Human factors in aviation crashes involving older pilots

BACKGROUND: Pilot errors are recognized as a contributing factor in as many as 80% of aviation crashes. Experimental studies using flight simulators indicate that due to decreased working memory capacity, older pilots are outperformed by their younger counterparts in communication tasks and flight summary scores. OBJECTIVE: This study examines age-related differences in crash circumstances and pilot errors in a sample of pilots who flew commuter aircraft or air taxis and who were involved in airplane or helicopter crashes. METHODS: A historical cohort of 3306 pilots who in 1987 flew commuter aircraft or air taxis and were 45-54 yr of age was constructed using the Federal Aviation Administration's airmen information system. Crash records of the study subjects for the years 1983-1997 were obtained from the National Transportation Safety Board (NTSB) by matching name and date of birth. NTSB's investigation reports were reviewed to identify pilot errors and other contributing factors. Comparisons of crash circumstances and human factors were made between pilots aged 40-49 yr and pilots aged 50-63 yr. RESULTS: A total of 165 crash records were studied, with 52% of these crashes involving pilots aged 50-63 yr. Crash circumstances, such as time and location of crash, type and phase of flight, and weather conditions, were similar between the two age groups. Pilot error was a contributing factor in 73% of the crashes involving younger pilots and in 69% of the crashes involving older pilots (p = 0.50). Age-related differences in the pattern of pilot errors were statistically insignificant. Overall, 23% of pilot errors were attributable to inattentiveness, 20% to flawed decisions, 18% to mishandled aircraft kinetics, and 18% to mishandled wind/runway conditions. CONCLUSIONS: Neither crash circumstances nor the prevalence and patterns of pilot errors appear to change significantly as age increases from the 40s to the 50s and early 60s.     

Factors associated with pilot error in aviation crashes

BACKGROUND: The importance of pilot error in aviation crashes has long been recognized. However, understanding and preventing pilot error remains the foremost challenge in aviation safety. OBJECTIVE: This study aims to identify pilot characteristics and crash circumstances that are associated with the presence of pilot error in a large sample of aviation crashes. METHODS: Different data files compiled by the National Transportation Safety Board for 329 major airline crashes, 1,627 commuter/air taxi crashes, and 27,935 general aviation crashes for the years 1983-96 were merged; and the presence of pilot error was analyzed in relation to variables indicating the characteristics of the pilot-in-command, crash circumstance, and aircraft. Multivariate logistic regression modeling was performed to assess the associations of individual variables with the likelihood of pilot error given a crash. RESULTS: Pilot error was a probable cause in 38% of the major airline crashes, 74% of the commuter/air taxi crashes, and 85% of the general aviation crashes. Among the factors examined, instrument meteorological condition and on-airport location were each associated with a significantly increased odds of pilot error. The likelihood of pilot error decreased as pilot certificate rating increased in commuter/air taxi and general aviation crashes. Neither pilot age nor gender was independently associated with the odds of pilot error. With adjustment for pilot characteristics and crash circumstances, flight experience as measured in total flight time showed a significant protective effect on pilot error in general aviation crashes. CONCLUSIONS: The prevalence and correlates of pilot error in aviation crashes vary with the type of flight operations. Adverse weather is consistently associated with a significantly elevated likelihood of pilot error, possibly due to increased performance demand.     

Correlates of pilot fatality in general aviation crashes

BACKGROUND: General aviation accounts for the majority of aviation crashes and casualties in the United States, and general aviation safety has not improved in the past decade. OBJECTIVE: This study identifies factors associated with pilot fatality in general aviation crashes. METHODS: We analyzed the National Transportation Safety Board's Factual Reports for all airplane and helicopter crashes of general aviation flights that occurred in North Carolina and Maryland during 1985 through 1994. Surviving pilots were compared with fatally injured pilots in relation to crash circumstances, and pilot and aircraft characteristics, at bivariate level and multivariate level. RESULTS: A total of 667 crashes resulted in 276 deaths and 368 injuries during the 10-yr period in the two states. Of the pilots-in-command involved in these crashes, 146 (22%) died. The case fatality rate for pilots was significantly higher in crashes that occurred between 6 p.m. and 5 a.m. (34%), away from airports (36%), with aircraft fire (69%), or in instrument meteorological weather conditions (IMC) (71%). Multivariate logistic regression revealed that the significant correlates of pilot fatality were aircraft fire [odds ratio (OR) 13.7, 95% confidence interval (CI) 6.9-27.2], off-airport location (OR 9.9, 95% CI 5.0-19.6), IMC (OR 9.1, 95% CI 4.3-19.6), nighttime (OR 2.2, 95% CI 1.3-3.7), and pilot age > or = 50 yr (OR 1.7, 95% CI 1.0-3.0). Pilot gender, flight experience, principal profession, and type of aircraft (airplane vs. helicopter) were not significantly associated with the likelihood of survival. CONCLUSIONS: The most important correlates of pilot fatality are variables likely related to increased impact forces. Better occupant protection equipment, such as air bag and crashworthy fuel system, are needed for general aviation aircraft.     

Suicide by aircraft: a comparative analysis

INTRODUCTION: Suicide is the 11 th leading cause of death in the United States. The objective of this study is to document the characteristics of aviation-related suicides and suicide attempts. METHODS: Aviation accidents reported by the National Transportation Safety Board between 1983 and 2003 were screened for cases in which suicide was listed as a probable cause. For each suicide case, two accidents were randomly selected as controls, matched on sex of pilot, type of flight, state, and year of occurrence. Mantel-Haenszel summary Chi-square tests were used to compare cases to controls. Conditional logistic regression modeling was performed to assess the association of pilot and flight characteristics with suicide-related crashes. RESULTS: During the 21-yr study period, 37 pilots committed or attempted suicide by aircraft, with 36 resulting in at least one fatality. All the cases were men and involved general aviation flights. Toxicological test results revealed that 24% of the cases had used alcohol and 14% had used other illicit drugs. Underlying factors included domestic and social problems (46%), legal trouble (40%), and pre-existing psychiatric conditions (38%). Compared with controls, suicide cases involved younger pilots (p < 0.05), were less likely to have another occupant (p < 0.0001), were more destructive to the aircraft (p < 0.0001), and were more likely to have occurred away from the airport (p < 0.0001). CONCLUSION: Aviation crashes caused by suicide differ from unintentional aviation accidents in pilot characteristics, crash circumstances, and outcomes.     

General aviation crashes involving military personnel as pilots.

BACKGROUND: Injuries sustained in off-duty activities are a major cause of mortality and morbidity among military personnel. Reducing these off-duty fatalities is a continuing priority of the military. METHODS: General aviation crashes recorded by the National Transportation Safety Board between 1983 and 1998 were analyzed for military pilots (n = 205) and other military personnel (n = 185), and compared with all other general aviation crashes (n = 32,807) to identify differences in the crash circumstances and sustained injury severities. RESULTS: During the 16-yr study period, a total of 45 military pilots and 52 other military personnel were fatally injured while flying general aviation flights. Military pilots who were involved in general aviation crashes were more likely to have advanced licenses and higher total flight times when compared with other military personnel and civilian pilots (p < 0.05). Among the three groups of pilots, other military personnel had the least flying time and the largest percentage of student/private licenses. Military personnel had significantly less time in type in the 90-d and 30-d periods preceding the crash compared with civilians (p < 0.05). Shoulder restraint usage was associated with less severe injuries for all groups. We estimate that general aviation deaths have cost the military at least $405 million since 1983. CONCLUSIONS: General aviation crashes are a costly source of mortality and morbidity for military personnel, particularly military pilots. Interventions aimed at improving safety of military personnel in the general aviation setting warrant special consideration.     

Gyroplane accidents 1985-2005: epidemiological analysis and pilot factors in 223 events

INTRODUCTION: Gyroplanes (autogyros) are regarded as a relatively safe and stable type of general-aviation aircraft. The U.S. Federal Aviation Administration categorizes them as sport pilot/light sport aircraft, and reports of gyroplane accidents are included in a publicly available database. We hypothesized that issues related to pilot experience and aircraft maintenance would affect the severity of accidents as indicated by aircraft damage and fatalities. METHODS: A search of the National Transportation Safety Board database for the period 1985-2005 yielded 223 reports of gyroplane accidents. Information from those reports was compiled and cross-referenced with pilot performance breakdowns and contextual information. The data was then analyzed using the Human Factors Analysis and Classification System. RESULTS: There was a strong effect of pilot experience on crash outcomes; compared to more experienced pilots, crashes involving pilots with less than 40 flight hours in the same make/model gyroplane were five times more likely to involve loss of control, twice as likely to destroy the aircraft, and four times more likely to involve fatalities. On the other hand, crashes involving pilots with more than 40 make/model hours were more likely to be related to perception-based performance breakdown. Maintenance issues were not found to play a significant role in this sample of crashes. CONCLUSION: The results support the hypothesis that pilot experience is a significant predictor of accident fatality in gyroplanes. Training that is adapted to the experience level of pilots as implemented in new FAA regulations for sport pilot and light sport aircraft (2004) may help to reduce the frequency and seriousness of gyroplane accidents.     

Case-control study of risk factors for fatal and non-fatal injury in crashes of civil aircraft

BACKGROUND: There have been few studies of the risk factors for fatal injury in air crashes, and none of risk factors for all serious injury (fatal and non-fatal). OBJECTIVE: To identify the potentially modifiable risk factors for fatal and non-fatal injury in civil aircraft crashes in New Zealand. METHODS: We analyzed records from all reported civil aircraft crashes in New Zealand (1988-1994). Air crash data from the official databases were merged with nationwide injury records and information obtained from coroners' files. Pilots-in-command who were fatally injured were compared with pilots-in-command who were not fatally injured using 50 variables covering pilot, aircraft, environmental, and operational characteristics. A second analysis compared pilots-in-command who were seriously injured (either fatally or non-fatally) with those who were involved in a crash but not hospitalized with an injury. Univariate and multivariate logistic regression analyses were then conducted to estimate the odds associated with each of the factors identified by the above comparisons. RESULTS: The most significant risk factors for fatal and serious injury were aerobatic flight, post-crash fire, not having a certificate of airworthiness, and off-airport location. Environmental characteristics including terrain type, wind, rain, and elevation of the crash site were identified as risk factors, as was being under instruction. Flying a twin-engine aircraft was a risk factor for fatal injury, while piloting a microlight aircraft was a risk factor for all serious injury. CONCLUSION: Environmental and operational factors, rather than pilot or aircraft characteristics, are the key determinants of the injury outcome of civil aircraft crashes.     

Cross-country VFR crashes: pilot and contextual factors

BACKGROUND: General Aviation (GA) cross-country crashes, particularly those involving weather, continue to be a major source of fatalities, with a fatality rate more than four times greater than for GA crashes in general. There has been much speculation and little solid evidence on the causes of these crashes. METHODS: We have designed a program of laboratory and database research into the causes of cross-country weather-related crashes including an analysis of air crashes in New Zealand between 1988 and 2000. There were 1308 reported occurrences in this period. We examined in detail 77 crashes where it could be determined that the aircraft was on a cross-country flight. RESULTS: In our first analysis we compared the characteristics of crashes that occurred in response to externally driven failures with crashes where the aircraft continued to be flown at the pilot's discretion up until the point of the crash. Clear differences were found for visibility, altitude, crash severity, and for several pilot characteristics. These differences are highly consistent with those found for previous research on pilot characteristics and crash involvement. In the second analysis we made comparisons between the weather-related and nonweather-related crashes in the discretionary control group and between subcategories of weather-related crashes. CONCLUSIONS: These data show that weather-related crashes occur further into the flight and closer to the planned destination than other kinds of cross-country crashes in GA. Pilots involved in these crashes are younger and have more recent flight time than pilots involved in other crashes. Their increased involvement cannot be explained simply by exposure (flight-time) but must be due to other factors.     

Ultralight aviation accident factors and latent failures: a 66-case study

INTRODUCTION: Little research has been done on ultralight aviation accidents. A better understanding of accident factors allows for better development of preventive measures. This study analyzes ultralight accidents with respect to significant factors related to active and latent failures, as referred to in Reason's Swiss cheese model of human error. METHODS: There were 66 National Transportation Safety Board (NTSB) ultralight accident reports that were examined from 1985-2004. Contextual information about active and latent failures was identified, compiled, and cross-referenced with pilot, aircraft, and accident information for evidence of significant relationships. RESULTS: Pilots with less than 40 make/model-specific flying hours were significantly more likely to crash fatally (chi2 = 9.07; p < 0.005; df=1) than other pilots and/or because of losing control (chi2 = 7.24; p < 0.05; df=1) than other accident causes. In contrast, pilots with 40 or more make/model-specific flying hours were significantly more likely to crash as a result of engine failure (chi2 = 9.33; p < 0.005; df = 1). Loss of control as an active failure was usually associated with such latent failures as strong winds and insufficient mission planning, while the engine failure accidents involved inadequate maintenance. CONCLUSION: In order to reduce accident prevalence, ultralight self-regulation organizations should focus on training in mission planning, aircraft familiarity, and proper maintenance procedures. Further research should concentrate on explaining the prevalence of the active and latent failures shown here and determining the effects of the FAA's new light sport aircraft category.     

Geographic variations in crash risk of general aviation and air taxis

INTRODUCTION: In the past decade, numerous studies have been published regarding the relationship of crash risk and mortality with respect to geographic area. METHODS: We analyzed data (United States only) regarding general aviation and air taxi crashes from 1992-94 from the National Transportation Safety Board. Data regarding 1993 flight hours and standard error were obtained from the Federal Aviation Administration (FAA) and pooled to provide a reliable estimate of crash risk for 1992-94. Calculations were determined for each state and FAA region. RESULTS: During the 3-yr period, the calculated United States crash rate is 8.9 crashes per 100,000 flight hours. The Alaskan and Northwest Mountain regions had the highest crash rates and fatal crash rates. DISCUSSION: This is the first study to report on geographical differences in rates determined as crashes per 100,000 pilot hours. It shows that even when the amount of flying is controlled for, crash rates and fatal crash rates are highest in mountainous regions. Our results indicate that aviation safety in mountainous regions deserves more attention.     

Alaska air carrier operator and pilot safety practices and attitudes: a statewide survey

INTRODUCTION: Aviation crashes are a leading cause of occupational fatalities in Alaska, with Alaskan pilots having nearly 100 times the fatality rate of U.S. workers overall. A survey was designed to study pilot and company practices and attitudes in order to develop intervention strategies that would reduce aviation fatalities. METHODS: Two surveys were administered: one of air carrier operators and one of active commercial pilots. Surveys from 153 air taxi and public-use operators were received at a 79% response rate. RESULTS: There are almost 2000 pilots employed in Alaska during peak season by air taxi operators and public agencies. Surveyed operators and pilots generally agreed that improved weather information and regional hazards training would be effective ways to prevent crashes. Operators were more in favor of operator financial incentives (p < 0.05) and better pre-employment hiring checks on pilots (p < 0.05) compared with pilots' survey responses. There were 48% of pilots of large operators and 73% of pilots of small operators who considered their jobs to be at least as safe as other jobs. CONCLUSIONS: The results of operator-pilot comparisons suggest that financial pressures on operators may influence their views on what measures would be effective in preventing crashes, and that Alaskan pilots underestimate their occupational fatality risk.     

Flight safety in Alaska: comparing attitudes and practices of high- and low-risk air carriers

INTRODUCTION: Aircraft operations are a vital component of the transportation system in Alaska. Between 1990-2002, a total of 481 people died in Alaska in aviation accidents. The purpose of this study was to examine the practices and attitudes of Alaska commuter and air taxi operators and their pilots as they relate to company fatal accident rates. METHODS: A case-control analysis based on accident statistics was performed, grouping operators and their pilots into cases and controls, based on operator fatal accident rates, during January 1990 to June 2001. Responses from two aviation safety surveys-one of air carrier operators and one of active commercial pilots-were compared between cases and controls. RESULTS: The average case pilot had less career flight experience than control pilots and worked 13 h x d(-1) and 81 h x wk(-10; that is, 1 h x d(-1) and 10 h wk-1 more than controls. Case operators were less likely to consider pilot fatigue a problem when scheduling flights (p = 0.05) and more likely to depend financially on timely delivery of bypass mail (p = 0.04). Case pilots were three times as likely as controls to fly daily into unknown weather conditions. Nearly 90% of case pilots reported that they never flew when so fatigued that they wanted to decline the flight, compared with 64% of control pilots (p = 0.01). CONCLUSIONS: Pilots of high-risk operators differed from those working for the other operators, both in experience and working conditions. The combination of pilot inexperience and longer work hours and workweeks may contribute to Alaska's high aviation crash rate.     

Pilot error in air carrier accidents: does age matter?

INTRODUCTION: The relationship between pilot age and safety performance has been the subject of research and controversy since the "Age 60 Rule" became effective in 1960. This study aimed to examine age-related differences in the prevalence and patterns of pilot error in air carrier accidents. METHODS: Investigation reports from the National Transportation Safety Board for accidents involving Part 121 operations in the United States between 1983 and 2002 were reviewed to identify pilot error and other contributing factors. Accident circumstances and the presence and type of pilot error were analyzed in relation to pilot age using Chi-square tests. RESULTS: Of the 558 air carrier accidents studied, 25% resulted from turbulence, 21% from mechanical failure, 16% from taxiing events, 13% from loss of control at landing or takeoff, and 25% from other causes. Accidents involving older pilots were more likely to be caused by turbulence, whereas accidents involving younger pilots were more likely to be taxiing events. Pilot error was a contributing factor in 34%, 38%, 35%, and 34% of the accidents involving pilots ages 25-34 yr, 35-44 yr, 45-54 yr, and 55-59 yr, respectively (p = 0.87). The patterns of pilot error were similar across age groups. Overall, 26% of the pilot errors identified were inattentiveness, 22% flawed decisions, 22% mishandled aircraft kinetics, and 11% poor crew interactions. CONCLUSION: The prevalence and patterns of pilot error in air carrier accidents do not seem to change with pilot age. The lack of association between pilot age and error may be due to the "safe worker effect" resulting from the rigorous selection processes and certification standards for professional pilots.     

General aviation accidents in degraded visibility: a case control study of 72 accidents

BACKGROUND: Accidents in degraded visibility continue to account for a disproportionately large number of fatal crashes and fatalities in general aviation (GA). However, the relatively small percentage of these crashes annually suggests that even a small reduction in the number of crashes can result in a large reduction in associated fatalities. OBJECTIVE: This study identified risk factors associated with GA accidents that occur in degraded visibility. METHODS: Data collected during 72 National Transportation Safety Board investigations of GA accidents were compared with data collected from 135 control flights matched on weather conditions, location, time, and rules of flight. Study variables included pilot demographics, experience, testing and accident histories, and the purpose and length of flights. Initial comparisons were conducted using Chi-square analyses, followed by the development of a logistic regression model. RESULTS: Univariate analyses identified significant differences in pilot age at accident, age at initial certification, certificate level, instrument rating, testing performance, and previous accident involvement. There were also significant differences between groups related to aircraft ownership, purpose of flight, and intended flight length. The logistic regression revealed significant increased risk for pilots initially certified after age 25 [odds ratio (OR) 4.5, 95% confidence interval (CI) 1.9-10.81, without instrument ratings (OR 4.8, 95% CI 1.8-12.8), with prior accidents/incidents (OR 3.1, 95% CI 1.1-8.7), and for intended flights longer than 300 nmi (OR 4.6, 95% CI 1.6-13.8). CONCLUSIONS: Certain risk factors (e.g., instrument rating) were consistent with previous research, while others (e.g., age at certification) raise questions for future research.     

Routine electroencephalograms of pilots later killed in crashes: a case-control study

BACKGROUND: The value of the electroencephalogram (EEG) as a screening tool in aviation medicine is subject to debate. We evaluated the use of periodic repetition of the EEG--after an initial EEG screening-in healthy, fully-licensed pilots to identify risk factors for fatal air crashes. METHODS: In a nested case-control study, we compared the EEG patterns of 33 pilots who died in military aviation accidents from 1990 to 2001 with the recordings of 66 controls. Cases and controls were matched for potentially confounding factors such as age, military membership, type of aircraft, and aeronautical experience. RESULTS: Both groups presented normal EEGs. EEG findings in cases and controls did not differ; nonspecific EEG abnormalities did not occur more frequently in those pilots who crashed. CONCLUSIONS: After an initial EEG screening, periodic repetition of the EEG is not a useful means to detect risk for fatal air crashes.     

The incidence, nature, and severity of injuries in New Zealand civil aviation

BACKGROUND: Strategies to improve aviation safety can be directed at the pre-crash, in-crash, or post-crash phases of aircraft crashes. For resources to be made available for in-crash interventions, and for these to be well designed, it is necessary in the first instance to have a detailed understanding of the injuries sustained in crashes. The purpose of this study was to describe the incidence, nature, and severity of injuries sustained in aircraft crashes and other related events in civil aviation in New Zealand. METHODS: National injury databases were searched for fatalities and hospitalizations sustained in aircraft crashes and related events, and cases were linked with Civil Aviation Authority accident records to identify the aircraft involved. Rates were based on estimates of total hours flown by active pilots. RESULTS: There were 104 fatalities identified for the period 1988-1992, giving a rate of 2.57 per 100,000 flight hours. There were 120 hospitalizations identified for the period 1988-1993, giving a rate of 2.45 per 100,000 flight hours. Most fatalities involved injury to multiple body regions, with at least one injury being sufficient in itself to cause death in 48% of cases. For hospitalizations, the lower extremities (23%), spine (20%), and head and face (18%) were the body regions most commonly injured, with fractures being predominant. While the majority of fatalities and hospitalizations occurred in fixed- and rotary-wing aircraft, the highest rates were for microlight and home-built aircraft. CONCLUSIONS: Different patterns of injury were evident for fixed- and rotary-wing aircraft. Reasons for these are suggested. Future research will seek to determine the relative risk associated with potentially modifiable risk factors.     

Antiemetic and sedative levels found together in 26 civil aviation pilot fatalities, 2000-2006

INTRODUCTION: Antiemetics and sedatives are two drug classes that may have harmful side effects when mixed. This is especially dangerous for pilots. Although many of these compounds are considered disqualifying by the FAA, their use occurs. Some pilots may be unaware of the danger of combining compounds from these two drug classes. Our laboratory was interested in evaluating the circumstances surrounding accidents in which the pilot was found positive for drugs from each of these two classes. METHODS: Epidemiological, toxicological, and aeromedical findings from pilots involved in such accidents were collected for the 7-yr period from 2000 through 2006. Case histories, accident information, and the probable cause of the aviation accidents were obtained from the National Transportation Safety Board (NTSB). Toxicological information was obtained from the Civil Aerospace Medical Institute's (CAMI's) Forensic Toxicology Research Laboratory. RESULTS: Of the 2184 fatal aviation accidents over this time period, 26 were found positive for compounds from both the antiemetic and the sedative drug classes. All 26 aircraft were operated under 14 CFR Part 91 as general aviation. All pilots were male; 21 tested positive for a disqualifying substance that may have affected their ability to control the aircraft. CONCLUSION: Although the percentage of accidents in which the pilot tests positive for a compound from each class is small, it is important for all pilots to understand the dangers of self-medicating and concomitant use of such substances. Under-reporting of medications by pilots during their certification process occurs; education is the key to preventing inadvertent drug-drug interactions.     

Glider accidents: an analysis of 143 cases, 2001-2005

INTRODUCTION: The majority of aviation crashes and casualties take place in general and sport aviation. Although gliding has gained popularity in recent decades, we could find no systematic analysis of glider accidents. This study determined factors associated with both non-fatal and fatal glider accidents to document their position within sport and general aviation accidents, and to suggest preventive measures and improvements. METHODS: We performed a retrospective review of glider accidents for the period 2001-2005 in the database maintained by the U.S. National Transportation Safety Board (NTSB). RESULTS: A total of 117 non-fatal and 26 fatal glider accidents were reported for the 5-yr period. Adverse weather was the cause in 20% of all non-fatal accidents, 60% of which occurred in the cruise phase. Logistic regression revealed that fatal accidents were predicted by pilot error, flight phase, and home-built aircraft. DISCUSSION: Factors contributing to glider crashes are specific to this type of sport aviation. Owners of home-built gliders should pay particular attention to the aircraft's specifications and design limits.