Waking Up To Fatigue

FAA Acting Administrator Robert Sturgell addressed the symposium, pointing out that current FAA regulations only include the mandatory scheduling of crew rest periods. There's nothing dealing with fatigue mitigation. "We like to think that not getting enough sleep, working tired, being a little drowsy---that they're just all part of how Americans live," Sturgell said. "We don't like to think that fatigue can be linked to catastrophe, but there's some truth in that." The controllers union has been telling the FAA for some time that fatigue is an issue among its members, some of whom have been working assigned overtime and six-day weeks for years. Sturgell didn't specify how the FAA will address fatigue---whether it will limit itself to being a facilitator for industry efforts or embark on a new round of rulemaking.

In its safety recommendation, the NTSB reported that it was investigating a February 13, 2008, incident in which both pilots on a Go!/Mesa Airlines regional jet fell asleep; the flight was out of radio contact with ATC for 18 minutes. Flight 1002 was a Bombardier CL-600 with two pilots, a flight attendant and 40 passengers. At 9:40 a.m., as the flight was crossing the Maui, Hawaii, the flight crew failed to respond to an ATC instruction. Despite continuous efforts by ATC to raise the flight, there was no reply. The airplane was heading southeast over the Pacific Ocean when the crew finally radioed ATC, which directed the flight back to its destination airport, General Lyman Field in Hilo, Hawaii.

The NTSB learned that the crew had been on duty for less than 4.5 hours when the incident occurred. The flight, however, took place on the third day of a trip sequence that involved numerous short flight segments and early starting times. The NTSB also says that one of the pilots was diagnosed with obstructive sleep apnea, a condition associated with poor sleep quality and daytime fatigue.

Since 1972, the Safety Board has issued 115 safety recommendations relating to fatigue in all modes of transportation. Fatigue in aviation was the subject of 32 such recommendations.

Upon investigating the October 19, 2004, crash of Corporate Airlines Flight 5966 at Kirksville, Miss., the NTSB concluded that fatigue likely contributed to the degraded performance of the pilots. Both pilots and 11 passengers were killed, while two passengers received serious injuries. The pilots had been on a nonprecision approach at night in IMC. They had been on duty for 14.5 hours and had conducted five previous landings in poor weather conditions.

In its investigation of a February 18, 2007, incident involving Delta Connection Flight 6448 at Cleveland Hopkins International Airport in Cleveland, Ohio, the NTSB found that the captain was suffering from fatigue. The Embraer regional jet landed on runway 28 when it was snowing; it didn't stop until it overran the runway's end, hit an ILS antenna and struck an airport perimeter fence.

The captain had been experiencing intermittent insomnia for several months. He told investigators that he had been unable to sleep the night before the accident and had been awake for all but one of the preceding 32 hours. Before the flight, he told the other pilot that he was tired, but he didn't tell take advantage of a policy allowing pilots to excuse themselves due to fatigue because he was afraid the company would fire him. According to the NTSB, "The administration of this policy didn't permit flight crew members to call in as fatigued without fear of reprisals."

As another example of why the FAA needs to act on fatigue management, the NTSB pointed to its investigation of the April 12, 2007, accident at Cherry Capital Airport in Traverse City, Mich., in which Pinnacle Airlines Flight 4712 ran off the runway in snow. None of the 52 people on board were hurt, but the CL-600 regional jet received substantial damage. The pilots had been on duty for 14 hours; they had conducted four previous landings in poor weather conditions. The cockpit voice recorder revealed that the pilots made several comments about being fatigued, and yawning sounds could be heard.

While the NTSB's safety recommendation made no mention of GA accidents involving fatigue, it remains just as important for general aviation as it is for the airlines. In some respects, dealing with fatigue is more difficult for GA pilots operating without any framework for duty hours, maximum flight times or required rest periods. Many GA pilots have to decide whether to fly after a long day of business meetings, a poor night's sleep, a long period of physical activity or some combination of these and other factors. Pressures to make the flight might be self-imposed or come from passengers and business associates. [See "Managing Fatigue" in the print edition of P&P October 2008.]

Take, for example, a PA30 Twin Comanche that crashed while making what should have been a routine approach. The pilot may not have been able to discern that fatigue was impairing his ability to function because he had been used to dealing with a number of troubling factors for an extended period of time.

At 6:19 p.m. on January 13, 2006, the airplane was descending for a landing at Visalia Municipal Airport in Visalia, Calif. The pilot had turned about a half-mile final and had been in radio contact via the local Unicom with a King Air that was on a four-mile final. The King Air pilot had radioed the accident pilot that there was plenty of room and he should go ahead and turn final for landing. The Twin Comanche struck the ground about 400 feet short of the approach end of runway 30. The commercial pilot and the three passengers were killed. The pilot had flown from Visalia to Byron Airport in Byron, Calif., to pick up two children; this was the return flight.

The pilot held a commercial certificate with ratings for single-engine and multi-engine airplanes and instruments. His logbooks couldn't be located, but on his application for an FAA medical about two years before the accident, he reported having 5,700 flight hours.

Investigators learned that the pilot had been traveling for work in the days preceding the accident. He took a commercial flight to Portland, Ore., and worked there before driving to Seattle, Wash., for two days of meetings. The night before the accident flight, he flew commercially from Seattle to California on a flight that was delayed an hour. He dropped off a friend in Fresno at 12:30 a.m., and arrived home at 1:30 a.m., the morning of the accident.

Toxicological testing indicated the presence of drugs usually used to control high blood pressure. Also, there was a high level of doxylamine, often used in over-the-counter sleep aids. The pilot hadn't reported using these drugs on his medical application. Investigators subpoenaed the pilot's personal medical records, which indicated that he had a history of lower back pain. On four different occasions, the pilot had complained to his doctor about experiencing difficulty sleeping due to the pain. The NTSB report suggested that the high level of doxylamine had likely accumulated due to daily use or use in excess of the maximum recommended dose.

The NTSB determined that the probable cause of this accident was the pilot's failure to maintain airspeed during the landing approach, which resulted in a stall and uncontrolled descent. Contributing factors included the pilot's impairment due to his prolonged use of a highly sedating over-the-counter sleep aid and the onset of fatigue due to lack of sleep.

Peter Katz is editor and publisher of NTSB Reporter, an independent monthly update on aircraft accident investigations and other NTSB news. To subscribe, write to: NTSB Reporter, Subscription Dept., P.O. Box 831, White Plains, NY 10602-0831.