- Email: [email protected]
Fatal Motor Vehicle Traffic Accidents Among Native Americans
Fatal Motor Vehicle Traffic Accidents Among Native Americans Martin C. Mahoney, PhD
An atypical number of motor vehicle-related deaths has been observed among Native American populations. Little is known about why Native groups exhibit increased mortality from this cause. To better understand factors influencing these occurrences, I examined fatal motor vehicle traffic accidents between 1980 and 1986 involving persons of Native American descent in New York State, exclusive of New York City. Data were obtained from a review of death certificates and coroner reports. Fatal motor vehicle collisions among Native Americans generally occurred among men (74%), in rural areas (55%),
between 9:00 P.M. and 3:00 A.M. (62%), and during weekends (78%). A sizable percentage of motor vehicle deaths resulted from collisions with pedestrians (28%). Death frequently occurred before the accident victim reached a hospital (74%). Median age at death for male victims was eight years greater than for female victims. These data suggest the need for immediate efforts focusing on primary prevention to minimize risk-taking behaviors likely to result in motor vehicle accidents. I discuss behavioral and environmental factors influencing these observations. [Am J Prev Med 1991;7:112-6]
Although substantial progress has been made in reducing deaths due to various diseases in Native American populations, similar progress has not been made in preventing deaths due to motor vehicle accidents. Higher motor vehicle-related mortality has been reported among Native American groups throughout the United States 1-4 and in Canada.5-7 During recent years, the age-adjusted death rate for motor vehicle accidents among Natives in Indian Health Service Areas has been more than double the overall U.S. rate. 3 Native Americans residing in New York State have also exhibited an atypical number of motor vehicle-related deaths. 8,9 The population in New York State provides an opportunity to study health outcomes among Native Americans. While most Native Americans reside in the southwestern United States, New York State has a sizable number of Native residents and ranks tenth in total Native population. Among states on the East coast, only North Carolina has a larger Native population. Little is known about why Native groups exhibit increased
mortality from motor vehicle traffic accidents (MVTAs). To better understand factors influencing these occurrences, in this article I characterize a series of fatal MVTAs occurring between 1980 and 1986, among persons of Native American descent in New York State, exclusive of New York City.
From the Division of Epidemiology, New York State Department of Health, Albany, New York, and the Department of Epidemiology, School of Public Health, State University of New York at Albany. Address reprint requests to Dr. Mahoney, Division of Epidemiology, Tower 565, New York State Department of Health, Albany, NY 12237-0683.
METHODS I reviewed death files for New York State, except for New York City, to identify all MVTA deaths (of various types, coded E810-E825) which occurred among persons of Native American descent between 1980 and 1986. The high accuracy of racial designation and other information reported on New York State death certificates has been demonstrated. 1 0 Demographic characteristics, cause of death, and any information relating to the MVT A was abstracted from the death certificate of each decedent. Location designation of the fatal accident was based on the population density of the town or village of occurrence. Rural areas were designated with population density of less than 188 persons per square mile. (This dichotomy apportions 20% of the total state population, exclusive of New York City, to rural areas.) I searched New York State Department of Motor Vehicle (DMV) files to document any history of alcohol-related traffic violations. Additionally, I requested coroner reports for all autopsied MVTA deaths.
112 American journal of Preventive Medicine, volume 7, number 2
RESULTS
A total of 57 motor vehicle-related deaths occurred among Native American residents of New York State, except New York City. Men accounted for 73.7% of these deaths. The National Highway Traffic Safety Administration has estimated the total
economic burden to society, including medical care, productivity losses, property damage, legal expenditures, emergency care, coroner/medical examiner costs, and the operation of government programs, resulting from fatal MVT As.11 Based on these figures, society, or more specifically, the Native American population in New York State, except New York City, experienced economic losses exceeding 20.4 million dollars between 1980
Table 1. Frequency of fatal motor vehicle accidents (E81Q-E825) among Native Americans, upstate New York, 1980-1986, by sex for selected variables Men (n = 42)
Variable Accident E812: E813: E814: E815: E816: E818: E819:
type Collision with another motor vehicle Collision with bicyclist Collision with pedestrian Collision with object on highway Resulting from loss of control, without collision on highway Other noncollision Unspecified nature
Women (n = 15)
Total (n
10 3 12 9 6 1 1
5 0 4 2 2 1
15 3 16 11 8 2 2
Age of decedenta 2". 23 years < 23 years
16 26
11 4
27 30
Location of accidentb Rural (less than 188 persons/sq. mile) All other areas (more than 188 persons/sq. mile)
20 20
10 5
30 25
Time of accidenrc 9:00 P.M.-3:00 3:00 A.M.-9:00 9:00 A.M.-3:00 3:00 P.M.-9:00
26 6 5 1
7 3 3 2
33 9 8 3
Day of accident Monday-Thursday Friday-Sunday
12 30
4 11
16 41
Identity of decedent Operator of motor vehicle Passenger in motor vehicle Pedestrian Bicyclist Unknown
14 12 12 3 1
5 5 4 0 1
19 17 16 3 2
Vehicle type (occupants of motor vehicles)d Car Pickup truck Motorcycle
21 4 2
7 3 0
28 7 2
Type of injury sustainede Cerebral trauma Cervical spine trauma Thoracic trauma Abdominal trauma Multiple trauma
20 3 11 2 11
10
30 3 15 3 14
A.M. A.M. P.M. P.M.
0 4 1 3
= 57)
•x2 = 4.18, df =
1; P = .04; odds ratio = 4.46, 95% confidence interval (1.05-20.37). 55, exact accident location unknown for two decedents. 'n = 53, time of accident unavailable for four decedents. dn = 36, excludes 16 pedestrian deaths, 3 bicyclist deaths, and two deaths for which vehicle type was unknown. •Based on injuries listed on death certificate; multiple injury notations were each counted as separate occurrences. bn =
Am] Prev Med 1991;7(2) 113
and 1986, or almost three million dollars annually, as a result of motor vehicle fatalities. As shown in Table 1, the most frequent type of fatal motor vehicle-related accident among Native Americans resulted from the collision of a motor vehicle with a pedestrian (28 .1 % ). Other frequent types of MVTAs involved collisions with another motor vehicle (26.3%) and collisions with other objects on highways (19.3%). During the period of study, the percentage of motor vehicle fatalities due to pedestrian-vehicle accidents was greater among Native Americans (28 .1 % ) than that observed in the general population of New York State, excluding New York City (18.7%). Among men, 16.9% of motor vehicle-related deaths in the general population resulted from pedestrian-vehicle collisions, compared to 28.6% of motor vehicle deaths in Native Americans (x 2 = 4.05, df = l; P = .044), whereas among women these percentages were 19.7% in the general population and 26.6% in Native Americans (x 2 = 0.30, df = l; P = .58). Marked differences in age at death were apparent between Native male and female victims with median age at death among women (17 years) substantially less than the median age at death observed among men (25 years). Male victims were four times more likely to be older than 23 at time of death (odds ratio = 4.4, 95% confidence intervals 1.05-20.37). Median age at death among occupants of motor vehicles was 25 years of age, while median age at death among pedestrians struck by a motor vehicle was 20 years of age. Fatal pedestrian-
vehicle accidents among men occurred at a median age of 22 years, compared to 15 years of age among women. Almost 55% of motor vehicle fatalities among Native Americans occurred in rural areas. This figure compares to 31 % of fatal motor vehicle accidents in the general population occurring in rural areas of New York State, except New York City. Fatal MVTAs among Native Americans occurred more frequently during April (21.1%) and June (12.3%). Sixty-two percent of the fatalities occurred between 9:00 P.M. and 3:00 A.M.; the most frequent hours of occurrence were 10:00 P.M. to 11:00 P.M., 1:00 A.M. to 2:00 A.M., and 2:00 A.M. to 3:00 A.M., with eight fatal MVTAs recorded during each of these time periods. Seventy-two percent of all fatal motor vehicle collisions among Native Americans happened on a Friday, Saturday, or Sunday. A greater percentage of fatalities among motor vehicle occupants (drivers and passengers) occurred on weekends (77.8%), compared to the percentage of fatal pedestrian collisions (56.3%) occurring on these days. Among persons who were occupants of a motor vehicle at the time of the accident, the majority were riding in a car (75.7%), whereas others were occupants of a pickup truck (18.9%) or were riding on a motorcycle (5.4%). Injuries commonly sustained as a result of motor vehicle accidents included cerebral trauma and thoracic trauma, as well as multiple trauma. Deaths resulting from motor vehicle accidents in this population typically occurred immediately following the collision. Almost 74% of Native American motor vehicle colli-
0.50 0.45 0.40 0.35
l
_J w 0.30 > w _J 0.25 _J 0 I 0 (.) 0.20 _J
:::::=:·:·
:.:::·:·:·
<(
0
0 0.15 0_J
••••••••••
c:J
0.10
l!tl+n+++H ±HVifl-I+'+
.f<;;;;>Hfo''4'IH'4:04
0.05 0.00
- DWAl 1
0
0 0
~ bicyclist
0
~
pedestrian
00
~ passenger
driver
Figure 1. Individual blood alcohol levels among Native American motor vehicle traffic accident victims, New York State (excluding New York City), 1980-1986. Based on toxicologic tests at time of autopsy (n=32). Blood alcohol tests were not available for all decedents due to lack of autopsy, lack of toxicologic testing, or unavailability of autopsy results. Assays negative for the presence of ethanol are indicated with zeros. 1 DWAI, driving while ability impaired (0.05%). 2 DWI, driving while intoxicated (0.10%).
114 American Journal of Preventive Medicine, volume 7, number 2
sion victims died before reaching a medical treatment center. Twenty-five persons were declared dead at the accident scene, while 17 of the 32 persons transported to a hospital were pronounced dead upon arrival, reflecting the serious nature of injuries sustained or possible delays in receiving emergency medical care. Alcohol is known to increase risk of injury, including risk of involvement in a fatal motor vehicle collision. Although DMV records do not provide information on alcohol impairment during the fatal episode, the records document any prior history of alcohol-related violations during the previous· 10 years. A total of 52 persons were old enough at time of death (?:: 16 years) to have obtained a license to operate a motor vehicle. Since deceased drivers are routinely purged from DMV files, only deaths occurring between 1983 and 1986 (n = 27) were included in the files at the time of the study. Records were available for 23 persons (85% ). Alcohol-related violations appeared on the records of seven individuals (30% ); all violations occurred among men. Three of the men with a history of alcohol-related traffic violations were killed in pedestrian accidents, with the remaining four dying as a result of motor vehicle collisions. I reviewed coroner reports to obtain information on the results of toxicologic testing for blood alcohol levels (BAC) at the time of the fatal MVT A. Not all victims were autopsied, and toxicology tests were not performed on all autopsied cases. Among Native American MVT A victims for whom BAC analyses were performed (n = 32), 77.3% had detectable levels of blood alcohol (ethanol). Many Native American MVTA victims were legally intoxicated (0.10% BAC) at the time of the fatal accident. Results of BAC testing appear in Figure 1.
DISCUSSION Nearly 55% of all motor vehicle fatalities among New York State Native Americans occurred in rural areas. Rural residence has been associated with various factors thought to increase risk of involvement in fatal motor vehicle accidents, including poor road maintenance, lack of guardrails, excessive rate of speed, and greater use of off-road vehicles.12 Rural residents are also less likely to wear seatbelts 12 and more likely to have an increased number of occupants in each vehicle. 13 The large number of fatalities occurring in rural areas may reflect a larger percentage of Native Americans residing in rural areas. According to U.S. Census figures for 1980, almost 78% of the Native population in New York State, excluding New York City, live in nonurbanized areas.14 Preadmission deaths have been reported to account for between 42% to 50% of all motor vehicle fatalities;I5 however, among this Native group, 73.7% of the victims were declared dead at the scene or upon arrival at a hospital. Since a larger percentage of accidents among Native Americans occurs in rural areas, lack of access to trauma treatment centers may explain the larger number of preadmission deaths observed in this population. Among this group, 43% of the Native victims were declared dead at the accident scene. It is possible that improved access to medical care might have improved the outcomes among the 30% of victims pronounced dead upon arrival at a medical treatment center. Injuries sustained in rural
areas may perhaps be more severe, due in part to greater rates of speed at time of collision and fewer roadside barriers. In this case, death would not correlate with access to medical care since few victims would survive long enough to receive medical treatment. Patterns of injury noted among this Native population replicate patterns of injury noted among other motor vehicle trauma victims, with head injuries a frequent occurrence.1°, 17 Alcohol is thought to be responsible for about 50% of all motor vehicle-related deaths.18-20 Among Native American MVTA victims for whom BAC levels were available, more than 77% had detectable levels of alcohol at the time of the fatal episode. The BAC levels were markedly elevated for many of these victims. An alarmingly large percentage of motor vehicle deaths among Native Americans in New York State result from pedestrian collisions (28.1 % ), an observation consistent with findings that pedestrian-motor vehicle accidents were a major cause of accidental deaths among a southwestern Indian tribe.4 Three individuals with a history of impaired motor vehicle operation were killed in pedestrian accidents; thus the role of alcohol abuse as a contributing factor in Native American pedestrian fatalities cannot be dismissed. However, pedestrian deaths could be reduced by environmental modifications, such as improvements in roadway line of sight, installation of sidewalks and guardrails, improved visibility from the lighting of roadways, and targeted educational programs.21 Determination of median age at death in this investigation for various causes of motor vehicle-related deaths can help target specific age groups for interventions. For example, school health education classes can include modules that (1) warn against the dangers of walking in roadways, (2) educate young drivers about the adverse consequences of excessive speed, (3) teach awareness of road conditions when driving, (4) advocate seatbelt use, (5) highlight bicycle safety rules (e.g. traffic regulations, helmet use, nightime riding), and (6) warn about the risks associated with operating motorcycles and other off-road vehicles. These data suggest a need for immediate efforts focusing on primary prevention. However, the reduction of motor vehiclerelated deaths among Native Americans is a complex issue, which will likely require a multi-faceted approach. Knowledge of the severity of incidence and societal costs resulting from MVTAs needs to be communicated to members of this population group. In addition, these Native groups need to be informed of ways to minimize risk-taking behaviors likely to result in motor vehicle accidents. Efforts to reduce motor vehicle-related deaths among Native Americans have been included in an Indian Health Service (IHS) goal to reduce the rate of motor vehicle fatalities 36% by 1990 (R. Smith, personal communication, July 1988). Health promotion programs developed for Native Americans should focus on increased use of seat belts and child safety seats. 22 As proposed by the IHS, reservation areas should implement curfew restrictions for teenaged motor vehicle operators (R. Smith, personal communication). Increased awareness of risktaking behaviors associated with motor vehicle accidents can be integrated with the promotion of healthy behaviors as a component of the IHS Community Injury Control Programs currently in operation. Activities of this IHS program include an injury prevention-specialist fellowship training program, an
Am j Prev Med 1991;7(2) 115
injury prevention campaign, and programs on pedestrian crash prevention, deterrence of drinking and driving, promotion of seat belt usage, roadway hazard identification, and child passenger protection. In addition, given the availability of financial resources, environmental modifications such as improved road maintenance, curve straightening, and the installation of guardrails and traffic control devices in rural areas may further serve to reduce motor vehicle deaths. This study included only fatalities that resulted from motor vehicle collisions and did not attempt to assess morbidity or societal costs associated with motor vehicle accidents among this population. Whereas this study provides important information on MVTA deaths among Native Americans during a seven-year period, the limited number of fatalities considered may limit the generalizability of these findings. Improvements in surveillance systems may help determine factors associated with survival of MVT As among this group. This study's findings emphasize the need to develop effective strategies to reduce motor vehicle-related deaths among Native American groups. A combination of behavioral and environmental modifications will prove valuable in achieving the IHS goal of reduced motor vehicle deaths among this population. Continued analysis of mortality statistics will aid in further defining risk factors and in measuring progress toward reducing motor vehicle-related deaths among Native Americans.
7. Young TK. Mortality pattern of isolated Indians in northwestern Ontario: a 10-year review. Public Health Rep 1983;98:467-75. 8. Michalek AM, Mahoney MC, Cummings KM, Hanley J, Snyder R. Mortality patterns among a Native American population, 1955-1984. NY State J Med 1989;89:557-61. 9. Mahoney MC, Michalek AM, Cummings KM, Nasca PC, Emrich LJ. Patterns of mortality among a northeastern Native American cohort, 1955-1984. Am J Epidemiol 1989;129:816-26. 10. Carucci PM. Reliability of statistical and medical information reported on birth and death certificates. Albany: New York State Department of Health Monograph No. 15, 1979. 11. National Highway Safety Administration, Office of Planning and Policy. The economic cost to society of motor vehicle accidents, 1986 addendum. September 1987, HS-807 195. 12. Baker SP, Whitfield RA, O'Neill B. Geographic variations in mortality from motor vehicle crashes. New Engl J Med 1987; 316: 1384-7. 13. Regional variation in mortality from motor vehicle accidents. Statistics Bull Metro Insurance Co 1987;68:26-31. 14. U.S. Bureau of the Census. General population characteristics, New York, PC80-1-B34. Washington, DC, 1982. 15. Frey CD, Huelke DF, Gikas PW. Resuscitation and survival in motor vehicle accidents. J Trauma 1969;9:292-310.
REFERENCES
16. Civil ID. Patterns of injury in motor vehicle trauma. N Zealand Med J 1986;99:905-6.
1. Baker SP, O'Neill B, Karpf RS. The injury fact book. Lexington, Massachusetts: Lexington Books, 1984.
17. Hutchinson TP, Harris RA. Recent trends in traffic injury. Injury 1978;10:133-7.
2. New Mexico Monthly Vital Statistics Report, February 1985. State Center for Health Statistics, New Mexico Health and Environment Department, Health Services Division, Vital Statistics Bureau.
18. Haddon W, Baker SP. Injury control. In: Clark DW, MacMahon B, eds. Preventive and community medicine, second edition. Boston: Little, Brown, 1981; 109-40.
3. Trends in Indian Health 1989. Rockville, Maryland: Indian Health Service, U.S. Department of Health and Human Services.
19. Waller JA. Injury: conceptual shifts and preventive implications. Ann Rev Public Health 1987;8:21-49.
4. Brown RC, Gurunanjappa BS, Hawk RJ, Bitsuie D. The epidemiology of accidents among the Navajo Indians. Public Health Rep 1970;80:8818.
20. National Highway Traffic Safety Administration, U.S. Department of Transportation 1985. National accident sampling system 1984, Report DOT HS 866 867. Washington, DC: NHTSA.
5. Hislop TG, Threlfall WJ, Gallagher RP, Band PR. Accidental and violent deaths among British Columbia Native Americans. Can J Public Health 1987;78:271-4.
21. Tranz RR, Christoffel KK. Pedestrian injury: the next motor vehicle injury challenge. Am J Disabled Children 1985;139:1187.
6. Mao Y, Morrison H, Semenciw R, Wigle D. Mortality on Canadian Indian Reserves 1977-1982. Can J Public Health 1986;77:263-8.
22. Sleet DA. Reducing motor vehicle trauma through health programming. Health Educ Q 1984; 11: 113-25.
116 American Journal
of Preventive
Medicine, volume 7, number 2
An atypical number of motor vehicle-related deaths has been observed among Native American populations. Little is known about why Native groups exhibit increased mortality from this cause. To better understand factors influencing these occurrences, I examined fatal motor vehicle traffic accidents between 1980 and 1986 involving persons of Native American descent in New York State, exclusive of New York City. Data were obtained from a review of death certificates and coroner reports. Fatal motor vehicle collisions among Native Americans generally occurred among men (74%), in rural areas (55%),
between 9:00 P.M. and 3:00 A.M. (62%), and during weekends (78%). A sizable percentage of motor vehicle deaths resulted from collisions with pedestrians (28%). Death frequently occurred before the accident victim reached a hospital (74%). Median age at death for male victims was eight years greater than for female victims. These data suggest the need for immediate efforts focusing on primary prevention to minimize risk-taking behaviors likely to result in motor vehicle accidents. I discuss behavioral and environmental factors influencing these observations. [Am J Prev Med 1991;7:112-6]
Although substantial progress has been made in reducing deaths due to various diseases in Native American populations, similar progress has not been made in preventing deaths due to motor vehicle accidents. Higher motor vehicle-related mortality has been reported among Native American groups throughout the United States 1-4 and in Canada.5-7 During recent years, the age-adjusted death rate for motor vehicle accidents among Natives in Indian Health Service Areas has been more than double the overall U.S. rate. 3 Native Americans residing in New York State have also exhibited an atypical number of motor vehicle-related deaths. 8,9 The population in New York State provides an opportunity to study health outcomes among Native Americans. While most Native Americans reside in the southwestern United States, New York State has a sizable number of Native residents and ranks tenth in total Native population. Among states on the East coast, only North Carolina has a larger Native population. Little is known about why Native groups exhibit increased
mortality from motor vehicle traffic accidents (MVTAs). To better understand factors influencing these occurrences, in this article I characterize a series of fatal MVTAs occurring between 1980 and 1986, among persons of Native American descent in New York State, exclusive of New York City.
From the Division of Epidemiology, New York State Department of Health, Albany, New York, and the Department of Epidemiology, School of Public Health, State University of New York at Albany. Address reprint requests to Dr. Mahoney, Division of Epidemiology, Tower 565, New York State Department of Health, Albany, NY 12237-0683.
METHODS I reviewed death files for New York State, except for New York City, to identify all MVTA deaths (of various types, coded E810-E825) which occurred among persons of Native American descent between 1980 and 1986. The high accuracy of racial designation and other information reported on New York State death certificates has been demonstrated. 1 0 Demographic characteristics, cause of death, and any information relating to the MVT A was abstracted from the death certificate of each decedent. Location designation of the fatal accident was based on the population density of the town or village of occurrence. Rural areas were designated with population density of less than 188 persons per square mile. (This dichotomy apportions 20% of the total state population, exclusive of New York City, to rural areas.) I searched New York State Department of Motor Vehicle (DMV) files to document any history of alcohol-related traffic violations. Additionally, I requested coroner reports for all autopsied MVTA deaths.
112 American journal of Preventive Medicine, volume 7, number 2
RESULTS
A total of 57 motor vehicle-related deaths occurred among Native American residents of New York State, except New York City. Men accounted for 73.7% of these deaths. The National Highway Traffic Safety Administration has estimated the total
economic burden to society, including medical care, productivity losses, property damage, legal expenditures, emergency care, coroner/medical examiner costs, and the operation of government programs, resulting from fatal MVT As.11 Based on these figures, society, or more specifically, the Native American population in New York State, except New York City, experienced economic losses exceeding 20.4 million dollars between 1980
Table 1. Frequency of fatal motor vehicle accidents (E81Q-E825) among Native Americans, upstate New York, 1980-1986, by sex for selected variables Men (n = 42)
Variable Accident E812: E813: E814: E815: E816: E818: E819:
type Collision with another motor vehicle Collision with bicyclist Collision with pedestrian Collision with object on highway Resulting from loss of control, without collision on highway Other noncollision Unspecified nature
Women (n = 15)
Total (n
10 3 12 9 6 1 1
5 0 4 2 2 1
15 3 16 11 8 2 2
Age of decedenta 2". 23 years < 23 years
16 26
11 4
27 30
Location of accidentb Rural (less than 188 persons/sq. mile) All other areas (more than 188 persons/sq. mile)
20 20
10 5
30 25
Time of accidenrc 9:00 P.M.-3:00 3:00 A.M.-9:00 9:00 A.M.-3:00 3:00 P.M.-9:00
26 6 5 1
7 3 3 2
33 9 8 3
Day of accident Monday-Thursday Friday-Sunday
12 30
4 11
16 41
Identity of decedent Operator of motor vehicle Passenger in motor vehicle Pedestrian Bicyclist Unknown
14 12 12 3 1
5 5 4 0 1
19 17 16 3 2
Vehicle type (occupants of motor vehicles)d Car Pickup truck Motorcycle
21 4 2
7 3 0
28 7 2
Type of injury sustainede Cerebral trauma Cervical spine trauma Thoracic trauma Abdominal trauma Multiple trauma
20 3 11 2 11
10
30 3 15 3 14
A.M. A.M. P.M. P.M.
0 4 1 3
= 57)
•x2 = 4.18, df =
1; P = .04; odds ratio = 4.46, 95% confidence interval (1.05-20.37). 55, exact accident location unknown for two decedents. 'n = 53, time of accident unavailable for four decedents. dn = 36, excludes 16 pedestrian deaths, 3 bicyclist deaths, and two deaths for which vehicle type was unknown. •Based on injuries listed on death certificate; multiple injury notations were each counted as separate occurrences. bn =
Am] Prev Med 1991;7(2) 113
and 1986, or almost three million dollars annually, as a result of motor vehicle fatalities. As shown in Table 1, the most frequent type of fatal motor vehicle-related accident among Native Americans resulted from the collision of a motor vehicle with a pedestrian (28 .1 % ). Other frequent types of MVTAs involved collisions with another motor vehicle (26.3%) and collisions with other objects on highways (19.3%). During the period of study, the percentage of motor vehicle fatalities due to pedestrian-vehicle accidents was greater among Native Americans (28 .1 % ) than that observed in the general population of New York State, excluding New York City (18.7%). Among men, 16.9% of motor vehicle-related deaths in the general population resulted from pedestrian-vehicle collisions, compared to 28.6% of motor vehicle deaths in Native Americans (x 2 = 4.05, df = l; P = .044), whereas among women these percentages were 19.7% in the general population and 26.6% in Native Americans (x 2 = 0.30, df = l; P = .58). Marked differences in age at death were apparent between Native male and female victims with median age at death among women (17 years) substantially less than the median age at death observed among men (25 years). Male victims were four times more likely to be older than 23 at time of death (odds ratio = 4.4, 95% confidence intervals 1.05-20.37). Median age at death among occupants of motor vehicles was 25 years of age, while median age at death among pedestrians struck by a motor vehicle was 20 years of age. Fatal pedestrian-
vehicle accidents among men occurred at a median age of 22 years, compared to 15 years of age among women. Almost 55% of motor vehicle fatalities among Native Americans occurred in rural areas. This figure compares to 31 % of fatal motor vehicle accidents in the general population occurring in rural areas of New York State, except New York City. Fatal MVTAs among Native Americans occurred more frequently during April (21.1%) and June (12.3%). Sixty-two percent of the fatalities occurred between 9:00 P.M. and 3:00 A.M.; the most frequent hours of occurrence were 10:00 P.M. to 11:00 P.M., 1:00 A.M. to 2:00 A.M., and 2:00 A.M. to 3:00 A.M., with eight fatal MVTAs recorded during each of these time periods. Seventy-two percent of all fatal motor vehicle collisions among Native Americans happened on a Friday, Saturday, or Sunday. A greater percentage of fatalities among motor vehicle occupants (drivers and passengers) occurred on weekends (77.8%), compared to the percentage of fatal pedestrian collisions (56.3%) occurring on these days. Among persons who were occupants of a motor vehicle at the time of the accident, the majority were riding in a car (75.7%), whereas others were occupants of a pickup truck (18.9%) or were riding on a motorcycle (5.4%). Injuries commonly sustained as a result of motor vehicle accidents included cerebral trauma and thoracic trauma, as well as multiple trauma. Deaths resulting from motor vehicle accidents in this population typically occurred immediately following the collision. Almost 74% of Native American motor vehicle colli-
0.50 0.45 0.40 0.35
l
_J w 0.30 > w _J 0.25 _J 0 I 0 (.) 0.20 _J
:::::=:·:·
:.:::·:·:·
<(
0
0 0.15 0_J
••••••••••
c:J
0.10
l!tl+n+++H ±HVifl-I+'+
.f<;;;;>Hfo''4'IH'4:04
0.05 0.00
- DWAl 1
0
0 0
~ bicyclist
0
~
pedestrian
00
~ passenger
driver
Figure 1. Individual blood alcohol levels among Native American motor vehicle traffic accident victims, New York State (excluding New York City), 1980-1986. Based on toxicologic tests at time of autopsy (n=32). Blood alcohol tests were not available for all decedents due to lack of autopsy, lack of toxicologic testing, or unavailability of autopsy results. Assays negative for the presence of ethanol are indicated with zeros. 1 DWAI, driving while ability impaired (0.05%). 2 DWI, driving while intoxicated (0.10%).
114 American Journal of Preventive Medicine, volume 7, number 2
sion victims died before reaching a medical treatment center. Twenty-five persons were declared dead at the accident scene, while 17 of the 32 persons transported to a hospital were pronounced dead upon arrival, reflecting the serious nature of injuries sustained or possible delays in receiving emergency medical care. Alcohol is known to increase risk of injury, including risk of involvement in a fatal motor vehicle collision. Although DMV records do not provide information on alcohol impairment during the fatal episode, the records document any prior history of alcohol-related violations during the previous· 10 years. A total of 52 persons were old enough at time of death (?:: 16 years) to have obtained a license to operate a motor vehicle. Since deceased drivers are routinely purged from DMV files, only deaths occurring between 1983 and 1986 (n = 27) were included in the files at the time of the study. Records were available for 23 persons (85% ). Alcohol-related violations appeared on the records of seven individuals (30% ); all violations occurred among men. Three of the men with a history of alcohol-related traffic violations were killed in pedestrian accidents, with the remaining four dying as a result of motor vehicle collisions. I reviewed coroner reports to obtain information on the results of toxicologic testing for blood alcohol levels (BAC) at the time of the fatal MVT A. Not all victims were autopsied, and toxicology tests were not performed on all autopsied cases. Among Native American MVT A victims for whom BAC analyses were performed (n = 32), 77.3% had detectable levels of blood alcohol (ethanol). Many Native American MVTA victims were legally intoxicated (0.10% BAC) at the time of the fatal accident. Results of BAC testing appear in Figure 1.
DISCUSSION Nearly 55% of all motor vehicle fatalities among New York State Native Americans occurred in rural areas. Rural residence has been associated with various factors thought to increase risk of involvement in fatal motor vehicle accidents, including poor road maintenance, lack of guardrails, excessive rate of speed, and greater use of off-road vehicles.12 Rural residents are also less likely to wear seatbelts 12 and more likely to have an increased number of occupants in each vehicle. 13 The large number of fatalities occurring in rural areas may reflect a larger percentage of Native Americans residing in rural areas. According to U.S. Census figures for 1980, almost 78% of the Native population in New York State, excluding New York City, live in nonurbanized areas.14 Preadmission deaths have been reported to account for between 42% to 50% of all motor vehicle fatalities;I5 however, among this Native group, 73.7% of the victims were declared dead at the scene or upon arrival at a hospital. Since a larger percentage of accidents among Native Americans occurs in rural areas, lack of access to trauma treatment centers may explain the larger number of preadmission deaths observed in this population. Among this group, 43% of the Native victims were declared dead at the accident scene. It is possible that improved access to medical care might have improved the outcomes among the 30% of victims pronounced dead upon arrival at a medical treatment center. Injuries sustained in rural
areas may perhaps be more severe, due in part to greater rates of speed at time of collision and fewer roadside barriers. In this case, death would not correlate with access to medical care since few victims would survive long enough to receive medical treatment. Patterns of injury noted among this Native population replicate patterns of injury noted among other motor vehicle trauma victims, with head injuries a frequent occurrence.1°, 17 Alcohol is thought to be responsible for about 50% of all motor vehicle-related deaths.18-20 Among Native American MVTA victims for whom BAC levels were available, more than 77% had detectable levels of alcohol at the time of the fatal episode. The BAC levels were markedly elevated for many of these victims. An alarmingly large percentage of motor vehicle deaths among Native Americans in New York State result from pedestrian collisions (28.1 % ), an observation consistent with findings that pedestrian-motor vehicle accidents were a major cause of accidental deaths among a southwestern Indian tribe.4 Three individuals with a history of impaired motor vehicle operation were killed in pedestrian accidents; thus the role of alcohol abuse as a contributing factor in Native American pedestrian fatalities cannot be dismissed. However, pedestrian deaths could be reduced by environmental modifications, such as improvements in roadway line of sight, installation of sidewalks and guardrails, improved visibility from the lighting of roadways, and targeted educational programs.21 Determination of median age at death in this investigation for various causes of motor vehicle-related deaths can help target specific age groups for interventions. For example, school health education classes can include modules that (1) warn against the dangers of walking in roadways, (2) educate young drivers about the adverse consequences of excessive speed, (3) teach awareness of road conditions when driving, (4) advocate seatbelt use, (5) highlight bicycle safety rules (e.g. traffic regulations, helmet use, nightime riding), and (6) warn about the risks associated with operating motorcycles and other off-road vehicles. These data suggest a need for immediate efforts focusing on primary prevention. However, the reduction of motor vehiclerelated deaths among Native Americans is a complex issue, which will likely require a multi-faceted approach. Knowledge of the severity of incidence and societal costs resulting from MVTAs needs to be communicated to members of this population group. In addition, these Native groups need to be informed of ways to minimize risk-taking behaviors likely to result in motor vehicle accidents. Efforts to reduce motor vehicle-related deaths among Native Americans have been included in an Indian Health Service (IHS) goal to reduce the rate of motor vehicle fatalities 36% by 1990 (R. Smith, personal communication, July 1988). Health promotion programs developed for Native Americans should focus on increased use of seat belts and child safety seats. 22 As proposed by the IHS, reservation areas should implement curfew restrictions for teenaged motor vehicle operators (R. Smith, personal communication). Increased awareness of risktaking behaviors associated with motor vehicle accidents can be integrated with the promotion of healthy behaviors as a component of the IHS Community Injury Control Programs currently in operation. Activities of this IHS program include an injury prevention-specialist fellowship training program, an
Am j Prev Med 1991;7(2) 115
injury prevention campaign, and programs on pedestrian crash prevention, deterrence of drinking and driving, promotion of seat belt usage, roadway hazard identification, and child passenger protection. In addition, given the availability of financial resources, environmental modifications such as improved road maintenance, curve straightening, and the installation of guardrails and traffic control devices in rural areas may further serve to reduce motor vehicle deaths. This study included only fatalities that resulted from motor vehicle collisions and did not attempt to assess morbidity or societal costs associated with motor vehicle accidents among this population. Whereas this study provides important information on MVTA deaths among Native Americans during a seven-year period, the limited number of fatalities considered may limit the generalizability of these findings. Improvements in surveillance systems may help determine factors associated with survival of MVT As among this group. This study's findings emphasize the need to develop effective strategies to reduce motor vehicle-related deaths among Native American groups. A combination of behavioral and environmental modifications will prove valuable in achieving the IHS goal of reduced motor vehicle deaths among this population. Continued analysis of mortality statistics will aid in further defining risk factors and in measuring progress toward reducing motor vehicle-related deaths among Native Americans.
7. Young TK. Mortality pattern of isolated Indians in northwestern Ontario: a 10-year review. Public Health Rep 1983;98:467-75. 8. Michalek AM, Mahoney MC, Cummings KM, Hanley J, Snyder R. Mortality patterns among a Native American population, 1955-1984. NY State J Med 1989;89:557-61. 9. Mahoney MC, Michalek AM, Cummings KM, Nasca PC, Emrich LJ. Patterns of mortality among a northeastern Native American cohort, 1955-1984. Am J Epidemiol 1989;129:816-26. 10. Carucci PM. Reliability of statistical and medical information reported on birth and death certificates. Albany: New York State Department of Health Monograph No. 15, 1979. 11. National Highway Safety Administration, Office of Planning and Policy. The economic cost to society of motor vehicle accidents, 1986 addendum. September 1987, HS-807 195. 12. Baker SP, Whitfield RA, O'Neill B. Geographic variations in mortality from motor vehicle crashes. New Engl J Med 1987; 316: 1384-7. 13. Regional variation in mortality from motor vehicle accidents. Statistics Bull Metro Insurance Co 1987;68:26-31. 14. U.S. Bureau of the Census. General population characteristics, New York, PC80-1-B34. Washington, DC, 1982. 15. Frey CD, Huelke DF, Gikas PW. Resuscitation and survival in motor vehicle accidents. J Trauma 1969;9:292-310.
REFERENCES
16. Civil ID. Patterns of injury in motor vehicle trauma. N Zealand Med J 1986;99:905-6.
1. Baker SP, O'Neill B, Karpf RS. The injury fact book. Lexington, Massachusetts: Lexington Books, 1984.
17. Hutchinson TP, Harris RA. Recent trends in traffic injury. Injury 1978;10:133-7.
2. New Mexico Monthly Vital Statistics Report, February 1985. State Center for Health Statistics, New Mexico Health and Environment Department, Health Services Division, Vital Statistics Bureau.
18. Haddon W, Baker SP. Injury control. In: Clark DW, MacMahon B, eds. Preventive and community medicine, second edition. Boston: Little, Brown, 1981; 109-40.
3. Trends in Indian Health 1989. Rockville, Maryland: Indian Health Service, U.S. Department of Health and Human Services.
19. Waller JA. Injury: conceptual shifts and preventive implications. Ann Rev Public Health 1987;8:21-49.
4. Brown RC, Gurunanjappa BS, Hawk RJ, Bitsuie D. The epidemiology of accidents among the Navajo Indians. Public Health Rep 1970;80:8818.
20. National Highway Traffic Safety Administration, U.S. Department of Transportation 1985. National accident sampling system 1984, Report DOT HS 866 867. Washington, DC: NHTSA.
5. Hislop TG, Threlfall WJ, Gallagher RP, Band PR. Accidental and violent deaths among British Columbia Native Americans. Can J Public Health 1987;78:271-4.
21. Tranz RR, Christoffel KK. Pedestrian injury: the next motor vehicle injury challenge. Am J Disabled Children 1985;139:1187.
6. Mao Y, Morrison H, Semenciw R, Wigle D. Mortality on Canadian Indian Reserves 1977-1982. Can J Public Health 1986;77:263-8.
22. Sleet DA. Reducing motor vehicle trauma through health programming. Health Educ Q 1984; 11: 113-25.
116 American Journal
of Preventive
Medicine, volume 7, number 2