Rear underride happens when a passenger vehicle strikes the back of a truck or trailer and slides underneath it. The danger comes from the mismatch between the height of the truck or trailer and the crash structure of the passenger vehicle. A passenger vehicle is designed to manage a frontal impact through its bumper, hood structure, engine compartment, airbags, and seat belts. But when the rear of a trailer sits above those structures, the trailer can enter the windshield, roof, and occupant compartment instead of allowing the vehicle's front end to absorb the crash. NHTSA and FMCSA describe underride as a crash in which any portion of a passenger vehicle slides under the body of a larger truck or trailer, and their law-enforcement guidance specifically asks officers to document the degree and location of underride because severe underride can involve occupant-compartment intrusion.[1]
A rear underride guard is the horizontal steel structure mounted near the back of many trailers. Under FMVSS No. 223, a rear impact guard is a device installed on or near the rear of a trailer or semitrailer so that, when the trailer is struck from behind, the guard limits the distance the striking vehicle's front end travels under the trailer.[2] The Insurance Institute for Highway Safety describes the same device in simpler terms: steel bars hanging from the backs of large trucks to prevent passenger vehicles from moving underneath in a rear crash.[3]
The point of the guard is not to make rear-end crashes harmless. It is to make them survivable more often. When the guard stays attached, sits low enough, and extends far enough across the rear of the trailer, it can help the striking vehicle engage its own crash-management structures. When the guard is missing, damaged, too high, too far forward, too narrow, or weakly attached, the trailer can bypass the passenger vehicle's frontal crash protection.
The issue exists within a larger pattern of crash risk around large trucks. NHTSA reported that 5,936 people were killed in crashes involving large trucks in 2022, and 70% of those killed were occupants of vehicles other than the truck.[4] Rear underride is only one subset of those crashes, but it shows how truck geometry can change the nature of a collision. A crash that might otherwise be a severe front-end impact can become a direct intrusion into the survival space of the people inside the passenger vehicle.
How FMVSS 223 and 224 Work Together
Federal rear underride protection is built around two standards. FMVSS No. 223 governs the rear impact guard itself. It sets requirements for the guard's strength, energy absorption, geometry, labeling, test procedures, and installation instructions. The regulation requires the guard's horizontal member to have a minimum vertical height at specified points and requires the guard to meet strength and energy-absorption requirements during testing.[5]
FMVSS No. 224 governs installation of the guard on the trailer or semitrailer. It applies generally to trailers and semitrailers with a gross vehicle weight rating of 4,536 kilograms or more, which is about 10,000 pounds. The stated purpose is to reduce deaths and serious injuries when light-duty vehicles crash into the rear of covered trailers and semitrailers.[6]
FMVSS 224 also defines important exemptions. Certain vehicles are excluded from the rear impact protection requirement, including pole trailers, pulpwood trailers, low-chassis vehicles, wheels-back vehicles, special-purpose vehicles, and road-construction controlled horizontal-discharge trailers.[7] Those exemptions matter in crash investigations because the first question is not simply whether the trailer had a guard. The first question is whether the trailer was required to have one under the standard that applied to that vehicle.
The installation details are just as important as the existence of the guard. FMVSS 224 sets limits for guard width, ground clearance, and rearward placement. Among other things, the guard's outermost surfaces must be close enough to the trailer's side extremities, its bottom edge generally may not be more than 560 millimeters above the ground, and the guard may not be set too far forward from the rear extremity of the trailer.[8] A guard that is strong in a laboratory but mounted too high, too narrow, or too far forward may not engage the striking vehicle in the way the standard assumes.
A separate in-service rule applies to motor carriers. Under 49 C.F.R. § 393.86, trailers and semitrailers manufactured on or after January 26, 1998, with a gross vehicle weight rating of 10,000 pounds or more must have a rear impact guard that met FMVSS 223 when the trailer was manufactured, installed so that the trailer met FMVSS 224 when it was manufactured.[9] That manufacturing-date language is critical. Investigators must identify the trailer's date of manufacture, determine which version of the standard applied, and then evaluate whether the trailer remained compliant after years of use, damage, repair, and maintenance.
Why the 2022 Update Matters
NHTSA strengthened the rear underride guard standard in 2022. The agency amended FMVSS 223 and 224 to require rear impact guards to provide sufficient strength and energy absorption when compact and subcompact passenger cars strike the rear of trailers at 56 kilometers per hour, or 35 miles per hour.[10] The rule became effective January 11, 2023, with a compliance date of July 15, 2024.[11]
The speed increase was significant because the prior federal framework was based on 30 mph impact testing. NHTSA explained that increasing the test speed from 30 mph to 35 mph requires the guard to manage 36% more crash energy.[12] That change brought the U.S. standard closer to Canada's rear impact guard requirements and reflected years of testing showing that stronger guards were feasible.
IIHS testing helped shape the public debate. IIHS evaluates rear underride guards in 35 mph tests using three overlap configurations: full-width overlap, 50% overlap, and 30% overlap.[13] The 30% overlap test is especially demanding because the passenger vehicle contacts the outer portion of the guard, where the guard may have less structural support. IIHS's testing program showed that some trailer manufacturers could build guards that performed well in more demanding offset impacts, while weaker designs allowed dangerous underride.[14]
The 2022 federal rule did not go as far as IIHS's full testing program. NHTSA adopted the 35 mph requirement for full-width and 50% overlap conditions but did not adopt a broad 30% low-overlap requirement.[15] That distinction should be clear in any article on rear underride guards. The rule strengthened the federal standard, but it did not require guards to prevent underride in every foreseeable offset, angled, or high-speed impact. That distinction also matters after a crash. A trailer may satisfy the federal standard and still be involved in a fatal underride crash if the impact speed exceeds the test condition, the overlap is too small, the angle is unusual, the guard is damaged, or the vehicle contacts a portion of the trailer that the standard does not fully address. Compliance is a minimum regulatory baseline. It is not proof that the guard was adequate for the actual crash conditions.
Maintenance Failures Can Turn a Guard Into a Hazard
Rear underride protection is not only a manufacturing issue. It is also a maintenance issue. A guard that was compliant when new can become ineffective after years of loading-dock impacts, corrosion, cracked welds, missing fasteners, bent horizontal members, or improper repairs.
The Technology & Maintenance Council's RP 732, Trailer Rear Impact Guard Repair Guidelines, treats rear impact guards as safety-critical components. RP 732 explains that the recommended practice applies to trailer rear impact guards on post-January 26, 1998 trailers and identifies them as "RIGs," or rear impact guards.[16] The same recommended practice states that FMVSS 223 and 224 regulate rear impact guard strength, location, energy absorption, dimensions, and testing.[17]
RP 732 recommends regular inspection for cracked welds, cracked or fractured vertical members, cracked or loose fasteners, loose fasteners attaching the guard to the trailer sill, bends, corrosion, cuts, punctures, tears, proper attachment to the trailer sill, rear cross-member condition, rear trailer sill condition, the last six feet of floor, and overall dimensional and structural integrity.[18] These are the physical signs that a rear guard may no longer be capable of performing as designed.
The repair guidance is cautious. RP 732 states that because each repair requires a case-by-case analysis, there is no way to be sure that any particular repair is sufficient. It recommends that cracked welds be repaired as soon as discovered; damaged, corroded, bent, cut, punctured, torn, or excessively worn components be discarded and replaced with manufacturer-approved replacement components; and deficiencies in dimensional or structural integrity be addressed by replacing the entire rear impact guard system with a manufacturer-approved system.[19]
FMCSA recognized a related inspection gap in 2021. Before the agency amended the annual inspection rules, rear impact guards were required under the operating rules but were not expressly listed in the minimum periodic inspection standards. FMCSA explained that this meant a commercial motor vehicle could pass an annual inspection even with a missing or damaged rear impact guard.[20] FMCSA then added rear impact guards to the annual inspection appendix. The current minimum periodic inspection standards identify rear impact guard violations such as a missing guard, insecure attachment, broken or missing fasteners, cracked welds, cracked or broken metal, damage that compromises attachment, excessive height above the ground, excessive rearward setback, insufficient width, and insufficient vertical height of the horizontal member.[21]
That update gives investigators a clearer maintenance framework. The post-crash question is not merely whether a rear guard existed. The question is whether it remained securely attached, correctly positioned, structurally intact, and dimensionally compliant. A guard bent upward from dock impacts, weakened by corrosion, or attached with missing fasteners may look like a guard in photographs while failing to provide meaningful underride protection.
One recent regulatory change should not be misread. In 2026, FMCSA removed the in-service certification-label requirement from § 393.86(a)(6). The agency explained that removing the in-service label requirement did not eliminate the manufacturing-stage FMVSS certification-label requirement and did not reduce the safety benefits of rear impact guards.[22] The practical point is that a missing or illegible label alone should not be treated as proof that the guard was structurally defective. Investigators still need measurements, attachment inspection, repair records, and trailer-manufacturer information.
Why Rear Underride Crashes Are Undercounted
Rear underride crashes are difficult to measure because official crash data is incomplete. GAO found that from 2008 through 2017, there were an average of about 219 reported fatalities per year from underride crashes involving large trucks. GAO also concluded that those fatalities were likely underreported because state and local crash forms varied, officers did not use a consistent definition of underride, some forms lacked underride fields, and officers received limited underride-specific training.[23]
That undercount matters because federal safety decisions rely on crash data. If underride is not consistently identified in police reports, then the national databases used by NHTSA, FMCSA, researchers, and policymakers may not show the full scope of the problem. GAO warned that incomplete underride data limits NHTSA's ability to understand underride crashes and evaluate safety standards.[24]
NHTSA and FMCSA later issued guidance aimed at improving underride reporting. That guidance tells officers to document underride in narratives, diagrams, and crash reports, and to use the term "underride" even when a state crash form does not include a dedicated underride field.[25] The guidance recognizes a practical problem: if the form does not ask the right question, the crash may never be counted as an underride crash.
NHTSA's crash studies show why better reporting matters. In a study of fatal rear impacts into trucks, NHTSA identified 532 fatalities among occupants of vehicles that struck the rear of trucks. Some underride was reported in 59% of those fatalities. Among light passenger vehicles, there were 403 fatalities, and some underride occurred in 260 of them, or nearly 65%.[26] A later NHTSA heavy-vehicle crash data study found rear underride in about 75% of rear-end struck crashes where the striking vehicle was a light vehicle. Windshield-level or more severe underride occurred in 36% of those crashes. The same study reported 724 fatalities among occupants of light vehicles in the analyzed underride and override crash population, with underride present in 70% of those fatalities.[27]
The data problem also affects trailer-specific evaluation. NTSB has noted that crash databases historically lacked trailer vehicle identification number and model-year information, making it difficult to evaluate how post-1998 rear underride standards performed in actual crashes.[28] Without accurate trailer age, configuration, and compliance information, investigators and regulators may know that a fatal underride occurred without knowing whether the trailer was built before or after a relevant standard, whether it was subject to FMVSS 224, or whether a newer guard failed.
How Investigators Reconstruct Rear Underride Crashes
A rear underride investigation starts with geometry. Investigators measure the guard's height from the ground, width across the trailer, distance from the side extremities, distance from the rear extremity, vertical height of the horizontal member, and attachment points to the trailer structure. Those measurements are compared against FMVSS 223, FMVSS 224, 49 C.F.R. § 393.86, and the standard that applied when the trailer was manufactured.[29]
The next step is structural inspection. Investigators look for cracked welds, torn parent metal, bent horizontal members, fractured vertical supports, rusted attachment points, missing bolts, replacement fasteners, dock-strike damage, non-original welds, repair plates, and deformation of the rear sill or cross members. RP 732 identifies those same general areas as inspection points and recommends manufacturer-approved repair or replacement when damage compromises structural or dimensional integrity.[30]
Crash reconstruction then asks how the passenger vehicle interacted with the guard. Investigators evaluate overlap, impact angle, speed, braking, lighting, roadway evidence, crush patterns, passenger-compartment intrusion, and whether the vehicle engaged the guard, the trailer body, or both. NHTSA's heavy-vehicle underride study used travel speed, skid distance, angle of impact, police reports, scene diagrams, narratives, photographs, and crash reconstructions to estimate impact speed and underride severity.[31]
The same NHTSA study illustrates why real-world crashes often exceed the assumptions built into guard standards. Estimated striking-vehicle speeds in rear underride crashes averaged nearly 60 mph in the study population, with many crashes occurring above the 30 mph or 35 mph conditions used in federal guard testing.[32] That does not make the guard irrelevant. It means investigators must distinguish between a guard that failed within its expected performance range and a crash that exceeded the guard's tested design conditions.
Electronic and physical evidence can help reconstruct the sequence. In NTSB's HIR-26-02 investigation involving fatal crashes with partial automation and stationary vehicles, investigators used vehicle data, damage patterns, laser scans, airbag control module information, and bulb-filament evidence to evaluate speed, braking, steering, lighting, and impact circumstances.[33] Those tools are not limited to underride cases, but they are especially useful when the central questions involve impact geometry, visibility, braking, and whether the striking vehicle had any meaningful opportunity to engage the guard.
Maintenance records often decide the deeper question. Annual inspection forms, repair invoices, trailer assignment logs, yard photographs, dock-damage reports, prior crash files, replacement-part records, and manufacturer instructions can show whether the rear guard had been damaged before the crash. A bent guard photographed months earlier, a repair invoice showing non-original components, or repeated dock damage can show that the post-crash condition was not sudden or unforeseeable.
The Remaining Gaps in Rear Underride Protection
Rear underride guards save lives when they are designed, installed, maintained, and engaged within their performance range. But the federal standard still has limits. NHTSA's 2022 rule strengthened guard performance to 35 mph in full-width and 50% overlap tests, but it did not require protection in every low-overlap, angled, or higher-speed rear impact.[34]
Single-unit trucks show another gap. NTSB's safety study on single-unit truck crashes found that these trucks create underride risks because of their height, stiffness, and configuration, but they have not historically been regulated in the same way as trailers and semitrailers.[35] In that study, NTSB found rear underride in more than 70% of injury and fatal crashes in which passenger vehicles collided with the rear of single-unit trucks, and the agency recommended that NHTSA require rear underride protection systems on newly manufactured single-unit trucks to reduce fatalities and serious injuries in these crashes.[36]
NTSB has also emphasized that underride protection must account for offset crashes and real-world guard attachment performance. In its 2014 safety recommendation letter, NTSB discussed the evolution from earlier rear-end protection rules to the post-1998 trailer standards and recommended stronger rear underride protection requirements for newly manufactured trailers with gross vehicle weight ratings over 10,000 pounds.[37]
The practical issue is that rear underride failures often happen in the gaps between design, maintenance, inspection, and reporting. A trailer may be old enough to fall under weaker requirements. A vehicle may be exempt. A guard may have been bent upward by years of dock impacts. A crash report may fail to identify underride. A label may be missing while the real issue lies in the guard's geometry or attachment. A guard may comply with the federal minimum while still failing to prevent underride in a high-speed or low-overlap crash.
That is why a rear underride crash investigation should not stop with the observation that a guard was present. The key questions are more specific: Was the trailer required to have a guard? Which standard applied when it was manufactured? Was the guard installed at the correct height, width, and rear setback? Was it structurally intact? Was it previously damaged? Were repairs manufacturer-approved? Did annual inspections identify guard defects? Did the crash fall within the guard's tested design envelope, or did it expose a known limitation in the federal standard?
Rear underride guards are one of the clearest examples of how a single truck component can determine whether a crash remains outside the passenger compartment or enters it. The law sets the baseline. Maintenance determines whether the guard remains capable of performing. Crash reconstruction determines whether the guard worked, failed, or was never positioned to protect the people in the striking vehicle.
Sources
- [1] NHTSA & FMCSA, Underride Crash Pamphlet.↩
- [2] 49 C.F.R. § 571.223 (FMVSS No. 223), Rear impact guards.↩
- [3] IIHS, Truck Underride.↩
- [4] NHTSA, Large Trucks: 2022 Data.↩
- [5] 49 C.F.R. § 571.223 (FMVSS No. 223).↩
- [6] 49 C.F.R. § 571.224 (FMVSS No. 224), Rear impact protection.↩
- [7] Id.↩
- [8] Id.↩
- [9] 49 C.F.R. § 393.86, Rear impact guards and rear impact protection.↩
- [10] Federal Motor Vehicle Safety Standards; Rear Impact Guards; Rear Impact Protection, 87 Fed. Reg. 42366 (July 15, 2022).↩
- [11] Id.↩
- [12] Id.↩
- [13] IIHS, Truck Underride.↩
- [14] Id.↩
- [15] 87 Fed. Reg. 42366 (July 15, 2022).↩
- [16] Technology & Maintenance Council, RP 732, Trailer Rear Impact Guard Repair Guidelines, submitted as attachment to FMCSA Docket No. FMCSA-2019-0211.↩
- [17] Id.↩
- [18] Id.↩
- [19] Id.↩
- [20] Parts and Accessories Necessary for Safe Operation; Rear Impact Guards and Rear Impact Protection, 86 Fed. Reg. 62105 (Nov. 9, 2021).↩
- [21] 49 C.F.R. Part 396, Appendix A, Minimum Periodic Inspection Standards.↩
- [22] Parts and Accessories Necessary for Safe Operation; Certification and Labeling Requirements for Rear Impact Protection Guards, 91 Fed. Reg. 7874 (Feb. 19, 2026).↩
- [23] GAO, GAO-19-264, Truck Underride Guards: Improved Data Collection, Inspections, and Research Needed.↩
- [24] Id.↩
- [25] NHTSA & FMCSA, Underride Crash Pamphlet.↩
- [26] NHTSA, Fatalities in Frontal Crashes Despite Seat Belts and Air Bags (rear impact fatality data).↩
- [27] NHTSA, Heavy Vehicle Rear and Side Underride Study.↩
- [28] NTSB, Safety Recommendation Letter H-14-001 through -007 (Apr. 3, 2014).↩
- [29] 49 C.F.R. § 571.223; 49 C.F.R. § 571.224; 49 C.F.R. § 393.86.↩
- [30] Technology & Maintenance Council, RP 732.↩
- [31] NHTSA, Heavy Vehicle Rear and Side Underride Study.↩
- [32] Id.↩
- [33] NTSB, HIR-26-02 (Mar. 31, 2026).↩
- [34] 87 Fed. Reg. 42366 (July 15, 2022).↩
- [35] NTSB, Safety Study SS-13/01, Crashes Involving Single-Unit Trucks That Resulted in Injuries and Deaths.↩
- [36] Id.↩
- [37] NTSB, Safety Recommendation Letter H-14-001 through -007 (Apr. 3, 2014).↩