WILDERNESS & ENVIRONMENTAL MEDICINE, xx, xxx (2012)

ORIGINAL RESEARCH

The Epidemiology of Caving Injuries in the United States

Alejandro C. Stella-Watts, MD; Christopher P. Holstege, MD; Jae K. Lee, PhD; Nathan P. Charlton, MD
From the Department of Emergence Medicine, University of Rochester, Rochester, NY (Dr Stella-Watts); and the Departments of Emergency
Medicine (Drs Holstege and Charlton) and Public Health Science (Dr Lee), University of Virginia, Charlottesville, VA.

Objective.—Caving is a demanding sport practiced throughout the world. Currently, there are no
collective data analyzing injury mechanism or type in these austere environments. This study is a
retrospective analysis of caving incidents documented by the National Speleological Society (NSS)—
American Caving Accidents (ACA) annual publication.
Methods.—This study retrospectively analyzes 877 incident reports collected between 1980 and
2008 by NSS-ACA. For each victim, the month, year, location, age, gender, incident type, injury zone
of the body, injury type, the result of the incident, and time intervals for rescue were extracted.
Results.—A total of 1356 victims were identified; 83% of victims were male, 17% were female.
Ages ranged from 2 to 69 years old, with an average of 27 years. The greatest number of events
occurred in summer months, peaking in July. The most common incident leading to traumatic injury
was a caver fall (74%), also contributing to 30% of caver fatalities. Lower extremities were most
commonly injured (29%), followed by the upper extremities and head (21% and 15%, respectively).
Fractures comprised 41% of injuries, followed by lacerations (13%), bruise, hematoma, and abrasions
(12%), and sprains and strains (7%).
Conclusions.—The majority of injuries were not life threatening; however, over the course of 28
years there were 81 documented fatalities. Similar to other studies of wilderness injuries, fractures, soft
tissue injuries, and lacerations were prominent in this study. In general, the overall precipitating event
leading to injuries is falling, leading to orthopedic trauma. To better prepare cave rescue teams we have
attempted to describe the characteristics of caving injuries in the United States.
Key words: caving, cave, injury, rescue, epidemiology

Introduction been mapped.4 Of these, there are a total of 498 miles of

Caving is a demanding activity practiced throughout the
United States and the world.1,2 In the United States alone
there are tens of thousands of known caves, and while the
exact numbers of caving visits are unknown, it is esti-
mated that at least 2 million people in the United States
visit caves each year.3 Although the majority of these
visits would be considered low risk (eg, guided caving
tours), the National Speleological Society has more than
10 000 members who are more likely involved in rigor-
ous caving excursions.2 Many states have specific data
pertaining to existing caves. For example, there are 4378
documented caves in Virginia, with only 1348 that have
Disclaimer: The authors have no conflicts of interest to declare.
No offprints are available from the authors.
Corresponding author: Nathan P. Charlton, MD, University of
Virginia, PO Box 800774, Charlottesville, VA 22903 (e-mail:
npc8a@virginia.edu).
surveyed passage, with the longest being 22.5 miles, and
the deepest cave is 1263 feet.4
Underground exploration has evolved into complex
excursions, going deeper into unexplored terrain. Some
of the most challenging projects involve weeks to
months in complete darkness thousands of meters under-
ground. This environment is dark and wet, requiring
climbing and ropework in muddy and slippery condi-
tions, and at times even underwater submersion.5,6 Be-
cause of the convoluted terrain, electronic communica-
tion is limited.
Although the caving community considers safety a
priority, there is an inherent level of risk involved.2
Rockfall, cold temperatures, and contaminated air are
just a few of the potential hazards.5,6 During the past
decade, there has been increasing research into appropri-
ate rescue equipment and extrication techniques within
cave environments.3,6 Although data are collected on
2 Stella-Watts et al

Table 1. Demographics of caving victims Methods

Demographics Number (%) This study is a retrospective analysis of caving incidents
in the United States from 1980 to 2008 collected and
Gender documented by the National Speleological Society
Male 967 (71.3) (NSS)—American Caving Accidents (ACA) annual pub-
Female 203 (15.0) lication.8 –22 These incident reports are voluntarily pro-
Unknown 186 (13.7) vided to the NSS by groups such as local caving grottos
Age (years) and rescue personnel, such as professional cave rescuers,
0–9 8 (0.06) emergency medical service (EMS) providers, and police.
10–19 258 (19.0)
The NSS requests that reports be submitted in their
20–29 228 (16.8)
standard format available on the NSS website.2 This
30–39 109 (8.0)
40–49 73 (5.4) form mandates date and place of incident, the mechanism
50–59 25 (1.8) of incident (using predefined terminology), and result of
60–69 6 (0.4) incident (broadly defined as fatality, injury, or no con-
⬎70 0 (0) sequence). Other details are optional including age, gen-
Unknown 649 (47.9) der, and injury type.2 These incident reports are compiled
Total victims 1356 and published in an annual NSS-ACA publication.
In this study a single author reviewed all incident
reports listed in each yearly publication. Data were ex-
yearly caving accidents, there are no collective data tracted using predefined variables for type of injury and
available analyzing injury mechanism or type to help body zones based on commonly found wilderness injury
establish appropriate medical protocols in these austere patterns.23 The mechanism of the incident was classified
environments.3,5,7 The purpose of this study is to de- based on the predefined terminology used on the NSS
scribe the demographics, mechanisms, and characteris- incident report form. For each victim, the month, year,
tics of caving injuries occurring within the United States. location, age, gender, incident type, injury zone of the

Figure 1. Monthly distribution of caving incidents.

US Caving Injuries 3

body, injury type, the result of the incident, and time Table 3. Mechanism of incident
intervals from rescue call to victim contact and from
Incident Number (%)
victim contact to cave exit were recorded. Victims were
included if they voluntarily entered a cave, an incident
occurred for which a report was generated, and the incident Caver fall 339 (24)
occurred within the United States. Incidents were excluded Mechanical falla 308 (22)
Equipment failure 19 (1)
if they did not occur within the United States or were the
Anchor failure 12 (1)
result of a victim entering a cave unintentionally.
It is important to note that caving incidents are com- Loss of cave integrity 152 (11)
plex, and often involve multiple factors and multiple Rockfallb 85 (6)
Entrapmentc 45 (3)
injuries. For each victim all injuries were recorded indi-
Buried 22 (2)
vidually; therefore, multiple body zones injured and in-
jury types were often included for the same caver. In Unable to exit cave 765 (54)
addition, multiple factors sometimes led to the inci- Stranded 185 (13)
Lost 178 (13)
dent, so multiple mechanisms of incident could be
Flooding 117 (8)
recorded for the same incident. Traumatic injuries Light failure 107 (8)
were separated into 10 injury zone categories. For Difficulty on rope 81 (6)
each injury zone, the injury was classified into 11 Stuck/wedged in rock 58 (4)
Exhaustion 33 (2)
Equipment problem 6 (⬍1)
Table 2. Yearly distribution of incidents
Cave environment 133 (9)
Year Incidents Cavers requiring rescue Hypothermia 45 (3)
Medical illness during cavingd 30 (2)
Poisoning by toxic gas 23 (2)
1980 24 43 Drowning 16 (1)
1981 23 52 Animal attack 10 (1)
1982 30 51 Illness acquired by cavinge 6 (⬍1)
1983 31 53 Burn (gas lamps) 3 (⬍1)
1984 31 38
1985 11 18 Injured while rescuing 13 (1)
1986 23 31 Harness hang syndrome 3 (⬍1)
1987 43 61
Cannot categorize 18 (1)
1988 38 54
1989 40 59 Total 1423
1990 49 110 a
1991 43 64 Human error; excludes equipment and anchor failure.
b
Injury caused by falling rocks; excludes being buried or entrapped.
1992 50 69 c
Trapped in a location by falling earth but no injury occurred.
1993 47 77 d
Medical condition such as seizure or heart attack occurring while
1994 not available not available caving.
1995 not available not available e
Illness such as infectious disease related to the cave environment.
1996 44 71
1997 35 52
1998 29 43 injury types. The rescue classification was identified
1999 42 67 for each victim to determine whether an incident re-
2000 36 48 sulted in needing rescue. Injury and medical out-
2001 30 40 comes, including the need for transport to medical
2002 30 40 care, were often unavailable.
2003 29 51
2004 20 25
2005 25 30 Results
2006 32 43
2007 23 36
From 1980 to 2008 a total of 877 incident reports were
2008 19 30 identified with a total of 1356 individual cavers involved
Total 877 1356 in these incidents. Results are divided into month of the
Average 32 50 year, demographics, incident type, injury zone of the
body, injury type, rescue classification, and time inter-
4 Stella-Watts et al

Table 4. Mechanism of medical problem victims and young adults (age ⬍20 years) constituted a
large percentage of the victims (37%).
Medical problem Number (%)
Yearly trends revealed an increased number of in-
cidents during summer months, with a peak in July
Traumatic injury 446 (72) (Figure 1). The average number of caving victims was
Animal attack 10 (2)
50 per year, with an average of 32 caving incidents per
Burn (gas lamps) 3 (⬍1)
Injured while rescuing 13 (2) year (Table 2).
New-onset medical conditions
Exhaustion 33 (5)
Hypothermia 45 (7)
MECHANISM OF INCIDENT
Medical illness during cavinga 30 (5)
Poisoning by toxic gas 23 (4) From the 1356 incident victims, there were 1423 identi-
Harness hang syndrome 3 (⬍1) fied inciting factors identified that led to the incident—
Drowning 16 (3) these were termed mechanism of incident and are listed
Illness acquired by cavingb 6 (⬍1)
in Table 3. Some of these incidents led to a caver not
Total 628
being injured but requiring rescue, whereas some led to
a
Medical condition such as seizure or heart attack occurring while illness such as hypothermia or traumatic injury. The most
caving. common incident leading to a rescue effort was being
b
Illness such as infectious disease related to the cave environment. unable to exit cave, such as being stranded or lost, (54%),
which rarely led to medical illness or injury. The second
vals from rescue call to victim contact and from victim most common mechanism of incident was caver fall,
contact to cave exit. which led to 24% of incidents, and was responsible for
74% of traumatic injuries. Only 628 incidents resulted in
a medical problem, listed in Table 4. Illness during
DEMOGRAPHICS AND OVERALL TRENDS
caving, such as heart attack or seizure, was an uncom-
Gender was documented for 1170 (86%) victims; age mon reason for needing rescue, only occurring in 2% of
was documented in 707 victims. Results are listed in total incidents (Table 3) and constituting 5% of all med-
Table 1. Age ranged from 2 to 69 years old. Pediatric ical problems (Table 4).

Table 5. Injury type by injury zone

Injury type
Fracture, Dislocation, Sprain/strain, Laceration, Amputation, Bruise/hematoma,
Injury zone number (%) number (%) number (%) number (%) number (%) number (%)

Head 20 (19) 0 (0) 0 (0) 42 (39) 0 (0) 6 (6)

Spine/back 31 (48) 0 (0) 3 (5) 2 (3) 0 (0) 9 (14)
Upper extremitya 51 (34) 32 (21) 5 (3) 19 (13) 1 (⬍1) 16 (11)
Shoulder to wrist 41 (36) 31 (27) 5 (4) 10 (9) 0 (0) 14 (12)
Hand 10 (28) 1 (3) 0 (0) 9 (25) 1 (3) 2 (6)
Lower extremityb 101 (48) 6 (3) 30 (14) 11 (5) 0 (0) 24 (11)
Hip to lower leg 56 (42) 5 (4) 11 (8) 10 (8) 0 (0) 21 (16)
Foot/ankle 45 (58) 1 (1) 19 (25) 1 (1) 0 (0) 3 (4)
Thorax 26 (53) 0 (0) 1 (2) 1 (2) 0 (0) 8 (16)
Abdomen 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (11)
Pelvis/genitalia 13 (56) 0 (0) 1 (4) 1 (4) 0 (0) 3 (13)
Otherc 2 (2) 0 (0) 0 (0) 0 (0) 0 (0) 4 (4)
Total 244 38 40 76 1 71
Percentaged 41 6 7 13 ⬍1 12
a
Upper extremity is further stratified into 2 categories: 1) shoulder to wrist, 2) hand.
b
Lower extremity is further stratified into 2 categories: 1) hip to lower leg, 2) foot and ankle.
c
Injury zone— other refers to when a caver is reported to suffer an injury such as a fracture or bruise without indicating a body zone.
d
Diagnosis percentage is calculated excluding injuries that did not report the diagnosis.
US Caving Injuries 5

INJURY ZONE AND INJURY TYPE Table 6. Rescue classification

Of the cavers needing rescue, a total of 459 incidents Rescue classification Number (%)
resulting in 720 distinct traumatic injuries were identified
and categorized into injury zones (Table 5). The most Fatality 81 (6)
common injury zone was the lower extremities, followed Injury and rescuea 351 (26)
by the upper extremities and head. Rescue, no injury 764 (56)
Injury type was also recorded for each injury zone. Injury, no rescue 92 (7)
From the data, classification of diagnosis was made for No injury, no rescueb 68 (5)
only 588 of the injuries (Table 5). The 132 injury zones Total 1356
for which the distinct injury was not reported were not a
Rescue refers to a caver requiring rescue assistance from EMS or
included in this calculation. Spine or back was the least cave rescuers.
accurately reported zone, with 16 victims (25%) not b
No injury, no aid refers to scenarios where an incident occurred
having a distinct injury listed. Of the injuries reported, and a caver decided to escape a cave only to realize there were no
fractures were the most common type of injury, making injuries sustained.
up 41% of all injuries (Table 5).
The head deserves specific mention as it is the only minor sprains or abrasions. Incidents in which no injury
injury zone that regularly receives protection from a or no rescue was recorded refers to when a potentially
helmet. Skull and facial fractures are less common, con- dangerous incident occurred, such as flooding or cave
tributing to 19% of head injuries, whereas lacerations collapse, where no one was injured or required external
constitute 39% of head injuries. rescue.

RESCUE CLASSIFICATION RESCUE TIME INTERVALS

All incident victims were assigned to 5 categories to
classify rescue requirement (Table 6). Including fatali-
ties, 88% of incidents required rescue from either other
climbers or rescue services. Some cavers were injured
but were able to exit without aid, such as those with
Although the NSS often did not have access to the time
intervals of rescues, some data were obtained. Only 256
(19%) of incidents reported the time interval between
when a rescue call was made (to either EMS or the
National Cave Rescue Commission) to when rescuers

Table 5. Continued

Injury type
Internal organ Burn, Minor injury not Diagnosis not Total number (including Total percent for
damage, number (%) number (%) specified, number (%) reported, number (%) diagnosis not reported) injury zone

5 (5) 3 (3) 9 (8) 22 (20) 108 15%

0 (0) 0 (0) 3 (5) 16 (25) 64 9%
0 (0) 8 (5) 4 (3) 14 (9) 150 21%
0 (0) 1 (⬍1) 3 (3) 9 (8) 114 16%
0 (0) 7 (19) 1 (3) 5 (14) 36 5%
1 (⬍1) 2 (1) 12 (5) 22 (10) 209 29%
1 (⬍1) 2 (2) 8 (6) 18 (13) 132 18%
0 (0) 0 (0) 4 (6) 4 (5) 77 11%
8 (16) 0 (0) 3 (6) 2 (4) 49 7%
7 (77) 0 (0) 0 (0) 1 (11) 9 1%
0 (0) 0 (0) 3 (13) 2 (9) 23 3%
3 (3) 1 (1) 45 (42) 53 (49) 108 15%
24 14 80 132 720 100%
4 2 14
6 Stella-Watts et al
Table 7. Fatalities
Variable Number (%)
Gender
Male 68 (84)
Female 13 (16)
Mechanism
Caver fall 24 (30)
Drowning 24 (30)
Jammed 15 (18)
Cardiac disease 10 (12)
Hypothermia 5 (6)
Other 3 (3)
Total 81
established contact with the victim. Of those, 113 (44%)
took 0 to 2 hours for rescuers to reach the victim, 143
(56%) took longer than 2 hours, and 13% required longer
than 24 hours. Individual cavers or caving groups who
were able to rescue themselves were not counted, as the
rescuers were already present at the time of the incident.
However, if a victim required aid from cavers of another
group, time was recorded if available.
The rescue duration, defined as the time between when
rescuers established contact with the victim to when the
victim reached the cave entrance, was documented in
444 (33%) of reported incidents. Of those reports, 190
(43%) required 0 to 2 hours for the victim to exit the
cave, whereas 244 (57%) required longer than 2 hours,
and 160 (36%) required longer than 4 hours.
An important time interval we were not able to ex-
trapolate was the time from when the incident occurred
to when a rescue call was made—an important variable
that can in some instances take hours or days, particu-
larly if the caver did not leave a trip plan with someone
on the surface.
FATALITIES
Fatality occurred in 81 (6%) individuals during the study
period (Table 7). The incidence of fatality was 3 deaths
per year. Water-related incidents (eg, drowning and
flooding) and caver falls tied for the highest percentage
of deaths. Medical illness (such as cardiac disease and
hypothermia) resulted in a minority of deaths compared
with accident-related deaths.
Discussion
Before this study, there had been no extensive analysis of
injuries involved in underground recreation and explora-
tion. The leading experts in speleology wrote much of
the previously published work on the medical aspects of
caving incidents; however, most recommendations were
anecdotal. Cave rescue is often a complex and time-
consuming endeavor. The time intervals between when
an incident occurs, a rescue call is made, and when
rescue teams establish contact with a victim can take
hours and even days, requiring numerous rescuers. Not
only do these rescuers need to prepare for injuries they
will encounter and anticipate further complications
throughout a rescue operation, rescuers must also protect
themselves from injuries. Understanding the etiology of
underground injuries and outcomes is of utmost impor-
tance when executing a cave rescue operation.
In this study the incidence of caving incidents was
reported as 32 per year, accounting for an average of 50
victims per year. This likely underestimates the true
number of incidents as this number takes into account
only those reported to the NSS. We would speculate that
incidents resulting in no injury or only minor injuries and
those requiring no rescue or rescue only from other
cavers would be those that are most likely underreported.
It was found that an increased number of incidents oc-
curred during summer months, with a peak in July. This
is likely related to the increased incidence of caving
excursions during the summer months.
Most injuries were not life threatening. However,
some fatalities did occur. Although the focus of this
study was not primarily on fatalities, caver fall, drown-
ing, becoming physically jammed, and cardiac disease
were common causes of fatal outcomes. With the excep-
tion of becoming jammed, these etiologies are consistent
with other studies on the epidemiology of mortality in
the wilderness.24,25 The total number of caving excur-
sions and total caving participants per year are unknown.
Because of this, the rate of total caving incidents per year
based on those reported to the NSS is likely an under-
estimate. However, because caving fatalities are often
better publicized and better documented, our calculated
rate of 3 deaths per year may be more accurate.
Similar to other epidemiologic studies of wilderness
injuries, fractures, soft tissue injuries, and lacerations
were prevalent in this study.23 Fractures were the most
common injury requiring rescue and were the most com-
mon injury for almost all injury zones, contributing to
48% of spine and back injuries, 48% of lower extremity
injuries, and 56% of pelvis and genitalia injuries. Lower
extremity, spine, and pelvis injuries are medically im-
portant as these victims are less likely to be able to
perform self-rescue and may require significant rescue
resources. Upper extremity injuries also constituted a
large number of injuries, and although in some situations
these victims may be more likely to self-rescue, difficult
terrain will pose a challenge to any victim with a signif-
US Caving Injuries
icant injury. Although generally not immediately life
threatening, these fractures can greatly impede rescue
and increase victim exposure to the environment. EMS
and rescue providers should be prepared with adequate
knowledge and supplies to splint and bandage the injured
victim and should be prepared for a completely depen-
dent victim requiring extrication. In addition, rescuers
should be able to perform spinal immobilization as head
and spine injury were common in this study.
Cavers can become trapped in the cave by injury,
illness, and many types of mechanical incidents. Of the
incidents reported, the majority of victims required some
form of aid to leave the cave. Because of the arduous
nature of caving, this is expected as any impairment in
performance or gear may lead to an insurmountable
obstacle. Any incident resulting in a caver being unable
to exit the cave will result in increased exposure to the
environment. With many caves harboring a cold and wet
environment, rescuers should be knowledgeable of the
signs and symptoms of hypothermia, and be prepared to
remove wet clothes and provide adequate warmth during
the rescue operation.
Burial occurred in 22 victims during this study. Rock-
fall and burial poses the potential problem of crush
injury, which, given the duration of most rescues, could
lead to rhabdomyolysis, crush syndrome, and rapid car-
diovascular collapse once the victim is freed. If avail-
able, rescuers should consider early administration of
intravenous fluids if crush syndrome or rhabdomyolysis
is of concern.26,27 Harness hang syndrome or suspension
trauma (cardiovascular shock induced by passively hang-
ing in a harness) could potentially result in similar pa-
thology; early intravenous fluids have recently been ad-
vocated for this syndrome as well.28 In this study
suspension trauma was rare with only 3 reported cases.
LIMITATIONS
There are a number of limitations that are present with
this analysis. Incident reports were often generated from
nonmedical personnel (eg, volunteer rescue groups, po-
lice, and fire fighters) and, therefore, the injury reported
in the incident report may not accurately reflect the final
medical diagnosis. As the reports are voluntary they
likely underreport the true incidence of caving accidents
and injuries. Gaps, including demographics and injury
descriptions, often existed in the reported data. In addi-
tion, we were not able to obtain data for the years 1994
and 1995.
The NSS-ACA report form requires that mechanism
of incident be recorded to submit an incident report;
however, the mechanism of incident does not necessarily
identify the mechanism of the injury. Therefore, the
7
mechanism of the injury could at times be misleading.
For example, equipment failure while on a rope can
cause a caver to fall and fracture a lower extremity; the
mechanism of incident for the fractured leg would then
be equipment failure, although the mechanism leading to
the injury was falling. As only 1 author recorded the
data, his interpretation of reported mechanism and injury
pattern in some instances may have resulted in misclas-
sification of the data.
This report does not include data from outside the
United States; therefore, injuries incurred by use of
equipment or techniques more common outside of the
United States may be missed. Additionally, large caving
expeditions are more common outside the United States;
therefore, these data may not include injuries more com-
mon on large expeditions. In this study it was not pos-
sible to identify the level of caver experience; therefore,
it was not possible to separate novice caver injuries from
those of experienced cavers. Novice cavers may have
different mechanisms of injury and injury patterns from
those of experienced cavers or those on caving expedi-
tions.
Conclusions
Underground exploration is a difficult undertaking re-
quiring experience, resources, and a high level of auton-
omy. When injury occurs in these austere environments,
rescue is difficult, requiring lengthy rescue times and
extensive resources. Cavers themselves are the key to
proper rescue. They must be well prepared, and make
smart decisions to orchestrate and initiate rescue efforts.
The NSS’s extensive and well-organized network of
grottos throughout every region has allowed them to
maintain documentation of underground injuries in the
United States for 3 decades. It is our intention that with
this initial evaluation and with continued reevaluation,
data collection, and medical knowledge of caving inju-
ries can be enhanced to provide greater benefit to rescu-
ers and cavers in need.
Acknowledgements
The authors are grateful for the help and expertise of Ray
Keeler, Editor of American Caving Accidents, and the
National Speleological Society.
References
1. International Union of Speleology. National and World
Region Speleological Organisations. Available at: http://
www.uis-speleo.org/people/natspel.html. Accessed De-
cember 23, 2011.
8 Stella-Watts et al
2. National Speleological Society. Available at: http://
www.caves.org/. Accessed September 1, 2011.
3. Hooker K, Shalit M. Subterranean medicine: an inquiry
into underground medical treatment protocols in cave res-
cue situations in national parks in the United States. Wil-
derness Environ Med. 2000;11:17–20.
4. The Virginia Speleological Survey. Available at http://
www.virginiacaves.org. Accessed September 1, 2011.
5. Hempel JC, Fregeau-Conover A, eds.On Call: A Complete
Reference for Cave Rescue. 2001 edition. Huntsville, AL:
The National Speleological Society Inc; 2001.
6. Hudson SE, McCurley LH. Caving and cave rescue. In:
Auerbach PS, ed. Wilderness Medicine: Management of
Wilderness and Environmental Emergencies. 5th ed. St
Louis, MO: Mosby Elsevier; 2007.
7. Ashford DA, Knutson RS, Sacks JJ. Injury among cavers:
results of a preliminary national survey. J Sports Med Phys
Fitness. 1999;39:71–73.
8. Knutson S. American caving accidents 1980 –1981. Natl
Speleol Soc News. 1983;41:99 –118.
9. Knutson S. American caving accidents 1982. Natl Speleol
Soc News. 1983;41:255–265.
10. Knutson S. American caving accidents 1983. Natl Speleol
Soc News. 1985;42:347–357.
11. Knutson S. American caving accidents 1984. Natl Speleol
Soc News. 1985;43:355–376.
12. Knutson S. American caving accidents 1985. Natl Speleol
Soc News. 1986 44:399 – 409.
13. Knutson S. American caving accidents 1986. Natl Speleol
Soc News. 1987;45:386 –398.
14. Knutson S. American caving accidents 1987. Natl Speleol
Soc News. 1988;46:475– 490.
15. Knutson S. American caving accidents 1988. Natl Speleol
Soc News. 1989;47:314 –334.
16. Knutson S. American caving accidents 1989. Natl Speleol
Soc News. 1990;48:331–349.
17. Knutson S. American caving accidents 1990. Natl Speleol
Soc News. 1990;48:330 –357.
18. Knutson S. American caving accidents 1992. Natl Speleol
Soc News. 1993;51:346 –370.
19. Putnam WO. American caving accidents 1996 –1998. Natl
Speleol Soc News. 2000;58:1–57.
20. Putnam WO. American caving accidents 2002–2003. Natl
Speleol Soc News. 2005;63:1– 48.
21. Putnam WO. American caving accidents 2004 –2005. Natl
Speleol Soc News. 2007;65:1– 45.
22. Keeler R. American caving accidents 2007–2008. Natl
Speleol Soc News. 2010;68:1– 44.
23. Forrester JD, Holstege CP. Injury and illness encountered
in Shenandoah National Park. Wilderness Environ Med.
2009;20:318 –326.
24. Goodman T, Iserson KV, Strich H. Wilderness mortalities:
a 13-year experience. Ann Emerg Med. 2001;37:279 –283.
25. Montalvo R, Wingard DL, Bracker M, Davidson TM.
Morbidity and mortality in the wilderness. West J Med.
1998;168:248 –254.
26. Better OS, Abassi ZA. Early fluid resuscitation in pa-
tients with rhabdomyolysis. Nat Rev Nephrol. 2011;7:
416 – 422.
27. Haojun F, Jianqi S, Shike H. Retrospective, analytical
study of field first aid following the Wenchean earthquake
in China. Prehosp Disaster Med. 2011;26:130 –134.
28. Mortimer RB. Risks and management of prolonged sus-
pension in an alpine harness. Wilderness Environ Med.
2011;22:77– 86.