Time to raise the roof.

The problem of roof crush has been largely ignored by government authorities, auto manufacturers, and the media. In light of the recent rollover tragedies, now is the time to take action.

A leading automotive safety expert has stated a basic truth: "[I]f the roof crushes over you, it's curtains."(1) Yet the problem of roof crush--roofs collapsing under the weight of a vehicle in a rollover--has been virtually ignored by government authorities, auto manufacturers, and the media. However, in light of the highly publicized controversy surrounding rollovers of Ford Explorers equipped with Firestone tires, the time may be right for a renewed call to action.

The rollover danger is highlighted by some startling statistics. In 1999, 23 percent of all car occupants and more than 60 percent of sport utility vehicle (SUV) occupants who died in crashes were involved in rollovers.(2) Since rollovers are the most frequent cause of roof crush, these statistics imply that roofs are not strong enough, especially on SUVs. Because SUVs have higher centers of gravity and greater weight, the impact from a rollover is more severe than in a regular car.(3)

The urgency of the problem is heightened by the continuing popularity of SUVs. Notwithstanding higher gas prices, General Motors expects to sell more light trucks than cars in 2001, until SUVs account for half of all vehicles on the road within a decade. Despite its bad tires and bad press, the Ford Explorer has been setting all-time sales records.(4)

Yet, the government has done virtually nothing to directly address roof strength since adopting a minimal standard in 1971. Federal Motor Vehicle Safety Standard (FMVSS) 216 only requires that force be applied to the front portion of the vehicle, with the windows up. No force need be applied to the B- and C-pillars (the metal columns surrounding the rear windows on cars) or the D-pillars (the rear roof-support columns on station wagons, vans, or SUVs). Light trucks, including SUVs, are required to meet only the minimum roof strength standard--an ability to bear 1 1/2 times their empty weight (with the windows up)--even though these vehicles experience a much higher rollover rate.(5)

Instead of toughening the standard, the government has regulated around the roof crush danger. For example, the National Highway Traffic Safety Administration (NHTSA) has recently established standards for upper interior head protection in an attempt to protect occupants when their heads hit the upper portion of the vehicle.(6)

Regarding the rollover danger, the agency has published a one- through five-star ratings system for rollover resistance.(7) The rating is based on a mathematical formula, which divides half the track width by the center of gravity height, called the static stability factor. This is intended to aid consumers in assessing rollover risk. As part of the Tread Act signed into law last November, Congress has ordered NHTSA to develop a dynamic rollover test program to rate the susceptibility of particular vehicles to rollover.(8)

Ironically, manufacturers initially opposed the government's roof strength standard, but now they embrace it. Because it is so easy to meet, the standard can be used by manufacturers essentially as a litigation shield. FMVSS 216 can fairly be described as minimal. No drop or dynamic rollover testing is required.

Moreover, to comply with this minimum standard, manufacturers rely on the strength provided by the vehicle's windshield and side window glass.

This may seem to be acceptable at first. However, in a rollover, the windshield frequently breaks just before the roof collapses, dramatically reducing roof strength. Yet the government does not require manufacturers to assess the vulnerability of the roof if the windshield breaks.(9)

Although rollovers and roof crush account for many fatalities, they also cause an even greater number of incapacitating injuries.(10) Rollovers of light trucks cause approximately 9,000 fatalities each year. Further, there are about 12,000 serious head injuries and 3,100 life-threatening neck injuries from rollovers. The resulting economic costs to society are staggering.(11) Cost-effective countermeasures are readily available, but without regulatory pressure, manufacturers have failed to make stronger and safer vehicle roofs.

Manufacturers have neglected this obligation despite the fact that the common law has long recognized that manufacturers have a duty to design vehicles so as to protect occupants in foreseeable crashes.(12) This includes fortifying the roofs. For example, in Dyson v. General Motors Corp., an occupant was severely injured when the roof collapsed into the seating area during a rollover. The court recognized the importance of roof strength in protecting occupants in rollovers: "[I]t is the obligation of an automobile manufacturer to provide more than merely a platform capable of transporting passengers from one point to another. The passengers must be provided a reasonably safe container within which to make the journey. The roof is a part of such container."(13)

Survival space

The goal of safe roof design is to maintain the occupant's "survival space," sometimes described as a "nonencroachment zone."(14) Manufacturers commonly advertise that their vehicles incorporate an internal safety cage designed to maintain the integrity of the passenger compartment in a crash. The roof designer's aim is to maintain a protective cocoon around the occupants so they avoid injurious contact with the vehicle structure. If the survival space is not maintained because the roof deforms and encroaches into this zone, severe injury or death can result.(15)

This is not a new concept. Industry representatives admitted in a 1990 study known as the Malibu II tests that "the absence of deformation [to the roof] may benefit belted occupants if it results in the belted occupant not contacting the roof."(16) This seems obvious. Yet, manufacturers do not design roofs to achieve this cocoonlike protection and maintain essential survival space. Rather, they design roofs to pass the minimum government standard set 30 years ago.(17)

The preamble to FMVSS 216 says that "the strength of a vehicle's roof affects the integrity of the passenger compartment and the safety of the occupants.... [I]t has been determined, therefore, that improved roof strength will increase occupant protection in rollover accidents."(18) It is not surprising that this goal has not been achieved, since the test does not assess what happens to the occupants during a rollover.

In the government test, the force is applied to the roof gradually, with the windows up, and only over the front portion of the vehicle. This obviously does not reflect the dynamic forces in an actual rollover.(19)

Moreover, it ignores occupant dynamics, or kinematics, during a rollover, which are just as significant as the forces on the vehicle. No anthropomorphic dummy is used to test compliance with FMVSS 216 in a rollover, in glaring contrast to the government's frontal-barrier and side-impact crash tests.

NHTSA has recognized the shortcomings of FMVSS 216 and recently considered updating it. But, so far, no revisions have been made.(20)

A question of causation 

Manufacturers have discovered that safety is a good sell. However, they have long denied that roof crush causes occupant injury in rollovers.(21) Manufacturers cite the Malibu I and II tests as demonstrating that, in a rollover, even seat-belted occupants are propelled toward the roof and suffer cervical injuries before the roof collapses. Thus, the companies argue that the roof crushes after the occupant has already sustained injuries.(22)

But manufacturers have long known that a strong roof is essential to protecting occupants in rollovers. In the 1950s and 1960s, General Motors, Chrysler, and Ford all conducted rollover tests to assess how the structure of their vehicles would protect occupants. When FMVSS 216 was adopted in the early 1970s, however, the manufacturers stopped dynamic rollover testing, claiming that the kinematics of the vehicle in a dynamic test could not be repeated.(23)

Also in the 1960s, various industry representatives acknowledged that roof intrusion during a rollover was a severe hazard to an occupant.(24)

In 1968, a Ford employee noted that roof crush posed a danger to occupants wearing seat belts:

   It is obvious that occupants that are restrained in upright positions are
   more susceptible to injury from a collapsing roof than unrestrained
   occupants who are free to tumble about the interior of the vehicle. It
   seems unjust to penalize people wearing effective restraint systems by
   exposing them to more severe rollover injuries than they might expect with
   no restraints.(25)

In the 1970s, several manufacturers, under contract with the Department of Transportation, developed experimental safety vehicles (ESVs) that addressed issues of crashworthiness, including the problem of roof crush. These vehicles were actually built and the safety features were tested, but the vehicles were not commercially produced. For example, Ford's vehicle had a roof and upper structure that was "designed to improve occupant protection in rollover accidents by increasing the resistance of the roof to crush."(26) The roof of the Ford ESV was strengthened by adding a rollbar, connected to the B-pillar, between the inner and outer panels of the roof. The A-, B-, and C-pillars were also reinforced.

The roof of the General Motors ESV consisted of structural box section members around the entire periphery of the roof. Additionally, a box section crossmember was added between the side rails connecting at the B-pillars. Then these structures were packaged between an inner and outer roof panel. With the adoption of FMVSS 216, manufacturers seem to have lost their incentive to make stronger roofs.

Defects and alternatives

Roof crush can result from a design defect, a manufacturing defect, or both. Design defects include structural engineering errors, such as cutouts, gaps, or holes at crucial load-bearing locations. For example, manufacturers have been widely criticized for using inherently weak, one-piece, open-windshield headers, side rails, and rear headers. An open piece roof section is connected on one side but is open, or not connected, at the other side. Because the roof section is open, it is not able to resist forces in a rollover accident as well as a closed section.

One-piece headers, in particular, have been identified as a weak link in what may otherwise be a strong roof system. Reinforced roof pillars and rails frequently transfer the forces of an impact to this weak link. Not surprisingly, this is the point at which the roof deforms and invades the survival space.

Manufacturing defects can also contribute to roof crush in a rollover. A common one is improperly made or nonfused welds, which separate on impact. Welds may also be improperly located or missing altogether.

Design alternatives to strengthen roofs and minimize structural intrusion into the occupants' survival space have been available for decades. These include the use of closed-box sections in the headers and side rails, rather than open, one-piece sections. Closed-box sections greatly increase resistance to roof crush.(27)

Also, the elimination of all unnecessary cutouts, gaps, and holes in the sheet metal help lessen the number of weak links where a structural failure might occur. Reinforcing the roof and increasing sheet metal thickness also increase roof strength.(28) The use of structural foam in the pillars and headers can double or even triple the strength of the component structure, a fact noted as early as 1973 by Japanese researchers developing an experimental safety vehicle for Nissan.(29)

Any combination of these design alternatives help to preserve the occupant survival space in a rollover.(30)

Automotive safety researchers have recently demonstrated the effectiveness of some of these countermeasures. They made modifications in the roof of a late-model SUV, including reinforcing and foam-filling the A-pillars, B-pillars, side headers, and front headers.(31) In an 18-inch drop test, these modifications reduced roof crush by 75 percent, compared with the same vehicle that did not have the modifications.(32) In the same test series, the non-modified survival space in vehicles was easily compromised.

The demonstration also showed that drop tests simulate the damage received in actual rollovers. For example, a 12-inch to 18-inch drop height placed the occupants at severe risk, but simple engineering countermeasures dramatically improved occupant protection.

Integrated approach

Automotive safety requires an integrated approach. A strong roof is essential to occupant protection in rollovers, but so are seat belts. Effective restraints can substantially reduce the risk of harmful contact with the roof and roof rails, provided, of course, that the roof does not collapse into the occupants' survival space. Yet many seat belts have improper geometry that lets occupants move toward the roof during a rollover.(33)

There is no standard governing how a restraint must perform in a rollover or when the roof or safety cage deforms. Indeed, NHTSA has recently rescinded a seat belt performance standard applicable to rollovers.(34)

Improved seat belt geometry can reduce excess occupant movement toward the roof in a rollover. In a 1996 study, an improved seat belt decreased movement toward the roof by as much as 75 percent. Better shoulder belts mounted to integrated seats are also effective in rollovers. Integrated seats have the seat belts mounted into the seats to provide a better fit for the occupant. Pretensioners, devices designed to take up seat belt slack during a collision, are also a highly desirable safety feature.(35)

In 1997, industry researchers conducted quasi-static rollover tests to determine head excursion with different restraint systems. They reported vertical excursion with an integrated seat, a pretensioner, and a combination of both.(36) A 2000 study reported that improved seat belt latch plates designed to restrict the webbing passing through the lap belt can reduce body excursion and provide better neck protection.(37)

We know these safety measures are possible because a number of production vehicles are already equipped with them. The Chrysler Sebring Convertible (1996-present), the 1998 Dodge Ram Quad Cab, and the 2001 Ford 150 extended-cab pickup trucks have integrated seats. Pretensioners have been included in production vehicles since the early 1990s.

In addition, Freightliner trucks, manufactured by a division of Chrysler, have used an occupant restraint system called Seat Pretensioner Activation for Crash Survival Enhancement (SPACE) since 1995. During a rollover or major frontal impact, SPACE pulls the driver away from the roof by automatically lowering the seat and tightening the seat belt.(38)

Great strides have been made in automotive safety in the last 10 to 20 years, and manufacturers now compete vigorously to market safer vehicles. Recent advances to help prevent rollovers include more stable suspensions, side-curtain air bags, special sensors to keep the bags inflated in case the vehicle rolls more than once, and pop-up rollbars in convertibles.

Despite these advances, manufacturers have long neglected the importance of a strong roof in a rollover. They continue to rely on a 30-year-old minimal government standard that has failed to keep pace with the huge growth in the number of SUVs on the road and the corresponding increase in rollover deaths and serious injuries. Effective alternative designs have long been available, but until they become the industry norm, many more motorists will be needlessly killed and injured in rollovers.

Notes

(1.) Curtis Rist, Roll Over, Newton, DISCOVER, Apr. 2001, at 49 (quoting Carl E. Nash).

(2.) Id.; see also Light Vehicle Rollover, Background on NHTSA's Activities in this Area, Slide 4, available at www.nhtsa.dot.gov (last visited June 28, 2001).

(3.) Rist, supra note 1.

(4.) Jim Henry, Light Trucks Lead Charge to Hot June, AUTOMOTIVE NEWS, July 9, 2001, at 37, 39.

(5.) 49 C.F.R. [sections] 517.216 (1971).

(6.) See FMVSS 201, Head Impact Protection, 49 C.F.R. [subsections] 571,572, 589 (1995). This standard requires head protection for the upper interior components of all passenger cars, multipurpose vehicles, trucks, and buses with a gross vehicle weight of 10,000 pounds or less (light vehicles). It will only be fully effective for cars manufactured after September 1, 2002.

(7.) Rollover Resistance, 66 Fed. Reg. 3388, 3395 (Jan. 12, 2001).

(8.) Patrick Ponticel, NHTSA Seeks Input for Dynamic Rollover Test Plan, AUTOMOTIVE ENGINEERING INT'L, July 2001, at 168; see also NHTSA, Rollover Resistance Ratings Information, available at http.www.nhtsa.dot.gov/hot/rollover (last visited Aug. 9, 2001).

(9.) Donald Friedman & Carl E. Nash, Advanced Roof Design for Rollover Protection, 17th Int'l Technical Conference on Enhanced Safety of Vehicles, Amsterdam, Netherlands, ESV Paper 01-512-W-94 (June 2001).

(10.) Glen C. Rains & Joseph N. Kanianthra, Determination of the Significance of Roof Crush on Head and Neck Injury to Passenger Vehicle Occupants in Rollover Crashes (Soc'y of Automotive Engineers Paper No. 950655, 1995); see Datalink, Light Vehicle Roof Crush Incidents and Outcomes, prepared for NHTSA Crashworthiness Staff, fig. 12 (Nov. 1992); Donald Friedman & Keith D. Friedman, Roof Collapse and the Risk of Severe Head and Neck Injury, 13th Experimental Safety Vehicle Conference, Paris, France, No. 91-S-0-11 (Nov. 1991).

(11.) LEON S. ROBERTSON, INJURY EPIDEMIOLOGY 3 (1992); see also Friedman & Friedman, supra note 10.

(12.) See Larsen v. Gen. Motors Corp., 391 F.2d 495, 502 (8th Cir. 1968); see also Polk v. Ford Motor Co., 529 F.2d 259, 264-65 (8th Cir. 1976); Arbet v. Gussarson, 225 N.W.2d 431, 436 (1975).

(13.) 298 F. Supp. 1064, 1073 (E.D. Pa. 1969).

(14.) E. Franchini, The Crash "Survival Space," (Soc'y of Automotive Engineers Paper No. 69005, 1969); see also Nat'l Highway Traffic Safety Admin., Comments of General Motors with Respect to Proposed Rule Making Roof Intrusion Protection, NHTSA Docket No. 02-06, Notice 4, Comment 015 (Apr. 5, 1971).

(15.) Brian Herbst et al., Strength Improvements to Automobile Roof Components (Soc'y of Automotive Engineers Paper No. 980209, 1998); Donald Friedman & Keith D. Friedman, Improved Vehicle Design for the Prevention of Severe Head and Neck Injuries to Restrained Occupants in Rollover Accidents, 16th Int'l Technical Conference on Enhanced Safety of Vehicles, Melbourne, Australia, U.S. Paper No. 96-S5-0-14 (1996); see also Rains & Kanianthra, supra note 10 (noting a relationship between roof intrusion and head injuries to belted occupants); Stephen R. Syson, Occupant to Roof Contact: Rollovers and Drop Tests, (Soc'y of Automotive Engineers Paper No. 950654, 1995) ("A reinforced roof structure and an advanced belt system can prevent serious neck injury in rollover collisions.").

(16.) G.S. Bahling et al., Rollover and Drop Tests--The Influence of Roof Strength on Injury Mechanics Using Belted Dummies (Soc'y of Automotive Engineers Paper No. 902314, 1990) (Malibu II).

(17.) 49 C.F.R. [sections] 517.216 (1971).

(18.) Id.

(19.) Herbst et al., supra note 15.

(20.) NHTSA, Upgrade of Rollover Crash Protection, B.01.13, available at www.nrd.nhtsa.dot. gov/departments/nrd-01/summaries/bo0113.html (last visited July 19, 2001).

(21.) See Edward A. Moffatt, Occupant Motion in Rollover Collision, Proceedings of the 19th Conf. of the Am. Ass'n for Auto. Med. (1975);see also Moffatt et al., Rollover Crash Tests--The Influence of Roof Strength as Injury Mechanisms (Soc'y of Automotive Engineers Paper No. 851734, 1985); Bahling et al., supra note 16; Moffatt et al., The Influence of Increased Roof Strength on Belted and Unbelted Dummies in Rollover and Drop Tests, 38 J. TRAUMA, INJURY, INFECTION & CRITICAL CARE 557 (1995).

(22.) Bahling et al., supra note 16.

(23.) R.A. Wilson & R.R. Gannon, Rollover Testing (Soc'y of Automotive Engineers Paper No. 720495, 1972).

(24.) Roy Hausler, Safety Design in the Automobile of the Future: An Industry View Point, TRAUMA & AUTOMOBILE 25 (William J. Curran & Neil L. Cheyat eds., 1966).

(25.) Memorandum from J.R. Weaver, Ford Motor Co., to H.G. Brilmeyer, Ford Motor Co. (July 8, 1968).

(26.) FORD EXPERIMENTAL SAFETY VEHICLE, FINAL REPORT, Contract No. DOT-0S-20005, at 29 (1973); see also GENERAL MOTORS EXPERIMENTAL SAFETY VEHICLE FINAL REPORT, Contract No. DOT-OS-00095 (June 30, 1972); CALSPAN/CHRYSLER RSV, PHASE III, FINAL DESIGN REPORT, Contract No. DOT HS-7-01551, at 91 (Dec. 1979).

(27.) See Herbst et al., supra note 15; see also GENERAL MOTORS EXPERIMENTAL SAFETY VEHICLE FINAL REPORT, supra note 26, at 38-41; Friedman & Friedman, supra note 10; Friedman & Nash, supra note 9.

(28.) Friedman & Friedman, supra note 10.

(29.) Id.; Herbst et al., supra note 15; Nagayuki Marumo & Teruo Maeda, Development of the Nissan Experimental Safety Vehicle, ESV Conference, at 61 (1973).

(30.) Herbst et al., supra note 15; Friedman & Friedman, supra note 10.

(31.) Herbst et al, supra note 15.

(32.) Id.

(33.) Friedman & Friedman, supra note 10.

(34.) Seat Belt Assemblies, 64 Fed. Reg. 27203, removed [para] S4.1(b) effective May 19, 1999.

(35.) Brian Herbst et al., The Ability of Three-Point Safety Belts to Restrain Occupants in Rollover Crashes, 16th Int'l Technical Conference on Enhanced Safety of Vehicles, Melbourne, Australia, Paper No. 96-S5-0-12 (1996); see also Friedman & Friedman, supra note 10.

(36.) James E Pywell et al., Characterization of Belt Restraint Systems in Quasistatic Rollover Tests (Soc'y of Automotive Engineers Paper No. 973334, 1997); see also Steven E. Meyer et al., Three-Point Restraint System Design Considerations for Reducing Vertical Occupant Excursion in Rollover Environments (Soc'y of Automotive Engineers Paper No. 2000-01-0605, 2000).

(37.) Casley C. Ward et al., Investigation of Restraint Function on Male and Female Occupants in a Rollover Event (Soc'y of Automotive Engineers Paper No. 2001-01-0177, 2001).

(38.) Jean L. Broge, Tubular Side Airbags from Freightliner, AUTOMOTIVE ENGINEERING INT'L, Feb. 2001, at 74.

James J. Murphy practices law in Milwaukee, Wisconsin. The author wishes to thank M. Josef Zimmermann and Thadd J. Llaurado, attorneys at his firm, for their assistance in preparing this article.