A crash-tested car harness. The maker states it is tested to a child-restraint standard; three attachment points spread crash forces across the chest.
Crash-tested to child standards
Three-point design
Vest-style comfort
FAA approved for air travel
Expensive
Difficult sizing
Bulky
Safety: 10/10 · Comfort: 8/10 · Fit: 7/10 · Value: 6/10
These are our own scores for the things that can be judged from the specification, the materials and how the product is built. They are not Amazon's customer rating.
A walking harness is built to withstand a dog pulling. A crash applies forces on a completely different order.
In a collision, an unrestrained dog becomes a projectile. That is a danger to the dog, and a serious danger to the people in the car. It is also why a walking harness clipped to a seatbelt is a poor idea: the hardware and the stitching are not designed for crash loads, the force concentrates on a narrow strap, and a harness that fails at the wrong moment leaves a loose dog in a crashed vehicle.
Car restraint is a separate product category from walking equipment, and they are not interchangeable.
This is the thing to understand before comparing products.
There is no mandatory government crash-test standard for pet restraints in most jurisdictions. Any manufacturer can print "crash tested" on a package without saying what test, at what speed, with what dummy weight, or whether it passed.
What distinguishes a serious product is the specifics: which standard was used, at what test speed, with what weight of test dog, and whether results are published. Independent testing by a third party is stronger than in-house testing.
Testing to a child-restraint standard is the strongest claim available in this category, because those standards are demanding and well defined. Where a manufacturer states that, it is worth reading exactly what they tested and how.
We have not conducted or verified crash testing ourselves, and we will not repeat a marketing claim as if we had.
Look for these, whatever the brand:
Wide, padded chest plate. Crash force must be spread across the sternum and chest, not concentrated on the neck or abdomen. A narrow strap concentrates force exactly where it does damage.
Structural webbing and hardware rated for load, with reinforced stitching at every attachment.
A short, fixed connection to the seatbelt system. A long tether lets the dog build up speed before the restraint engages, which converts a restrained dog into a partially unrestrained one.
Rear-seat use. Front airbags are dangerous to dogs, as they are to children.
A correctly built restraint fitted loosely is not a restraint.
Measure girth around the deepest part of the chest and follow the manufacturer's sizing chart. Do not guess between sizes.
Two fingers under any strap, no more.
The chest plate sits over the breastbone, not on the throat and not across the front of the shoulders.
Re-check after a few journeys, and seasonally as coat thickness and body shape change.
Keep the tether short enough that the dog cannot reach the front seats or leave the seat.
A crash-tested crate, secured. For larger dogs this is often the better answer, and an unsecured crate is itself a projectile — it must be strapped down or wedged in the load space.
A rigid barrier behind the rear seats. Keeps the dog out of the passenger compartment; does not restrain the dog within the boot.
A booster seat for a small dog, provided it uses a proper harness rather than a clip to a collar.
A collar clipped to a seatbelt. Never. In a crash this concentrates the entire force on the dog's neck.
In some jurisdictions, an unrestrained animal in a vehicle can contribute to a driving offence, and insurers have declined claims where an unrestrained animal was a factor. Check your local rules.
Beyond the law: a loose dog is a distraction, and a dog in a footwell or on a lap is a hazard before any collision happens.
No. Walking harnesses are built to withstand pulling, not crash forces, and the hardware and stitching are not rated for them. In a collision an unrestrained or inadequately restrained dog becomes a projectile — a danger to itself and to the people in the car.
Less than it sounds, because there is no mandatory government standard for pet restraints in most jurisdictions. Any manufacturer can print it without saying what test, at what speed, with what dummy weight, or whether it passed. Look for the specifics, and for independent third-party testing rather than in-house.
Measure girth around the deepest part of the chest and follow the sizing chart rather than guessing. Two fingers under any strap, no more. The padded chest plate should sit over the breastbone, not on the throat or across the front of the shoulders, and the tether should be short enough that the dog cannot leave the seat.
The rear seat, never a front seat with an active airbag — airbags are dangerous to dogs as they are to children. For a large dog, a crash-tested crate secured in the load space is often the better system; an unsecured crate is itself a projectile.
Yes, even if it looks undamaged. Webbing and stitching are load-bearing and a single impact compromises them — the same rule applies to child car seats for the same reason.