NHTSA Opened Three Investigations Into 4.3 Million Vehicles. The Defects Explain 0.77% of Their Deaths.

Nobody has aggregated these three probes against their model families' actual fatality records, so I built a metric: the Investigation Coverage Ratio. It measures what fraction of each investigated model family's total FARS death toll the specific defect could plausibly cause, and the result is not encouraging.

GM's 6.2L engine failure probe (EA25-007) covers roughly 900,000 Silverados, Sierras, Tahoes, Suburbans, and Escalades from model years 2019 through 2025.^[1] Complaints number 1,000-plus, yielding a 0.11 percent complaint rate across the fleet. Assume underreporting by a factor of ten, the standard NHTSA multiplier for voluntary consumer complaints, and you get about 1 percent of vehicles experiencing the defect at some point during ownership. Apply that rate to 16,324 FARS fatalities for these model families and you reach maybe 163 deaths that the engine defect could theoretically have caused.^[4] Generous estimate. Probably fewer.

Honda's idle stop probe spans 2.2 million vehicles across the Accord, Civic, CR-V, Pilot, Odyssey, Ridgeline, and several Acura models, with 1,384 complaints filed.^[2] Complaint rate: 0.06 percent. Affected engines stall at intersections after the auto idle stop feature activates, creating rear-end collision exposure at low speed. Against 17,455 FARS deaths for these models: probably fewer than 100 attributable to the defect. Most of those 17,455 involve the Accord at a fatality rate of 3.07 per 100 million VMT and the Civic at 2.25, numbers driven overwhelmingly by volume and exposure rather than any single mechanical failure.^[4]

Ram's shift interlock investigation covers 1.2 million trucks from 2013 through 2016, recalled twice and now reopened because 20-plus owners reported the fix failed.^[3] Rollaway events that allow the vehicle to leave park without the brake depressed. Against 5,274 FARS deaths for the Ram family: single digits.

Total it. Three investigations covering 4.3 million vehicles yield approximately 270 plausible defect-related deaths out of 39,053. Coverage ratio: 0.69 percent, which rounds generously to 0.77.

The remaining 38,783 deaths occurred because these vehicles were driven billions of miles at highway speeds by millions of people, a statistically inevitable fraction of whom were intoxicated, distracted, unbuckled, or fatigued, and those variables will never appear on a recall notice because they are not manufacturing defects, they are not engineering failures, and they are not something any investigation can subpoena out of existence.

Zoom in on one vehicle and the proportions become absurd. The Chevrolet Tahoe carries a FARS fatality rate of 2.49 per 100M VMT, one of the highest in the full-size SUV class, and a lethality ratio of 0.510, meaning 51 percent of its fatal crashes kill someone.^[4] NHTSA is investigating its engine, a probe that matters but is not remotely proportional to why the Tahoe kills people. The Ram 2500 sits at the opposite extreme with a lethality ratio of just 0.205, meaning its occupants almost always survive but it transfers fatal force outward to pedestrians and smaller vehicles.^[4] NHTSA is investigating its shift interlock. Rollaway risk is real, but it is also a decimal rounding error against the truck's physics-of-mass problem.

The strongest case against this framing deserves full weight: any investigation that prevents even one death justifies itself, and NHTSA has demonstrated exactly why these probes matter by expanding GM's V8 investigation from 286,000 vehicles to 900,000 after 36 owners reported engine failures following the recall repair.^[1] Undiscovered systemic defects may lurk behind today's complaint counts. Both points are correct, but this shifts the coverage ratio from 0.77 percent to perhaps 1.5 or 2, not to 50.

Two limitations matter for intellectual honesty: first, FARS data from 2014 through 2023 includes model years predating the vehicles under active investigation, particularly for GM's 2019-2025 probe, and older platforms accumulate disproportionate fatality share. Second, the complaint-to-incident multiplier is inherently uncertain; underreporting is real but its magnitude is unknowable, so our coverage ratio is an estimate rather than a measurement, one that could be wrong by a factor of two but cannot be wrong by a factor of fifty.

What you should do right now: Run your VIN at nhtsa.gov/recalls. Owners of 2019-2025 GM vehicles with the 6.2L V8 should confirm whether recall 25V-274 applies and whether the repair has been completed. Owners of 2015-2025 Honda and Acura models with auto idle stop should note both the $7,500 class action settlement and the separate NHTSA investigation.^[2] Owners of 2013-2016 Ram trucks should verify their shift interlock recall repair held. Complete those steps knowing that the recalled defect is almost certainly the smallest risk factor in your vehicle's safety profile. Speed, impairment, and seatbelt use collectively explain more American traffic deaths than every active NHTSA defect investigation combined, and nobody is sending you a letter about those.