Regulatory Evaluation | National Highway Traffic Safety Administration(NHTSA)

Federal Motor Vehicle Safety Standard 214, “Side Impact Protection” was amended to assure occupant protection in a 33.5 mph crash test and phased-in to new passenger cars during model years 1994-1997. A Thoracic Trauma Index, TTI(d) is measured on Side Impact Dummies seated adjacent to the impact point. Manufacturers upgraded side structures and affixed padding in cars to improve TTI(d). Later, they installed two types of side air bags – torso bags and head air bags – for additional occupant protection in cars and LTVs. Statistical analyses of 1993-2005 crash data from the Fatality Analysis Reporting System (FARS) and the General Estimates System (GES) estimate fatality reductions for these technologies.

TTI(d) improvements, torso bags and head-curtain air bags could have saved an estimated 2,934 lives in calendar year 2003 if every car and LTV on the road had been equipped with them.

Side air bags with head protection, such as torso bags with head curtains reduce fatality risk in side impacts by an estimated 24 percent for the nearside occupant, the person seated adjacent to the struck side of the vehicle. That benefit adds to the effect of improved side structures and padding built into passenger cars during the 1980s and 90s that had already reduced fatality risk for nearside occupants by 33 percent in 2-door cars and 17 percent in 4-door cars.

In 2003, over 9,000 fatalities, approximately 29 percent of all occupant fatalities in cars and LTVs (light trucks and vans – i.e., pickup trucks, sport utility vehicles, minivans and full-size vans) began with a side impact. The side of a vehicle, especially the door area adjacent to the occupant is intrinsically a vulnerable spot: there is limited space and structure between the occupant and the outside. Side impacts can also be difficult to avoid. Even the most prudent driving on our part cannot eliminate the risk that another vehicle will fail to yield, run a red light or turn without warning across our path.

Since the 1970’s, the National Highway Traffic Safety Administration (NHTSA), the manufacturers and others in the safety community have worked hard to reduce fatality risk in side impacts, especially for the most vulnerable occupant, the “nearside” occupant: the driver in a left-side impact and the right-front passenger in a right-side impact. The effort resulted in the four tangible improvements in side impact protection that are evaluated in this report:

1. Upgrading the side structure of passenger cars to slow down and reduce the extent of door intrusion into the passenger compartment after a side impact. Improvements include redesigning or strengthening the beams that horizontally reinforce the doors; the pillars, sills, and roof rails that surround the doors; and the cross-members or seat structures that resist lateral crush.

2. Installation of thick, energy absorbing padding within the door structure to reduce the probability of occupant injury after the door interior contacts the occupant.

3. Torso air bags that deploy from the seat or the door to provide an energy-absorbing cushion between the occupant’s torso and the vehicle’s side structure. Torso air bags cover a much larger impact area and absorb more energy than padding.

4. Head-protection air bags that complement the torso bags by cushioning head impacts with the side structure and possibly barring occupant ejection through side windows. Head protection may consist of:

a. “Torso/head combination bags” that deploy from the seat to protect the torso but also extend upward far enough to protect the head impact zones around the side window, or

b. “Head curtains” or “inflatable tubular structures” that drop down from the roof rail into the side-window area, separately from the torso bags.

During the 1980’s, NHTSA and the safety community developed a procedure for assessing injury risk in side impacts, including:

· A crash test configuration simulating a severe intersection collision in which a fast-moving vehicle strikes a slow-moving vehicle in the door, at a right angle.

· A Thoracic Trauma Index (TTI) that predicts the severity of thoracic injuries when occupants’ torsos contact the interior side surface of the struck vehicle.

· A Side Impact Dummy (SID) on which TTI can be reliably measured in side impact tests. The injury score measured on the dummy is called TTI(d).

In 1990 NHTSA amended Federal Motor Vehicle Safety Standard (FMVSS) 214, Side Impact Protection for passenger cars, adding a 33.5 mph impact by an MDB into the side of the car and limiting TTI(d) for a SID in the nearside position up to a maximum of 90 in 2-door cars and 85 in 4-door cars. The requirement was phased-in to passenger cars during model years 1994 to 1997 and subsequently extended to LTVs, effective in model year 1999, limiting TTI(d) to 85.

The manufacturers redesigned structures and/or affixed padding to substantially reduce average TTI(d) during and, to some extent, even before the 1994-1997 phase-in of FMVSS 214. But their actions varied from model to model. Many 2-door cars, with their long, vulnerable door areas, received extensive structural reinforcement or other redesign, whereas some of the heavier 4-door cars and most LTVs needed little or no change to meet FMVSS 214. In many cars, manufacturers improved TTI(d) well beyond the NHTSA requirements.

Manufacturers have continued to improve side impact protection by installing side air bags and/or upgrading side structures as they redesigned their cars. Torso bags first appeared on production vehicles in 1996 and head-protection air bags in 1998. By model year 2003, nearly 30 percent of new cars were equipped with torso bags and nearly 20 percent with head-protection air bags. NHTSA does not require side air bags, but encourages all improvements to side impact protection, including side air bags, by informing consumers about the performance of new vehicles. The agency’s New Car Assessment Program (NCAP) includes a rating system of one star (worst) to five stars (best) on a side impact test. Buying a Safer Car brochures specify what make-models are equipped with torso and/or head air bags. The information is available to consumers on the agency’s web site, www.safercar.gov.

TTI(d) performance at the 33.5 mph test speed of FMVSS 214 demonstrates how much cars have improved over the years. In 2-door cars, TTI(d) for front-seat occupants has improved, on the average, from 114 in baseline 1981-1985 models to 44 in models equipped with side air bags and meeting FMVSS 214: amazing progress on a difficult safety problem.

This report investigates if the improvements in side impact protection have saved lives in actual crashes, based on statistical analyses of crash data. The Government Performance and Results Act of 1993 and Executive Order 12866 require agencies to evaluate the benefits of their existing regulations. The statistical analyses use calendar year 1993-2005 crash data from the Fatality Analysis Reporting System (FARS) and the General Estimates System (GES) of the National Automotive Sampling System (NASS). The analyses are divided into two main sections:

· Effect of TTI(d) improvements by structure and padding (without side air bags) on the fatality risk of front-seat occupants (drivers and right-front passengers) in passenger cars. Many of the improvements date to the mid-1990s. By now, the cars have been on the road for nearly a decade. While there is a fair amount of uncertainty, the results are essentially final in the sense that most of the eventual data are already in hand.

o A parallel analysis for compact pickup trucks did not show a statistically significant effect.

· Effect of side air bags – torso bags and/or head-protection air bags – for front-seat occupants of cars and LTVs. Side air bags, especially head air bags began to appear in large numbers only after 2000. Analyses already show statistically significant results, but more data are on the way. The findings of this report will be updated periodically during the next five years.

o Side air bags are principally designed to protect nearside occupants but might conceivably also benefit farside occupants: the driver in a right-side impact and the right-front passenger in a left-side impact. Statistical analyses separately focus on nearside and farside occupants.

The main findings of this report are that structural improvements and padding for cars, and side air bags for cars and LTVs have significantly reduced occupants’ fatality risk. The two types of side air bags – torso bags and head-protection air bags – make substantial and complementary contributions to fatality reduction for nearside occupants. Head curtains (or inflatable tubular structures) also appear to have a significant benefit for farside occupants of passenger cars. The public will obtain the most protection if they have all of these improvements: structures and padding that meet or exceed the requirements of FMVSS 214, torso bags and head curtains. The combined effects are impressive, amounting to a 42 percent cumulative fatality reduction in 2‑door cars, and a 30 percent reduction in 4-door cars.

The risk of chest injury in a side impact is measured on a specially designed side impact dummy during a crash test in the FMVSS 214 configuration, a 33.5 mph impact by a moving deformable barrier into the side of the test vehicle. Accelerations measured on the upper and lower ribs and lower spine are combined into a Thoracic Trauma Index for the dummy - TTI(d). TTI(d) gauges occupants’ injury risk in nearside impacts: the lower the TTI(d), the lower the risk of injury. Reductions in the average TTI(d) of the many vehicles NHTSA has tested over the years demonstrate improved safety in side impacts.

· TTI(d) for front-seat occupants in the FMVSS 214 test configuration, by model year, averaged:

· In 2-door cars, TTI(d) improved by 45 units since 1981-1985 without side air bags and an additional 25 units with side air bags, for a total of 70. Average performance was originally much worse than the FMVSS 214 requirement and is now much better.

· In 4-door cars, TTI(d) improved by 22 units since 1981-1985 without side air bags and an additional 15 units with side air bags, for a total of 37. Average performance was once about the same as the FMVSS 214 requirement and is now much better.

· TTI(d) performance used to be much worse in 2-door cars than in 4-door cars; it is now nearly the same.

o 56 percent of cars received substantial structural modifications, usually accompanied with padding.

· This report identifies 15 make-models that substantially improved TTI(d), by a known amount, without side air bags: from an average of 85 to 62, a 23-unit improvement. Fatality risk of nearside front-seat occupants in multivehicle crashes decreased by a statistically significant 18 percent in these models (90 percent confidence bounds, 7 to 28 percent).

· For passenger cars with TTI(d) in the below-90 range, each unit improvement of TTI(d) without side air bags is associated with an estimated 0.863 percent fatality reduction for nearside occupants in multivehicle crashes (confidence bounds, 0.33 to 1.46 percent).

o The fatality reductions for nearside occupants in single-vehicle crashes and for farside occupants were not statistically significant.

· For pre-FMVSS 214, 2-door cars with TTI(d) in the 90+ range, each unit improvement of TTI(d) was associated with an estimated 0.927 percent fatality reduction for all occupants in side impacts (confidence bounds, 0.52 to 1.33 percent).

· In 2-door cars, the cumulative effect of reducing TTI(d) from 114 (1981-1985 baseline) to 69 (post-FMVSS 214 without side air bags) is a 33 percent fatality reduction for nearside occupants in multivehicle crashes (confidence bounds, 18 to 47 percent).

· In 4-door cars, the cumulative effect of reducing TTI(d) from 85 (1981-1985 baseline) to 63 (post-FMVSS 214 without side air bags) is a 17 percent fatality reduction for nearside occupants in multivehicle crashes (confidence bounds, 7 to 27 percent).

· TTI(d) improvement by structures and padding in passenger cars saved an estimated 803 lives in calendar year 2003.

· If every passenger car on the road in 2003 had been equipped with these improvements, they would have saved an estimated 1,143 lives.

· Torso bags plus head protection in passenger cars reduces the fatality risk of nearside front-seat occupants in single- and multivehicle crashes by a statistically significant 24 percent (90 percent confidence bounds, 4 to 42 percent).[1]

o The data also show a statistically significant fatality reduction in LTVs and suggest that the effectiveness may be the same as in cars.

o The available data do not show a difference in fatality reduction between the two types of head air bags: head curtains (or inflatable tubular structures) and torso/head combination bags.

· Torso bags alone reduce the fatality risk of nearside occupants in passenger cars by an estimated 12 percent (confidence bounds, -3 to +23 percent).

· Through 2005, there were few vehicles equipped with head curtains only (no torso bags): not enough for a separate statistical analysis. However, the preceding results suggest that torso bags and head air bags are both effective in nearside impacts and make approximately equal contributions to fatality reduction.

· Specific mechanisms whereby side air bags mitigate injuries in farside impacts have not yet been widely demonstrated or quantified by testing.

· Nevertheless, statistical analyses of FARS and GES data show significant reductions of fatality risk for head curtains plus torso bags in farside impacts to passenger cars.

· Furthermore, analyses of life-threatening injuries to farside occupants in passenger cars without side air bags suggest that head curtains or inflatable tubular structures could have benefited unrestrained occupants – or even belted drivers if no passenger had been sitting between them and the right side of the car – because:

o Head curtains would have deployed and covered areas responsible for a large proportion of the life-threatening injuries, and

o In most of those impacts, the head curtains would still have been at least partially inflated at the time the farside occupant contacted them.

· A 24 percent fatality reduction is estimated (same as for nearside occupants) for head curtains plus torso bags in farside impacts to passenger cars – for unrestrained occupants and for belted drivers riding alone in the front seat.

· With the limited crash data available to date, no consistently significant fatality reduction was found and, for now, none is claimed in farside impacts for:

· Head curtains reduced the risk of fatal occupant ejection in side impacts by a statistically significant 30 percent.

o Through model year 2003, head air bags in passenger cars were only designed to deploy in side impacts. Head curtains with rollover sensors began to appear in selected LTVs during mid-model year 2002. Crash data were not sufficient to evaluate to what extent this promising technology reduces ejections in rollover crashes.

· Side air bags could have saved an estimated 1,791 lives in calendar year 2003 if every passenger car and LTV on the road had been equipped with head curtains (or inflatable tubular structures) plus torso bags and if every LTV on the road had been equipped with torso bags plus head protection. However, the number of lives saved if all vehicles on the road were to have side air bags in a future year would be smaller than 1,791, since:

o The long-term shift of the on-road fleet from cars to LTVs will reduce the number of potentially fatal side impacts because LTVs are less vulnerable, when struck in the side, than cars.

o The increasing proportion of vehicles equipped with Electronic Stability Control will further reduce the number of potentially fatal side impact and rollover crashes by preventing these crashes altogether.

The estimation of future lives saved is beyond the scope of this report, but will be addressed in NHTSA’s forthcoming Final Regulatory Impact Analysis to add a pole test to FMVSS 214.

· Side impact protection could have saved an estimated 2,934 lives in calendar year 2003 if every car on the road had been equipped with head curtains, torso bags and FMVSS 214 side structures/padding, and if every LTV on the road had been equipped with torso bags plus head protection.

· Relative to 1981-1985 baseline cars, the combination of head curtains, torso bags and FMVSS 214 side structures/padding reduces fatality risk of drivers and right-front passengers in all side impacts by:

· In LTVs, torso bags plus head protection reduce fatality risk of drivers and right-front passengers in all side impacts by 15 percent.

[1] A small portion of this effectiveness may actually be due to energy-absorbing materials (other than air bags) installed to meet the FMVSS 201 upgrade of head-impact protection. NHTSA will evaluate FMVSS 201 in the future; this report only addresses its interaction with side air bags. In many make-models, the introduction of head air bags coincided with FMVSS 201 certification; nevertheless, the energy-absorbing materials remained largely unchanged in the year that head air bags were introduced, and for that reason could not have accounted for a large portion of the fatality reduction for those make-models in that year.