Controlled flight into terrain accidents among commuter and air taxi operators in Alaska

Background: Between 1990 and 1998, aviation accidents in Alaska caused 100 occupational pilot deaths (equivalent to 430/100,000pilots/year, approximat ely 86 times the overall U.S. worker fatality rate). Although Alaskan geogr aphy and climate increase aviation risks, many accidents were attributed to pilot error. While most accidents occurred during takeoff/landing, most fa talities resulted from Controlled Flight Into Terrain (CFIT). The purpose o f this study was to examine risk factors for:CFIT. Methods: Using National Transportation Safety Board airplane accident data we identified CFIT from fight phase and event description fields, and calculated odds ratios for CF IT/non-CFIT accidents for visual conditions, aircraft features, and pilot e xperience. Results: Between 1991 and 1998, 351 single aircraft commuter and air taxi accidents occurred in Alaska; 59 (17%) were CFIT. Of 140 total fa talities, 82 (59%) occurred in 30 CFIT accidents. There was a twelve-fold r isk for death in CFIT vs, non-CFIT accidents (OR = 12.42, 95% CI = 8.19-18. 88). Accidents while flying Visual Flight Rules (VFR) into poor visibility were more likely CFIT than non-CFIT (Odds ratio = 46.06, Confidence Interva l = 19.32-112.46), and caused 37% of all deaths. Additionally, flights in I nstrument Meteorological Conditions (IMC) were 47 times more likely to be C FIT than non-CFIT. No risk for CFIT was shown for fight hours, number of en gines, passenger presence, or pilot age. All CFIT were attributed to pilot error, often for continuing VFR into poor visibility. Conclusion: CFIT caus ed most aviation deaths. Further research into human factors contributing t o CFIT is needed. Implementation of global positioning, ground-proximity/av oidance technology, might reduce CFIT incidence.