NHTSA Report Number DOT HS 809 833 |
October 2004 |
Lives Saved by the Federal Motor Vehicle Safety Standards and Other Vehicle Safety Technologies, 1960-2002 - Passenger Cars and Light Trucks - With a Review of 19 FMVSS and their Effectiveness in Reducing Fatalities, Injuries and Crashes
Charles J. Kahane, Ph.D
Abstract
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 1990’s. 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:
- Review the effectiveness estimates in past evaluations of safety technologies for cars and LTVs, describing how they work, and the history of the FMVSS that regulate them.
- Develop a model that uses Fatality Analysis Reporting System (FARS) data and these past effectiveness estimates to calculate how many lives the following technologies have saved, individually and in combination, in each year from 1960 to 2002:
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 (1990’s) 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 driver’s life. If the 3-point belts, a 1970’s 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 1960’s 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 1960’s 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 ˆ
‚ belt use O
‚ keeps
‚ increasing O
‚ O
‚ + air bags
‚
20000 ˆ + more vehicles O
‚ and VMT O
‚ O
‚ O
‚
‚ O
‚
15000 ˆ O
‚
‚ O
‚
‚ O O
‚ O O O
‚
10000 ˆ O BUCKLE
‚
‚ O UP
‚
‚ LAWS
‚ O
‚
5000 ˆ 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 ˆO O O O O O O
Šˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆ||ƒˆƒ
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 LTV’s 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 1980’s, both trend lines were fairly close together, and both moved up or down in response to baby boomers starting to drive (1960’s), energy crisis (1970’s) and recession (early 1980’s). From the mid 1980’s, 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:
- Not much effect before the FMVSS.
- Steady growth in the early-to-mid 1970’s as the early FMVSS phased in.
- A slowdown in 1978-82, when belt use declined prior to national buckle-up campaigns.
- The largest gains came with the buckle-up laws in the mid-to-late 1980’s.
- Steady progress since the late 1980’s thanks to continued increases in belt use, air bags and other recent FMVSS.
TABLE 2: ACTUAL OCCUPANT FATALITIES, POTENTIAL FATALITIES WITHOUT
THE VEHICLE SAFETY TECHNOLOGIES, AND LIVES SAVED IN CARS/LTVs
CY |
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 ˆ
‚
‚ potential P
55000 ˆ fatalities P
‚ P
‚ P P P
50000 ˆ P keep rising
‚ P
‚ P
45000 ˆ P P P
‚ P P P P
‚ A A P P P P
40000 ˆ 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 ˆ 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 ˆ A A A A A A A
‚A A
‚ The FMVSS hold the line!
25000 ˆ
‚
‚
20000 ˆ
‚
‚
15000 ˆ
‚
‚
10000 ˆ
‚
‚
5000 ˆ
‚
‚
0 ˆ
‚
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
FIGURE 3: PERCENT OF POTENTIAL FATALITIES SAVED BY VEHICLE SAFETY TECHNOLOGIES, 1960-2002
45 ˆ
‚
‚ O
‚ more belt use O
40 ˆ O
‚ + air bags O
‚ O
‚ O
35 ˆ O
‚ O
‚ O
‚ O
30 ˆ O
‚ O
‚
‚ O
25 ˆ O O
‚
‚ O
‚ O
20 ˆ
‚ buckle-up
‚ O laws
‚
15 ˆ
‚ O
‚ O
‚ O O O O
10 ˆ O O declining belt use
‚ O O
‚ O early
‚ O non-belt
5 ˆ O FMVSS
‚ O O phase in
‚ O O
‚ O O O
0 ˆO 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
Estimates of lives saved by each FMVSS Car/LTV safety technologies saved an estimated 24,561 lives in 2002, comprising 14,175 car occupants, 10,331 LTV occupants and 56 pedestrians, bicyclists and motorcyclists saved by car/LTV braking improvements. Table 3 shows how many lives were saved by the individual FMVSS – i.e., by the technologies associated with each FMVSS:
- Safety belts are by far the most important occupant protection, saving an estimated 14,570 lives: over half the total of 24,561. The estimate includes safety belts of all types (3-point, lap-only, automatic), at all designated seat positions. Safety belts are designed to keep occupants within the vehicle and close to their original seating position, provide “ride-down” by gradually decelerating the occupant as the vehicle deforms and absorbs energy, and, if possible, prevent occupants from contacting harmful interior surfaces or one another (however, NHTSA recommends correctly installed, age-appropriate safety or booster seats for child passengers until they are at least 8 years old, unless they are at least 4’9" tall). Safety belts are especially important in LTVs, where a large proportion of unrestrained fatalities are ejections and/or rollover crashes; belts saved 6,872 lives in LTVs, nearly two-thirds of the 10,331 LTV occupants saved.
- Frontal air bags saved 2,473 lives in 2002, when 63 percent of cars and LTVs on the road were equipped with driver or dual air bags. Benefits can be expected to grow in future years as the on-road fleet approaches 100 percent air bag equipped. Air bags have significant benefits in frontal and partially frontal impacts for nearly all occupants age 13 and older, including the oldest drivers and passengers, by providing energy absorption and ride-down and by preventing head contacts with the windshield or windshield header. Risk from air bags to child passengers age 12 and younger can be eliminated by riding in the back seat, correctly restrained – or by turning off the on-off switch in pickup trucks where children cannot ride in a back seat correctly restrained.
- Energy-absorbing steering assemblies meeting FMVSS 203 and 204 are an important “built-in” safety technology that saved an estimated 2,657 lives in 2002. In the 1960’s, they were the first basic protection for drivers in frontal crashes, designed to cushion their impact into the steering assembly. Today, the combination of energy-absorbing columns, safety belts and air bags provides far better protection for the driver in frontal crashes.
- Improvements to door locks, latches and hinges, generally implemented by manufacturers in the 1960’s and regulated by industry standards subsequently incorporated into FMVSS 206, saved 1,398 lives in 2002. They reduce the risk of occupant ejection by keeping doors closed in rollover crashes.
TABLE 3: ESTIMATES OF LIVES SAVED BY SAFETY TECHNOLOGIES IN 2002
FMVSS & Safety Technology
|
Car Occupants
|
LTV Occupants
|
Pedestrians
Bicyclists
Motorcyclists
|
TOTAL
|
105: Dual master cylinders & front disc brakes
|
288
|
199
|
56
|
538
|
108: Conspicuity tape for heavy trailers
|
91
|
68
|
|
159
|
201: Voluntary mid/lower instrument panel improvements
|
631
|
299
|
|
930
|
203/204: Energy-absorbing steering assemblies
|
1,660
|
997
|
|
2,657
|
206: Improved door locks
|
704
|
694
|
|
1,398
|
208: Safety belts – all types, all seat positions
|
7,699
|
6,872
|
|
14,570*
|
208: Frontal air bags
|
1,642
|
831
|
|
2,473*
|
212: Adhesive windshield bonding
|
229
|
118
|
|
347
|
213: Child safety seats
|
223
|
112
|
|
335*
|
214: Side door beams & voluntary (pre-1994) TTI(d) reductions
|
848
|
146
|
|
994
|
216: Roof crush strength (eliminate true hardtops)
|
161
|
|
|
161
|
Total
|
14,175
|
10,331
|
56
|
24,561
|
- Side door beams in cars and LTVs meeting the original static crush test of FMVSS 214, plus modifications to 2-door cars in anticipation of the dynamic test requirement later added to FMVSS 214 saved 994 lives in side impacts. Side door beams are primarily effective in side impacts with fixed objects, such as trees or poles.
- Improvements to mid- and lower instrument panels, not actually required by FMVSS 201 but historically and functionally associated with that standard to some extent, saved an estimated 930 right-front passengers in frontal crashes in 2002. Instrument panels were redesigned, using energy-absorbing materials, to decelerate occupants at a safe rate and keep them in an upright position during frontal crashes.
- Car/LTV braking improvements directly or indirectly associated with FMVSS 105 include dual master cylinders and front disc brakes. By eliminating brake failure or helping cars and LTVs stop more effectively, they saved 538 lives in 2002, including 56 pedestrians, bicyclists or motorcyclists
- Adhesive windshield bonding saved 347 lives in 2002 by keeping the windshield attached to the vehicle in severe impacts and preventing occupant ejection via the windshield portal. FMVSS 212 regulates windshield mounting.
- Child safety seats meeting FMVSS 213 saved an estimated 335 young passengers in 2002. Child safety seats are the basic protection system for passengers who are too small to obtain full benefits from safety belts. Newborns should start with rear-facing infant seats, graduate to forward-facing toddler seats, booster seats and, finally, when they are at least 9 years old or 4’9” tall, to adult safety belts.
- FMVSS 216, roof crush strength, is associated with the redesign of true hardtops as pillared hardtops or sedans during the 1970’s. True hardtops had no B-pillars to support the roof, making it more susceptible to crush in a rollover. If cars were still built that way there would have been 161 additional fatalities in 2002.
- FMVSS 108 requires red-and-white conspicuity tape on heavy truck trailers. The tape reflects another vehicle’s headlights strongly and it is highly visible in the dark. Although this device is furnished on heavy trailers, not cars or LTVs, it is the occupants of cars and LTVs who primarily benefit by avoiding collisions with the trailers. The tape saved an estimated 159 car and LTV occupants in 2002.
Table 4 shows cumulative lives saved from 1960 through 2002: 232,255 car occupants and 94,117 LTV occupants, plus 2,180 pedestrians, bicyclists and motorcyclists saved by car/LTV braking improvements, for an estimated total of 328,551. Safety belts (168,524) accounted for more than half the total. Air bags, one of the most recent technologies, had saved 12,074 lives by the end of 2002, and child safety seats, 5,954. The “built in” non-belt technologies regulated by the remaining nine FMVSS in Table 4 (105, 108, 201, 203/204, 206, 212, 214 and 216) add up to 142,000 lives saved; energy-absorbing steering assemblies, improved door locks, and voluntary instrument panel improvements saved the most lives.
TABLE 4: ESTIMATES OF LIVES SAVED BY SAFETY TECHNOLOGIES IN 1960-2002
FMVSS & Safety Technology
|
Car Occupants
|
LTV Occupants
|
Pedestrians
Bicyclists
Motorcyclists
|
TOTAL
|
105: Dual master cylinders & front disc brakes
|
7,993
|
2,880
|
2,180
|
13,053
|
108: Conspicuity tape for heavy trailers
|
683
|
422
|
|
1,105
|
201: Voluntary mid/lower instrument panel improvements
|
16,670
|
4,373
|
|
21,043
|
203/204: Energy-absorbing steering assemblies
|
41,545
|
11,472
|
|
53,017
|
206: Improved door locks
|
19,504
|
9,398
|
|
28,902
|
208: Safety belts – all types, all seat positions
|
109,519
|
59,004
|
|
168,524
|
208: Frontal air bags
|
8,770
|
3,304
|
|
12,074
|
212: Adhesive windshield bonding
|
5,248
|
1,462
|
|
6,710
|
213: Child safety seats
|
4,854
|
1,100
|
|
5,954
|
214: Side door beams & voluntary (pre-1994) TTI(d) reductions
|
14,002
|
701
|
|
14,703
|
216: Roof crush strength (eliminate true hardtops)
|
3,466
|
|
|
3,466
|
Total
|
232,255
|
94,117
|
2,180
|
328,551
|
Comments on some assumptions in the model The effectiveness estimates used in the model derive from past NHTSA evaluations. Estimates were based on statistical analyses of crash data, comparing fatality risk in vehicles built just before and just after make-models were equipped with the technology; the reductions were statistically significant, and the analyses attempted to statistically control for factors other than the technology by using double-pair comparison, control groups, logistic regression, or other techniques. However, in the preparation of this report, the estimates in past evaluations were generally not updated with data that subsequently had become available.
The basic assumption of the model is that FARS fatality cases with a safety technology are evidence of additional crashes where that technology saved lives: if there are 100 belted fatalities in a type of crash where statistical analysis shows 50 percent belt effectiveness, there must have been another 100 people in potentially fatal crashes who were saved by the belt. This is a leap of faith to the extent that we cannot identify those 100 specific crashes were occupants were “saved by the belt” – we assume they must exist, based on our effectiveness estimate.
The model simulates “removing” safety equipment from a modern vehicle one piece at a time, starting with the most recent technology and working backward. Some of these technologies were introduced at about the same time, and it is not always obvious which was first: for some of the earliest ones, there is limited written information, and the people who worked on them have long since retired. A case could be made for changing the order of “removing” the technologies. The model would still produce the same estimate of overall lives saved, but the allocation among the FMVSS could change.
The model assumes that the belt use of fatally injured occupants (not survivors) on FARS is accurately reported. NHTSA has long believed this to be true, based on statistical analyses comparing FARS data with belt use observed in surveys. In the future, conceivably, event data recorders could provide more direct evidence on belt use in crash data files.
Finally, when the model says vehicle safety technology saved 328,551 lives, it means there would have been that many additional fatalities in 1960-2002 if everything else had stayed the same: the same increase in VMT from 1960-2002, the same driving behaviors. It is somewhat of a paper estimate. If safety belts and the other modern vehicle safety technologies had never been invented, and if occupant fatalities had continued climbing toward 57,000 instead of remaining near 32,000, as shown in Table 2, the public might have demanded much stronger regulation of drivers (e.g., licensing) or the infrastructure (e.g., speed limits). Consumers might purchase a different mix of vehicles (e.g., larger cars) and some people might be more reluctant to travel during the riskiest hours (e.g., weekend nights). Those measures might have prevented at least some of the additional 328,551 fatalities – but surely not as efficiently, and with as little impairment of driving enjoyment and mobility as the vehicle safety technologies.
The complete report is available here in pdf format. Part-1 Part-2
Return to Regulatory Evaluation Page
|