicle factors associated with the serious and fatal crashes of older rural drivers. Accid Anal Prev. Jan 2013;50:768-775.
(7.) Helmkamp JC, Aitken ME, Graham J, Campbell CR. State-specific ATV-related fatality rates: an update in the new millennium. Public Health Rep. Jul-Aug 2012;127(4):364-374.
(8.) Hall AJ, Bixler D, Helmkamp JC, Kraner JC, Kaplan JA. Fatal All-Terrain Vehicle Crashes. Injury Types and Alcohol Use. Am J Prev Med. 2009;36(4):311-316.
(9.) Denning G, Jennissen C, Harland K, Ellis D, Buresh C. All-Terrain Vehicles (ATVs) on the Road: A Serious Traffic Safety and Public Health Concern. Traffic Inj Prev. 2013;14(1):78-85.
(10.) Denning GM, Harland KK, Ellis DG, Jennissen CA. More fatal all-terrain vehicle crashes occur on the roadway than off: Increased risk-taking characterises roadway fatalities. Inj Prev. 2013;19(4):250-256.
(11.) National Highway Traffic Safety Administration. Fatality Analysis Reporting System Analytical User's Manual 1975-2012. www-nrd.nhtsa.dot.gov/Pubs/811855.pdf. Published January 2014. Accessed August 5, 2014.
(12.) United States Department of Agriculture. 2013 Urban influence codes, http://www.ers.usda.gov/data-products/urban-influence-codes.aspx Updated July 8, 2013. Accessed August 5, 2014.
(13.) United States Census Bureau. Population and housing u nit estimates. http://www.census.gov/popest/index.html. Updated June 25, 2015. Accessed November 5, 2015.
(14.) Federal Highway Administration Highway Statistics Series, http://www.fhwa.dot.gov/policyinformation/statistics.cfm. Updated November 13, 2015. Accessed November 14, 2015.
(15.) Adekoya N, Majumder R. Fatal traumatic brain injury. West Virginia, 1989-1998. Public Health Rep. Sep-Oct 2004:119(5):486-492.
(16.) Carr AM, Bailes JE, Helmkamp JC, et al. Neurological Injury and Death in All-terrain Vehicle Crashes in West Virginia: A 10-year Retrospective Review. Neurosurgery. 2004;54(4):861-867.
(17.) Helmkamp JC. A comparison of state-specific all-terrain vehicle-related death rates, 1990-1999. Am J Public Health. Nov 2001;91(11):1792-1795.
(18.) Helmkamp JC, Aitken ME, Lawrence BA. ATV and bicycle deaths and associated costs in the United States, 2000-2005. Public Health Rep. May-Jun 2009;124(3):409-418.
(19.) Helmkamp JC, Furbee PM, Coben JH, Tadros A. All-terrain vehicle-related hospitalizations in the United States, 2000-2004. Am J Prev Med Jan 2008:34(1):39-45.
(20.) Centers for Disease Prevention and Control. Alcohol and other drug use among victims of motor-vehicle crashes--West Virginia, 2004-2005. MMWR. Dec 2006:55(48): 1293-1296.
(21.) West Virginia Department of Health and Human Resources. Health risks: the Appalachian lifestyle, http://www.wvdhhr.org/bph/hsc/pubs/other/appalachian/app_risk.pdf. Accessed October 4, 2015.
(22.) Anderson RWG, Searson DJ. Use of age--period--cohort models to estimate effects of vehicle age, year of crash and year of vehicle manufacture on driver injury and fatality rates in single vehicle crashes in New South Wales, 2003-2010. Accid Anal Prev. 2015:75:202-210.
(23.) Lecuyer JF, Chouinard A. Study on the effect of vehicle age and the importation of vehicle 15 years and older on the number of fatalities, seriouc injuries and collisions in Canada. Paper presented at: Canadian Multidisciplinary Road Safety Conference XVI, June 11-14 2006: Winnipeg Manitoba, www.tc.gc.ca/media/documents/roadsafety/paper_39_lecuyer.pdf Accessed November 1, 2015.
(24.) Carr BG, Branas CC, Metlay JP, Sullivan AF, Camargo CA, Jr. Access to emergency care in the United States. Ann Emerg Med. Aug 2009;54(2):261-269.
(25.) Desta F, Colbert JJ, Rentch JS, Gottschalk KW. Aspect induced differences in vegetation, soil, and microclimate characteristics of an Appalachian watershed. Castanea. 2004;69(2):92-108.
(26.) Yu R, Xiong Y, Abdel-Aty M. A correlated random parameter approach to investigate the effects of weather conditions on crash risk for a mountainous freeway. Transp Res Part C: Emerg Technol. 2015:50:68-77.
(27.) Langley RL Higgins SA, Herrin KB. Risk factors associated with fatal animal-vehicle collisions in the United States, 1995-2004. Wilderness Environ Med. Winter 2006;17(4):229-239.
(28.) Biondi KM, Belant JL, Martin JA, Devault TL, Wang G. White-tailed deer incidents with US civil aircraft. Wildlife Society. 2011; 35(3):303-309.
(29.) Cairney P, Bennett P. An exploratory study of surface characterisitcs and crash occurence on selected roads in Australia. https.//www.arrb.com.au/admin/file/content13/c6/ARR382_web.pdf. Published May 2013. Accessed November 1, 2015.
(30.) Saelens BE, Sallis JF, Frank LD. Environmental correlates of walking and cycling: findings from the transportation, urban design, and planning literatures. Ann Behav Med. Spring 2003;25(2):80-91.
(31.) Parks SE, Housemann RA. Brownson RC. Differential correlates of physical activity in urban and rural adults of various socioeconomic backgrounds in the United States. J Epidemiol Community Health. Jan 2003:57(1):29-35.
(32.) Eyler AA, Brownson RC, Bacak SJ, Housemann RA. The epidemiology of walking for physical activity in the United States. Med Sci Sports Exerc. Sep 2003;35(9):1529-1536.
(33.) Wilcox S, Castro C, King AC. Housemann R, Brownson RC. Determinants of leisure time physical activity in rural compared with urban older and ethnically diverse women in the United States. J Epidemiol Community Health. Sep 2000;54(9):667-672.
(34.) Federal Highway Administration. Factors contributing to pedestrian and bicycle crashes on rural highways, http://www.fhwa.dot.gov/publications/research/safety/10052. Published June 2010. Accessed November 1, 2015.
Kenneth B. Plants, DVM, MPH
WVU School of Public Health, Department of Epidemiology
Toni Marie Rudisill, MS, PhD
WVU Injury Control Research Center
Motao Zhu, MD, MPH, PhD
WVU School of Public Health, Department of Epidemiology
Corresponding Author: Kenneth B. Plants DVM, MPH, WVU School of Public Health, Department of Epidemiology, PO Box 9190, Morgantown, WV 26506. Email: firstname.lastname@example.org.
Table 1: Incidence of traffic fatality for West Virginia and the rest of the United States by resident population, licensed drivers, registered vehicles, and miles travelled, 2008-2012(a) United States West Virginia Numerator Fatalities 168,618 1,723 Denominator Number Rate Number Rate Resident population 1.54 B 11.0 9.25 M 18.7 Licensed drivers 1.05 B 16.1 6.34 M 27.3 Registered vehicles 1.31 B 13.0 7.31 M 23.6 Vehicle miles traveled 14.82 T 0.00115 98.00 B 0.00176 Numerator Fatalities Denominator Rate ratio (95% CI) Resident population 1.71 (1.63-1.79)(*) Licensed drivers 1.70(1.62-1.78)(*) Registered vehicles 1.83(1.74-1.92)(*) Vehicle miles traveled 1.55(1.48-1.62)(*) (a): Abbreviations: B=billion, CI=confidence interval, M=million, T=trillion; rates are per 100.000 (*) specifies statistical significance, p < 0.05; US rates served as the referent Table 2: Incidence of traffic fatality per capita by sex, age and rurality for West Virginia and the rest of the United States, 2008-2012 United States West Virginia Characteristic Number of Rate per 100,000 Number of Rate per Fatalities Fatalities 100,000 Sex Male 119,014 15.71 1,239 27.40 Female 49,550 6.34 486 10.39 Age (years) 0-15 7,265 2.38 52 3.27 16-24 35,526 16.42 354 29.75 25-64 98,467 12.11 1,017 20.61 65+ 27,070 13.43 302 20.38 Rurality Urban 120,604 9.21 915 16.25 Rural 35,673 21.65 857 24.03 Characteristic Rate Ratio (95% CI) Sex Male 1.74(1.65-1.84)(*) Female 1.64(1.50-1.79)(*) Age (years) 0-15 1.37(1.05-1.80)(*) 16-24 1.81 (1.63-2.01)(*) 25-64 1.70(1.60-1.81)(*) 65+ 1.52(1.35-1.70)(*) Rurality Urban 1.77(1.65-1.88)(*) Rural 1.11 (1.04-1.19)(*) (*) specifies statistical significant at p<0.05; US rates served as referent group Table 3: Incidence of traffic fatality per capita by person and transportation type, WV and US, 2008-2012 United States West Virginia Fatalities Rate per Fatalities Rate per N 100,000 N 100,000 Automobile Driver 85,860 5.59 1,105 11.94 Automobile Passenger 32,297 2.11 345 3.73 Motorcycle Drivers 23,731 1.54 169 1.82 Pedestrians 21,461 1.43 98 1.06 Bicyclists 3,360 0.22 6 0.06 Truck Drivers 2,606 0.17 27 0.29 ATV Drivers 1,546 0.10 91 0.98 Bus Riders 160 0.01 0 0.00 Crude Rate Adjusted Rate Ratio (95% CI) Ratio (95% Cl)(a) Automobile Driver 2.14 (2.01-2.27)(*) 2.02(1.90-2.15)(*) Automobile Passenger 1.77(1.59-1.97)(*) 1.74(1.56-1.94)(*) Motorcycle Drivers 1.18(1.02-1.38)(*) 1.19(1.01-1.39)(*) Pedestrians 0.74(0.61-0.91)(*) 0.74(0.61-0.91)(*) Bicyclists 0.30(0.13-0.66)(*) 0.30(0.13-0.66)(*) Truck Drivers 1.72(1.18-2.51)(*) 1.72(1.18-2.52)(*) ATV Drivers 9.77(7.91, 12.08)(*) 10.08(8.06-12.63)(*) Bus Riders NA NA (a): Rate ratios for bicyclists and pedestrians were adjusted for rurality of the collision. Other person/transportation ratios were adjusted for rurality of the collision and whether the decedent was a resident of the state where their collision occurred based on their vehicle registration, (*) specifies statistical significant at p<0.05; US rates served as referent group
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Original Research|
|Author:||Plants, Kenneth B.; Rudisill, Toni Marie; Zhu, Motao|
|Publication:||West Virginia Medical Journal|
|Date:||Mar 1, 2017|
|Previous Article:||A focused observation tool using dreyfus stages of skill acquisition as an evaluative scale.|
|Next Article:||Metastatic renal cell carcinoma of the head and neck: A case series.|