|NHTSA Report Number DOT 807 067||February 1987|
Charles J. Kahane, Ph.D.
Federal Motor Vehicle Safety Standard 207 specifies strength requirements for automotive seats and their attachment assemblies, so as to minimize the possibility of their failure by forces acting on them as a result of vehicle impact. The standard includes a requirement that front seats with folding seatbacks be equipped with a locking device, designed to limit the forward motion of the seatback in a collision and to keep the seatback away from front seat occupants. These seat back locks were installed in domestic two-door passenger cars during 1967-68. The objectives of this preliminary evaluation are to determine if seat back locks are effective in reducing deaths or injuries and to measure the actual cost of the locks. The evaluation is based on statistical analyses of Washington, Texas, New York, Fatal Accident Reporting System and Multidisciplinary Accident Investigation data (with special emphasis on back seat occupants, frontal crashes, and crashes involving occupant ejection or vehicle fire), sled test analyses and a cost study of production lock assemblies. It was found that the locks hold seatbacks in place in crashes when the back seat is unoccupied, but locks or other seat components often separate at moderate crash speeds when there are unrestrained back seat occupants. No statistically significant injury or fatality reductions were found for seat back locks in any of the accident data files or in the sled tests. The locks add about $14 (in 1985 dollars) to the lifetime cost of owning and operating a car.
Federal Motor Vehicle Safety Standard 207, which took effect for passenger cars on January 1, 1968, is one of the National Highway Traffic Safety Administration's initial vehicle regulations. It specifies strength requirements for automotive seats and their attachment assemblies, so as to "minimize the possibility of their failure by forces acting on them as a result of vehicle impact." Many of the specific requirements of Standard 207 are based on the Society of Automotive Engineers' Recommended Practice J879, which had already been in place since November 1963. The one significant difference between Standard 207 and Recommended Practice J879 - the one tangible modification of vehicle seating systems in model years 1967-68 - was the introduction of seat back locks in the folding front seatbacks of passenger cars with two doors.
In a two-door car without seat back locks, the seatback may fold over without restraint in a frontal crash and press on the backs of front seat occupants while they are impacting the steering assembly or instrument panel, possibly increasing the severity of those impacts. The purpose of seat back locks is to limit the forward motion of folding seatbacks in crashes and keep the seatback away from the front seat occupant at the time they impact the steering assembly or instrument panel. Seat back locks are potentially even more important when there are unrestrained back seat passengers in the car (although, actually, this is not a crash situation addressed by Standard 207). If the locks hold the seatback upright and in place, they can help prevent back seat passengers from being thrown into the front half of the compartment, where they might injure a front seat occupant, impact the windshield or instrument panel, or be ejected from the car through open doors or windows.
Current imports as well as older domestic cars have manually operated seat back locks which must be disengaged each time a back seat passenger enters or leaves the car. Current domestic vehicles have an automatic lock which allows the seatback to move freely except during impacts or other sudden decelerations. A possible side effect of manual seat back locks is that they might slow down back seat passengers' egress from the car in emergency situations such as fires or immersion in water.
Executive Order 12291 (February 1981) requires agencies to evaluate their existing regulations. The objectives of an evaluation are to determine the actual benefits - lives saved, injuries prevented, damage avoided - and costs of safety equipment installed in production vehicles in connection with a standard. This report is a preliminary evaluation of seat back locks for two-door passenger cars - i.e., the specific piece of safety equipment installed in connection with Standard 207.
The cost of seat back locks was estimated by analyzing the components of the locks that were actually installed in six production cars. It is estimated that manual seat back locks add $12.83 (in 1985 dollars) to the lifetime cost of purchasing and operating a two-door car. Automatic locks add $14.14 to lifetime cost. In 1985, 3.2 million two-door cars were sold with automatic locks and 1.3 million with manual locks; thus the total consumer cost of the locks was about $62 million.
The effectiveness of seat back locks was estimated by analyzing sled test results and highway accident data. There were 28 frontal sled tests with an average of three instrumented, unrestrained dummies per test: 14 runs with a seat assembly that included seat back locks and 14 matching runs under identical conditions and with the same type of seat, except that the locks were disabled to allow free movement of the seatback as in a pre-standard car. Well over a million accident cases from Washington State, Texas, New York, the Fatal Accident Reporting System, Multidisciplinary Accident Investigation and the National Crash Severity Study were statistically analyzed. The analysis strategy is to calculate the reduction in casualty risk in two-door cars in the model years immediately after the introduction of seat back locks; since many safety devices besides the locks were installed in those years, the reduction in the two-door cars is compared to the corresponding reduction in four-door cars of the same makes, models and model years.
Since four-door cars received the same safety equipment as two-door cars except seat back locks. they act as a control group for the purpose of this evaluation. They are not always a perfect control group, as will be documented in the evaluation.
The effectiveness analysis addresses two questions. Do seat back locks prevent the loading of the seatback on the front seat occupants and retain back seat passengers within the rear half of the passenger compartment? Do seat back locks reduce deaths and injuries? The answer to the first question is fundamentally, "No." Specifically, when there are unrestrained back seat passengers, seat back locks or other seating system components were torn loose in every sled test at 26.5 mph and in a large percentage of frontal crashes of moderate severity on the Multidisciplinary Accident Investigation file:
|Frontal Delta V (mph)||Percent of Cars with Seats Torn Loose|
|Less than 10||6|
|10 - 14.9||21|
|15 - 19.9||35|
|20 - 29.9||47|
|30 or more||88|
At lower speeds, even when seating components were not torn loose, seatbacks were deflected forward to the point where they could not be relied on to keep back seat passengers in-the rear half of the passenger compartment and away from front seat occupants. Seat back locks did perform better when the back seat was unoccupied: all seats remained intact in the sled tests, while 12 percent of seats tore loose at 20-25 mph and 26 percent at 25-30 mph in the accident file. Nevertheless, when the back seat is unoccupied, the sled tests showed that, even without seat back locks, the seatback only makes a minimal contribution to impact forces on front seat occupants.
Under these circumstances, it is unreasonable to expect seat back locks to have much effect on fatality or injury rates. Indeed, no significant casualty reduction was found in the sled tests or any of the accident data files, in spite of the large samples of data analyzed and the strenuous efforts to avoid biases in the analyses. Positive results were not obtained for any seating position (front seat occupants with nobody sitting behind them, front seat occupants with somebody sitting behind them, back seat passengers) or any specific injury type. No benefits were found in frontal crashes, or for restrained occupants, or unrestrained occupants.
One consistent statistical finding was that occupant ejection decreased dramatically in four-door cars in the model years that seat back locks were first installed in two-door cars, but that such a reduction did not take place in the two-door cars. Analyses of the National Crash Severity Study and the Multidisciplinary file suggest that these effects may largely if not entirely be due to changes in door locks and latches that merely coincided with the introduction of seat back locks. These door latch modifications greatly improved door integrity in crashes for four-door cars, but not for two-door cars. In other words, four-door cars are not an appropriate control group for the analysis of ejections. Nevertheless, the possibility cannot be ruled out that seat back locks may slightly increase front seat occupants' risk of ejection, because they prevent seatbacks from folding over and blocking avenues of ejection.
The Fatal Accident Reporting System data at least appear to be consistent with a tentative conclusion that manual seat back locks may quite possibly have increased the risk. as feared, of back seat passengers dying in vehicle fires, but the increase may have been offset by a decrease of front seat occupants' risk in those accidents. None of the results on vehicle fires, however, were statistically significant.
This evaluation is called a "preliminary" one because, statistically speaking, it is difficult to prove definitively that a safety device is not effective. That is because the absence of statistically significant positive findings does not constitute proof that a safety device is completely ineffective. There might still be some narrowly defined crash types where the locks are, in fact, beneficial but there are simply not enough data of that type of crash for a statistically meaningful effectiveness analysis. Specifically, it seems logical that SBL might be beneficial for children riding in the back seat in moderate speed frontal crashes (Delta V 20 mph or less): the SBL may withstand the relatively light impact load of a child and the seatback may retain the child within the safer rear half of the passenger compartment, whereas in a car without SBL, the child might be propelled over the folded seatback and contact the windshield, header, instrument panel or a front seat occupant. Even though the analyses of this evaluation do not show any fatality or injury reductions with SBL for child back seat occupants, the sample size for this limited crash situation is too small for the results to be convincing evidence that, in fact, SBL are not effective there. Similarly, it seems logical that SBL might account for a modest reduction of nonfatal injuries in low to moderate speed frontal crashes (Delta V 15 mph or so), especially when there are adults in the back seat; as above, the available data did not show a significant positive effect but do not preclude the possibility that there is, in fact, a modest positive effect. These crash situations could be possible topics for further study.
What has been accomplished. though. is that a large number of accident cases from various files were analyzed by techniques that are believed to be unbiased. If seat back locks had a positive overall effect even as small as a few percent, it should have appeared in at least some of the analyses, but no such effects were found. The frequent tearing loose of the locks at Delta V over 15 mph when there are back seat occupants and the minimal amelioration of occupant trajectories, as observed in the sled tests, even when the locks remain intact are important additional justifications for a preliminary conclusion, based on available data, that seat back locks have generally not been effective in reducing deaths or injuries.