Trends in Large Truck Crashes

DOT HS 808 690

Large trucks (commercial vehicles weighing more than 10,000 lbs.) have historically made an important contribution to the growth of the national economy by facilitating the distribution of a large portion of the nation's products. However, their contribution to this progress has not come without a price in terms of traffic crashes, injuries, and fatalities.

Because of their size, weight, and the amount of travel, large trucks play a major role in both theoccurrence and consequences of traffic crashes. Large trucks account for about 3 percent of motor vehicles involved in police-reported crashes of all severities, about 8 percent of vehicles in total fatal crashes, andare associated with about 12 percent of the annual total traffic fatality count. Large trucks are alsoestimated to account for about 7 percent of the total number of vehicle miles traveled each year.

This study examines two aspects of trends in the involvement of large trucks in traffic crashes. First,general trends in vehicle registrations, travel, and crashes for large trucks during the years 1975-1995 areexamined and contrasted with other major types of vehicles. Second, using three years of detailed crashdata, trends in the involvement of drivers by age group are studied in an attempt to determine if drivers of acertain age appear to be more prone to being involved in crashes with large trucks, or more likely to die inthese crashes.

The data sources for the study are:

* Driver Licenses - Federal Highway Administration (FHWA), 1993-1995

* Vehicle Registrations - FHWA, R. L. Polk, 1975-1995

* Average Annual Travel - National Personal Transportation Study (NPTS), 1995

* Traffic Crashes - General Estimates System (GES), 1993-1995

* Fatal Crashes and Fatalities - Fatality Analysis Reporting System (FARS), 1975-1995

General Trends in Large Trucks

Motor vehicle travel has increased steadily over time due to increases in both the number of vehicles andthe amount of average annual travel. Between 1975 and 1995 the number of registered vehicles hasincreased from 126,153,000 to 197,065,000, an increase of 56 percent. The amount of total vehicle milestraveled during the same period has also increased, from approximately 1,328 billion vehicle miles of travel(VMT) to 2,403 billion miles, an increase of 81 percent. While both counts have increased gradually overtime, the changes have not been uniform across the major classes of vehicles, i.e., passenger cars, lighttrucks and vans, motorcycles, and large trucks. Total traffic fatalities and large truck-related fatalitieshave also experienced changes during the same period of time, but the changes do not follow the samepattern of increases found in the number of vehicles registered and their annual travel.

These changes are presented in more detail in the following sections of the paper for two specific purposes. The first purpose is to document how changes in large truck registration and travel compare to those ofother major types of vehicles, while the second purpose is to document what changes have occurred in thenumber and type of people killed in crashes involving large trucks during the same period. The results willfocus on the changes in both the total fatality toll in large truck crashes and the partitioning of fatalitiesbetween the occupants of the truck and the occupants of other vehicles.

Vehicle Registrations

The number of total registered vehicles has grown steadily during the 1975-1995 period, from 126 millionin 1975 to about 197 million in 1995, a 56 percent increase. Table 1 presents historical data on vehicleregistration for the four major classes of vehicles, i.e., passenger cars, light trucks and vans, motorcycles,and large trucks. It is evident from this table that the growth pattern has not been the same for these fourclasses of vehicles.

The number of passenger car registrations increased from 95 to 123 million between 1975 and 1995, anincrease of 30 percent. The increase in passenger car registrations has been fairly uniform over the yearsand at a lower rate than for all vehicles combined. Passenger cars accounted for about 75 percent of allregistered vehicles in 1975, but their proportion has declined to about 63 percent of all registered vehiclesin 1995.

The class of vehicles which has shown the largest increase is that of light trucks and vans. About 20million light trucks and vans were registered in 1975 and this has increased to over 62 million registrationsin 1995, a three-fold increase. Total registrations for passenger cars and light trucks and vans combinedincreased by 60 percent during the 21-year period, which matches the growth of all vehicles combined.

Large truck registrations were at the 5.4 million level in 1975, increased to 5.9 million by 1979, declinedsteadily to 5.3 million by 1986, and have increased by over 30 percent in the last 10 years, to a total of 6.9million. This last increase is almost twice as large as for all other classes combined.

Motorcycle registrations follow yet another pattern. From the almost 5 million level in 1975, motorcycleregistrations increased to over 5.8 million in 1981. The number of registered motorcycles has declinedsince 1981 and reached the 3.7 million level in 1995, a 35 percent decline during this last period.

Table 1 VEHICLE REGISTRATIONS, 1975 - 1995

Vehicle Miles Traveled

The amount of vehicle miles traveled for each class of vehicle is estimated and published by FHWA eachyear. Table 2 presents a summary of the travel estimates for the four classes of vehicles for the period 1975-1995. As seen in Table 2, all classes of vehicles experienced appreciable increases in the number ofvehicle miles traveled each year. Passenger car, light truck and van, and large truck travel increasedthrough the years but at different rates. For example, during the 21-year period, vehicle miles traveledincreased by about 50 percent for passenger cars, more than doubled for large trucks, and more than tripledfor light trucks and vans. Motorcycle travel increased by about 70 percent during the period.

Table 2 - VEHICLE MILES OF TRAVEL, 1975-1995 (in Millions)

YEAR	PASSENGER CARS	LIGHT TRUCKS AND VANS	MOTORCYCLES	LARGETRUCKS
1975	1,030,376	204,274	5,629	81,330
1976	1,070,667	233,382	6,003	86,070
1977	1,102,726	257,108	6,349	95,021
1978	1,136,459	289,463	7,158	105,731
1979	1,111,705	293,840	8,637	109,004
1980	1,107,056	295,475	10,214	108,491
1981	1,120,126	307,044	10,690	108,702
1982	1,149,375	323,022	9,910	106,880
1983	1,190,076	335,590	8,760	113,163
1984	1,224,812	358,106	8,784	123,927
1985	1,245,837	387,800	9,086	126,580
1986	1,274,668	415,593	9,397	130,141
1987	1,326,907	443,872	9,506	135,601
1988	1,381,270	487,450	10,024	141,397
1989	1,411,131	520,977	10,371	148,318
1990	1,424,615	554,661	9,557	149,810
1991	1,410,934	595,619	9,178	150,729
1992	1,436,449	642,583	9,557	152,803
1993	1,445,314	675,450	9,906	159,904
1994	1,460,673	712,229	10,251	170,415
1995	1,499,643	668,341	9,531	173,326
Percent Change 1975-1995	+45%	+227%	+69%	+113%

Large Truck Fatalities

Since 1975, the National Highway Traffic Safety Administration has maintained a reporting system forfatal traffic crashes, the Fatality Analysis Reporting System (FARS). FARS contains data on a census offatal traffic crashes within the 50 states, the District of Columbia, and Puerto Rico (although Puerto Ricois not included in the national totals). To be included in FARS, a crash must involve a motor vehicletraveling on a trafficway customarily open to the public, and must result in the death of an occupant of avehicle or a nonmotorist within 30 days of the crash. Tables 3, 4, and 5 contain FARS data on theinvolvement of large trucks in fatal crashes.

Table 3 - TRAFFIC FATALITIES, 1975 - 1995

Table 3 contains the number of all traffic fatalities for the years 1975 - 1995 along with the number of fatalities in crashes involving a large truck. The fourth column in Table 3 represents the percentage of total fatalities represented by large truck crashes. From Table 3, it can be seen that the number of fatalitiesin large truck crashes appears to be highly correlated to the total number of fatalities, and that both countsshow a poor relationship with their respective travel estimates. This seems to indicate that the trend inlarge truck-related fatalities, which account for approximately 12 percent of the total fatalities, hasfollowed the same pattern as the trend for total fatalities, thus indicating little relative change since 1975.

Table 4

LARGE TRUCK OCCUPANT FATALITIES, 1975 - 1995

Table 4 and Figure 1 reflect the number of large truck occupant fatalities. Fatalities in Table 4 areseparated into two groups, those occurring in single-vehicle crashes and those in crashes in which anothervehicle was involved. Table 4 indicates that, after a rise in the late seventies, the number of large truckoccupant fatalities has decreased steadily. The number of large truck occupant fatalities decreased from1,432 in 1979 to 644 in 1995. This decrease is substantial in comparison to the smaller change in the totalnumber of truck-related fatalities during the same period. For the entire period, 1975-1995, twice as manytruck occupant fatalities occurred in single-vehicle crashes as in crashes with other vehicles. Given thelarge change in the number of total truck occupant fatalities over the years, it is surprising that this ratiohas remained almost constant.

The finding that large truck occupant fatalities have declined while no appreciable change has occurred inthe total number of truck-related fatalities, has led to the compilation of Table 5, which presents thenumber of fatalities in crashes between a large truck and another vehicle. Large truck occupant fatalitiesresulting from collisions involving more than one large truck have been excluded. By separating thesefatalities into two groups, large truck occupants and other vehicle occupants, it is possible to determinehow the ratio of these two counts has changed over the years.

Table 5 - OCCUPANT FATALITIES IN

LARGE TRUCK-OTHER VEHICLE CRASHES, 1975 - 1995

From Table 5, it can be seen that the ratio of occupant fatalities in other vehicles to large truck occupantfatalities has changed significantly, from a value of 16 in 1975 to 30 or greater since 1990. The increase inthis ratio is entirely due to a decrease in the number of truck occupant fatalities, since fatalities in the othervehicles have not changed appreciably.

Finally, Figure 2 suggests that the ratio did not increase gradually but appears to have occurred at twospecific points in time, i.e., in 1980 and again in 1990. While it is not the purpose of this paper to explainwhy this change occurred, one might speculate that it might be due to an increased disparity in the size andweight of the two types of vehicles, and to the better protection for the truck driver due to improved crashworthiness of the tractor.

Figure 2

Trends in Large Truck Crashes by Age of Driver

This part of the study focuses on drivers of passenger vehicles (passenger cars, light trucks, and vans)involved in collisions with another vehicle, with focus on large trucks. These collisions are separated intotwo groups: crashes between passenger vehicles and large trucks are in the first group, and multi-vehiclecrashes not involving a large truck are placed in the second group. The second group includes mostlycollisions between passenger vehicles. Data from the three most recent years available, 1993 - 1995, havebeen compiled to determine if differences exist among groups of drivers of passenger vehicles in these twotypes of crashes, by the age of the driver.

The questions being asked are:

* Are drivers of a certain age more likely to be involved in a large truck crash?

* Does the age of the driver appear to have any bearing on the fatal consequences of large truck crashes?

Both questions are addressed by comparing the statistics of the two groups of crashes during the period1993 - 1995. Three years of data are used in order to obtain more stable results, especially for those agegroups with smaller numbers of licensed drivers. The 1995 NPTS survey is the source of drivers' annualtravel and estimates derived from that survey have been utilized in compiling travel estimates for 1993through 1995.

In analyzing crashes between passenger vehicles and large trucks, it is necessary to provide a reasonablebasis for comparison. Since passenger car crashes with large trucks by definition involve two or morevehicles, a logical contrasting counterpart would be all multiple-vehicle crashes that do not involve a largetruck. By far the largest proportion of these crashes involve only passenger vehicles colliding with oneanother. This study, therefore, analyzes only drivers of passenger vehicles and separates them into twogroups. One group consists of those drivers involved in crashes with "non-large-truck" vehicles, mostlydrivers involved in two passenger vehicle crashes. The other group consists of drivers of passengervehicles involved in crashes with large trucks.

Four basic measures are examined in this portion of the study:

Age distribution of drivers in police-reported crashes;

Age distribution of driver fatalities;

Driver Fatality Rate (per crash) by age group of driver; and

Driver Fatality Rate (per VMT) by age group of driver.

By comparing these measures for the two groups of crashes, it may be possible to determine whetherdifferences exist among the various age groups of drivers in both the occurrence and the fatal consequencesof these two types of crashes.

Large Truck Crashes and Fatalities

In analyzing crash and fatality data, it is important to take into account any differences that may exist inthe amount and type of travel performed by various groups of drivers. In this case, it would by helpful toknow whether some driver age groups of passenger vehicles are more likely to share the same drivingenvironment as large trucks. Unfortunately, no such measure is available that would give some indicationof disparity, among drivers of passenger vehicles, in the likelihood of experiencing a crash with a largetruck. The results that follow are based only on crash data, and could, therefore, change if measures ofexposure to large trucks become available.

Data from NHTSA's General Estimates System (GES) are used to study possible differences in driverinvolvement in large truck crashes by age of the driver. GES, operational since 1988, collects data using anationally representative probability sample of all police-reported crashes. To be eligible for inclusion inthe GES sample, a police accident report (PAR) must be completed for the crash and the crash mustinvolve at least one motor vehicle traveling on a trafficway and result in property damage, injury, or death. As the data in GES are obtained from a sample of all police-reported crashes, the GES statistics areestimates of persons injured and injury crashes and are subject to sampling and nonsampling errors. ( Formore information, see National Accident Sampling System GES Technical Note, DOT HS 807-796.)

GES data for the period 1993 - 1995 estimate that approximately 26 million drivers of passenger vehicleswere involved in collisions with another passenger vehicle. An additional 900,000 drivers of passengervehicles were involved in collisions with large trucks during this period. Table 6 contains the number ofpassenger vehicle drivers involved in the two groups of crashes together with their respective agedistributions, based upon GES. Younger drivers, i.e., 16 to 20 years of age, account for a significantlylower percentage of the drivers involved in large truck crashes as compared to other multiple-vehiclecrashes. Drivers aged 75 and above account for the same proportion in both groups of crashes.

Drivers of Passenger Vehicles Involved in Crashes

(GES 1993, 1994, 1995 Combined)

Figure 3 presents a graph of the number of drivers involved in the two types of> crashes. (The scale fordrivers in all crashes has been reduced by a factor of 30 in order to make the comparison between the twosets of counts easier to interpret.) A ratio of the two counts is also shown in Figure 3, which could beinterpreted as the odds of a driver of a passenger vehicle being involved in a collision with a large truck. The ratio is lowest for the youngest group of drivers, rises steadily until age 45, remains at that valuethrough age 74, and drops for older drivers.

Figure 3

Table 7

Driver Fatalities in Passenger Vehicles

(GES 1993, 1994, 1995 Combined)

Figure 4 shows the number of driver fatalities in the two types of crashes. The ratio of the two, alsoshown, reflects the odds of a passenger vehicle driver being killed in a collision with a large truck. Theratio is at the highest level, 0.34, for drivers between the ages of 16 and 64 combined, and declines steadilyfor the older groups, reaching a value of 0.23 for drivers 80 and over. These statistics on driver crashesand fatalities indicate that the youngest and the oldest age groups of drivers are, relative to the remainingage groups, involved in a lower proportion of large truck crashes, and that drivers older than 64 years ofage account for a lower proportion of fatalities in large truck collisions than they do in crashes with anotherpassenger vehicle.

Large Truck Fatality Rates and Driver Age

Another way to assess the effect of driver age in large truck crashes is by comparing the risk of a driverfatality in the two types of crashes. There are various ways of determining fatality risk depending on thechoice of measure. In this case, two measures of risk are considered, i.e., the number of fatalities per crashand the number of fatalities per vehicle miles traveled (VMT).

The fatality rate per crash is provided by the ratio of the number of fatalities from Table 7 and the numberof drivers involved from Table 6. The rate obtained for each of the two groups of crashes shows a clearpattern for the age of involved drivers. For both groups, the fatality rate per crash remains almost flat forall ages under 55 and increases rapidly for older drivers. For the purpose of this study the focus is on theratio of the fatality rate in crashes with large trucks vs. crashes with another passenger vehicle. In general,it is known that the fatality risk varies by driver age in both types of crashes, and that the fatality risk forthe driver of a passenger vehicle is greater, about nine times as great, in crashes with a large truck as incollisions between passenger vehicles. What is of interest here is to determine if this increased risk isconstant across all age groups.

Figure 5 shows how the fatality rate per crash increases rapidly for drivers in the older age groups. Thisincrease, although at different levels, is present in both types of crashes. The ratio of the two rates, alsoshown in Figure 5, provides a measure of the greater severity of crashes with large trucks for each agegroup. The results show that crashes with large trucks are 13 times as lethal for the youngest group ofdrivers, and that the disparity decreases gradually to a value of less than 7 for the oldest drivers. Thisindicates that the risk of being fatally injured in large truck crashes, as compared to collisions betweenpassenger vehicles, increases by a greater amount for the younger drivers than for the older group. Theseresults also show that the risk of being fatally injured in a crash increases with the age of the driverinvolved, but the presence of a large truck in the crash results in a greater increase in the fatality risk foryoung drivers than the much older ones.

Figure 5

The other measure or risk, fatality rate per VMT, requires the use of travel estimates over the three yearperiod. Table 8 contains these estimates based on the 1995 NPTS survey, combined with driver licensingdata. The analysis yields a pattern for the ratio which, although different than that shown in Figure 5, stillconfirms that the increase in the fatality risk for large truck crashes is not as large for the older drivers asfor younger> drivers. Figure 6 indicates that the ratio of the two rates remains at the same level for driversless than 65 years of age and drops gradually for older groups.

Table 8 - Estimated (1993-1995) Travel by Driver Age

(Derived from NPTS, 1995)

Figure 6

These results support two major conclusions. The first states that older drivers do not seem to be overinvolved in collisions with large trucks. This may be due to the possibility that older drivers, because oftheir driving patterns, are less exposed to conflicts with large trucks. Older drivers may be less likely toshare the same driving environment of drivers of large trucks, i.e. type of highways, time of day, etc. Thesecond conclusion is that the increase in the risk of fatal injury, per crash and per VMT, to drivers ofpassenger vehicles, when they collide with a large truck as compared to when they collide with anotherpassenger vehicle, is considerably lower for older drivers than for much younger drivers. This result doesnot reflect a lower risk for older drivers in large truck collisions, it reflects only the fact that the greatersurvival probability that young drivers have relative to older drivers in passenger vehicle collisions isappreciably reduced in collisions with large trucks.

Summary of Findings and Conclusions

Large trucks account for about 3.5 percent of all vehicles and for approximately 7 percent of all motorvehicle travel. Large trucks account for about 3 percent of all vehicles involved in crashes, are involved inless than 6 percent of all reported crashes in which 12 percent of all traffic fatalities occur.

The trend in the number of fatalities in crashes involving large trucks has followed essentially the samepattern as the trend for total traffic fatalities. These crashes have accounted for essentially the samepercentage of the total fatalities during the 1975-1995 period (about 12 percent).

The trend in the number of fatalities in crashes involving large trucks does not appear to be similar to thetrend in travel for large trucks. Large truck travel has more than doubled during the 1975 - 1995 periodwhile the number of truck-related fatalities has not changed appreciably during the same period.

Although the number of large truck occupant fatalities has varied from a high of 1,432 in 1979 to a low of585 fatalities in 1992, almost two thirds of these fatalities occurred in single-vehicle crashes in each yearduring the 1975 - 1995 period.

In crashes between passenger vehicles and large trucks, the ratio of occupant fatalities has increased from16 to 1 in the late seventies to about 32 to 1 in the early nineties. This change reflects only minor changesin passenger vehicle occupant fatalities but an almost 50 percent reduction in the number of large truckoccupant fatalities. This change could possibly be associated with the decrease in the size and weight ofpassenger vehicles, and to improvements to the crashworthiness of the truck cab.

In assessing the effects of the age of drivers of passenger vehicles involved in large truck crashes, it isnecessary, when selecting a basis for comparison, to take into account that these are multiple-vehiclecrashes, and therefore should be compared only to other types of multiple-vehicle crashes.

In comparing the age distribution of drivers involved in crashes with large trucks to the distribution of thoseinvolved in crashes among passenger vehicles, the results show that younger drivers are under-representedin large truck crashes while the remaining age groups show no differences.

Being involved in a collision with a large truck increases the probability of fatal injury to the driver of apassenger vehicle by nine fold, on the average, when compared to collisions between passenger vehicles. This increase in risk is not uniform across all ages. The number of fatalities per crash is 13 times as greatfor the youngest age group of drivers in large truck crashes and 7 times as great for the oldest group than inmultiple vehicle crashes between passenger vehicles. The ratio of the fatality rate per vehicle miles oftravel, between the two types of crashes, also declines for older age groups.

These findings on driver age and involvement in crashes with large trucks are based on crash data andgross estimates of travel and do not take into account that older drivers, because of their driving patterns,may encounter relatively fewer large trucks in their travel. Furthermore, crashes of older drivers with largetrucks may occur under more forgiving conditions, i.e., lower speeds, daytime hours, urban settings, etc. More detailed studies, based on finer measures of exposure, are needed in order to better understand theunderlying statistics on large truck crashes.

For copies of the full report, please call (202) 366-4198 or toll free, 1-800-934-8517. Questions regarding the report may be directed to Ezio Cerrelli at (202) 366-5358.