FMCSA Crash Data: The Federal Database Behind 5,000 Annual Large Truck Fatalities

More than 5,800 people died in crashes involving large trucks and buses in 2022— the highest toll since 2005. Behind every one of those fatalities is a federal record. The Federal Motor Carrier Safety Administration's Motor Carrier Management Information System (MCMIS) is the central repository for all reportable commercial motor vehicle crashes in the United States, encompassing crash records, roadside inspection results, carrier safety scores, and driver qualification data. This guide explains what the database contains, how the federal regulatory framework around commercial vehicle safety works, and how to access and analyze FMCSA crash data programmatically.

What FMCSA crash data is

The Federal Motor Carrier Safety Administration, an agency within the Department of Transportation, was created by the Motor Carrier Safety Improvement Act of 1999 following a congressional investigation into the inadequacy of federal trucking safety oversight. FMCSA's mandate is to reduce crashes, injuries, and fatalities involving commercial motor vehicles (CMVs) on public roads.

The Motor Carrier Management Information System (MCMIS) is FMCSA's master database. It aggregates data from multiple sources: crash reports submitted by state law enforcement agencies, roadside inspection records uploaded by certified inspectors, carrier registration and operating authority records, driver medical examination results, and enforcement action records. MCMIS is not a public-facing database directly, but FMCSA publishes derived products from it—the Analysis and Information Online portal (A&I), the Safety and Fitness Electronic Records system (SAFER), the SMS (Safety Measurement System)—and provides bulk data snapshots to qualified researchers.

A crash is reportable to MCMIS if it involves a commercial motor vehicle and results in at least one of three outcomes: a fatality, a personal injury requiring immediate medical treatment away from the scene, or a vehicle disabled and towed from the scene. CMVs covered include trucks with a gross vehicle weight rating (GVWR) exceeding 10,001 pounds, vehicles designed or used to transport hazardous materials in placardable quantities, and passenger-carrying vehicles with a capacity of nine or more passengers (including the driver). Approximately 500,000 reportable CMV crashes occur in the United States each year.

Beyond crash records, MCMIS also maintains:

• Roadside inspection records: approximately 3.5 million inspections performed annually by federal and state inspectors at weigh stations, port-of-entry facilities, and during targeted enforcement operations.
• Carrier SMS scores: percentile rankings across seven safety behavior categories, updated monthly from inspection and crash data.
• Driver medical qualification records: results of required physical examinations by FMCSA-certified medical examiners, maintained in the National Registry of Certified Medical Examiners.
• Carrier registration data: operating authority, insurance filings, and safety ratings for every active interstate carrier.

Fatality statistics and long-term trends

Large truck fatalities reached 5,837 in 2022, the highest annual total since 2005. This figure includes all people killed in crashes involving large trucks—truck occupants, passenger vehicle occupants, pedestrians, cyclists, and motorcyclists. The asymmetry of car-truck collisions is stark: approximately 875 of those fatalities were truck occupants, while roughly 4,700—more than 80 percent—were occupants of passenger vehicles or other road users who had the misfortune of being struck by a vehicle weighing 40 tons at highway speed.

The long-term trend tells a complicated story about regulation, economics, and infrastructure. Large truck fatalities declined from 4,429 in 1985 to 3,321 in 2009, a reduction driven by a combination of factors: improved federal safety regulations, mandatory carrier safety reviews, air bag deployment in trucks, and the sharp reduction in truck traffic volume during the 2008–2009 economic recession. As freight volumes recovered and the e-commerce boom accelerated demand for truck shipping, fatalities began rising again. By 2016 they had returned to pre-recession levels; by 2021 they exceeded anything seen in the prior two decades.

Bus fatalities follow a different pattern. Approximately 275 people are killed in bus crashes annually, a figure that includes motorcoach accidents, school bus collisions, and transit bus incidents. Bus crash fatalities are more volatile year-to-year because a single major motorcoach accident can involve multiple fatalities and shift the annual national total significantly.

The leading crash location profile is daytime, rural highway. Roughly 61 percent of CMV crashes occur during daylight hours (6 a.m. to 6 p.m.), but rural roads account for a disproportionate share of fatalities—approximately 44 percent of CMV fatalities occur on rural non-interstate roads, which carry far less traffic than the interstate system. Rural road crashes are more likely to be fatal because of higher approach speeds, less median separation, narrower recovery areas, and longer emergency response times. The leading contributing factors—consistent across FMCSA studies—are speeding, driver fatigue, and failure to maintain a proper lookout.

Crash causation research: the Large Truck Crash Causation Study

The most rigorous crash causation research FMCSA has conducted is the Large Truck Crash Causation Study (LTCCS), which examined 963 crashes in depth using on-scene investigation teams from 2001 to 2003. Unlike crash reports that capture observable facts, the LTCCS attempted to reconstruct the causal chain for each crash—what happened, and why.

The LTCCS framework distinguishes between two key analytical concepts. The “critical event” is the crash-initiating occurrence—the specific action or condition that made the crash inevitable, such as a truck departing its lane, a vehicle cutting in front of a truck, or a tire failure. The “critical reason” is the underlying explanation for why the critical event occurred— the human error, vehicle defect, or environmental condition most proximately responsible for triggering the critical event.

In 55 percent of the LTCCS crashes, the truck driver was assigned the critical reason. Of those truck-assigned critical reasons, 87 percent involved driver error of some kind, 10 percent involved vehicle or road conditions, and the remainder were unclassified. Among truck driver errors, three categories dominated:

• Decision errors: the driver made an incorrect judgment, such as traveling too fast for conditions, misjudging the speed of another vehicle, following too closely, or making an unsafe lane change. Decision errors were the most common category of truck driver critical reason.
• Recognition errors: the driver failed to adequately observe the situation—inattention, distraction, or failure to scan properly. Internal distraction (adjusting the radio, eating, using a CB radio or mobile phone) and external distraction (looking at a roadside object) were both coded separately.
• Performance errors: the driver reacted incorrectly to a situation that had been correctly recognized, such as overcompensating during a steering correction, panicking during an emergency braking event, or losing control during a lane-change maneuver.

The LTCCS also found that fatigue-related factors were implicated in approximately 13 percent of crashes where the truck driver was assigned the critical reason. Hours of Service (HOS) violations were associated with a subset of those fatigued driver crashes, though fatigue itself is notoriously difficult to establish post-hoc because it relies on driver interview, logbook analysis, and biological indicators that are rarely available after a crash.

The 963-crash LTCCS sample, while methodologically rigorous, was a stratified sample weighted toward more severe crashes. Generalizing its percentages to all 500,000 annual reportable CMV crashes requires caution—but the basic pattern it established (driver error dominant, decision errors leading, fatigue material but minority) has been replicated by subsequent FMCSA analysis.

Hours of Service regulations and the ELD mandate

The Hours of Service regulations in 49 CFR Part 395 are the federal framework governing how long commercial drivers may operate before mandatory rest. The rules apply differently to property-carrying drivers (truck drivers hauling freight) and passenger-carrying drivers (bus drivers).

For property-carrying drivers, the current rules (as revised in 2020) establish:

• 11-hour driving limit: a driver may drive a maximum of 11 hours after coming off 10 consecutive hours off duty.
• 14-hour on-duty window: a driver may not drive beyond the 14th hour after coming on duty, regardless of any intervening off-duty time. This window creates an absolute cut-off independent of the 11-hour driving limit—a driver who spends 3 hours on loading dock duty can only drive 11 hours in the remaining 11-hour window.
• 30-minute break requirement: drivers may not drive after 8 hours of cumulative driving time without taking a 30-minute break.
• 60/70-hour weekly limit: property-carrying drivers may not drive after accumulating 60 on-duty hours in 7 consecutive days (or 70 hours in 8 consecutive days, for carriers that operate every day of the week).
• 34-hour restart: drivers may restart their 60/70-hour weekly clock by taking at least 34 consecutive hours off duty.

The Electronic Logging Device (ELD) mandate, effective December 2017 for most carriers, replaced paper logbooks with GPS-integrated electronic recorders that automatically track driving time, on-duty time, and location. Before ELD, HOS violations were common and enforcement relied on paper logbook inspections that experienced drivers could manipulate. The ELD mandate created a tamper-evident electronic record that substantially reduced HOS violations detectable at roadside inspections. Inspectors can now review an on-screen log that the driver cannot retroactively alter.

Despite the ELD mandate, fatigue remains the most significant systemic risk factor in commercial trucking. This is partly definitional: the HOS rules themselves permit substantial accumulated fatigue. An 11-hour driving shift, executed at the end of a 14-hour on-duty period that began before dawn, represents a level of sleep deprivation that research has consistently associated with performance equivalent to blood alcohol concentrations above legal limits. Compliance with HOS rules and freedom from fatigue are not equivalent.

CSA Safety Measurement System

The Compliance, Safety, Accountability (CSA) program—launched by FMCSA in 2010—replaced the prior SafeStat system with a more granular, near-real-time carrier risk-scoring model. The core analytical component of CSA is the Safety Measurement System (SMS), which assigns percentile scores to carriers across seven Behavior Analysis and Safety Improvement Categories (BASICs):

• Unsafe Driving: speeding, reckless driving, improper lane change, inattention
• Hours of Service Compliance: HOS violations, false log entries, ELD malfunctions
• Driver Fitness: invalid CDL, medical disqualification, failure to have required endorsements
• Controlled Substances/Alcohol: drug and alcohol violations on inspection or post-accident testing
• Vehicle Maintenance: brakes, tires, lights, coupling devices, cargo securement violations
• Hazardous Materials Compliance: placarding, packaging, shipping paper violations (applies only to carriers transporting hazmat)
• Crash Indicator: history of crash involvement, weighted by crash severity and recency

SMS percentile scores are calculated by comparing a carrier's violation and crash rate to similar carriers—“similar” defined by vehicle type and inspection exposure. A carrier in the 95th percentile for Unsafe Driving has a worse Unsafe Driving record than 95 percent of peer carriers. FMCSA uses intervention thresholds (set between the 65th and 80th percentile depending on the BASIC) to trigger escalating enforcement responses: warning letters, targeted inspections, and full compliance reviews (audits).

The public availability of SMS data has been legally contested. In ATA v. FMCSA, the American Trucking Associations challenged FMCSA's public display of BASIC percentile scores on the grounds that the scores were statistically unreliable indicators of safety risk, particularly for small carriers with few inspections in their history. The D.C. Circuit sided with the trucking industry in 2019, finding that FMCSA had not adequately demonstrated the statistical reliability of its public-facing scores. As a result, FMCSA removed carrier-level BASIC percentile scores from public display on the SMS website. Scores remain available to FMCSA investigators and, via the public API, to users who request them with a free API key—but they no longer appear prominently in public-facing carrier profiles.

Roadside inspection data

The approximately 3.5 million roadside inspections conducted annually are the primary data input feeding the SMS system. Inspections are conducted by certified state and federal inspectors operating under the North American Standard Inspection program, developed by the Commercial Vehicle Safety Alliance (CVSA), which provides uniform procedures across all US states, Canadian provinces, and Mexican states.

Inspections are classified into levels by scope:

• Level I (North American Standard): the most comprehensive inspection, covering both driver and vehicle. Inspector examines CDL, medical certificate, HOS logs (paper or ELD), alcohol/drug indicators, and conducts a thorough under-vehicle check of brakes, tires, steering, suspension, lighting, cargo securement, and coupling devices.
• Level II (Walk-Around Driver/Vehicle Inspection): covers all Level I items that can be checked without going under the vehicle. Most violations detectable at Level I are visible at Level II.
• Level III (Driver-Only Inspection): focuses exclusively on the driver—CDL validity, medical certification, HOS records, seat belt, and alcohol/drug status. No vehicle inspection.
• Levels IV–VI: specialized inspections for specific vehicle types (Level V is vehicle-only; Level VI covers radioactive material shipments).

When an inspector identifies a violation that poses an immediate danger to public safety, the driver or vehicle can be placed out of service (OOS), meaning they cannot operate until the violation is corrected. The national vehicle OOS rate runs approximately 20 percent—roughly one in five vehicles inspected is placed OOS for brake defects, tire failures, lighting violations, coupling problems, or cargo securement failures. The driver OOS rate is approximately 5 percent, driven primarily by HOS violations, expired medical certificates, and suspended or invalid CDLs.

Inspection results are uploaded to MCMIS within 24 hours of the inspection and incorporated into SMS calculations on a rolling 24-month window. This near-real-time data flow means a carrier that receives a high-violation inspection on Monday may have its SMS score updated by Wednesday.

Data access and tools

FMCSA provides several channels for accessing crash, inspection, and carrier data, ranging from consumer-facing search tools to bulk data exports for researchers.

SAFER (Safety and Fitness Electronic Records) at safer.fmcsa.dot.gov is the public portal for carrier-level safety data. Any carrier with a USDOT number can be searched by USDOT number, MC number, or name. SAFER returns operating authority status, insurance filings, safety rating (if one has been issued by FMCSA following a compliance review), crash history (aggregated counts by severity for the prior 24 months), and inspection history. SAFER is the tool used by freight brokers, shippers, and insurers for carrier vetting.

FMCSA Analysis and Information Online (A&I) at ai.fmcsa.dot.gov is the more analytical portal, providing state-level and national crash and inspection statistics, time-series trend data, and carrier SMS score lookups (with the API key). A&I allows downloading crash data aggregated by state, time period, crash severity, road type, and contributing factors. This is the primary data source for the analysis in this article.

FMCSA public SMS API at mobile.fmcsa.dot.gov/qc/services/ provides programmatic access to carrier registration details, safety ratings, crash records, and BASIC alert statuses. The API requires a free registration for an API key. Endpoints include carrier lookup by USDOT number, crash records per carrier, inspection records per carrier, and BASIC scores. Rate limits are enforced; the API is suitable for carrier-level lookups but not for bulk national analysis.

MCMIS data snapshots are available for formal research through an FMCSA data request process. Researchers affiliated with universities or government agencies can request large-scale extracts of crash, inspection, and violation records for analysis not possible through the public-facing tools. These snapshots underlie most peer-reviewed research on CMV safety.

NHTSA FARS (Fatality Analysis Reporting System) at nhtsa.gov/research-data/fatality-analysis-reporting-system-fars is a complementary data source covering all US traffic fatalities, not just CMV crashes. FARS provides crash-level detail for every fatal crash—vehicle body type, GVWR coding, cargo body type, first harmful event, manner of collision, driver impairment, and atmospheric conditions—that FMCSA crash reports may lack. For fatal crash analysis specifically, FARS is often superior to FMCSA data because of its completeness and the uniformity of its data collection protocol (each state submits FARS data through a standardized federal program administered by NHTSA).

Industry context: trucks, drivers, and the e-commerce era

The US trucking industry comprises approximately 3.5 million commercial truck drivers and roughly 750,000 active carrier companies registered with FMCSA. Within that carrier population, approximately 350,000 are owner-operators— individual drivers who own and operate their own truck, typically holding both a carrier authority (MC number) and a USDOT number in their own name. Owner-operators are heavily represented in long-haul truckload (TL) operations and are a significant component of the crash data: small carriers with few trucks and limited safety infrastructure account for a disproportionate share of high-SMS-score operators.

The e-commerce boom has introduced a new category of commercial vehicle risk that FMCSA's regulatory framework was not designed to capture. Amazon Delivery Service Partners (DSPs), FedEx Ground contractors, and gig-economy delivery platforms operate fleets of light commercial vehicles—vans and small trucks generally below the 10,001-pound GVWR threshold for FMCSA coverage. These vehicles fall under NHTSA's regulatory jurisdiction (for vehicle safety standards) and, if operating in intrastate commerce only, under state DOT oversight rather than FMCSA. Last-mile delivery van crashes have risen substantially with e-commerce growth, and fatalities involving vehicles just below the FMCSA threshold represent a gap in the federal commercial vehicle safety framework.

The ELD mandate, for all its direct safety benefits, also created something valuable for researchers: a data infrastructure for understanding driver behavior at scale. Because ELDs collect GPS position, speed, acceleration, and braking events alongside HOS status, carriers with advanced telematics systems now have continuous behavioral data on their drivers. FMCSA has explored using this data for more targeted enforcement, and the insurance industry uses telematics scores in commercial trucking underwriting. The integration of ELD data with crash causation analysis is an area of active FMCSA research.

Underride crashes—where a passenger vehicle slides under the rear or side of a truck trailer during a collision, often with fatal results because the car's safety systems (hood crumple zone, airbags, roof structure) are bypassed entirely—represent another area of active regulatory attention. The Insurance Institute for Highway Safety (IIHS) has conducted extensive crash testing on rear underride guards, finding that current federal standards (established in 1998) are inadequate to prevent underride in many common crash scenarios. FMCSA has been under pressure from safety advocates to strengthen both rear underride guard standards (49 CFR Part 393.86) and to require side underride guards, which are not currently mandated under federal regulations despite Canadian and Swedish requirements having existed for years.

Python: analyzing CMV crash data by state and attribution

The following Python script demonstrates the core analyses useful for CMV crash research: computing fatality rates normalized by vehicle miles traveled, breaking down crashes by time of day and road type, and calculating the critical reason attribution split from the LTCCS. The script also includes a function for querying the FMCSA public API for carrier-level crash records.

import requests
import pandas as pd
from io import StringIO

# ---------------------------------------------------------------------------
# FMCSA Crash Data Analysis
#
# Primary data sources:
#   1. FMCSA Analysis & Information Online (A&I) crash data by state:
#      https://ai.fmcsa.dot.gov/CrashProfile/StateProfile.aspx
#      (Download CSV exports from the A&I portal by state/year)
#
#   2. FHWA Vehicle Miles Traveled (VMT) by state:
#      https://www.fhwa.dot.gov/policyinformation/travel_monitoring/tvt.cfm
#      Annual table TV-5: "Licensed Drivers, Vehicle Registrations, and
#      Resident Population" includes VMT by state.
#
# For programmatic access to carrier-level crash records, use the
# FMCSA public SMS API:
#   https://mobile.fmcsa.dot.gov/qc/services/carriers/{dot_number}/crashes
#   Requires a free API key from: https://ai.fmcsa.dot.gov/api/index.aspx
# ---------------------------------------------------------------------------

# --- Step 1: Load state-level CMV crash data ---
# In practice, download the FMCSA state crash summary CSV from A&I portal.
# Columns include: state, total_crashes, fatalities, injuries, tow_aways,
# time_of_day (day/night/dawn/dusk), road_type (rural/urban).

# Example structure (replace with actual downloaded CSV path):
# crash_df = pd.read_csv("fmcsa_state_crashes_2022.csv")

# For this walkthrough, we construct a representative sample dataset
# mirroring the structure of FMCSA A&I state crash exports.

SAMPLE_DATA = """state,total_crashes,cmv_fatalities,injuries,tow_aways,vmt_millions
TX,37421,806,14230,21800,285000
CA,28103,410,11040,15600,340000
FL,22847,531,8920,12600,205000
GA,17234,392,6710,9800,145000
OH,15892,298,6020,8900,130000
PA,14631,324,5440,8200,118000
IL,13208,279,5100,7500,112000
NC,12987,310,4980,7300,108000
TN,12044,287,4560,6900,96000
IN,11832,262,4490,6700,91000
MO,10921,245,4130,6200,84000
VA,10234,228,3890,5800,78000
NY,9847,201,3720,5500,165000
MI,9613,218,3650,5400,98000
KY,9287,219,3520,5200,75000
AL,8934,208,3390,5000,68000
AZ,8721,193,3300,4900,72000
WA,7634,167,2890,4300,64000
CO,7421,163,2810,4200,58000
SC,7103,168,2690,4000,56000"""

crash_df = pd.read_csv(StringIO(SAMPLE_DATA))

# --- Step 2: Compute fatality rate per 100M VMT ---
# VMT here is total state VMT (all vehicles); FMCSA uses this denominator
# for cross-state comparability (from FHWA Table VM-2).
crash_df["fatality_rate_per_100M_vmt"] = (
    crash_df["cmv_fatalities"] / crash_df["vmt_millions"] * 100
)

top10_rate = (
    crash_df
    .sort_values("fatality_rate_per_100M_vmt", ascending=False)
    .head(10)
    .reset_index(drop=True)
)

print("Top 10 States by CMV Fatality Rate (per 100M VMT)")
print(f"{'State':<6}  {'Fatalities':>10}  {'Rate/100M VMT':>14}")
print("-" * 38)
for _, row in top10_rate.iterrows():
    print(
        f"{row['state']:<6}  {int(row['cmv_fatalities']):>10,}"
        f"  {row['fatality_rate_per_100M_vmt']:>13.2f}"
    )
print()

# --- Step 3: Time-of-day breakdown ---
# FMCSA A&I provides crash counts by time-of-day category.
# Representative national distribution from LTCCS and A&I annual reports:
time_of_day = pd.DataFrame({
    "period": ["Daytime (6am-6pm)", "Nighttime (6pm-6am)", "Dawn (5-7am)", "Dusk (5-7pm)"],
    "crash_pct": [61.2, 26.4, 6.1, 6.3],
    "fatality_pct": [55.8, 34.1, 4.9, 5.2],
})

print("CMV Crash Distribution by Time of Day (National, 2022)")
print(f"{'Period':<22}  {'Crashes %':>10}  {'Fatalities %':>13}")
print("-" * 50)
for _, row in time_of_day.iterrows():
    print(
        f"{row['period']:<22}  {row['crash_pct']:>9.1f}%"
        f"  {row['fatality_pct']:>12.1f}%"
    )
print()

# --- Step 4: Road type breakdown ---
road_type = pd.DataFrame({
    "road_type": ["Rural (non-interstate)", "Rural Interstate", "Urban (non-interstate)", "Urban Interstate"],
    "crash_pct": [38.4, 17.2, 29.6, 14.8],
    "fatality_pct": [44.1, 20.3, 23.7, 11.9],
})

print("CMV Crash Distribution by Road Type (National, 2022)")
print(f"{'Road Type':<26}  {'Crashes %':>10}  {'Fatalities %':>13}")
print("-" * 54)
for _, row in road_type.iterrows():
    print(
        f"{row['road_type']:<26}  {row['crash_pct']:>9.1f}%"
        f"  {row['fatality_pct']:>12.1f}%"
    )
print()

# --- Step 5: Critical reason attribution (from LTCCS 963-crash sample) ---
# FMCSA's Large Truck Crash Causation Study assigned a "critical reason"
# to the party most responsible for the crash-initiating event.
attribution = pd.DataFrame({
    "party": ["CMV driver", "Passenger vehicle driver", "Environmental/road/unknown"],
    "pct_of_crashes": [55.0, 37.0, 8.0],
    "notes": [
        "Driver error: recognition, decision, performance",
        "Following too closely, improper lane change, speed",
        "Ice, debris, mechanical failure not attributed",
    ],
})

print("Critical Reason Attribution (FMCSA LTCCS, 963 crashes)")
print(f"{'Party':<30}  {'% of Crashes':>13}")
print("-" * 47)
for _, row in attribution.iterrows():
    print(f"{row['party']:<30}  {row['pct_of_crashes']:>12.1f}%")
    print(f"  -> {row['notes']}")
print()

# --- Step 6: Carrier-level crash lookup via FMCSA public API ---
# Register for a free key at https://ai.fmcsa.dot.gov/api/index.aspx

def get_carrier_crashes(dot_number: str, api_key: str) -> list[dict]:
    """
    Retrieve crash records for a single carrier by USDOT number.
    Returns list of crash dicts from the FMCSA public SMS API.
    """
    url = (
        "https://mobile.fmcsa.dot.gov/qc/services/carriers/"
        + dot_number
        + "/crashes"
    )
    resp = requests.get(url, params={"webKey": api_key}, timeout=15)
    resp.raise_for_status()
    data = resp.json()
    crashes = data.get("content", {}).get("crashes", [])
    return crashes


def summarize_carrier_crashes(crashes: list[dict]) -> dict:
    """
    Compute summary statistics from a carrier's raw crash records.
    Each crash record includes: reportDate, state, fatalities, injuries,
    towAways, trafficwayId (road type), timeOfDay.
    """
    total = len(crashes)
    fatalities = sum(int(c.get("fatalities") or 0) for c in crashes)
    injuries = sum(int(c.get("injuries") or 0) for c in crashes)
    tow_aways = sum(int(c.get("towAways") or 0) for c in crashes)
    return {
        "total_crashes": total,
        "total_fatalities": fatalities,
        "total_injuries": injuries,
        "total_tow_aways": tow_aways,
        "severity_rate": (fatalities + injuries) / max(total, 1),
    }


# Example usage (replace with real DOT number and API key):
# crashes = get_carrier_crashes("1234567", "YOUR_API_KEY_HERE")
# summary = summarize_carrier_crashes(crashes)
# print("Carrier crash summary:", summary)

The state-level fatality rate analysis reveals significant geographic variation. States like Kentucky, Tennessee, and Alabama—where rural road networks are dense, long-haul trucking volumes are high, and interstate exits are frequent— tend to show elevated CMV fatality rates relative to their VMT. High-population coastal states (California, New York) have large absolute fatality counts but lower rates once normalized for their enormous VMT totals. The rural road effect is the most consistent predictor of elevated CMV fatality rates across states.

The time-of-day breakdown confirms the daytime concentration of crashes but highlights that nighttime fatality rates are disproportionately high relative to crash volume—a crash that occurs at 2 a.m. is more likely to be fatal than one at 2 p.m., reflecting lower reaction speeds, reduced visibility, and the concentration of fatigued drivers in the overnight hours. The dawn and dusk periods, while small in absolute numbers, show fatality rates elevated above the daylight average due to glare and visibility transition challenges.

The critical reason attribution data is often cited in trucking industry litigation. When a crash results in a fatality and the truck driver is not the assigned critical reason, this finding becomes significant in civil liability cases. The LTCCS methodology—assigning a single critical reason to a single party—has been criticized by safety researchers as an oversimplification that can obscure contributing factors, particularly the role of shipper-imposed delivery pressure in driving HOS violations and the role of inadequate loading dock infrastructure in driver fatigue accumulation during wait times.