At least 189 animals died on U.S. commercial flights between 2005 and 2011. Of those, 98 were brachycephalic dogs — more than half.
But that number is a share of reported deaths, not a rate of harm. No per-animal risk can be calculated because airlines did not have to report how many animals they carried until 2015. The 98 count is evidence of a blind spot, not proof of danger.
More than half of every reported animal death on U.S. flights in one six-year window belonged to a single family of breeds. That is the number people remember: 98 of 189. It gets recited as if it were a risk statistic.
It is not.
What the 98 figure actually provides is an inventory of a missing denominator. That gap is the real story, because it is why the 2015 reporting expansion matters more than any individual death.
Why 98 deaths cluster around one skull shape
The dogs behind the 98 share an anatomy, not a behavior. Brachycephalic breeds — bulldogs and pugs are the usual examples — have compressed skulls that narrow the airway.
Under stress, that structure makes breathing harder. In a hold or a cabin, the margin for error shrinks.
What the official record does not supply is the detail that risk-based coverage tends to invent. The historical publication does not break out bulldogs from pugs. It does not offer consolidated tarmac temperatures, flight durations, or apron times for the incidents.
Those conditions are absent from the underlying data, not merely underreported.
An earlier official compilation shows what the record does contain. The publication Deceased Canines in Transit: May 2005–May 2010, issued by the United States. Office of Aviation Enforcement and Proceedings, records 122 dog deaths, 22 deaths of other pets, and 88 pets lost or injured. Even the official five-year window is a count of incidents, and it closed before the 2015 rule required carriers to say how many animals they moved.
Flight deals
most people never see
Our AI monitors 150+ airlines for pricing anomalies that traditional search engines miss. Air Traveler Club members save $650 per trip per person on average: see how it works.
Each deal saves 40–80% vs. regular fares:
The denominator the rule wouldn’t ask for
U.S. airlines have long owed the Department of Transportation monthly reports on pet death, injury, and loss during air transport. Those filings capture date and time, a written account, cause, and any corrective step taken.
What they did not capture for nearly a decade was the total number of animals carried. Without that total, the 98 deaths cannot produce a death rate for brachycephalic dogs.
Everything changed with a single final rule. Published by the U.S. Department of Transportation at 79 FR 37938, it took effect January 1, 2015. For the first time, covered carriers had to disclose how many animals they transported each year.
| Period or requirement | Reported information | Total animals transported |
|---|---|---|
| May 2005–May 2010 official DOT publication | 122 dog deaths; 22 other pet deaths; 88 lost or injured | Not reported |
| 2015 reporting expansion | Monthly incident reports plus annual totals | Required annually |
| Delta 2015 annual report | Transport figure first visible | 98,779 |
| 2023 reporting records | Carrier-level reports identified | Carrier totals; no national series verified |
| Source: U.S. Department of Transportation; U.S. Government Publishing Office (GovInfo); U.S. Department of Transportation Office of Aviation Enforcement and Proceedings | ||
Delta’s 98,779 transported animals in calendar year 2015 made the point concrete. One carrier-year was now visible. What still was not visible was a national breed-adjusted series.
The post-2015 rule collects transport totals, not brachycephalic-specific denominators.
The Department’s long-standing position is that in-transit deaths are an extremely small fraction of animals flown. For the pre-2015 era, that claim rested on assertion. The numbers to test it had not been collected.
The rule is about reporting, not about which dogs fly
The current machinery sits in Part 235 of Title 14, published by the U.S. Government Publishing Office / GovInfo. It applies to scheduled domestic and international passenger service flown by U.S. carriers operating at least one aircraft designed for more than 60 passenger seats.
Reporting runs two ways: monthly incident reports, plus an annual report listing incidents and total animals transported.
The change widened what carriers must disclose, but it did not move the custody boundary. An animal the carrier never takes custody of may never enter the dataset at all, even if it rides in the cabin. Foreign air carriers also sit outside the U.S. reporting duty.
The dataset itself landed at DOT in a roundabout way. An earlier arrangement envisioned filings going to USDA’s Animal and Plant Health Inspection Service.
A February 2005 technical amendment changed that, sending carrier submissions straight to the Department of Transportation. The aviation regulator has owned the published record since the start.
That matters for the transpacific traveler. A U.S.-carrier leg falls inside the reporting regime. The foreign-carrier leg may not.
The transparency you get on one half of a routing is not automatically present on the other.
Why an overhead bin became a reporting blind spot
The rule tracks a narrow slice of what happens to animals in transit. That narrow slice is the difference between a statistic and a story.
The 2018 United flight from Houston to New York is the concrete example. A French bulldog — the owner was Catalina Robledo — was placed in an overhead bin at crew instruction.
The flight ran about 3.5 hours. The dog did not survive. A federal agency investigation followed.
The incident received wide attention in part because it involved a brachycephalic dog in the cabin. The reporting rule runs on carrier custody, not on species: the record covers animals the airline held, so a death at a seatback falls outside it. The rule creates its own blind spots.
What it means for a pet owner at booking
For a U.S. carrier, the annual report is the one place you can see how many animals it moved and how many incidents it reported. Read it alongside the airline’s own breed and cargo rules. Those carrier rules, not the federal totals, decide whether your dog flies.
Whether a particular dog should fly is a veterinary question, not a data question. The rest — cabin versus cargo, and what the airline’s current policy says — is a booking decision the historical record does not make for you.
On a transpacific itinerary, the two legs can come with two different sets of rules. Check both carriers’ policies before you book, not just the one that sells you the ticket.
Key terms
- Denominator
- A denominator is the total an incident count is measured against — here, the number of animals an airline carried rather than the number that died. Without it, a death toll can be expressed only as a share of reported deaths, never as a rate per animal flown. Because the rule produces carrier-level totals rather than industry-wide ones, a single airline’s exposure can be sized up even though the national breed comparison remains out of reach.
- Brachycephalic
- Brachycephalic describes a dog bred for a short, compressed skull, with bulldogs and pugs as the usual examples. The shortened airway makes breathing harder under stress, which is why these breeds appear so heavily in air-transport death reports. Because carriers set their own breed and cargo rules on top of the federal reporting duty, whether a flat-faced dog flies at all is an airline decision rather than a regulatory one.
- 14 CFR Part 235
- Part 235 is the section of the Code of Federal Regulations that tells U.S. air carriers how to report incidents involving animals during air transport. Its reporting changes, published at 79 FR 37938, took effect January 1, 2015, and cover scheduled passenger service flown by carriers with at least one aircraft designed for more than 60 passenger seats. Because the rule was written to collect information rather than to set safety standards, its filings can show that an incident happened without settling whether the dog should have been aboard.
- Carrier custody
- Carrier custody is the point at which an airline takes responsibility for an animal, typically when a pet is handed over at check-in or shipped as cargo. Under Part 235, the reporting clock attaches to that handover, and filings describe the date and time, cause, and any corrective step taken for animals the carrier held. That boundary explains why the federal record can count a death in the hold while missing one that happens at a passenger’s seat on the same flight.
Questions? Answers.
Which dog breeds cannot fly on planes?
The verified historical data does not establish a list of banned breeds. The deaths cluster among brachycephalic or flat-faced dogs, with bulldogs and pugs as cited examples, but the record reviewed does not establish proportions between them or impose a breed ban. Check the airline’s current breed policy instead.
What airlines fly brachycephalic dogs?
U.S. carriers flying scheduled passenger service and operating at least one aircraft designed for more than 60 passenger seats are subject to Part 235 and may carry brachycephalic dogs under their own policies. Foreign air carriers are not subject to the U.S. reporting requirement, so their policies sit outside the American dataset.
Can the DOT figures tell me my dog’s odds of dying on a flight?
No. The 2005–2011 count predates the requirement to report total animals transported, so it cannot produce a per-animal risk rate. Since 2015, carrier-level transport totals have been reported, but a national breed-adjusted comparison of brachycephalic versus non-brachycephalic rates was not verified in the material reviewed here.
Which pets fall outside the DOT incident-reporting rules?
A pet riding at a passenger’s seat outside carrier custody can fall outside the reporting duty. Foreign air carriers are also outside the U.S. reporting requirement, so on an itinerary with a foreign-carrier segment those incidents are not counted in the American dataset.
