| NHTSA Report Number DOT HS 809 833 | October 2004 |
The complete report is available here in pdf format.
Part-1 Part-2
The National Highway Traffic Safety Administration (NHTSA) began to evaluate its Federal Motor Vehicle Safety Standards (FMVSS) in 1975. By October 2004, NHTSA had evaluated the effectiveness of virtually all the life-saving technologies introduced in passenger cars, pickup trucks, sport utility vehicles and vans from about 1960 up through the later 1990's. A statistical model estimates the number of lives saved from 1960 to 2002 by the combination of these life-saving technologies. Fatality Analysis Reporting System (FARS) data for 1975-2002 document the actual crash fatalities in vehicles that, especially in recent years, include many safety technologies. Using NHTSA's published effectiveness estimates, the model estimates how many people would have died if the vehicles had not been equipped with any of the safety technologies. In addition to equipment meeting specific FMVSS, the model tallies lives saved by installations in advance of the FMVSS, back to 1960, and by non-compulsory improvements, such as the redesign of mid and lower instrument panels. FARS data have been available since 1975, but an extension of the model allows estimates of lives saved in 1960-1974.
Vehicle safety technologies saved an estimated 328,551 lives from 1960 through 2002. The annual number of lives saved grew quite steadily from 115 in 1960, when a small number of people used lap belts, to 24,561 in 2002, when most cars and light trucks were equipped with numerous modern safety technologies and belt use on the road achieved 75 percent.
EXECUTIVE SUMMARY
The National Highway Traffic Safety Administration (NHTSA) began to evaluate the effectiveness of its Federal Motor Vehicle Safety Standards (FMVSS) in 1975. By October 2004, NHTSA had evaluated virtually all the life-saving technologies introduced in passenger cars and in LTVs (light trucks and vans i.e., pickup trucks, sport utility vehicles, minivans and full-size vans) from about 1960 up through the later 1990s. The agency is now ready to estimate the number of lives saved from 1960 to 2002, year-by-year, by the combination of all these life-saving technologies, and by each individual technology.
Past evaluation reports estimated the effectiveness of a safety technology a percentage reduction of fatalities by statistically analyzing crash data on vehicles produced just before vs. just after a make-model received that technology. Effectiveness, if accurately estimated, should not change much over time. But the benefits of a technology the absolute number of lives saved in a year readily change from year to year depending on the number of vehicles equipped with the technology, their mileage and the crash-involvement rate of the driving population (exposure). This report will:
Heavy
FMVSS: Safety Technologies Cars LTVs Trucks
105: Dual master cylinders & front disc brakes[1] X X
108: Conspicuity tape for heavy trailers X[2]
(201) Voluntary mid/lower instrument panel improvements X X
203/204: Energy-absorbing steering assemblies X X
206: Improved door locks X X
208: Lap belts X X
3-point belts X X
2-point automatic belts X
Voluntary NCAP-related improvements for belted occs. X
Frontal air bags X X
212: Adhesive windshield bonding X X
213: Child safety seats X X
214: Side door beams X X
Voluntary (pre-1994) side impact protection in 2-door cars X
216: Roof crush strength (eliminate true hardtops) X
In addition to safety equipment installed to meet specific FMVSS, the model tallies lives saved by installations in advance of the FMVSS, and by non-compulsory improvements, as shown in the preceding list, such as the redesign of mid and lower instrument panels and modifications to improve performance on the New Car Assessment Program. The model includes car/LTV occupants saved by car/LTV technologies or child safety seats (99 percent of the total), plus pedestrians/bicyclists/ motorcyclists saved by car/LTV brake improvements, and car/LTV occupants saved by conspicuity tape on heavy trailers.
The model does not include technologies so recent that NHTSA has not yet evaluated them based on statistical analysis of crash data, such as the dynamic-test standard for side impact protection (1994-97 phase-in), or head air bags. The study is limited to technologies in cars and LTVs, or that save lives of car/LTV occupants; for example, motorcycle helmets are not included. It is limited to vehicle technologies. It does not estimate the effects of behavioral safety programs (such as the reduction of impaired driving) except, of course, to the extent that programs to increase belt use have contributed greatly to the number of lives saved by belts; roadway and traffic engineering improvements; and shifts in the vehicle fleet e.g., from large to small cars, or from cars to LTVs. The model is limited to estimating fatality reduction by the safety technologies: NHTSA does not have enough building blocks (evaluation results) to develop estimates for the numbers of nonfatal injuries prevented over the years.
How the model works Consider 1,000 cases of driver fatalities in directly frontal multivehicle crashes in cars with 1960 technology: no energy-absorbing steering columns, all drivers unbelted, no air bags. A NHTSA evaluation estimates that energy-absorbing columns reduce fatalities of drivers in frontal crashes by 12.1 percent. Thus, if these cars had been equipped with them, there would have been only 879 fatalities, a saving of 121 lives. Another evaluation estimates that 3-point belts, in cars with energy-absorbing columns, reduce drivers fatality risk by 42 percent in these types of crashes. If the cars had been equipped with 3-point belts in addition to energy-absorbing columns, and the drivers had buckled up, the 879 fatalities would have diminished to 510, saving another 369 lives. A third evaluation estimates that air bags reduce fatality risk by 25.3 percent for belted drivers in these types of crashes, in cars with energy-absorbing columns. Air bags would have cut the 510 fatalities down to 381, saving another 129 lives.
The model uses 1975-2002 FARS data and performs the same calculations in reverse order: e.g., there might be 381 actual FARS cases of 3-point-belted driver fatalities in directly frontal multivehicle crashes in model year 1999 cars, all of which are equipped with air bags and energy-absorbing columns. If air bags, the most recent (1990s) safety technology, had been removed from the cars, fatalities would have increased to 510. In other words, there must have been 129 potentially fatal collisions in these model year 1999 cars that did not become FARS cases because air bags saved the drivers life. If the 3-point belts, a 1970s technology, had also been removed from the cars, and the drivers had been unbelted, the fatalities would have increased to 879. Finally, if the energy-absorbing columns, a 1960s technology, had been replaced by rigid columns, degrading these cars all the way back to a 1960 level of safety, fatalities would have increased to 1,000. The three technologies, in combination, saved 619 lives: 129 by air bags, 369 by 3-point belts and 121 by energy-absorbing columns. In summary, FARS cases of fatalities in vehicles equipped with modern safety technologies constitute evidence of an even larger number of fatalities that would have occurred without those technologies. This approach, based on reverse chronological order is not the only one that could have been used in the model; however, alternative approaches would have generated the same estimate of overall lives saved in 1960-2002, differing only in how they allocated that total among the individual safety technologies.
FARS data have been available since 1975, but the FMVSS date back to January 1, 1968, and some technologies were introduced before that. An extension of the model allows estimates of lives saved in 1960-1974.
Lives saved in 1960-2002 Safety technologies saved an estimated 328,551 lives from 1960 through 2002. Table 1 shows that the annual number of lives saved grew quite steadily from 115 in 1960, when a small number of people used lap belts, to 24,561 in 2002, when most cars and LTVs were equipped with numerous modern safety technologies and belt use on the road achieved 75 percent. (Safety belt use continued to increase after 2002, and reached 80 percent in 2004.)
Figure 1 tracks the benefits of vehicle safety technologies. Fewer than 1,000 lives per year were saved in 1960-67. Starting in 1968, vehicles incorporating most of the safety improvements of the 1960s superseded older vehicles; lives saved reached 4,000 in 1978, but remained at that level for 6 years as belt use temporarily declined. The greatest increase, from 4,835 in 1984 to 11,265 in 1988, came with buckle-up laws. Since 1988, continued increases in belt use, air bags and other recent technologies, and a steadily escalating base of more vehicles and more VMT (vehicle miles of travel) have helped the fatality reduction grow steadily, exceeding 15,000 in 1994 and 20,000 in 2000, reaching 24,561 in 2002.
TABLE 1: LIVES SAVED BY VEHICLE SAFETY TECHNOLOGIES, 1960-2002
(Car and LTV occupants saved,
plus non-occupants and motorcyclists saved by car/LTV brake improvements)
LIVES
CY SAVED
1960 115
1961 117
1962 135
1963 160
1964 203
1965 251
1966 339
1967 509
1968 816
1969 1,179
1970 1,447
1971 1,774
1972 2,226
1973 2,576
1974 2,518
1975 3,058
1976 3,240
1977 3,671
1978 4,040
1979 4,299
1980 4,539
1981 4,455
1982 4,057
1983 4,248
1984 4,835
1985 6,389
1986 8,523
1987 9,973
1988 11,265
1989 11,487
1990 11,711
1991 12,194
1992 12,483
1993 13,796
1994 15,154
1995 16,117
1996 17,813
1997 18,560
1998 19,380
1999 19,942
2000 21,789
2001 22,605
2002 24,561
===========
328,551
FIGURE 1: LIVES SAVED BY VEHICLE SAFETY TECHNOLOGIES, 1960-2002
25000
belt use O
keeps
increasing O
O
+ air bags
20000 + more vehicles O
and VMT O
O
O
O
15000 O
O
O O
O O O
10000 O BUCKLE
O UP
LAWS
O
5000 O
energy O O O O O O
crisis O O declining belt use
O O O
O early non-belt
O O O technologies phase in
O O
0 O O O O O O O
||||||||||||||||||||||||||||||||||||||||||
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2
9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 0 0
6 6 6 6 6 7 7 7 7 7 8 8 8 8 8 9 9 9 9 9 0 0
0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2 4 6 8 0 2
CALENDAR YEAR
Car/LTV occupants: actual fatalities, potential fatalities and percent
saved Among the 328,551 lives saved
in 1960-2002, 326,371 were occupants of cars and LTVs. (The remaining 2,180 were pedestrians,
bicyclists and motorcyclists who avoided fatal impacts by cars or LTVs because
dual master cylinders or front disc brakes improved the car or LTVs braking
performance.) The sum of the actual
fatalities and the lives saved is the number of fatalities that potentially
would have happened if cars and LTVs still had 1960 safety technology and
nobody used safety belts. Table 2 shows
1,443,030 actual car/LTV occupant fatalities in 1960-2002; without the 326,371
lives saved, there would have been 1,796,401 potential fatalities. Actual car and LTV occupant fatalities only
increased from 28,183 in 1960 to 32,737 in 2002. Without the vehicle safety technologies and increases in belt
use, they would have more than doubled, from 28,298 in 1960 to 57,242 in 2002.
Figure 2 compares the trends in actual and potential fatalities. Up to the early 1980s, both trend lines
were fairly close together, and both moved up or down in response to baby
boomers starting to drive (1960s), energy crisis (1970s) and recession (early
1980s). From the mid 1980s, vehicle
safety made a big difference. Potential
fatalities kept rising as registered vehicles and VMT increased in an affluent
society. But increased belt use, air
bags and other vehicle safety technologies held the line on actual fatalities
at about 32,000 a year.
The overall, combined effectiveness of the vehicle safety technologies is the percent of potential fatalities that were saved, as shown in the right column of Table 2. The effectiveness grew in every year from 1960 to 2002, from a humble 0.40 percent in 1960 to a very substantial 42.81 percent fatality reduction in 2002. Figure 3 charts the trend, showing:
TABLE 2: ACTUAL OCCUPANT FATALITIES, POTENTIAL FATALITIES WITHOUT
THE VEHICLE SAFETY TECHNOLOGIES, AND LIVES SAVED IN CARS/LTVs
|
|
Actual |
W/o Safety Techs |
Lives Saved |
Percent Saved |
|
1960 |
28,183 |
28,298 |
115 |
0.40 |
|
1961 |
28,087 |
28,204 |
117 |
0.41 |
|
1962 |
30,544 |
30,679 |
135 |
0.44 |
|
1963 |
32,664 |
32,823 |
159 |
0.49 |
|
1964 |
35,603 |
35,805 |
202 |
0.56 |
|
1965 |
36,518 |
36,767 |
249 |
0.68 |
|
1966 |
39,130 |
39,465 |
334 |
0.85 |
|
1967 |
39,327 |
39,826 |
499 |
1.25 |
|
1968 |
41,019 |
41,818 |
799 |
1.91 |
|
1969 |
42,117 |
43,273 |
1,156 |
2.67 |
|
1970 |
39,556 |
40,972 |
1,415 |
3.45 |
|
1971 |
38,916 |
40,651 |
1,735 |
4.27 |
|
1972 |
40,103 |
42,281 |
2,178 |
5.15 |
|
1973 |
38,739 |
41,258 |
2,520 |
6.11 |
|
1974 |
31,145 |
33,608 |
2,463 |
7.33 |
|
1975 |
31,361 |
34,355 |
2,995 |
8.72 |
|
1976 |
32,222 |
35,398 |
3,176 |
8.97 |
|
1977 |
33,173 |
36,772 |
3,599 |
9.79 |
|
1978 |
34,988 |
38,951 |
3,964 |
10.18 |
|
1979 |
35,108 |
39,325 |
4,217 |
10.72 |
|
1980 |
35,097 |
39,554 |
4,456 |
11.27 |
|
1981 |
33,911 |
38,284 |
4,373 |
11.42 |
|
1982 |
29,855 |
33,834 |
3,979 |
11.76 |
|
1983 |
29,209 |
33,384 |
4,176 |
12.51 |
|
1984 |
30,177 |
34,935 |
4,758 |
13.62 |
|
1985 |
30,044 |
36,357 |
6,314 |
17.37 |
|
1986 |
32,380 |
40,827 |
8,447 |
20.69 |
|
1987 |
33,306 |
43,203 |
9,898 |
22.91 |
|
1988 |
34,217 |
45,407 |
11,190 |
24.64 |
|
1989 |
33,709 |
45,127 |
11,418 |
25.30 |
|
1990 |
32,830 |
44,470 |
11,640 |
26.18 |
|
1991 |
30,928 |
43,060 |
12,131 |
28.17 |
|
1992 |
29,542 |
41,966 |
12,424 |
29.60 |
|
1993 |
30,182 |
43,917 |
13,735 |
31.27 |
|
1994 |
30,979 |
46,075 |
15,096 |
32.76 |
|
1995 |
32,057 |
48,113 |
16,056 |
33.37 |
|
1996 |
32,534 |
50,289 |
17,755 |
35.31 |
|
1997 |
32,501 |
51,003 |
18,502 |
36.28 |
|
1998 |
31,940 |
51,263 |
19,323 |
37.69 |
|
1999 |
32,151 |
52,038 |
19,887 |
38.22 |
|
2000 |
32,234 |
53,968 |
21,734 |
40.27 |
|
2001 |
32,009 |
54,558 |
22,548 |
41.33 |
|
2002 |
32,737 |
57,242 |
24,506 |
42.81 |
|
|
1,443,030 |
1,769,401 |
326,371 |
|
FIGURE 2: ACTUAL VS. POTENTIAL CAR/LTV OCCUPANT FATALITIES
(A = actual fatalities; P = potential fatalities without the vehicle safety technologies)
60000
potential P
55000 fatalities P
P
P P P
50000 P keep rising
P
P
45000 P P P
P P P P
A A P P
P P
40000 P A A P
A P P P
FMVSS start to
baby- A A
A P P make a big difference
boomers A energy P
recession P
35000 start A crisis P P
A A A P A
A P A
A P P A A A A A A
driving A A A A
A A A A A A A
30000
A
A A A A A A
A A
The FMVSS hold the line!
25000
20000
15000
10000
5000
0
1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 2
2
9
9 9 9 9 9
9 9 9 9 9
9 9 9 9 9
9 9 9 9 0
0
6
6 6 6 6
7 7 7 7 7
8 8 8 8 8
9 9 9 9 9
0 0
0
2 4 6 8 0
2 4 6 8 0
2 4 6 8 0
2 4 6 8 0
2