category: literaturenote citekey: kmeturbanruraldifferencesmotor2006 title: Urban–rural differences in motor vehicle crash fatality and hospitalization rates among children and youth authors: "Kmet, Leanne; Macarthur, Colin" year: 2006 date: 2006-00-00 2006 doi: 10.1016/j.aap.2005.07.007 publication: "Accident Analysis & Prevention" url: "https://doi.org/10.1016/j.aap.2005.07.007" zotero_key: UZF259QW zotero_storage: EQPWF8CF collections: transpordi_planeerimine folder: 001_artiklid/Road Safety Indicators firstAuthor: "Kmet, Leanne"

status: converted

Accident Analysis and Prevention 38 (2006) 122–127

Urban–rural differences in motor vehicle crash fatality and hospitalization rates among children and youth

Leanne Kmet a, Colin Macarthur ^b,^∗

^a Department of Community Health Sciences, University of Calgary, 3330 Hospital Drive NW, Calgary, Alta., Canada T2N 4N1 ^b University of Toronto, Department of Paediatrics, Bloorview MacMillan Children's Center, Bloorview Research Institute, 150 Kilgour Road, Toronto, Ont., Canada M4G 1R8

Received 29 April 2005; received in revised form 29 July 2005; accepted 29 July 2005

Abstract

This population-based study examined motor vehicle crash hospitalization rates and death rates among children and youth in rural and urban areas of the province of Alberta, Canada. Using police report data (1997–2002, inclusive), average annual motor vehicle crash hospitalization and death rates among those 0–19 years of age were calculated for rural and urban regions. Across all age and sex strata examined, both the hospitalization and the fatality rates were significantly higher in rural compared with urban regions. After adjusting for age, sex and calendar year, the relative risk of a motor vehicle crash hospitalization (rural versus urban) was 3.0 (95% CI: 2.8, 3.2). After adjusting for age, sex and calendar year, the relative risk of a motor vehicle crash fatality was 5.4 (95% CI: 4.2, 6.9). Motor vehicle crash injury hospitalization and fatality rates among children and youth in the province of Alberta are considerably higher in rural areas compared with urban areas. There is a need to identify social, demographic and environmental driving hazards associated with the rural environment. © 2005 Elsevier Ltd. All rights reserved.

Keywords: Child; Motor vehicle; Mortality; Hospitalization; Urban–rural

1. Introduction

In Canada, motor vehicle traffic crashes are the leading cause of injury mortality and a leading cause of injury hospitalization among children and youth (Child Injury Division—Bureau of Reproductive and Child Health, 1999). Such injuries accounted for 574 deaths and 6035 hospitalizations in the 0–19 year age group in Canada in 1997 (Child Injury Division—Bureau of Reproductive and Child Health, 1999). Motor vehicle traffic crashes accounted for approximately 33%, 50% and 75% of all unintentional injury deaths among those 0–4, 5–14 and 15–19 years of age, respectively (Child Injury Division—Bureau of Reproductive and Child Health, 1999).

Examination of geographic variation in motor vehicle crash injury rates may contribute to the effective targeting of injury countermeasures. The objective of this populationbased study was to examine motor vehicle crash hospitalization and death rates among children and youth in rural and urban areas of the province of Alberta, Canada.

2. Materials and methods

2.1. Motor vehicle crash records

Data on motor vehicle crash injuries resulting in hospitalization or death among children and youth (0–19 years) in the province of Alberta from January 1, 1997 to December 31, 2002 were obtained from police reports compiled in the Alberta Collision Information System. Legislation in Alberta requires that any traffic collision resulting in death, injury or property damage exceeding CDN\$ 1000 be reported to police. Fatal injuries were defined as those resulting in death within 30 days of the collision. (It has been estimated that only 2% of all motor vehicle crash-related fatalities occur

^∗ Corresponding author. Tel.: +1 416 424 3862; fax: +1 416 425 9332. E-mail address: cmacarthur@bloorviewmacmillan.on.ca (C. Macarthur).

after 30 days (Mao et al., 1997).) The analysis was restricted to occupants of privately operated vehicles injured on public roadways.

2.2. Urban/rural classification

During the study period, Alberta had 17 geographically defined Regional Health Authorities, each of which assumed local responsibility for injury prevention. For each motor vehicle crash injury, the crash location was determined by linking the police detachment responding to the crash to the respective Health Authority. The two regions with the largest population densities (about 160 persons/km2) were defined as 'urban'. The remaining 15 regions (all with population densities <11 persons/km2) were defined as 'rural'.

2.3. Rate calculations and comparisons

Age- and sex-specific population estimates for each Regional Health Authority were obtained from the Alberta Health Care Population Registry. Assuming a Poisson distribution of both the hospitalizations and the deaths (appropriate given rare events in sufficiently large populations (Selvin, 1995)), data from the 6-year study period were used to calculate average annual motor vehicle crash hospitalization and death rates for urban and rural regions, along with 95% Poisson confidence intervals. Rates were also calculated by age group (0–14, 15–19 years) and sex (male, female). Using these rates, the relative risk of a motor vehicle crash injury hospitalization (rural versus urban) or death (rural versus urban) along with 95% confidence intervals were calculated for each age- and sex-specific stratum. Poisson regression analysis was then used to calculate adjusted relative risks (rural versus urban), specifying either the number of deaths or the number of injury hospitalizations as the dependent variable and including population size (by age group and sex), location (rural versus urban) and calendar year as independent variables. All analyses were conducted using STATA 6.0 (College Station, TX).

3. Results

A total of 4660 children and youth (0–19 years) in Alberta were involved in motor vehicle crashes resulting in hospitalization or death during the study period. Injuries to pedestrians (n = 440), bicyclists (n = 201), motorcyclists (n = 123), and injuries involving farm vehicles (n = 2), construction vehicles (n = 2), public transportation (n = 11), off-road snow, or all-terrain vehicles (n = 118) were excluded. Injuries for which the sex of the victim (n = 5), the type of vehicle involved (n = 5) or the crash location (n = 3) were not documented were also excluded, leaving 383 fatalities and 3367 hospitalizations for analysis.

There was homogeneity of motor vehicle traffic crash fatality and hospitalization rates between the two urban regions and also across the 15 rural regions. Therefore, the data from the two urban regions were combined as were the data from the 15 rural regions. The pooled data were then used to estimate summary urban and rural rates for Alberta.

As shown in Table 1, the motor vehicle traffic crash fatality rates were significantly higher – across all age and sex strata – in rural Alberta, compared with urban Alberta. The relative risk of a motor vehicle traffic crash fatality (rural versus urban) – after adjusting for age, sex, and calendar year – was 5.4 (95% confidence interval (CI): 4.2, 6.9).

Similar findings were noted for motor vehicle traffic crash hospitalization rates. As shown in Table 2, hospitalization rates were significantly higher – across all age and sex strata – in rural Alberta, compared with urban Alberta. The relative risk of a motor vehicle traffic crash hospitalization (rural versus urban) – after adjusting for age, sex, and calendar year – was 3.0 (95% CI: 2.8, 3.2).

To assess whether the higher relative risks in rural versus urban areas were influenced by multiple casualty crashes, we also determined rates based on the number of crashes. Data from 1997 (85 fatalities, 537 hospitalizations) were excluded from these calculations as case numbers (i.e., unique crash identifiers) were not available. The remaining 298 deaths resulted from 259 motor vehicle crashes. Forty-six of these occurred in urban areas, compared with 213 in rural areas, yielding average annual fatal motor vehicle crash rates (per 100,000 population) of 1.9 and 11.3, respectively. The relative risk (rural versus urban) of a fatal motor vehicle crash was 5.9 (95% CI: 4.3, 8.3). The 2830 injury hospitalizations occurred as a result of 693 urban crashes and 1503 rural crashes, yielding average annual crash rates (per 100,000 population) of 29.0 and 79.7, respectively. The relative risk (rural versus urban) of a motor vehicle crash resulting in hospitalization was 2.8 (95% CI: 2.5, 3.0).

Finally, since a single crash could result in both fatalities and injury hospitalizations, the data were combined and the number of severe crashes (i.e., crashes resulting in death and/or hospitalization) was determined. There were 724 severe urban crashes and 1625 severe rural crashes, yielding average annual crash rates (per 100,000 population) of 30.3 and 86.2, respectively. The relative risk (rural versus urban) of a severe motor vehicle crash was 2.8 (95% CI: 2.6, 3.1).

As would be expected, in both the rural and urban areas, motor vehicle crash injury rates (deaths and hospitalizations) were considerably higher among youth 15–19 years of age, compared with younger children (see Tables 1 and 2). In both areas, the rates were highest among 15–19-year-old males. For example, in rural areas, males in this age group were ten times more likely than younger children of both sexes to die in motor vehicle crashes.

As shown in Table 3, rural adolescent males were significantly more likely to be involved in a motor vehicle crash resulting in death or hospitalization, compared with rural adolescent females. Urban data showed an increased relative risk (male versus female) for motor vehicle crash fatalities among adolescents only.

Table 1 Fatal motor vehicle crash injury rates by rural and urban location among children and youth (0–19 years) in Alberta, Canada (1997–2002)

Age (years)	Sex	Rural		Urban		Relative risk (rural vs.
		No. of deaths	Fatal injury ratea (95% CI)b	No. of deaths	Fatal injury ratea (95% CI)b	urban) (95% CI)b
0–14	Male	44	5.1 (3.7–6.9)	9	0.8 (0.4–1.6)	6.2 (3.0–12.7)
	Female	40	4.9 (3.5–6.7)	7	0.7 (0.3–1.4)	7.3 (3.3–16.3)
15–19	Male	142	47.7 (40.2–56.2)	41	11.1 (8.0–15.1)	4.3 (3.0–6.1)
	Female	85	29.9 (23.9–37.0)	15	4.3 (2.4–7.1)	7.0 (4.0–12.1)

^a Average annual rate, expressed as deaths/100,000.

Table 2 Motor vehicle crash injury hospitalization rates by rural and urban location among children and youth (0–19 years) in Alberta, Canada (1997–2002)

Age (years)	Sex	Rural		Urban		Relative risk (rural vs.
		No. of hospital admissions	Ratea (95% CI)b	No. of hospital admissions	Ratea (95% CI)b	urban) (95% CI)b
0–14	Male	309	35.9 (32.0–40.2)	111	10.1 (8.3–12.2)	3.5 (2.9–4.4)
	Female	273	33.5 (29.7–37.7)	114	10.9 (9.0–13.1)	3.1 (2.5–3.8)
15–19	Male	1004	337.2 (316.6–358.7)	395	107.3 (97.0–118.5)	3.1 (2.8–3.5)
	Female	786	276.5 (257.5–296.5)	375	107.1 (96.5–118.5)	2.6 (2.3–2.9)

^a Average annual rate, expressed as injury hospitalizations/100,000.

Table 3 Relative risk (male vs. female) of motor vehicle crash injury (death and hospitalization) by rural and urban location among children and youth (0–19 years) in Alberta, Canada (1997–2002)

| Age (years) | Location | Relative risk: male vs. female (95% CI)a | | | |-------------|----------|------------------------------------------|-----------------|--| | | | Death | Hospitalization | | | 0–14 | Rural | 1.0 (0.7–1.6) | 1.1 (0.9–1.3) | | | 0–14 | Urban | 1.2 (0.5–3.3) | 0.9 (0.7–1.2) | | | 15–19 | Rural | 1.6 (1.2–2.1) | 1.2 (1.1–1.3) | | | 15–19 | Urban | 2.6 (1.4–4.7) | 1.0 (0.9–1.2) | |

^a 95% Poisson confidence interval.

4. Discussion

From 1997 to 2002, rural children and youth in Alberta, Canada were five times more likely than their urban counterparts to die in a motor vehicle crash and three times more likely to be admitted to hospital following a motor vehicle crash. These increases in risk (rural versus urban) were consistent across age groups and sex.

An increased risk of fatal and severe motor vehicle crash injuries among children and youth in rural (versus urban) environments has been previously documented (Hwang et al., 1997; King et al., 1994; Lapidus et al., 1998; Niemcryk et al., 1997; Svenson et al., 1996a,b). For example, a Nevada study examining severe motor vehicle crash injuries (Niemcryk et al., 1997) showed that rates among children (0–14 years) were three times higher in rural areas, compared with urban areas. In contrast, however, to the two- to three-fold increased risk of fatal crash involvement in rural (versus urban) areas reported by others (Hwang et al., 1997; King et al., 1994; Svenson et al., 1996a,b), we observed a five-fold increase in risk.

The main strength of our study was the use of populationbased data collected over a 6-year period, resulting in a large sample size and stable estimates of motor vehicle crash injury rates. The same data source was used to identify motor vehicle crash injuries among both urban and rural children. When estimating relative risks (rural versus urban), Poisson regression analyses were used to control for confounding by age, sex and calendar year.

The study, however, also had several limitations. For example, police data are known to underestimate the number of motor vehicle crash injuries. Comparisons of police, hospital and vital statistics data show that between 85% and 99% of crash fatalities are captured by police records (Aptel et al., 1999; Rosman and Knuiman, 1994). Capture of non-fatal crash injuries requiring hospitalization is lower, ranging from 45% to 67% (Alsop and Langley, 2001; Aptel et al., 1999; Cryer et al., 2001; Rosman, 2001; Rosman and Knuiman, 1994). Further, some (Alsop and Langley, 2001; Rosman, 2001), but not all (Aptel et al., 1999; Rosman and Knuiman, 1994) studies indicate that police underestimation of nonfatal injuries is greater among children and youth compared with adults. In addition, a single study reported greater underestimation in rural areas compared with urban areas (Aptel et al., 1999). Taken together, these data suggest that both the absolute injury rates and the relative risks (rural versus urban) in our study may be conservative estimates, particularly for injury hospitalizations.

^b 95% Poisson confidence interval.

Our study also had other limitations. Injury rates were calculated using the number of injuries occurring in a given region in the numerator and regional population estimates in the denominator. As such, each injury included in the numerator did not necessarily represent an injury to an 'at risk' individual in the denominator. While rates calculated in this manner succeed in identifying high risk locations, they are subject to error. For example, in areas with many visitors, motor vehicle crash rates may be exaggerated. Conversely, in areas where residents frequently travel outside the area, the rates may be underestimated.

Further, as our data did not identify injured individuals by place of residence, our analysis was not specific to Alberta residents. While estimates of non-fatal motor vehicle crash injuries to out-of-province residents are not available in Alberta, data from the provincial Vital Statistics Death Registry indicate that out-of-province residents account for only 8% of all motor vehicle crash deaths in the province each year. Therefore, even if out-of-province deaths were systematically distributed to particular geographic areas, this bias would not have materially altered our relative risk estimates (rural versus urban) for crash fatalities.

Finally, data on all variables of interest were not available to us. Regional and individual-level information on variables such as vehicle travel speed, type of collision, environmental conditions, alcohol use and safety restraint use would have permitted an exploration of the predictors contributing to the rural–urban differences in motor vehicle crash injury rates. In general, there is limited population-based information on the urban–rural distribution of motor vehicle crash injury risk factors.

The higher motor vehicle crash injury rates observed in rural Alberta may reflect greater exposure to vehicle travel by individuals in outlying areas. With respect to travel to school, an observational study of 4000 children attending randomly selected elementary schools in South Carolina found that in both urban and suburban areas, 44% of children were driven to and from school in private vehicles (Sirard et al., 2005). Similarly, in a population-based survey in Georgia, 43% of parents of children 5–15 years reported driving their children to and from school most of the time (Anon., 2002). While the percentage of children who traveled in private vehicles or buses was not reported by place of residence, the percentage of children who regularly walked to school was similar in rural and urban areas (17% and 19%, respectively) (Anon., 2002).

While the frequency of travel to and from school by private vehicle may be similar in rural and urban areas, distances traveled (to and from school, and overall) are likely greater in rural areas. For example, a survey of drivers in Ontario, Canada, found that rural drivers, compared with urban drivers, spend more time on the road (55 min per day versus 47 min per day, respectively) and drive longer distances (44 km per day versus 31 km per day, respectively) (Chipman et al., 1993). In an ecologic analysis of United States Federal Highway Administration data, estimates of vehicle miles traveled per capita (in thousands) ranged from 10.0 to 52.5 in rural areas (median of 15.0) and from 5.3 to 12.2 in urban areas (median of 8.0) (Clark and Cushing, 2004). Regardless, even a two-fold urban–rural differential in average exposure in Alberta would not explain the magnitude of the urban–rural differences in crash injury rates that were observed.

In addition to absolute differences in distances traveled, it is important to note that a mile of rural travel is not necessarily equivalent to a mile of urban travel. For example, both crash and crash injury risks are influenced by traffic flow characteristics such as traffic volume, vehicle density and the traffic volume to road capacity ratio (Lord et al., 2005). However, the combined contribution of these factors to road safety has not been well established and the need for separate analyses of urban and rural roadways has been noted (Lord et al., 2005). Similarly, road safety features such as traffic control devices, graded curves, lighting and divided traffic streams are, for the most part, less common in rural compared with urban areas, which likely contributes to urban–rural differences in crash and crash injury risks. For example, in the United States, head-on collisions account for 12% of all rural fatal crashes but only 7% of all urban fatal crashes (National Highway Traffic Safety Administration, 2002). A before-andafter analysis found that the installation of rumble strips (i.e., raised or grooved patterns that produce audible or tactile warnings when crossed by vehicle tires) along the centerlines of undivided, two-lane rural roads was associated with a 14% reduction in all injury crashes and a 25% reduction in head-on and opposing direction sideswipe injury crashes (Persaud et al., 2004).

In an ecological study, which considered police reported crashes and self-reported estimates of vehicle miles traveled in the U.S., fatal crash incidence density (i.e., fatal crashes/vehicle miles traveled) was determined for rural and urban areas (Zwerling et al., 2005). Fatal crash incidence density was calculated as the product of crash incidence density (all crashes/million vehicle miles traveled), the crash injury rate (injury crashes/all crashes) and the injury fatality rate (fatalities/injury crashes). While crash incidence density and crash injury rates were generally similar or slightly higher (rural versus urban) among drivers of both sexes and all ages, injury fatality rates were consistently 2–3.5 times higher in rural areas. Accordingly, the authors suggested that differences in crash and post-crash factors may contribute more to urban–rural differences in fatality risks than differences in driving exposure. For example, in the U.S., about 80% of vehicles involved in fatal rural crashes must be towed from the scene, compared with 66% of vehicles involved in fatal urban crashes (National Highway Traffic Safety Adminis tration, 2002), suggesting higher crash speeds and/or more severe crashes in rural areas. Delayed access to emergency care and inadequate advanced trauma care, known to be predictive of worse outcomes among crash survivors (Grossman et al., 1997; Muelleman and Mueller, 1996), may also contribute to the higher crash fatality rates observed in rural areas. In general, rural (versus urban) crash victims experience longer discovery times, less advanced on-scene care, and longer transport times to hospital (Esposito et al., 1995; Grossman et al., 1997; Svenson et al., 1996a,b). In Alberta, there are only two trauma centers (located in the urban areas of Calgary and Edmonton) to serve the entire province, which spans over 660,000 km2.

Lastly, some data indicate less compliance with road safety behaviors in rural areas. A lower prevalence of seat belt use has been observed in rural areas of Alberta compared with the entire province. In 1999, the prevalence of seat belt use among drivers and front seat passengers of light-duty vehicles was about 69% in rural areas (Alberta Occupant Restraint Program, 2002), compared with 89% for the entire province (Transport Canada, 1999). By 2001, the prevalence in rural areas had increased to 76% (Alberta Occupant Restraint Program, 2002) but remained below the provincial average of 85% (Transport Canada, 2001). While Canadian data specific to children are not available, a U.S. population-based observational survey of restraint use at controlled intersections during daylight hours found that restraint use among child passengers 0–4 years was less common in rural areas compared with urban areas (72% and 96%, respectively) (National Highway Traffic Safety Administration, 2003).

Impaired driving citations are issued at a three-fold higher rate in rural (versus urban) areas in Alberta (Slavik, 1996). A survey of Canadian police officers (Jonah et al., 1999); however, found that 70% of such citations were issued as a result of an impaired driver attracting attention, e.g., by driving erratically or following a collision attended by police. Only 11% were issued during focused enforcement campaigns. Therefore, given the low probability of encountering a police officer while driving, impaired driving citations likely underestimate the true impaired driving rate, and such underestimation is probably greater in remote rural areas compared with urban areas.

A recent survey of Australian drivers showed that 33% admitted to exceeding posted speed limits by 10 km/h or more in urban zones with 60 km/h limits, and 46% admitted to this practice in rural zones with 100 km/h limits (Mitchell-Taverner et al., 2003). Furthermore, drivers residing in some rural areas were more likely than other drivers (15% versus 10%) to report speeds of at least 120 km/h in areas with posted speed limits of 100 km/h (Mitchell-Taverner et al., 2003). These higher speeds likely contribute to the higher injury crash rates in rural areas. A case–control study in rural Australia (Kloeden et al., 2001) compared the pre-crash travel speeds of passenger vehicles involved in severe or fatal crashes with control vehicles (traveling in the same direction, at the same location, time of day, day of week and time of year) and found that travel speeds 10 km/h above the average speed of non-crash vehicles more than doubled the risk of injury crash involvement. Travel speeds 20 km/h above the average speed resulted in a near six-fold increase in risk.

The logistical difficulties and financial constraints faced by police agencies responsible for enforcing traffic laws on low volume rural roads may contribute to these urban–rural differences in unsafe driving behaviors. For example, despite a lower self-reported prevalence of speeding in urban areas, urban drivers in Australia were more likely than rural drivers (20% versus 15%) to have received a citation for speeding in the previous two years (Mitchell-Taverner et al., 2003).

5. Conclusion

This study showed that motor vehicle crash fatality and injury hospitalization rates among children and youth in the province of Alberta were five times higher and three times higher, respectively, in rural compared with urban areas. Possible reasons for the elevated rates observed in rural areas include greater exposure to motor vehicle travel, less safe driving environments, reduced access to advanced trauma care and less compliance with road safety behaviors. There is a need to identify social, demographic and environmental driving hazards associated with the rural environment.

Acknowledgements

The authors would like to thank Ms. L. Owens and Ms. M. Anderson (Transportation Safety Services Division, Alberta Transportation) and Mr. L. Svenson (Alberta Health and Wellness) for their assistance in data collection and interpretation.

References

Alberta Occupant Restraint Program, 2002. Rural seat belt use observational study 1999 and 2001: Summary report. Edmonton, Alberta Centre for Injury Control and Research, Alberta's Regional Health Authorities, Transport Canada (available: http://www.med.ualberta.ca/ acicr/download/2001seat.pdf).

Alsop, J., Langley, J., 2001. Under-reporting of motor vehicle traffic crash victims in New Zealand. Accid. Anal. Prev. 33, 353–359.

Anon., 2000. School transportation modes—Georgia. MMWR Morb. Mortal Wkly. Rep. 51, 704–705.

Aptel, I., Salmi, L.R., Masson, F., Bourde, A., Henrion, G., Erny, P., 1999. Road accident statistics: discrepancies between police and hospital data in a French island. Accid. Anal. Prev. 31, 101–108.

Child Injury Division—Bureau of Reproductive and Child Health, 1999. Canadian Injury Data: Mortality—1997 and Hospitalizations— 1996–97. Ottawa, Laboratory Centre for Disease Control, Health Protection Branch, Health Canada (available: http://www.phac-aspc.gc.ca/ injury-bles/cid98-dbc98/injury e.pdf).

Chipman, M.L., MacGregor, C.G., Smiley, A.M., Lee-Gosselin, M., 1993. The role of exposure in comparisons of crash risk among different drivers and driving environments. Accid. Anal. Prev. 25, 207–211.

Clark, D.E., Cushing, B.M., 2004. Rural and urban traffic fatalities, vehicle miles and population density. Accid. Anal. Prev. 36, 967–972.

Cryer, P.C., Westrup, S., Cook, A.C., Ashwell, V., Bridger, P., Clarke, C., 2001. Investigation of bias after data linkage of hospital admissions data to police road traffic crash reports. Inj. Prev. 7, 234–241.

Esposito, T.J., Maier, R.V., Rivara, F.P., Pilcher, S., Griffith, J., Lazear, S., Hogan, S., 1995. The impact of variation in trauma care times: urban versus rural. Prehospital Disaster Med. 10, 161–166.

Grossman, D.C., Kim, A., Macdonald, S.C., Klein, P., Copass, M.K., Maier, R.V., 1997. Urban–rural differences in prehospital care of major trauma. J. Trauma 42, 723–729.
Hwang, H.C., Stallones, L., Keefe, T.J., 1997. Childhood injury deaths: rural and urban differences, Colorado 1980–8. Inj. Prev. 3, 35–37.
Jonah, B., Yuen, L., Au-Yeung, E., Paterson, D., Dawson, N., Thiessen, R., Arora, H., 1999. Front-line police officers' practices, perceptions and attitudes about the enforcement of impaired driving laws in Canada. Accid. Anal. Prev. 31, 421–443.
King, W.D., Nichols, M.H., Hardwick, W.E., Palmisano, P.A., 1994. Urban/rural differences in child passenger deaths. Pediatr. Emerg. Care 10, 34–36.
Kloeden, C.N., Ponte, G., McLean, A.J., 2001. Travelling Speed and the Risk of Crash Involvement on Rural Roads. Adelaide, Road Accident Research Unit, Adelaide University (available: http://www.atsb.gov.au/ road/rpts/cr204/index.cfm).
Lapidus, G., Lerer, T., Zavoski, R., Banco, L., 1998. Childhood injury in Connecticut. Conn. Med. 62, 323–331.
Lord, D., Manar, A., Vizioli, A., 2005. Modeling crash-flow-density and crash-flow-V/C ratio relationships for rural and urban freeway segments. Accid. Anal. Prev. 37, 185–199.
Mao, Y., Zhang, J., Robbins, G., Clarke, K., Lam, M., Pickett, W., 1997. Factors affecting the severity of motor vehicle traffic crashes involving young drivers in Ontario. Inj. Prev. 3, 183–189.
Mitchell-Taverner, P., Zipparo, L., Goldsworthy, J., 2003. Survey on Speeding and Enforcement. Surry Hills, Taverner Research Company (available: http://www.atsb.gov.au/road/pdf/cr214a final report.pdf).
Muelleman, R.L., Mueller, K., 1996. Fatal motor vehicle crashes: variations of crash characteristics within rural regions of different population densities. J. Trauma 41, 315–320.
National Highway Traffic Safety Administration, 2002. Traffic safety facts, 2002: Rural/Urban Comparison. National Center for Statistics and Analysis, Washington, DC (available: http://www-nrd.nhtsa.dot. gov/pdf/nrd-30/NCSA/TSF2002/2002RuralUrban.pdf).
National Highway Traffic Safety Administration, 2003. Research Note: the Use of Child Restraints in 2002. National Center for Statistics and Analysis, Washington, DC (available: http://www.nhtsa.dot.gov/ CPS/ChildRestraints/ChildRestraints.html).
Niemcryk, S.J., Kaufmann, C.R., Brawley, M., Yount, S.I., 1997. Motor vehicle crashes, restraint use, and severity of injury in children in Nevada. Am. J. Prev. Med. 13, 109–114.
Persaud, B.N., Retting, R.A., Lyon, C.A., 2004. Crash reduction following installation of centerline rumble strips on rural two-lane roads. Accid. Anal. Prev. 36, 1073–1079.
Rosman, D.L., 2001. The Western Australian Road Injury Database (1987–1996): ten years of linked police, hospital and death records of road crashes and injuries. Accid. Anal. Prev. 33, 81–88.
Rosman, D.L., Knuiman, M.W., 1994. A comparison of hospital and police road injury data. Accid. Anal. Prev. 26, 215–222.
Selvin, S., 1995. Poisson regression analysis. In: Practical Biostatistical Methods. Duxbury Press, Toronto, pp. 455–496.
Sirard, J.R., Ainsworth, B.E., McIver, K.L., Pate, R.R., 2005. Prevalence of active commuting at urban and suburban elementary schools in Columbia, SC. Am. J. Public Health 95, 236–237.
Slavik, W., 1996. Alberta Profile: Social and Health Indicators of Alcohol and Drug Abuse, 1996 Update Now Including Regional Health Authorities. Alberta Alcohol and Drug Abuse Commission, Edmonton, Alberta.
Svenson, J.E., Nypaver, M., Calhoun, R., 1996a. Pediatric prehospital care: epidemiology of use in a predominantly rural state. Pediatr. Emerg. Care 12, 173–179.
Svenson, J.E., Spurlock, C., Nypaver, M., 1996b. Factors associated with the higher traumatic death rate among rural children. Ann. Emerg. Med. 27, 625–632.
Transport Canada, 1999. Results of Transport Canada's June 1999 Survey of Seat Belt Use in Canada. Road Safety Programs Branch, Transport Canada, Ottawa (available: http://www.tc.gc.ca/roadsafety/ tp2436/cl9907/menu.htm).
Transport Canada, 2001. Results of Transport Canada's July 2001 Survey of Seat Belt Use in Canada. Road Safety Programs Branch, Transport Canada, Ottawa (available: http://www.tc.gc.ca/roadsafety/ tp2436/rs200107/en/menu.htm).
Zwerling, C., Peek-Asa, C., Whitten, P.S., Choi, S.-W., Sprince, N.L., Jones, M.P., 2005. Fatal motor vehicle crashes in rural and urban areas: decomposing rates into contributing factors. Inj. Prev. 11, 24–28.

kmeturbanruraldifferencesmotor2006.md 35 KB Historik Rå