That's the stat for all cars and trucks, but drone delivery isn't going to get commuters off the road. How many deaths are due to package delivery vehicles? More specifically, how many are due to the vehicles that will actually be replaced by drones?
That's the number to beat, and even a single collision with an airplane would most likely make up the yearly allotment before we even get into "package falls on someone's head", "drone runs out of battery over the freeway", and all the other failure modes a drone has.
That's why it's important to regularize drone traffic before it get out of hand. Safe flyways to get to neighborhoods. Delivery routes that find residential homes less busy (e.g. when school is in session).
On the subject of vehicles-causing-accidents, it's possible delivery trucks trying to beat a deadline may be over-represented in statistics. Speeding, driving aggressively, parking all over the place. So there's that.
Drones are point to point delivery, which is why they're currently used for transporting high priority items like cooked food and blood transfusions. If drones were legalized in more places, they'd substitute for quick trips to the store. This would be safer for everyone involved and reduce pollution.
Planes fly at much higher altitudes than delivery drones, so the only collision worries would be near airports. The FAA already bans drones near airports and requires special authorization to fly a drone within miles of major airports. Even if you get approval, you can't exceed 400 feet above ground level.
That's the number to beat, and even a single collision with an airplane would most likely make up the yearly allotment before we even get into "package falls on someone's head", "drone runs out of battery over the freeway", and all the other failure modes a drone has.