Rail safety recommendations tied to major crashes have repeatedly stalled at the federal level, according to a new analysis that examined what happens to guidance issued after investigations by the National Transportation Safety Board.
The Howard Center for Investigative Journalism at the University of Maryland analyzed data from the NTSB—an agency that investigates major accidents and recommends safety improvements—and found that the Federal Railroad Administration fully implemented only five of 81 recommendations issued from 2015 through 2024. The center said the overall record reflects an extended pattern in which industry opposition meets federal inaction, followed by congressional delays and then more tragedy.
The center’s findings draw attention to two broad areas that investigators and regulators have long linked to accidents: track condition and track inspection, and employee scheduling and fatigue. The analysis said the federal regulator has left many safety recommendations outstanding, including measures aimed at requiring track replacement tied to rail wear limits and rules intended to address crew fatigue.
The analysis also highlighted an ongoing gap between investigations and regulatory outcomes in the context of a 2021 derailment of Amtrak’s Empire Builder passenger train in Joplin, Montana. The NTSB said the 2021 crash involved worn-down track defects owned by BNSF Railway, and that a company inspector likely missed defects due to workload. The NTSB also cited the Federal Railroad Administration’s failure to set rules both for track replacement and for workloads after previous rail accidents.
On the Federal Railroad Administration’s implementation status, the analysis cited FRA spokesman Warren Flatau, who said the agency had 102 open NTSB recommendations as of late September, including some dating to 1998. Flatau said the FRA was taking action or planning to take action on more than 70% of them and that the agency planned no immediate action on the rest, generally because officials had determined they were infeasible or could be addressed under existing regulations. The FRA did not respond to further requests for details on what actions it is taking.
The Howard Center’s report connected the derailment’s issues to a longer regulatory history. It cited an earlier NTSB-backed effort to regulate how much rail wear is allowed before replacement—after investigators concluded that dangerously worn-down rail contributed to a 1980 derailment in Muldraugh, Kentucky, involving vinyl chloride. Investigators concluded that track defects collectively contribute to about a quarter of rail accidents in the United States, and the Howard Center said that since the Muldraugh derailment, 44 people have died and another 2,300 have been injured in nearly 15,000 main-line accidents it tied to track defects using FRA accident data.
The analysis said the Montana derailment occurred on a long-distance route carrying more than 300,000 passengers each year between Chicago and the Pacific Northwest. It reported that NTSB officials said the rail wear limits recommended in earlier years would have required BNSF to replace rail involved in the crash. It also said a BNSF inspector drove over the track in a company vehicle twice during the week of the derailment and documented concerns, but did not get out to walk the tracks—something the NTSB said likely would have identified defects—and that there had been no reported walking inspection in nearly a year.
The center said the FRA and rail regulators have also faced persistent scrutiny around inspection methods. It noted that the Association of American Railroads, which represents major U.S. freight railroads, has pushed to reduce required human inspections in favor of additional automated inspections. The analysis said the FRA granted an AAR request on Dec. 5, allowing participating railroads to cut weekly visual inspections in half if they comply with specified safety and reporting conditions, and the NTSB said automated inspections are a supplement rather than a replacement because neither method is 100% effective.
The analysis contrasted that view with rail workers’ accounts of the practical impact on inspectors. It quoted rail inspector Lance Marston, who said: “There’s just no replacement for getting out there and putting boots on the ground.” It also cited BNSF track inspector Christie Lee, who said automation had reduced staffing and increased workloads, describing how the territory he covers changed from earlier coverage with more inspectors to later coverage “solely on two guys, two people, to ride the whole thing.”
The report also examined fatigue-related accidents and how regulators and railroads have treated crew scheduling and rest. It described a 2004 crash in Macdona, Texas, in which a Union Pacific engineer, operating on less than two hours’ sleep, hit a BNSF train and helped lead to the release of chlorine gas that suffocated the conductor and two residents and left 30 people in severe respiratory distress. The NTSB said the engineer’s fatigue—linked to a chronic lack of sleep, disrupted sleep cycle, and long work shifts—contributed to the fatal mistake.
The Howard Center said investigators found regulators failed to act on safety recommendations that would have addressed crew scheduling practices, even after Congress mandated fatigue risk management plans under the Rail Safety Improvement Act. It reported that Congress required fatigue plans and other measures, including screening employees for sleep disorders and providing fatigue training, but said the FRA finished approving fatigue plans for all railroads only this year after more than 15 years.
In the fatigue context, the analysis said rail unions argued that punitive attendance policies discourage taking personal rest days. It also cited former NTSB investigator Rick Narvell, who said, “They just want bodies on the trains, and honestly I don’t think they care if they’re fatigued or not.”
The Howard Center’s analysis further described the rail industry’s efforts to delay or narrow adoption of safety technology, using the long fight over positive train control as an example. It said the NTSB first recommended the FRA consider requiring the technology in 1970 and that Congress ultimately stepped in after several early-2000s tragedies, setting a 2015 deadline—then later extending it after railroads lobbied to postpone. The analysis said railroads eventually reached full compliance in 2020, but that by then a tally inside the NTSB cited more preventable accidents and casualties than the schedule initially targeted.
The center said the pattern continues as the industry pursues regulatory changes. It described how, when the Trump administration invited public input this spring on how to reduce federal bureaucracy, the Association of American Railroads submitted a spreadsheet itemizing 80 regulations for modification or repeal and laid out a deregulatory agenda aimed at reducing federal oversight. The analysis said the list included repealing a requirement for minimum two-person crews and reducing administrative requirements for fatigue management plans, and that in a letter AAR said its goal was to cut regulatory red tape in a “manner that is consistent with railroad safety.”
The three major freight railroads included in the analysis—BNSF, Norfolk Southern and Union Pacific—did not dispute the report’s facts or findings, the Howard Center said. All sent statements saying they are committed to safety.
University of Maryland reporters Mary Burke, Taylor Nichols, Adriana Navarro and April Quevedo contributed reporting and data analysis for this story.