In-flight medical incapacitation and impairment of airline pilots

Abstract

Medical incapacitation in the cockpit is rare, although it is a concern that has been the subject of several investigations over the years. With recent heightened interest in this problem, it seemed worthwhile to review all relevant scientific literature on the topic. Medline, PsychLit, the Aerospace Database, and other online databases were searched for studies of pilot in-flight medical incapacitation and impairment. The search revealed 13 articles during the years from 1968 to 2000. The studies represented several different approaches and were divided into five categories as follows: in-flight medical events, career termination, simulator data, questionnaires, and epidemiological analysis. The articles based on in-flight medical events showed that the leading causes of those episodes were myocardial infarctions, cardiac arrhythmias, and epileptic seizures. Few of the other types of studies used data from actual in-flight medical occurrences, instead relying on indirect measures such as career termination due to permanent medical grounding, loss of licensure insurance, or general epidemiological data to estimate the frequency of in-flight medical events. The reviewed studies provided only limited information on the frequency and categories of in-flight medical events and did not include incapacitation rates, making meaningful comparison between studies difficult. Future research needs to be based on actual in-flight medical events, and should be normalized to a useful denominator, such as flight time, to allow for meaningful comparison between studies.

Full text

In-Flight Medical Incapacitation
and Impairment of U.S. Airline
Pilots: 1993 to 1998
Charles A. DeJohn 1
Alex M. Wolbrink 2
Julie G. Larcher 1
1Civil Aerospace Medical Institute
Oklahoma City, OK 73125
2American Airlines
Dallas Fort Worth Airport, TX 75261
Civil Aerospace Medical Institute
Federal Aviation Administration
Oklahoma City, OK 73125
October 2004
Final Report
This document is available to the public
through the National Technical Information
Service, Springeld, Virginia 22161.
Ofce of Aerospace Medicine
Washington, DC 20591
DOT/FAA/AM-04/16

NOTICE
This document is disseminated under the sponsorship of
the U.S. Department of Transportation in the interest of
information exchange. The United States Government
assumes no liability for the contents thereof.

i
Technical Report Documentation Page
1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.
DOT/FAA/AM-04/16
4. Title and Subtitle 5. Report Date
October 2004 In-Flight Medical Incapacitation and Impairment of U.S. Airline Pilots:
1993 to 1998
6. Performing Organization Code
7. Author(s) 8. Performing Organization Report No.
DeJohn CA1, Wolbrink AM2, Larcher JG1
9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)
1FAA Civil Aerospace Medical Institute
P.O. Box 25082
Oklahoma City, OK 73125
2American Airlines
Dallas Fort Worth
Airport, TX 7526
11. Contract or Grant No.
12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered
Office of Aerospace Medicine
Federal Aviation Administration 14. Sponsoring Agency Code
800 Independence Ave., S.W.
Washington, DC 20591
15. Supplemental Notes
16. Abstract
Although it is not known when the first accident due to pilot in-flight medical incapacitation occurred, a
recent survey showed that almost one-third of all pilots who responded had experienced an incapacitation
requiring another crewmember to take over their duties, with safety of flight significantly threatened in 3% of
cases. The importance of in-flight medical incapacitation and impairment can be better understood when it is
realized that each in-flight medical incapacitation or impairment could potentially lead to an aircraft accident.
We studied in-flight medical incapacitations and impairments in U.S. airline pilots from 1993 through 1998.
We defined in-flight medical incapacitation as a condition in which a flight crewmember was unable to
perform any flight duties and impairment as a condition in which a crewmember could perform limited flight
duties, even though performance may have been degraded. We found 39 incapacitations and 11 impairments
aboard 47 aircraft during the six-year period. All pilots were males. The average age for incapacitations was
47.0 years (range 25 to 59 years). The average age for impairments was 43.3 years (range 27 to 57 years). The
in-flight medical event rate was 0.058 per 100,000 flight hours. The probability that an in-flight medical event
would result in an aircraft accident was 0.04. Incapacitations significantly increased with age, with more
serious categories in the older age groups. The most frequent categories of incapacitation were loss of
consciousness, cardiac, neurological, and gastrointestinal. Safety of flight was seriously impacted in seven of the
47 flights and resulted in two non-fatal accidents.
17. Key Words 18. Distribution Statement
Incapacitation, Impairment, Aircraft Accidents,
Diversions
Document is available to the public through the
National Technical Information Service,
Springfield, Virginia 22161
19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of Pages 22. Price
Unclassified Unclassified 30
Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

iii
ACKNOWLEDGMENTS
We gratefully acknowledge Connie Peterman of Advancia™ for her techni-
cal support as the programmer for the Civil Aerospace Medical Institute Inca-
pacitation Database and the hours spent in preparing and executing the many
queries necessary to extract the data used in preparing this paper.

v
TABLE OF CONTENTS
Page
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
In-Flight Medical Event Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Career Termination Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Simulator Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Questionnaire Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Epidemiological Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Summary of Incapacitation Study Methodologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Frequency and Rate of In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Probability of an Accident Due to an In-Flight Medical Event. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Age and Gender Distribution of In-Flight Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Categories of In-Flight Medical Incapacitations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Categories of In-Flight Medical Impairments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Fatal In-Flight Medical Events. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Safety of Flight and In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
In-Flight Medical Events and Similar Medical Histories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Diversions Resulting From In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Aeromedical Certication Actions Resulting From In-Flight Medical Events. . . . . . . . . . . . . . . . . . . . . . . 8
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Frequency and Rate of In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Probability of an Accident Due to an In-Flight Medical Event. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Age Distribution of In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Categories of In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Fatal In-Flight Medical Events. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Safety of Flight and In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
In-Flight Medical Events and Similar Medical Histories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Diversions Resulting From In-Flight Medical Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Aeromedical Certication Actions Resulting From In-Flight Medical Events. . . . . . . . . . . . . . . . . . . . . . 12
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
APPENDIX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1
Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A-1

1
IN-FLIGHT MEDICAL INCAPACITATION AND IMPAIRMENT
OF U.S. AIRLINE PILOTS: 1993 TO 1998
INTRODUCTION
The rst fatal aircraft accident occurred in 1909, and
by the end of 1910 there were 38 aviation fatalities. For
some time, it was believed the rst accidents attributed
to pilot in-ight medical incapacitation occurred in
1911; however, after reviewing these cases, Parmet and
Underwood-Ground (28) believed that they were the
result of pilot error. Therefore, the date of the rst aircraft
accident attributed to pilot in-ight medical incapacita-
tion is still unknown.
Over the years there have been many studies of airline
pilot medical incapacitation (4, 6, 8, 19, 23, 24, 29, 30).
Most studies can be classied as either direct studies of in-
ight medical events, career termination studies, simulator
studies, questionnaire studies, or epidemiological studies.
In addition, Li (22) recently performed a comprehensive
review of pilot-related factors in aircraft accidents.
In-Flight Medical Event Studies
In 1969, Buley (4) summarized three sets of airline pilot
incapacitation data. First, he reported on the progress of
a collaborative study initiated by the International Civil
Aviation Organization (ICAO) and performed by the
International Federation of Air Line Pilots Associations
(IFALPA) and the International Air Transport Association
(IATA). Buley examined in-ight deaths of airline pilots
between 1961 and 1968. He found 17 reported cases
of airline pilot deaths, all resulting from heart disease.
Five of the 17 cases ended in aircraft accidents, four of
which were fatal, resulting in 148 fatalities. Another ve
cases nearly resulted in accidents. Of the 17 events, seven
occurred on the ramp, ve while enroute, four during
approach, and one during landing roll-out. Buley next
reviewed 42 cases of non-fatal in-ight incapacitation
in pilots of IATA-member airlines between 1960 and
1966. In 24 of the 42 cases, causal organic disease was
diagnosed. The most common categories of incapacitation
were epileptiform manifestations (6), coronary occlusions
(4), and renal/ureteric colic (4).
In 1975, Raboutet and Raboutet (30) reviewed 17
incidents of sudden incapacitation in French professional
civil pilots between 1948 and 1972. They found ve cases
of myocardial infarction, three cases of angina pectoris,
two cases of ischemic heart disease, two cases of epilep-
tic seizures, and one case each of diabetes, pulmonary
embolism, cerebral vascular accident, atrial brillation,
and intestinal hemorrhage of unknown etiology. For-
tunately, none of the cases resulted in aircraft accidents
or the death of the pilot, and only two cases resulted in
the complete incapacitation of the pilot due to epileptic
seizures. Raboutet and Raboutet stated that for an in-
capacitation accident to occur, the incapacity must: (1)
affect the pilot at the controls, (2) be sudden, (3) be total,
and (4) take place during a critical phase of ight.
Chapman reviewed IATA data and found 208 in-ight
medical incapacitations between 1965 and 1977, which
included 13 cardiac cases, or one cardiac incapacitation
per year. He calculated the probability of an accident
due to cardiac incapacitation to be about 10-10, assum-
ing: (1) one accident per 400 incapacitations (i.e., one
accident every 400 years), (2) 600 ying hours per pilot
per year, and (3) subtle incapacitation during a critical
phase of ight (7).
Martin-Saint-Laurent and associates (23) found ten
cases of sudden in-ight incapacitation out of a popula-
tion of 1,800 Air France pilots and ight engineers from
1968 to 1988. The most common causes of in-ight
incapacitation were cardiac arrhythmias (2) and epileptic
seizures (2). Two out of the ten ights diverted. The two
pilots who suffered epileptic seizures and one pilot from
the arrhythmia group who had an in-ight episode of
atrial brillation (followed by a cerebral vascular accident
with hemiparetic and epileptic seizure on the ground fol-
lowing the ight) were all permanently grounded. Five
others were temporarily grounded. No action was taken
against one pilot who experienced hypoxia and another
who experienced CO2 intoxication.
Career Termination Studies
Preston (29) followed 1,000 United Kingdom airline
pilots and found that 73 were permanently grounded
for medical causes between 1954 and 1965. Of the 73
pilots, the most common causes for loss of employment
were psychiatric (36), cardiovascular (8), respiratory (6),
and diabetes (4). Preston attributed the low incidence of
cardiovascular groundings to possible Anglo-Saxon racial
differences between this group of pilots and other pilots,
and the high incidence of psychiatric groundings to poor
pilot selection procedures.

2
3
Lane (21) examined IATA loss of license insurance data
and estimated the overall in-ight incidence of in-ight
incapacitation to be 0.06 per 1,000 pilots per year.
Kulak, Wick, and Billings (20) performed a similar
study of career termination due to loss of licensure insur-
ance in members of the U.S. Airline Pilots Association
from 1955 through 1966. They found 891 cases of career
termination: 229 due to accidents and 662 the result of
disease. The rate of death and disability due to accidents
was 2.07 per 1000 pilots per year, while the rate for disease
was 8.05 per 1000 pilots per year. Although the overall
rate for cardiovascular disease was only 2.91 per 1000
pilots per year, the age specic rate ranged from zero for
pilots under 30 years of age to 27.33 for pilots between
55 and 58 years of age. Flying accidents accounted for the
majority of career terminations for all age groups. Using
the incapacitation incidence rates for termination due to
disease by age and the age distribution of active ALPA
pilots, the authors estimated the probability of serious in-
ight incapacitation by age. Their estimates ranged from
1 per 58,000 pilots for the 30-to 34 -year age group to 1
per 3,500 pilots for the 55-to 59- year age group.
Simulator Studies
Harper, Kidera and Cullen performed two simulator
studies, one dealing with obvious and maximal loss of
function (14), and the other with subtle or partial loss
of function (13). Although operationally interesting,
discoveries were made. For example, the mean time to
detect subtle incapacitation was 1.5 minutes, and 25%
of the simulator sessions ended in “aircraft accidents”
(13). The studies, however, were not designed to address
medical causes of in-ight incapacitation.
Chapman (7) analyzed more than 1,300 simulator
exercises in which the subtle incapacitation of the ying
pilot was programmed to occur at a critical phase of ight.
Two protocols were used. The rst involved 500 exercises
where major aircraft system failures were simulated as
part of the drill. In 485 of the 500 cases, it was deter-
mined there would not have been any danger to an actual
aircraft. In 15 cases, it was believed that safety of ight
would have been at risk. In eight cases, it was considered
that aircraft accidents would have resulted. The second
protocol involved 800 exercises without simultaneous
aircraft system failures. In this series, only ten out of the
800 were felt to have represented a risk to safety of ight,
and in two cases, the observers felt that an aircraft accident
would have resulted. Again, these studies did not address
medical causes of in-ight incapacitation.
There are inherent problems when simulator results
are used to predict in-ight outcomes. Crews that
“crashed” in the Harper, Kidera, and Cullen (13) study,
for example, stated, “We wouldn’t let it happen in a real
airplane.” Also, in addition to the obvious difculty of
attempting to predict possible in-ight outcomes from
simulated data in the Chapman study, there was the
added drawback of foreknowledge by the subjects, since
they knew there would be an incapacitation at some time
during each drill.
Questionnaire Studies
Buley (4) reviewed the results of a questionnaire ad-
ministered to pilots of IFALPA-member associations in
1967, in which 27% of approximately 5000 respondents
reported about 2,000 incidents of signicant in-ight
incapacitation. Safety of ight was affected in 4% of
cases. Almost one-half of reported incapacitations oc-
curred in the enroute phase of ight. Unfortunately,
the IFALPA questionnaire was administered to actively
ying airline pilots; therefore, pilots with more serious
medical conditions, who may have suffered more severe
types of incapacitation, had been previously eliminated
and were not part of the study. This skewed the data by
eliminating the potential for reporting more serious medi-
cal conditions while including the less serious conditions
reported by actively ying pilots. In addition, most of the
conditions addressed in the questionnaire were temporary
in nature and would usually result in pilot impairment
(i.e., nausea, vomiting, indigestion, etc.) rather than to-
tal incapacitation and would not likely be addressed by
medical certication.
In 1971, Lane (21) updated the 1967 IFALPA ques-
tionnaire data, analyzed by Buley (4), with IATA data for
1962 through 1968 that was provided to ICAO. Lane
added 51 additional, non-accident cases to Buley’s origi-
nal 17 cases, for a total of 68 cases. He then calculated
the probability of an incapacitating event resulting in an
accident would be 5/68, or 0.074.
In 1991, James and Green (19) replicated Lane’s
1967 IFALPA survey with similar results. Of 1,251
respondents, 29% reported at least one incident of in-
ight incapacitation severe enough to require another
crewmember to assume their duties. The most common
causes of incapacitation were gastrointestinal (58.4%),
earache due to a blocked ear (13.9%), and faintness or
general weakness (8.5%). The most common phases
of ight where incapacitations occurred were enroute
(42.1%), followed by climb (18.4%), descent (17.3%),
and on the ramp (11.4%). Safety of ight was felt to
be potentially affected in 45% of cases and denitely
affected in 3% of cases. Of those reporting that safety of
ight had been affected, 43% stated the incapacitation
event placed the remaining aircrew under maximum
workload. As with the 1967 Buley study, the question-
naire was administered to actively ying airline pilots,
again eliminating the potential for reporting more

2
3
serious medical conditions. In addition, the study did
not provide incapacitation rates which would allow for
comparison with similar studies.
Epidemiological Studies
Castello-Branco and associates (6) found 13 deaths and
eight medical incapacitations in a longitudinal evaluation
of deaths and disease in 408 active Portuguese airline pilots
between 1945 and 1983. The most common causes of
death were accidents, myocardial infarcts, and cancer. By
relating the number of deaths and incapacitations with
the number of pilots at risk, they calculated incidence
rates by age group. Death and incapacitation rates ranged
from zero per 100 pilots at risk for the 20 to 24 year age
group, to 3.64 per 100 pilots at risk for the 55 to 59 year
age group. Although this is an excellent longitudinal study
of airline pilots for an extended period and provides valu-
able insight into causes of death and disease, it does not
directly reect causes of in-ight incapacitation.
Summary of Incapacitation Study Methodologies
Most previous studies we reviewed did not use data
from actual in-ight medical events. Instead, indirect
measures, such as career termination due to permanent
medical grounding or loss of licensure insurance data and
general epidemiological data, were used to approximate
the frequency of in-ight medical events. These studies
provided information on the frequency and categories of
in-ight medical events; however, they did not include
incapacitation rates, making meaningful comparison
between studies difcult. Although in-ight medical
incapacitation rates can be inferred, these data are not
directly based on in-ight medical events.
The objective of this study was to provide incapacitation
rates that could be easily compared with similar studies
of in-ight medical incapacitation.
METHODS
Details of aircrew in-ight medical events aboard U.S.
airlines between 1993 and 1998 were collected by the
Federal Aviation Administration (FAA), Civil Aerospace
Medical Institute’s (CAMI’s), Aerospace Medical Research
Team (AMRT) and stored in a Microsoft Access 2000®
Database (Version 9.0). The ofcial method of case no-
tication was through the use of a Medical Case Alert
Form (Appendix A, Table A-1); however, many cases
were discovered through the FAA Administrator’s Daily
Bulletin, telephone calls, news media, periodic searches
of the National Transportation Safety Board (NTSB) and
FAA accident databases, and direct interaction between
the CAMI and the NTSB.
Event data included incident, operational, pilot, and
nal disposition information. Incident information in-
cluded a brief narrative and/or full report of the event
when available, including injuries and/or fatalities. Op-
erational information included the date, time, location,
type of operation, accident or incident classication, phase
of ight, airline, ight number, aircraft number, type of
aircraft, origin, destination, and diversion details. Pilot
information included the pilot’s name, social security
number, age, gender, class of medical certicate, pilot
certicate number, FAA medical history, aircrew posi-
tion, occupation, and employer. NTSB numbers were
recorded, and a unique CAMI Incapacitation Database
number was also assigned to each case. The authors
could not independently verify the validity of much of
the information collected from sources outside of CAMI,
including aircrew statements, airline records, and hospital
records, etc. In most cases, this information had to be
accepted without conrmation.
Cases were reviewed by the authors and classied as
either an “impairment” or “incapacitation.” Individuals
were classied as impaired when they could still perform
limited in-ight duties, such as read checklists or perform
radio communications, even though their performances
may have been degraded. Examples of impairments in-
clude food poisoning, the use of monovision contact
lenses, fatigue, and kidney stones. Individuals were con-
sidered incapacitated when they could no longer perform
any in-ight duties. Examples of incapacitation include
heart attacks and epileptic seizures.
Cases were also classied as “possible,” “probable,” or
“certain” depending on the degree of condence in the
supporting evidence. For airline events, the other aircrew
members were witnesses to the occurrence, and reports
were required by the airline. In addition, in cases serious
enough to require further evaluation and treatment, the
hospital record provided additional conrmation.
Each case was also assigned to one of several broad
medical categories. Incapacitation categories included loss
of consciousness, cardiac, neurological, gastrointestinal,
urological, vascular, medication, hypoxia, decompression
sickness, and injury. Impairment categories included
respiratory, cardiac, gastrointestinal, infectious disease,
vision, and reaction to medication.
RESULTS
Frequency and Rate of In-Flight Medical Events
We found 39 incapacitations and 11 impairments of
U.S. airline pilots on 47 ights during the period 1993 to
1998 (More than one pilot was affected on three ights.
See Table A-2, Case Summaries, in Appendix A.). During

4
5
this period, U.S. airlines ew a total of 85,732,000 revenue
passengers hours (26); therefore, the rate of in-ight in-
capacitations and impairments was 0.04549 per 100,000
hours (95% CI 0.04545, 0.04553) and 0.01283 per
100,000 hours (95% CI 0.01281, 0.01285), respectively.
A summary of the in-ight medical events is contained
in Appendix A, Table A-2.
Probability of an Accident Due to an In-Flight
Medical Event
There were two non-fatal aircraft accidents due to
the in-ight medical impairment of the pilots. One was
caused by the pilot’s visual impairment due to the use
of monovision contact lenses during an approach. The
other was caused by ight crew fatigue. Combining the
39 incapacitations and 11 impairments gives 50 in-ight
medical events on 47 ights; therefore, the probability
that an in-ight medical event would be associated with
an accident was 2 out of 50 events, or 0.04.
There were 54,295,899 ights and 217 accidents
involving U.S. Part 121 scheduled and non-scheduled
airlines between 1993 and 1998 (25, 27). The probability
of an aircraft accident for a pilot experiencing an in-ight
medical event is summarized in Table 1.
The “law of rare events” states that the total number of
events of interest will assume (approximately) the Poisson
distribution if: (a) the event may occur in any of a large
number of trials, but (b) the probability of occurrence in
any given trial is small. Examples of events that follow a
Poisson distribution are doctor visits, absenteeism in the
workplace, mortgage pre-payments, loan defaults, bank
failures, insurance claims, and aircraft accidents (5). As-
suming that accidents involving pilot in-ight medical
events can be described using a Poisson distribution,
there is a statistically signicant difference between the
proportion of accidents given an in-ight medical event,
compared with the proportion of accidents in the absence
of such an event (z = 3.08, p < 0.001).
Age and Gender Distribution of In-Flight Medical
Events
All pilots who had an in-ight event were males. The
average age for incapacitations was 47.0 years (range 25
to 59 years) while the average age for impairments was
43.3 years (range 27 to 57 years). Figure 1 shows the age
distribution for the percentage of U.S. airline pilots hav-
ing in-ight incapacitations and impairments between
1993 and 1998 based on the average age distribution for
professional pilots (from a recent FAA
study; Appendix A, Fig. A-1) (3).
Examination of Figure 1 suggests
an increase in in-ight medical inca-
pacitations with increasing U.S. airline
pilot age.
Table 1. The Probability of an Accident Given an In-Flight Medical Event.
Accident
Yes No Total
Yes 2 48 50
No 215 54,295,634 54,295,849
In-Flight
Event Total 217 54,295,682 54,295,899
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
25-29
30-34
35-39
40-44
45-49
50-54
55-59
Pilot Age Group (Years)
Percent of Incapacitated/
Impaired Pilots
Incapacitations
Impairments
Figure 1. Percent of U.S. airline pilot in-flight incapacitations and impairments as a
function of age (1993 to 1998).
Figure 1. Percent of U.S. airline pilot in-ight incapacitations and impairments as a function of
age (1993 to 1998).

4
5
Figure 2 shows the percent of pilot in-ight medical
incapacitations by age group. A linear regression of the
data shown in Figure 2 indicates that the percentage of
in-ight incapacitations increased with pilot age group
(R2 = 0.69, p < 0.01).
Categories of In-Flight Medical Incapacitations
All 39 in-ight medical incapacitations were classied
as “certain.” The most frequent categories were loss of
consciousness (LOC) (9), gastrointestinal (GI) (6), neu-
rological (6), cardiac (5), and urological (3). Of the nine
loss of consciousness cases, four were caused by vasovagal
syncope, one was the result of neurogenic syncope, one
was due to pain secondary to a duodenal bulb ulcer,
one was the result of decompression sickness, and two
were due to unknown causes. The six gastrointestinal
cases included two cases of cholelithiasis, two cases of
intestinal gas expansion with altitude, and two cases of
possible food poisoning. Four of the six neurological cases
were grand-mal seizures, one was an alcohol withdrawal
seizure, and one was a petit mal seizure. Three of the ve
cardiac cases were fatal myocardial infarctions, and one
was a fatal dysrhythmia, while one cardiac case involved
a non-fatal coronary spasm. All three urological cases
involved renal lithiasis.
Figure 4 shows the most frequent categories of in-
ight medical incapacitation by age. Examination of the
gure suggests an increase in incapacitations with pilot
age. Also, the data in Figure 4 suggest that more serious
categories, such as loss of consciousness secondary to
ulcers, cardiac events like myocardial infarctions, and
neurological seizures occurred more frequently in older
y = 0.0189x - 0.0071
R2 = .6926
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0.18
25-29
30-34
35-39
40-44
45-49
50-54
55-59
Age Groups (Years)
Percent of Pilots
Figure 2. Age distribution of in-flight medical incapacitations of U.S. airline
pilots from 1993 to 1998.
Figure 2. Age distribution of in-ight medical incapacitations of U.S. airline pilots from
1993 to 1998.
Figure 3. Frequent categories of in-flight medical incapacitation.












Figure 3. Frequent categories of in-ight medical incapacitation.

6
7
pilots. At the same time, the less serious medical categories,
such as gastrointestinal events due to gas expansion and
food poisoning and loss of consciousness due to vasovagal
syncope, occurred more frequently in younger pilots.
Less frequent causes of in-ight medical incapacita-
tion included hypoxia (2), diabetes (1), decompression
sickness (1), vascular (1), reaction to medication (1) and
traumatic injury (1). The two hypoxia cases occurred on
the same ight when the ight engineer inadvertently
opened the outow valve at altitude, accidentally depres-
surizing the aircraft. The diabetes case involved a second
ofcer who had had two hypoglycemic episodes within
a three-month period, one at the gate resulting in his
removal from the aircraft prior to ight and another in-
ight. The decompression case occurred when a DC-8
cargo captain ordered his crew to continue to climb to
33,000 feet after it was determined that the aircraft could
not be pressurized. Interestingly, the captain ordered the
aircrew to use oxygen, believing that it would provide
protection against decompression sickness at that altitude.
The vascular case involved an Airbus captain with a his-
tory of chronic, controlled atrial brillation who decided
to discontinue his digoxin and propranolol on his own
and developed a temporoparietal cerebral infarct during
landing. One pilot suffered heart palpitations that were
attributed to an herbal medication he was taking for
weight control; another suffered an injury when hydraulic
uid came into contact with his eye during an aircraft
pre-ight inspection.
Categories of In-Flight Medical Impairments
All of the 11 in-ight medical impairments were
classied as “certain.” Categories of in-ight medical
impairment included respiratory (4), fatigue (2), vision
(2), cardiac (1), gastrointestinal (1), and infectious dis-
ease (1). Three of the four respiratory cases occurred on
the same taxiing DC-8 aircraft when carbon dioxide
poisoning, caused by fumes from dry ice carried in the
cargo compartment, impaired the captain, rst ofcer
and jumpseater. The fourth respiratory case was due to
barosinusitis. The two fatigue cases occurred on a DC-8
ight, resulting in an accident with three serious injuries.
The captain had been awake 40 of the previous 66 hours
and the rst ofcer for 47 of the previous 66 hours prior
to the accident. Of the two vision cases, one occurred
when a B-737 captain looked directly into a laser that
appeared to be tracking the aircraft from the ground,
resulting in a temporary loss of night vision. The other
occurred when an MD-88 struck the approach lights
during an approach to a landing because of the captain’s
use of monovision contact lenses, resulting in three minor
injuries on evacuation of the aircraft. The cardiac case
involved a 57-year-old L-1011 captain who had been
experiencing retro-sternal chest pain since the previous
day, which became more continuous in ight and began
radiating to his left jaw and arm. Although the clini-
cal impression of the attending physician was unstable
angina, his electrocardiogram was normal on physical
examination. One Embraer-120 captain was diagnosed
with viral gastroenteritis and secondary dehydration, and
an MD-88 rst ofcer was found to have a viral infection
that led to a vasovagal response.
Fatal In-Flight Events
Four pilots died as a result of their in-ight incapaci-
tating event; however, no passenger deaths resulted from
these incapacitations. No pilot deaths resulted from in-
ight medical impairments. The mean age of the four
0
1
2
3
Frequency
25-29
30-34
35-39
40-44
45-49
50-54
55-59
Age Groups (Years)
LOC
Cardiac
Neurological
GI
Urological
Figure 4. Most frequent categories of in-flight medical incapacitation by age.
Figure 4. Most frequent categories of in-ight medical incapacitation
by age.

6
7
pilot fatalities was 53 (range 48 to 56 years). All four
deceased pilots were pronounced dead because of cardiac
events after being transported to the hospital. Three of the
four deaths resulted from myocardial infarctions (MIs),
while one was the result of a cardiac dysrhythmia. Two
of the three pilots who suffered MIs and the pilot who
died as a result of a fatal dysrhythmia had cardiac medi-
cal histories that were documented in the pilot’s FAA
medical record. As a result, pathology codes, history
codes, or both were assigned by the FAA prior to their
in-ight medical events. Two of the three ights where
the pilots suffered an MI and the one ight where the
pilot suffered an arrhythmia diverted to alternate airports
because of the in-ight medical events; however, the one
ight that did not divert when the rst ofcer suffered
an MI was inbound to their nal destination at the
time. Cardiopulmonary resuscitation was attempted in
all cases. Safety of ight was seriously affected only once
temporarily when the rst ofcer’s foot became lodged
against the rudder pedal when he stiffened, requiring the
captain to apply opposite rudder pressure until the foot
could be dislodged.
Safety of Flight and In-Flight Medical Events
Safety of ight is negatively affected during any in-
ight medical event; however, we considered safety of
ight to be a factor during an event only when there
was imminent danger of an aircraft accident resulting
from the medical event. We found that on seven of the
47 ights it was seriously affected. The mean age of the
seven pilots involved in ights where safety of ight was
seriously affected was 48.4 years (range 42-56, SD =
4.5), and the mean age of the 41 pilots who were not
involved in ights where safety of ight was seriously
affected was 45.7 years (range 25-59, SD = 10.7). There
was no signicant difference in the mean age between
the two groups. As previously discussed, two of the seven
ights ended in aircraft accidents. The seven cases are
summarized below.
• A 45-year-old B-737 rst ofcer experiencing an
alcohol withdrawal seizure suddenly screamed, ex-
tended his arms up rigidly, pushed full right rudder,
and slumped over the yoke during an approach. The
aircraft descended to 1,000 feet above ground level in
an uncoordinated turn to 25 degrees angle of bank
before ight attendants could pull the rst ofcer
off the controls, allowing the captain to recover the
airplane. “Mayday” calls were made, and the captain
executed a missed approach before making a successful
landing.
• When a 48-year-old DC-9 rst ofcer’s foot
became lodged against a rudder pedal after he
stiffened during a heart attack, the captain had to
apply opposite rudder to control the aircraft until
the foot could be dislodged.
• A 44-year-old ight engineer and a 42-year-old
captain lost consciousness when the ight engineer
accidentally turned off a ow pack with the cargo
heat outow valve open, depressurizing their B-727 at
33,000 feet. The captain and ight engineer regained
consciousness only after the rst ofcer donned his
oxygen mask and made an emergency descent.
• A 49-year-old captain stiffened so violently during
an epileptic seizure after landing that he suffered a
fractured shoulder and a lumbar compression fracture.
At the same time, he applied such force to the rudder
pedals that he caused the aircraft to turn sharply and
stop suddenly. The rst ofcer had to remove the cap-
tain from the controls to taxi the aircraft to the gate.
• During an approach own at higher than normal
airspeed, a 56-year-old A-300 captain suffering a ce-
rebral infarction did not ask for the landing gear to be
extended and simply nodded agreement when the rst
ofcer questioned him about it. After touchdown, the
captain used reverse thrust for longer than required
and applied full take-off power on the taxiway. After
the rst ofcer reduced power, the captain again ap-
plied take-off power, and the rst ofcer shut down
the engines and called for assistance.
• When a 48-year-old MD-88 captain wearing mo-
novision contact lenses attempted to make a visual
approach over water under reduced lighting condi-
tions in rain and fog, he perceived the aircraft to be
higher than it actually was. This resulted in a steeper
than normal nal approach, causing the aircraft to
strike the approach lights. Although no one was hurt
on impact, three passengers received minor injuries
during the evacuation following the accident.
• A cargo DC-8 crashed on approach because the air-
crew’s judgment, decision-making, and ying abilities
were impaired by fatigue. The 50-year-old captain had
been awake for 40 of the previous 66 hours, and the
54-year-old rst ofcer had been awake for 47 of the
previous 66 hours prior to the accident. The captain
entered an approach turn stall and failed to recover,
resulting in the accident, which caused serious injuries
to himself, the rst ofcer, and second ofcer.
In-Flight Medical Events and Similar Medical
Histories
There are times when the airman’s FAA medical record
contains coding that is similar to the category assigned to
an in-ight event. While we found no pilots whose in-
ight medical impairments were categorized similarly to

8
9
the codes assigned in their FAA medical history, nine of
39 incapacitations were categorized similarly to the codes
assigned in the pilot’s FAA medical record, as shown in
Table A-3.
Diversions Resulting From In-Flight Medical Events
A ight diversion occurs whenever the aircraft lands at
a destination other than the originally intended airport.
Nineteen of the 39 ights involving incapacitated pilots
and three of the 11 ights with impaired pilots diverted
because of the in-ight medical event. The odds of a ight
diverting is, therefore, 0.48 (95% CI 0.33, 0.64) in the case
of an in-ight medical incapacitation and 0.27 (95% CI
0.01, 0.54) in the event of an impairment. In three of the
19 diversions (16%) for incapacitated pilots, the aircrew-
member did not survive, while in all three diversions for
impaired pilots the affected crewmember did survive.
Of the 19 diversions for incapacitated pilots, three were
classied as cardiac cases, three as gastrointestinal, three as
epileptic seizures, two as hypoxia, and eight due to other
causes. Two of the three cardiac cases were fatal heart attacks,
1 in a 48-year-old pilot and the other in a 56-year-old, and
one was a fatal arrhythmia in a 55-year-old pilot. One of the
gastrointestinal cases was due to food poisoning, 1 the result
of intestinal gas expansion with an increase in altitude, and
the third was suspected to be peritonitis. The two hypoxia
cases occurred on the same ight when the ight engineer
accidentally depressurized the aircraft at 33,000, and both
the captain and ight engineer temporarily lost conscious-
ness. The rst ofcer eventually donned his oxygen mask
and made an emergency descent. Eight cases classied as
“other” included decompression sickness, unknown loss
of consciousness, cholelithiasis, renal lithiasis, muscle
cramps, chest pain of unknown origin, vertigo secondary
to labyrinthitis, and vasovagal syncope.
The three diversions for impaired pilots included 1 case
of cardiac chest pain in a 57-year-old pilot due to unstable
angina, 1 case of viral infection leading to a vasovagal syn-
copal response in a 51-year-old, and 1 case of barosinusitis
during climb in a 43-year-old.
Aeromedical Certication Actions Resulting From
In-Flight Medical Events
The 39 in-ight medical incapacitations led to 16
medical certicate denials (Table A-2, cases 2, 5, 7, 9, 12,
13, 18, 21, 37, 38, 39, 89, 96, 100, 175, and 178). In
addition, a special-issuance code was assigned to 1 pilot
(Table A-2, case 2), 16 history codes were assigned to ten
pilots (Table A-2, cases 2, 5, 11, 12, 27, 32, 33, 73, 82,
and 175), 37 pathology codes were assigned to 22 pilots
(Table A-2, cases 2, 5, 7, 9, 10, 11, 12, 13, 14, 18, 21, 29,
31, 32, 37, 38, 39, 73, 95, 96, 175, and 178), and four
EKG codes were assigned to 1 pilot (Table A-2, case 12).
Also, 11 pilot medical certicates were reafrmed (Table
A-2 ,cases 14, 20, 33, 34, 35, 39, 81, 82, 87, 95, and 98),
ve pilots were eventually re-certied (Table A-2, cases 2,
5, 9, 18, and 73), and in two cases, no codes needed to be
assigned (Table A-2, cases 16 and 17).
The 11 in-ight medical impairments led to three deni-
als (Table A-2, cases 19, 218, and 219). In addition, six
pathology codes were assigned to three pilots (Table A-2,
cases 19, 218, and 219), three medical certicates were
reafrmed (Table A-2, cases 30, 74, and 86), and in ve
cases, no codes were assigned (Table A-2, cases 83, 84, 85,
97, and 148).
Figure 5 shows the trend in the number of aeromedical
certication actions per in-ight medical event (incapacita-
tions and impairments) for all types of actions, including
the assignment of special issuance codes, electrocardiogram
codes, history codes, pathology codes, as well as certicate
denials, re-certication, and reafrmation. As shown in the
gure, there was a signicant decrease in the number of
aeromedical certication actions per event between 1993
and 1998 (p<0.05).
Figure 5. The trend of FAA actions per in-ight medical events.
Figure 5. The trend of FAA actions per in-flight medical events.
y = -0.8343x + 1667.5
R2 = 0.7115
0
1
2
3
4
5
6
7
1993 1994 1995 1996 1997 1998
Year
FAA Actions/Event

8
9
DISCUSSION
Frequency and Rate of In-Flight Events
We found 39 in-ight medical incapacitations for a
rate of 0.045 per 100,000 hours and 11 impairments
for a rate of 0.013 per 100,000 hours on 47 U.S. airline
ights between 1993 and 1998. It is interesting that there
were approximately four times as many incapacitations
as impairments, since impairments were generally less
serious events and could be expected to occur at least as
frequently as incapacitations. This is probably a report-
ing phenomenon. Incapacitated pilots were generally
more seriously ill than impaired pilots. Consequently,
incapacitations resulted in more declared emergencies,
ight diversions, ambulance requests, and hospitaliza-
tions than impairments. Therefore, an incapacitation
would probably have been better documented than an
impairment. Since many of the impairment cases were
less well documented, any record would have depended
on pilot self-reporting, something that would probably
be avoided by most pilots.
Ironically, only two accidents resulted from pilot in-
ight medical impairments, while no incapacitations
resulted in accidents. This may have been because of
the insidious nature of the two impairment accidents.
When a dramatic incapacitating event, such as a heart
attack or epileptic seizure occurs, it is often obvious and
can be dealt with by the unaffected crewmember. In the
two impairments that ended in aircraft accidents, the
pilots were probably not aware there was a problem. In
one case, the pilot normally ew with monovision contact
lenses. In the other, the pilots were probably aware they
were fatigued but were not cognizant of how seriously it
was impacting their performance at the time.
Martin-Saint-Laurent et al. (23) studied sudden
in-ight incapacitation in Air France pilots and ight
engineers from 1968 to 1988 and reported an incapacita-
tion incidence of 0.044 per 100,000 ight hours. While
this is very close to the rate found for incapacitations
in this study, the Martin-Saint-Laurent et al. study
included categories that would have been classied as
impairments in this study; therefore, a more appropriate
comparison would be to include incapacitations and
impairments together. Combining incapacitations and
impairments gives a total of 48 in-ight medical events
and a rate of 0.059 per 100,000 ight hours, which is
only slightly higher than the rate in the Martin-Saint-
Laurent et al. study.
A review of all U.S. Air Force (USAF) accidents coded
for incapacitation, preexisting disease, or other acute ill-
nesses between 1978 and 1987 yielded an incapacitation
rate of 0.019 per 100,000 ight hours (24). This rate is
less than half the rate in the Martin-Saint-Laurent (23)
study and this study; however, the USAF study involved
military pilots who may have been younger and in better
physical condition, and it was restricted to events that
resulted in aircraft accidents. We found that incapacita-
tions rarely resulted in accidents; in fact, there were no
accidents among the Martin-Saint-Laurent et al. cases and
two accidents in this study, neither of which was fatal.
Two additional studies dealing with U.S. airline
pilots were based on loss of licensure data, rather than
in-ight events, and reported incapacitation rates based
on the number of pilots incapacitated per year instead
of ight time (20, 21); therefore, their results could not
be compared with the results of this study in a mean-
ingful way.
The two major airline pilot incapacitation studies that
dealt with in-ight medical events provide only qualita-
tive results (4, 19), which do not allow for meaningful
comparison with other quantitative studies.
Probability of an Accident Due to an In-Flight
Medical Event
Froom reported that less than 1% of all aircraft accidents
are due to pilot in-ight incapacitation (11). Lane calcu-
lated a probability of 5/68 or 0.074 (21). However, Lane
included categories that were classied as impairments in
our study. Combining incapacitations and impairments
in this study, we calculated the probability of an accident
to be 2/50 or 0.04, about half of the probability found
in the Lane study. The difference in proportions may be
due to differences in the types of aircraft operations in
the two studies. While we have included only U.S. airline
pilot in-ight medical events, only two of the ve ights
in the Lane study were airline ights, and one of those
was a positioning ight. Also, two of the ve accidents
in the Lane study involved cargo aircraft, and one was a
Department of Defense charter ight.
The proportion of military pilot in-ight medical in-
capacitations leading to accidents was much higher, prob-
ably due to the difference in the operational environment.
Rayman reported that the probability of an accident was
20/146, or 0.14, in one study of Air Force pilot incapaci-
tations from 1970 to 1980 (32), and 28/59, or 0.47, in
another Air Force study from 1966 to 1971 (31).
Although we found a statistically signicant difference
between the chance of an accident (given there was an in-
ight medical event), compared with ights where there
was no in-ight medical event, this difference must be
interpreted in terms of its operational signicance. The
in-ight events in the two accidents are not representa-
tive of most other in-ight events. In neither case was
the ight crew acutely affected by a medical condition

10
11
in the same sense as a pilot who suffers chest pain from a
heart attack or abdominal pain from a kidney stone, for
example. These events might have ultimately led to an ac-
cident because they did not represent dramatic events that
could have been detected and dealt with by the unaffected
pilot. Some authors have made clear distinctions between
obvious and subtle incapacitation (7, 14). Raboutet and
Raboutet even asserted that an incapacitation needed to
be complete for an accident to result (30). However, it is
easy to imagine scenarios that involve: (1) incapacitation
of the non-ying pilot, which increases the workload on
the ying pilot to an unsafe level, (2) the subtle, insidious
incapacitation of a crewmember that is not apparent until
a critical phase of ight, (3) the partial incapacitation of
a crewmember that degrades performance to an unsafe
level, and (4) the incapacitation of a crewmember dur-
ing a non-critical phase of ight that continues into a
critical phase, resulting in an unacceptable increase in
workload for the ying pilot. Any of these situations
could result in an aircraft accident without meeting the
Raboutets’ criteria. In addition, Crowley (8) also found
the conditions described by Raboutet and Raboutet overly
restrictive. Our ndings suggest that a subtle, unperceived
impairment might be more dangerous than an obvious,
complete incapacitation. Although safety of ight was
severely affected in seven of the 47 ights studied, ac-
cidents resulted in only two of those ights. In the other
ve ights, where the incapacitating event was not subtle,
the unaffected pilot was able to recognize the emergency
and prevent an accident, even in those situations where
the event occurred on short nal.
Age Distribution of In-Flight Medical Events
Some studies examining the Age 60 Rule for airline
pilots have focused on pilot performance and aircraft
accident data (3, 16, 17, 18). Hyland et al. studied all
accidents involving pilots with Class I medical certi-
cates and found a decrease in accident rate with age (16,
17, 18). Broach et al. found a “U”-shaped curve with a
decrease in accidents with increasing age, followed by a
slight increase in older age groups for professional pilots
holding Class I or II medical and ATP or Commercial
pilot certicates (3).
Other researchers have suggested that replacement of
older, experienced pilots by younger, inexperienced pilots
could adversely affect ight safety, and it may be preferable
to grant waivers to experienced pilots with an increased
incidence of disease-related, in-ight sudden incapacita-
tion than to replace them with younger, inexperienced
pilots (2, 11). Froom reported that inexperienced pilots
have a two to three time increased incidence of pilot error-
related accidents and cautioned that the estimated risk of
in-ight medical incapacitation needed to be balanced
by a consideration of pilot experience (11).
We found a signicant increase in the percentage of
incapacitations with age among the most frequent cat-
egories of in-ight medical incapacitation for U.S. airline
pilots; however, we did not nd a signicant difference
between the mean age of pilots where safety of ight was
seriously impacted and those pilots where safety of ight
was not seriously at issue.
Categories of In-Flight Medical Events
Although many studies have dealt with pilot medical
incapacitation, few have analyzed in-ight medical events
(6, 7, 13, 14, 19, 21, 29). Studies of in-ight medical
events had different results or categorized cases differently,
making comparison between studies difcult; however,
we found no signicant differences between four in-ight
studies (4, 24, 31, 32) when they were compared by
categorizing cases with the same classication scheme
used in this study and compared using a Krushal-Wallis
ANOVA.
Fatal In-Flight Medical Events
Only three in-ight medical incapacitation studies
reported fatalities (24, 30, 31); however, it was not al-
ways clear when fatalities occurred or how many fatalities
occurred in several other studies. Rayman reported 24
fatalities in a six-year study of sudden in-ight incapaci-
tation in USAF pilots, or four fatalities per year (31). In
addition, in a ten-year study of in-ight incapacitation
in USAF pilots, McCormick and Lyons (24) found one
pilot was fatally injured in the crash of his single-seat
aircraft after suffering a myocardial infarction in-ight,
for a rate of 0.1 fatalities per year, while Raboutet and
Raboutet (30), in a 25-year investigation of professional
French pilots, found that one pilot suffered a massive
pulmonary embolism in-ight and died about one month
post-crash, yielding a rate of 0.04 fatalities per year.
We found four deaths in our six-year study, which
equates to 0.67 fatalities per year. The wide range in
fatality rates per year from 0.04 to four in other studies
can be partially explained by the fact that pilot fatalities
resulting from in-ight medical incapacitation are very
rare, random events; therefore, exposure time should be
considered in evaluating them. Fatality rates based on
ight-time exposure were not provided in earlier studies;
therefore, comparisons were not possible. Accounting
for ight-time exposure, we calculated a fatality rate of
0.00467 per 100,000 ight hours (95% CI 0.00465,
0.00468).

10
11
Safety of Flight and In-Flight Medical Events
Chapman (7) analyzed over 1,300 simulator exercises
using two protocols. In the rst protocol, the authors de-
termined that safety of ight was at risk in 15 out of 500
(3%) of the cases, and it was felt that an accident would
have resulted in eight (1.6%) cases. In the second protocol,
ten out of 800 (1.25%) were felt to have represented a risk
to safety of ight; in two (0.25%), the authors believed
aircraft accidents would have resulted.
A survey by the International Federation of Airline
Pilots Associations conducted by Bennett (1) showed
that the pilots surveyed considered that safety was only
signicantly threatened in 3% of the incidents because
there was time available to warn the other pilot of the
problem .
Buley (4) reviewed IFALPA in-ight incapacitation
questionnaire data and reported that 40% of responding
aircrew members felt that safety of ight was not affected,
56% believed it was potentially affected, and 4% were
convinced that safety of ight was actually affected.
Our data showed that safety of ight was severely
impacted in 15% (7 out of 47) of the ights we studied.
This gure is higher than those reported by Chapman,
Bennett, or Buley. Differences in the percentage of
ights where safety of ight was impacted might be due
to differences in methods among studies. In our study,
we reviewed actual in-ight events and judged whether
the circumstances would have severely impacted safety
of ight. In the studies conducted by Chapman, the
researchers were required to determine if what was done
in a ight simulator would have affected safety of ight
in the aircraft under similar circumstances. The Buley
study was the only one that collected safety of ight
data directly from aircrew members; however, Buley’s
study would have relied on the ability of pilots to recall
details of events that may have occurred many years
earlier. Stone and Shiffman (34) recently reported that
retrospective assessments may be prone to recall bias and
distortion. In addition, Hunter (15) recently reported that
exposure to hazardous aviation events may be associated
with risk misperception by pilots. These factors suggest
why pilots in the Buley study might have perceived the
risk to safety of ight for a past event as being less than
it actually was.
In-Flight Medical Events and Similar Medical
Histories
There are times when the airman’s FAA medical record
contains coding that is similar to the category assigned
to an in-ight event. We found no pilots whose in-ight
medical impairments were categorized similarly to the
codes assigned in their FAA medical record. However,
nine of 39 incapacitations were categorized similarly to
the codes assigned in the pilot’s FAA medical record as
shown in Table A-3. It should be noted that the similari-
ties between the assigned incapacitation categories and
the corresponding FAA medical codes do not necessarily
imply a cause-and-effect relationship and do not, in any
way, suggest the airman should not have been medically
certied because of the documented pre-existing condi-
tion. For example, an airman with hypertension controlled
with medication may be assigned a pathology code of
485 and issued a valid medical certicate. If the airman
then suffers a stroke in-ight, it might be argued that
there could have been a relationship between the airman’s
hypertension and the stroke, since hypertension is a risk
factor for stroke. However, the stroke could also have
resulted from an undiagnosed cerebral aneurism.
In addition, an FAA medical code does not necessarily
imply the airman should have been disqualied. Airmen
often have codes assigned (pathology codes, history codes,
or EKG codes) to indicate the presence of medical condi-
tions that are not disqualifying.
Diversions Resulting From In-Flight Medical
Events
Diversions for medical purposes represent a signicant
problem for commercial air carriers (9). Delay to origi-
nal destination, passenger inconvenience, increased risk
to safety, and cost factor into the complexity of aircraft
diversions (12). The exact cost of a medical diversion typi-
cally ranges between approximately $3,000- $100,000,
depending upon whether fuel needs to be dumped before
landing and whether or not passenger overnight accom-
modations are arranged (33). Landing weight is also a
consideration, and valuable fuel may have to be jettisoned
to attain a suitable landing weight for a premature touch-
down. While it is more difcult to put a dollar amount
on safety of ight, this is perhaps the most important
consideration in any diversion situation. If the unaffected
pilot is forced to proceed via an unplanned route to an
unexpected destination and perform an unfamiliar instru-
ment approach, this could reduce the margin of safety
still further in an already hazardous situation.
Nineteen of the 39 ights involving incapacitated pilots
and three of the 11 ights with impaired pilots diverted.
The diversion rate for all in-ight medical events was 22
out of 47, or 46.8% of ights. This is over twice the 20%
diversion rate reported by Martin-Saint-Laurent et al.
(23) in a study of in-ight incapacitation in commercial
aviation in Air France pilots and ight engineers from
1968 to 1988. Since the methodologies of the two studies
are very similar, the difference in percentages of diverted
ights are probably due to other factors. One reason

12
13
might be differences in the corporate culture between U.S.
airlines in the 1990s and Air France in the 1960s that
might have inuenced the ight crews’ decision whether
to divert. Another reason could be differences in the type
of operation between the U.S. domestic airlines and Air
France. For example, if many of the Air France ights
were international ights, diversions for medical events
may not have been practical because continuing to the
destination would often be as appropriate.
Aeromedical Certication Actions Resulting From
In-Flight Medical Events
The pathology, history, and EKG codes assigned as a re-
sult of FAA Aeromedical Certication Division (AMCD)
action become part of the airman’s FAA medical record
and have been incorporated into the new computerized
Document Imaging and Workow System (DIWS). It
must be noted that more than one type of action was
taken in some cases. For example, a pilot may have been
assigned a pathology code, been denied as a result of an
in-ight medical event, and then eventually re-certied
with a special issuance medical certicate.
Although there was a signicant decrease in the number
of certication actions per in-ight medical event with
time, it did not appear to reect any change in FAA
aeromedical certication policy, nor did it appear to be
a function of the frequency of in-ight medical events.
In fact, the frequency of events increased with time, and
the effect of frequency was accounted for in Figure 5.
One possible explanation for the trend was the chance
distribution of events. An examination of the events
showed those which occurred between 1993 through
1995 generally required more certication actions, while
the events from 1996 to 1998 generally required fewer
certication actions. Between January 1, 1993, and De-
cember 31, 1995, there were two cases of fatigue resulting
in an aircraft accident with serious injuries, an alcohol
withdrawal seizure, three cardiac events, three neurologi-
cal events, a vasovagal syncopal episode, two episodes of
loss-of-consciousness, and two miscellaneous events. The
average number of FAA certication actions per event for
this group was 3.6. In contrast, between January 1, 1996,
through December 31, 1998, there were two neurological
cases, three cardiac cases, four cases of vasovagal syncope,
ve gastrointestinal cases, three cases of carbon dioxide
poisoning, two cases of hypoxia, three cases of renal
lithiasis, two cases of unknown losses of consciousness,
and two miscellaneous cases. In addition, there was one
case each in the following categories: endocrine, vascular,
vision, trauma, laser illumination blindness, medication
use, and respiratory. The mean number of FAA aero-
medical certication actions per event for the post-1996
group was 1.7, which is about half that of the pre-1996
group. Therefore, the decreasing trend in the number of
certication actions with time appears to have been due
to random causes and not any purposeful change in FAA
aeromedical certication policy.
In-ight medical incapacitation and impairment cases
are regularly reviewed at CAMI medical staff meetings
attended by physicians from the Aerospace Medical Cer-
tication and the Aerospace Medical Research Divisions.
Details of the in-ight event, as well as the pilot’s medical
history, are evaluated in deciding what action to take and
which codes to assign. This process is very important
because it may affect future aeromedical certication deci-
sion-making, should the airman develop future medical
problems or experience another in-ight event.
It is important to note the FAA actions taken in the
39 in-ight medical incapacitation cases and 11 impair-
ment cases we studied represent action after the fact. In
many instances, the FAA Aerospace Medical Certica-
tion Division denies medical certication applications for
airline pilots, possibly preventing many serious in-ight
events. One FAA study (10) reported an overall denial
rate for medical reasons of 4.3 per 1,000 active aviators
for calendar years 1987 and 1988. The highest age-specic
denial rate was for the 55-to-59 age group, and the most
signicant causes for denial for all age groups were coronary
artery disease (8.5%), disqualifying medications (6.2%),
psychoneurotic disorders (6.1%), myocardial infarction
(5%), and disturbance of consciousness (4.4%).
CONCLUSIONS
In-ight medical events in U.S. airline pilots were very
rare; resulting aircrew deaths were even more rare, and re-
sulting aircraft accidents were extremely rare. Fortunately,
in the six years and nearly 86 million ight hours covered
by this study, there were no passenger fatalities caused by
pilot in-ight medical events. The two aircraft accidents
resulted in serious injuries to three aircrew members and
minor injuries to three passengers.
One study, focusing on professional pilot perfor-
mance and aircraft accident data, found a curvilinear
relationship between pilot age and performance (3).
Still, other researchers have argued that replacement
of older, experienced pilots by younger, inexperienced
pilots could adversely affect ight safety. We found a
signicant increase in the percentage of incapacitations
with age. However, there was no difference between the
mean age of pilots involved in ights where safety of
ight was seriously affected and the mean age of pilots
not involved in such ights.

12
13
Although there was a signicant difference between
the probability of an aircraft accident, given an in-ight
medical event, this result must be interpreted in its
operational context. Both accidents involved the subtle
impairment of the pilot in ways that are not classically
thought of as medical incapacitation, and it may be that
subtle impairment of a pilot is more dangerous than
obvious medical incapacitation.
The most important factor that appears to be respon-
sible for the exceptionally good U.S. airline safety record
associated with in-ight medical incapacitations is the
presence of a second pilot. In ve out of the seven cases
where safety of ight was considered to be severely im-
pacted, the aircraft was taken over by the unaffected pilot
who made a successful landing. In the two cases where
the affected pilot remained at the controls after subtle
impairment, both resulted in an aircraft accident.
Aeromedical studies on incapacitation have been few
in number, retrospective, and less detailed than most
other scientic studies. There is a lack of high-quality
data, which has led to a lack of adequate research and
inadequate information and recommendations. To be
most valuable, future research needs to be based on ac-
tual in-ight medical events and should be normalized
to a useful denominator, such as ight time, to allow
for meaningful comparison between studies. Since the
most frequent categories of incapacitation were loss of
consciousness, cardiac, neurological, and gastrointestinal
(occurring mostly in older pilots), future research should
be directed toward these areas.
REFERENCES
1. Bennett G. Medical-cause accidents in commercial
aviation. Euro Soc Card 1992; 13 (Supplement
H):13-15.
2. Bennett G. Pilot incapacitation and aircraft accidents.
Euro Heart J 1988; 9 (Supplement G):21-4.
3. Broach D, Joseph, KM, Schroeder DJ. Pilot age and
accident rates report 3: An analysis of professional
air transport pilot accident rates by age. Oklahoma
City, OK: Federal Aviation Administration, Civil
Aerospace Medical Institute. 2003: OAM Research
Task AAM-00-A-HRR-520.
4. Buley LE. Incidence, causes and results of airline pilot
incapacitation while on duty. Aero Med 1969; 40
(1):64-70.
5. Cameron A, Trivedi P (1988). Regression analysis of
count data. New York, NY: Cambridge University
Press.
6. Castelo-Branco A, Cabral-Sa A, Borges JC. Compara-
tive study of physical and mental incapacities among
Portuguese airline pilots under and over age 60.
Aviat Space Environ Med 1985; 56:752-7.
7. Chapman P. The consequences of in-ight incapaci-
tation in civil aviation. Aviat Space Environ Med
1984; 55:497-500.
8. Crowley J, Campbell J, Johnson P, Powell JA. Coro-
nary artery disease and in-ight incapacitation.
U.S. Army Medical Department Journal 2000;
(July-September):17-22.
9. Davies G, Degotardi P. In-ight medical emergencies.
Aviat Space Environ Med 1982; 53(7):694-700.
10. Downey L, Dark S. Medically disqualied airline
pilots in calendar years 1987 and 1988. Washing-
ton, DC: Federal Aviation Administration, Ofce of
Aviation Medicine. 1990; DOT/FAA/AM-90/5.1
11. Froom P, Benbassat J, Gross M, et al. Air accident,
pilot experience, and disease-related in-ight sud-
den incapacitation. Aviat Space Environ Med 1988;
59:278-81.
12. Goodwin T. In-ight medical emergencies: An over-
view. BMJ 2000; 321:1338-41.
13. Harper C, Kidera G, Cullen J. Study of simulated air-
line pilot incapacitation: Phase II. Subtle or partial
loss of function. Aero Med 1971; 42 (9):946-8.
14. Harper C, Kidera G, Cullen J. Study of simulated
airline pilot incapacitation; Phase 1 – obvious and
maximal loss of function. Aero Med 1970; 41 (10):
1139-42.
15. Hunter DR. Risk perception and risk tolerance in
aircraft pilots. Washington, DC: Ofce of Aerospace
Medicine. 2002; DOT/FAA/AM-02/17.1
16. Hyland D. Experimental evaluation of aging and
pilot performance. 7th International Symposium
on Aviation Psychology. 1993; Columbus, OH.
17. Hyland D, Kay E, Deimler J, et al. Age 60 rule re-
search, Part II: Airline pilot age and performance – A
review of the scientic literature. Washington, DC:
Federal Aviation Administration, Ofce of Aviation
Medicine. 1994; DOT/FAA/AM-94/21.1
1This publication and all Ofce of Aerospace Medicine technical
reports are available in full-text from the Civil Aerospace Medical
Institute’s publications Web site: http://www.cami.jccbi.gov/aam-
400A/index.html

14
18. Hyland D, Kay E, Deimler J. Age 60 rule research, Part
IV: Experimental evaluation of pilot performance.
Washington, DC: Federal Aviation Administration,
Ofce of Aviation Medicine. 1994; DOT/FAA/
AM-94/23.1
19. James M, Green R. Airline pilot incapacitation survey.
Aviat Space Environ Med 1991; 62:1068-72.
20. Kulak LL, Wick RL, Jr., Billings CE. Epidemiological
study of in-ight airline pilot incapacitation. Aero
Med 1971; 42 (6):670-2.
21. Lane JC. Risk of in-ight incapacitation of airline
pilots. Aero Med 1971; 42 (12):1319-21.
22. Li G. Pilot-related factors in aircraft crashes: A review
of epidemiologic studies. Aviat Space Environ Med
1994; 65:944-52.
23. Martin-Saint-Laurent A, Lavernhe J, Casano G, et
al. Clinical aspects of in-ight incapacitations in
commercial aviation. Aviat Space Environ Med
1990; 61:256-60.
24. McCormick TJ, Lyons TJ. Medical causes of in-ight
incapacitation: USAF experience 1978-1987. Aviat
Space Environ Med 1991; 62:884-7.
25. NTSB. http://www.ntsb.gov/NTSB/query.asp#query
_start 10/24/02.
26. NTSB. Accidents and accident rates by NTSB clas-
sication, 1982 through 2001, for U.S. air carriers
operating under 14 CFR 121, http://www.ntsb.gov/
aviation/Table2.htm on 10/30/02.
27. NTSB. Accidents, fatalities, and rates, 1982 through
2001, for U.S. air carriers operating under 14 CFR
121, scheduled and nonscheduled service (air-
lines), http://www.ntsb.gov/aviation/Table5.htm
on 2/24/02.
28. Parmet A, Underwood-Ground K. Reported in-ight
incapacitation: The early birds of 1911. Aviat Space
Environ Med 1987; 58:276-8.
29. Preston FS. Twelve year survey of airline pilots. Aero
Med 1968:312-4.
30. Raboutet J, Raboutet P. Sudden incapacitation en-
countered in ight by professional pilots in French
civil aviation, 1948-1972. Aviat Space Environ Med
1975; 46 (1):80-1.
31. Rayman R. Sudden incapacitation in ight: 1 Jan 1966
- 30 Nov 1971. Aero Med 1973; 44 (8):953-5.
32. Rayman R, McNaughton GB. Sudden incapacitation:
USAF experience, 1970-80. Aviat Space Environ
Med 1983; 54 (2):161-4.
33. Rodriguez A, Golaszewski R. (2004). Some prelimi-
nary evidence on the costs to airlines and passengers
of schedule disruptions: working paper. Jenkintown,
PA, GRA Inc.
34. Stone A, Shiffman S. Capturing momentary, self-
report data: A proposal for reporting guidelines.
Ann Behav Med 2002; 24 (3):236-43.

A1
0
5
10
15
20
25
LE29
30-34
35-39
40-44
45-49
50-54
55-59
GE60
Age Group (Years)
Percent of Pilots
Figure A-1. Average age distribution of U.S. professional pilots from 1993 to 1998 (3). Inclusion
criteria included: air transport pilots or commercial pilots, class-I or -II medical certificate,
occupation professional pilot, employer CFR Part 121 or 135 operator, 200 recent flight hours and at
least 1500 total flight hours.
APPENDIX A

A2
A3
Table A-1. Medical Case Alert Form.
MEDICAL CASE ALERT
(Including incapacitations, special medical circumstances, etc.)
1. AIRCRAFT
ACCIDENT/INCIDENT
1A. DATE 1B. TIME 1C. LOCATION 1D.
NUMBER/TYPE
1E. NUMBER OF PERSONS ON
BOARD
1F. PILOT FATALITY
YES NO
1G. COPILOT FATALITY
YES NO
2A. OCCUPANT
STATUS X Pilot Cabin Crew
CoPilot Passenger
2B. FULL NAME 2C. SEX
Male Female
3. MEDICAL CLASS
I II III
3A. SOCIAL SECURITY NUMBER (Airmen Only)
3B. ANY KNOWN MEDICAL CONDITIONS (ex: SODA, SI, Undisclosed medication or condition, Path Codes)
4. INJURY STATUS
5. ESTIMATED ROLE OF
INCAPACITATION NONE POSSIBLE CERTAIN
6. ESTIMATED ROLE OF
IMPAIRMENT NONE POSSIBLE CERTAIN
7. POSSIBLE FACTORS INVOLVED IN
THE MISHAP
7a. Medical condition
7b. Medication
7c. Alcohol
7d. Marijuana/Illicit drug
7e. Carbon Monoxide
7f. Pesticide
7g. Hypoxia
7h. Spatial disorientation
7i. Fatigue
7j. Emotional Stress
7k. Other (specify)
8. TOXICOLOGICAL
DATA
PERFORMED BY:
Address 1:
Address 2:
City: State: Zip Code:
Phone:
FINDINGS:
9. AUTOPSY DATA
PERFORMED BY:
Address 1:
Address 2:
City: State: Zip Code:
Phone:
COMMENTS: Pending
10. FAA/NTSB IIC
FEEDBACK
PROVIDED BY:
Address 1:
Address 2:
City: State: Zip Code:
Phone:
11. NARRATIVE COMMENTS: (Elaborate on any of the above, or other significant factors. Use a separate sheet if additional space is needed.)
Medical person completing form: (Name) (Phone) DATE:

A2
A3
Table A-2. Case Summaries.
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
2 1994 45 Yes No Incapacitation ALCOHOL
WITHDRAWAL
SEIZURE
The first officer suddenly screamed, extended his arms up rigidly,
pushed full right rudder and slumped over the yoke. Aircraft
descended to 1,000 feet before flight attendants pulled the F/O off
the controls. "Mayday" calls were made and the captain made a full
missed approach before making a successful normal landing.
Oxygen was administered for about 7 minutes. F/O had another
grand mal seizure at the hospital.
3 1994 52 No Yes Incapacitation CARDIAC Aircraft was on approach with the first officer flying when his left
arm slid off the throttles and the airman lay back in the seat. CPR
was performed. It appeared to be a heart attack.
4 1994 55 No Yes Incapacitation CARDIAC The captain became limp at the controls. CPR was performed. The
second officer moved into the captain's seat, the F/O declared an
emergency and the flight diverted where an uneventful, overweight,
landing was made. The captain was transported to hospital where
he was pronounced dead.
5 1995 59 No No Incapacitation VASOVAGAL The captain lost consciousness and slumped across the center
pedestal. Oxygen was administered and the captain regained
consciousness in 30 to 40 seconds. However, he passed out again,
and recovered again. First officer landed the aircraft; captain taxied
to the gate.
6 1995 48 Yes Yes Incapacitation CARDIAC The first officer complained of heartburn, profuse sweating,
tingling in both of his arms and nausea. His appearance was
described as ashen gray. The captain assumed control. Symptoms
passed and he resumed flying. First officer complained that the
heartburn pains were returning so the captain assumed control and
diverted.
First officer eventually lost consciousness, began twitching and
stiffened, and loudly exhaled. CPR was performed. The first
officer's left leg had became lodged against the left rudder when he
stiffened, requiring the captain to apply right rudder to control the
aircraft until the F/O’s leg was dislodged.

A4
A5
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
7 1995 57 No No Incapacitation NEUROLOGICAL During descent from cruise, the captain failed to respond to a
heading from Air Traffic Control and did not respond to the first
officer. What was described as a grand mal seizure followed. The
F/O landed the aircraft and the captain was taken to local area
hospital. (Hospital suspected a brain tumor of the left temporal
lobe).
9 1995 56 No No Incapacitation LOSS OF
CONSCIOUSNESS
GASTROINTESTINAL
The captain became nauseated, stomach became bloated and
uncomfortable then he lost consciousness. The flight was diverted.
Upon landing the captain was taken to hospital.
10 1995 41 No No Incapacitation MISCELLANEOUS The first officer experienced severe back pain soon after take-off.
The captain declared an emergency, and the flight returned to its
place of origin.
11 1996 37 No No Incapacitation GASTROINTESTINAL The first officer became medically incapacitated inflight, and the
flight diverted. The F/O was admitted to hospital for treatment of
possible peritonitis. He remained in hospital for two days and was
sent home by train. The first officer had had an appendectomy two
months previous to the incident.
12 1996 56 No No Incapacitation VASOVAGAL
GASTROINTESTINAL
The captain experienced severe abdominal pains and became
medically incapacitated in flight. He was taken to hospital, where
workups including CTscan, gastro consultation, and treadmill tests
concluded event as probable gastroenteritis with vasovagal
response. He had a near-syncopal episode during treadmill. Neuro
and gastrointestinal work-ups were within normal limits.
13 1996 43 No No Incapacitation ENDOCRINE The second officer became hypoglycemic. He was on oral anti-
hypoglycemic agents and had a similar episode six weeks prior at
the gate. This resulted in the airman being removed from the
aircraft prior to the flight. Medical certificate was surrendered by
the airman.
14 1996 47 No No Incapacitation VASOVAGAL
GASTROINTESTINAL
Take-off was aborted due to the first officer’s incapacitation. F/O
had slumped over in the cockpit but "recovered" as they were
taxiing back to the gate. He appeared pale and was sweating
profusely. The F/O was taken to hospital and held overnight. The
airman reported eating a bad sandwich the day before.

A4
A5
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
16 1996 44 Yes No Incapacitation HYPOXIA The aircraft experienced decompression at 33,000 feet. The first
officer made an emergency descent. It appears the flight engineer
may have turned off the right flow pack with the cargo heat outflow
valve open. Cabin altitude increased toward 33,000 ft and the first
officer donned his mask. The captain, flight engineer, and flight
attendant lost consciousness but regained during descent. Oxygen
masks deployed for all passengers. The captain took over the
aircraft and landed.
17 1996 42 Yes No Incapacitation HYPOXIA The aircraft experienced decompression at 33,000 feet. The first
officer made an emergency descent. It appears the flight engineer
may have turned off the right flow pack with the cargo heat outflow
valve open. Cabin altitude increased toward 33,000 ft and the first
officer donned his mask. The captain, flight engineer, and flight
attendant lost consciousness but regained during descent. Oxygen
masks deployed for all passengers. The captain took over the
aircraft and landed.
18 1996 31 No No Incapacitation VASOVAGAL While the aircraft was climbing through 28,000 feet the first officer
went limp and slumped to left side of seat. Attempts to revive him
were unsuccessful. After approximately 20 to 30 seconds he started
to stir a little, then abruptly started flailing about. The flailing lasted
about 10 seconds, knocking off the captain's glasses and turning on
the deice switches on the overhead panel. The F/O then came to.
Over a period of about five minutes the airman regained full
consciousness, but was drenched in sweat, and “as white as my
shirt.” Airman became fully coherent with complete situational
awareness. The flight then proceeded normally to the alternate
airport. Initial evaluation was negative and electrocardiogram was
normal.
19 1996 57 No No Impairment CARDIAC In-flight, the captain began experiencing chest pain. The airman
reported experiencing intermittent retrosternal chest pain since the
previous day, which was becoming more continuous with radiation
to his left jaw and left arm. The flight diverted, the captain taxied
the aircraft to the gate. The clinical impression was chest pain with
possible unstable angina.

A6
A7
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
20 1996 41 No No Incapacitation MISCELLANEOUS The captain experienced chest pains during flight. A nurse who was
on board stated that he appeared pale and anxious; the flight
diverted. Upon landing the airman was taken to the hospital for
treatment. The captain was admitted and kept overnight for
observation. Electrocardiogram was performed; he was without
chest pain upon entry to the emergency room, stress test performed
and enzymes tested. All studies were normal.
21 1996 56 Yes No Incapacitation VASCULAR A scheduled domestic passenger flight was stopped on the taxiway
after landing when the captain became incapacitated.
The first officer stated that the captain was flying the aircraft, and
during the approach the captain did not ask for the landing gear to
be extended. The approach to the runway was flown at a higher
than normal speed and after touchdown the captain used reverse
thrust for a longer than normal time. After exiting the runway onto
the taxiway the captain applied takeoff engine power. The first
officer closed the engine power levers, the captain again tried to
apply takeoff engine power. The F/O realized the captain was
incoherent and closed the engine power levers and shut down the
engines. He then called for assistance from the flight attendants and
asked the air traffic controller to send out rescue personnel.
27 1997 37 No No Incapacitation GASTROINTESTINAL While enroute, the first officer experienced severe stomach pain.
The captain decided to divert. Paramedics took the first officer to
hospital. Diagnosis: severe intestinal gas blockage.
29 1997 54 No No Incapacitation TRAUMATIC INJURY While the first officer was performing the pre-flight check, a drop
of hydraulic fluid (Skydrol) entered into his eye. The airman was
then taken to the local hospital where the eye was washed out.
30 1997 34 No No Impairment LASER ILLUMINATION
BLINDNESS
The captain was on the controls when he noticed a green light
illuminating the aircraft. He reported that his exposure to the light
caused a minimal, yet persistent loss of night vision.
31 1997 58 No No Incapacitation UROLOGICAL First officer declared an emergency due to the incapacitation of the
captain. An onboard doctor determined it was a stomach related
problem. The captain was vomiting, had pain in stomach, and was
very weak. The first officer landed the aircraft. The captain was
taken to hospital, and kidney stones were passed.

A6
A7
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
32 1997 51 No No Incapacitation GALL BLADDER Enroute, the first officer developed severe abdominal pains. The
captain radioed that the F/O had "gas pain with significant
discomfort" and requested paramedics meet the aircraft, an
emergency was declared. The F/O was taken to hospital. Diagnosis
was gallbladder attack; subsequently surgery was performed.
33 1997 53 No No Incapacitation GASTROINTESTINAL The captain became ill with flu-like symptoms, nausea, vomiting
and diarrhea during a transatlantic flight. An onboard physician
said the captain was suffering from acute gastroenteritis, secondary
to food
p
oisonin
g
.
34 1997 54 No No Incapacitation GALL BLADDER The flight declared an emergency and made an unscheduled landing
because the flight engineer had a suspected heart attack. The airman
was transported to hospital in stable condition. Working diagnoses
was cholelithiasis.
35 1997 30 No No Incapacitation MEDICATION An emergency was declared after the first officer experienced chest
pains. The pilot was taken to the local hospital. Preliminary reports
indicated the chest pains were not cardiac related.
37 1998 38 No No Incapacitation NEUROLOGICAL
(SEIZURE)
The first officer was on a break, and asleep when a flight attendant
noticed he was bleeding from his mouth and tongue. The airman
was awakened and appeared disoriented. It was assumed the F/O
had a seizure while sleeping. The airman was taken to a hospital
and had a seizure while havin
g
an ECG.
38 1998 48 No No Incapacitation UNKNOWN LOSS OF
CONSCIOUSNESS
The captain experienced severe abdominal pain during flight. He
collapsed in his seat and was unresponsive. The captain reported
eating at a Cuban restaurant. Complaints were diarrhea, vomiting,
sweating, panting, and abdominal pain. The hospital found him to
be dehydrated and administered approx 3 liters of fluid.
39 1998 50 No No Incapacitation UNKNOWN LOSS OF
CONSCIOUSNESS
The first officer was found unresponsive and an onboard physician
was called to attend him. He appeared to have had a seizure and
appeared to be unconscious for 30 minutes. He walked through the
airport and refused medical attention from the paramedics who had
been sent to meet him at the gate; he also refused to be checked at
a hospital.

A8
A9
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
72 1998 56 No Yes Incapacitation CARDIAC The first officer told his captain that he was not feeling well shortly
before suffering an apparent heart attack. He was observed
slumped over in his seat. The flight diverted. CPR was given.
Paramedics were waiting at the gate six minutes after the F/O
collapsed, but they were unable to revive him.
73 1998 50 No No Incapacitation UROLOGICAL During flight, the captain suffered severe back pain. The flight
diverted. The pain went away after landing but returned on the way
to the hospital. An IVP at the hospital revealed dilation at the left
utethral-vesticular junction, most likely due to a kidney stone.
74 1998 51 No No Impairment VASOVAGAL The first officer had what was thought to be indigestion before
departure. After the F/O stretched and raised an arm, the airman
stated that it felt "heavy" and that a weakness was felt in the arm.
The F/O broke out into a cold sweat.
81 1998 47 No No Incapacitation UROLOGICAL The first officer was having chest and abdominal pains. Hospital
diagnosed kidney stones.
82 1998 56 No No Incapacitation CARDIAC The captain became intensely nauseated during the flight and
developed severe anterior chest heaviness, along with aching
discomfort without radiation. The captain broke out into a sweat
and it appeared the airman was having a heart attack. The captain
was treated onboard by two physicians; given four tablets of
nitroglycerin that produced decreased chest discomfort but gave
him a headache.
83 1998 55 No No Impairment CARBON DIOXIDE
POISONING
The aircraft was taxiing to takeoff when all four occupants became
short of breath. All occupants donned oxygen masks, and the
captain taxied the airplane back to the ramp. The crew were
transported to hospital.
Diagnosis was physical impairment resulting from an
accumulation in the cockpit of carbon dioxide fumes produced by
dry ice, a hazardous material, carried in the main cargo
compartment.

A8
A9
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
84 1998 28 No No Impairment CARBON DIOXIDE
POISONING
The aircraft was taxiing to takeoff when all four occupants became
short of breath. All occupants donned oxygen masks, and the
captain taxied the airplane back to the ramp. The crew were
transported to hospital.
Diagnosis was physical impairment resulting from an
accumulation in the cockpit of carbon dioxide fumes produced by
dry ice, a hazardous material, carried in the main cargo
compartment.
85 1998 27 No No Impairment CARBON DIOXIDE
POISONING
The aircraft was taxiing to takeoff when all four occupants became
short of breath. All occupants donned oxygen masks, and the
captain taxied the airplane back to the ramp. The crew were
transported to hospital.
Diagnosis was physical impairment resulting from an
accumulation in the cockpit of carbon dioxide fumes produced by
dry ice, a hazardous material, carried in the main cargo
compartment
86 1998 29 No No Impairment GASTROINTESTINAL
DEHYDRATION
The captain had been feeling ill all day. On approach he began to
vomit, which continued for a few minutes. Local clinic diagnosed
gastroenteritis.
87 1998 31 No No Incapacitation MISCELLANEOUS The first officer experienced dizziness when getting up to go for
aspirin for an earache. LOC was initially reported, although this
was not confirmed.
89 1998 49 Yes No Incapacitation NEUROLOGICAL
(SEIZURE)
The aircraft had just landed when the captain apparently had a
seizure episode for over a minute and a half. The captain’s body
became stiff, the back arched, and the captain bit his tongue, and
dislocated/fractured the left shoulder, also sustaining a lumbar
compression fracture. The captain caused the aircraft to turn right
and come to a sudden stop due to the stretched position. The
airman regained consciousness shortly thereafter. The first officer
removed the captain from the controls and taxied to the gate.
95 1998 25 No No Incapacitation LOSS OF
CONSCIOUSNESS
CARDIAC
During flight the F/O lost consciousness for less than one minute.
After regaining consciousness, the F/O was able to fully perform
some duties and the flight continued to destination.

A10
A11
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
96 1994 57 No No Incapacitation NEUROLOGICAL
(SEIZURE)
During a transatlantic flight, the captain lost consciousness. The
aircraft diverted. The captain was in the jump seat at the time, the
episode was described as an out-of-body sensation with the head
jerking to the right. The captain passed out and became wedged in
between the cockpit seats. The captain was unresponsive with arms
folded across the chest. The captain was stiff, and did bite the
tongue. However, he regained his senses in less than one minute
and did not describe any significant postictal phase.
97 1998 43 No No Impairment RESPIRATORY On departure, the F/O experienced increasing pain and pressure in
the sinuses and right inner ear. The pain became worse as the
aircraft ascended; pressure increased and was accompanied by
numbness. Flight diverted, the pain subsided on descent and was
tolerable at sea level.
98 1998 34 No No Incapacitation GASTROINTESTINAL The flight diverted after the F/O became ill. It was reported the F/O
had flu-like symptoms, cramps and vomiting, but did not lose
consciousness.
100 1994 59 No No Incapacitation MISCELLANEOUS Inflight, the captain’s performance was poor and inattentive, with
portions of his speech being unrecognizable. The first officer
assumed all flight deck duties but the captain wanted to participate
and to avoid a confrontation the F/O allowed the captain to assist,
but the duties became very difficult for the captain to accomplish.
After landing, the captain assisted in parking the aircraft but was
unable to respond intelligently.
148 1996 48 Yes No Impairment VISION The airplane struck the approach light structure and the end of the
runway deck during the approach. Because of the captain's use of
monovision contact lenses, the airman was unable to overcome the
visual illusions resulting from the approach over water in limited
light. These illusions led the captain to perceive that the airplane
was higher than it was during the visual portion of the approach,
and thus, to his unnecessarily steepening the approach during the
final 10 seconds before impact.

A10
A11
Case Summaries
Shaded sequential cases occurred on the same flight (i.e. cases 16-17, 83-85, and 218-219)
Case Year Age Safety of
Flight
Issue
Deceased Event Category CAMI Narrative
175 1993 33 No No Incapacitation NEUROLOGICAL
(SEIZURE)
First officer experienced LOC in-flight, and experienced feeling
disorientated, presyncopal and "numb all over" for 10 seconds
prior to the LOC. The F/O was witnessed to have a 5-minute
episode of tonic-clonic convulsions, with a postictal state
accompanied by confusion.
178 1994 52 No No Incapacitation LOSS OF
CONSCIOUSNESS
DECOMPRESSION
SICKNESS
During climbout, the crew was unable to pressurize the airplane.
The crew donned oxygen masks and climb was continued to
FL330.
Shortly after level-off, the captain became incapacitated from
decompression sickness. The flight diverted.
218 1993 54 Yes No Impairment FATIGUE Additional factors contributing to the cause were the inadequacy of
the flight and duty time regulations applied to 14 CFR, PART 121,
Supplemental Air Carrier, International Operations, and the
circumstances that resulted in the extended flight/duty hours and
fatigue of the flight crew.
219 1993 50 Yes No Impairment FATIGUE Additional factors contributing to the cause were the inadequacy of
the flight and duty time regulations applied to 14 CFR, PART 121,
Supplemental Air Carrier, International Operations, and the
circumstances that resulted in the extended flight/duty hours and
fatigue of the flight crew.