As an industry, we like to point to the relatively good safety record of flight instruction, and with some justification: The fatal accident rate on instructional flights is less than half that of general aviation as a whole. Even so, the long-term average hovers around two a month. Over 20 years, 478 fatal accidents in civilian flight instruction killed a total of 909 students, CFIs, and passengers.
Of course, the great majority of these (87 percent) involved airplanes, which accounted for 90 percent of the casualties. Helicopters made up most of the rest, although there were about a dozen in other categories of aircraft (chiefly gliders and gyrocopters). By far the largest share of those in airplanes (almost 30 percent) were what the Air Safety Institute classifies as “maneuvering accidents,” a catch-all category that includes everything from low-altitude buzzing to planned aerobatics to turns in the traffic pattern. (Mid-air collisions were a distant second at 13 percent, while mechanical problems were way down the list at 6 percent. That last might help reassure prospective students who are taken aback by the first sight of a well-used training aircraft.)
Because the “maneuvering” category is so diverse, it’s worth breaking down further, and some of the details might be surprising. The largest share of fatal maneuvering accidents (once again about 30 percent) came while practicing emergency procedures, and almost half of those (17 of 36) occurred in twins, which only accounted for 15 percent of fatal instructional accidents overall. Clearly, the precautions most operators require while teaching single-engine instrument approaches and VMC demonstrations are justified by the record. Simulated engine failures in singles have been a problem, too, especially when the power was pulled close to the ground. Airplanes from a Piper Tomahawk to a Pilatus PC-12 were destroyed attempting to practice a turn back to the field following a power loss during initial climb. The old caution about not allowing a simulated emergency to become the real thing still bears repeating, not to mention evaluation during instructor check flights.
Almost as common were inadvertent stalls outside of emergency drills, which came in several flavors. All told, there were 34, of which a discouraging 19 occurred during attempted low passes or the pull-ups that followed them. (Remember, these were instructional accidents, and only four were on student solos. In 15 out of 19, the CFI on board was supposed to have been providing adult supervision.) Another 10 involved steep turns at low altitudes, while just five took place during prescribed airwork.
Two dozen fatal accidents involved attempted aerobatics. Two-thirds of them were during intentional aerobatic practice, mostly in more-or-less suitable airplanes (although one freight operator did lose two pilots and a Beech 99 after the crew attempted a barrel roll during a six-month proficiency check). Eight other pilots pulled their airplanes apart attempting maneuvers for which they weren’t designed or approved. Again, three-quarters of those flights were under the command of a CFI.
In 20 accidents, low-altitude flights were suddenly interrupted by something solid, like trees, wires, or hillsides. Fourteen of these occurred in daylight VMC, and four more in VMC at night. Finally, six fatal accidents were the result of ill-advised canyon flights in which the airplanes were unable to either turn around or out-climb the terrain ahead. Three of those flights were specifically described as instruction in mountain flying.
So while flight training’s safety record is good, there would seem to be a few clear opportunities for improvement. Messing around at low altitudes caused more fatal accidents than mechanical failures and fuel mismanagement combined. Instructors who are rigorous in their adherence to the FAR minimum-altitude rules stand a much better chance of inspiring similar respect in their students. And operators do well to be attentive to the ways in which their CFIs prepare for, execute, and recover from simulated emergency training.