Understanding Combat Casualty Care Statistics
John B. Holcomb, MD, Lynn G. Stansbury, MD, Howard R. Champion, FRCS, Charles Wade, PhD, and
Ronald F. Bellamy, MD
Maintaining good hospital records
during military conflicts can provide
medical personnel and researchers with
feedback to rapidly adjust treatment
strategies and improve outcomes. But to
convert the resulting raw data into
meaningful conclusions requires clear
terminology and well thought out equa-
tions, utilizing consistent numerators
and denominators. Our objective was to
arrive at terminology and equations that
would produce the best insight into the
effectiveness of care at different stages
of treatment, either pre or post medical
treatment facility care. We first clarified
ously difficult circumstances and involve degrees and contexts
of injury and care unfamiliar to many practitioners, civilian or
military. Further, the ready availability of raw battle casualty
data on the Internet invites misinterpretation by those not famil-
iar with its pitfalls. Much of the potential for such misinterpre-
tation boils down to familiar epidemiologic problems of consis-
tency of numerators and denominators.
The United States Department of Defense (DoD) maintains
two Internet Websites providing information on battle casualties.1,2
The Defense link website has data on return to duty casualties
Operations and Reports (DIOR)2provides information from the
current and past conflicts in sufficient detail for calculation of
proportional mortality (that is, the fraction of an exposed group—
three essential terms: 1) the case fatality
rate (CFR) as percentage of fatalities
among all wounded; 2) killed in action
deaths among all seriously injured (not
returning to duty); and 3) died of
wounds (DOW) as percentage of deaths
following admission to a medical treat-
ment facility among all seriously injured
(not returning to duty). These equations
were then applied consistently across
data from the WWII, Vietnam and the
current Global War on Terrorism. Us-
ing this clear set of definitions we used
the equations to ask two basic questions:
What is the overall lethality of the bat-
tlefield? How effective is combat casu-
alty care? To answer these questions
with current data, the three services
have collaboratively created a joint the-
ater trauma registry (JTTR), cataloging
all the serious injuries, procedures, and
outcomes for the current war. These def-
initions and equations, consistently ap-
plied to the JTTR, will allow meaningful
comparisons and help direct future re-
search and appropriate application of
J Trauma. 2006;60:397–401.
ccurate understanding of the epidemiology and outcome
of battle injury is essential to improving combat casualty
care, but combat trauma data are acquired under notori-
those injured in combat—who die, expressed as a percent), sug-
gesting that battle mortality for injured United States forces has
presented on these three conflicts are comparable. They are not.
However, they do provide a basis for illustrating the major pitfalls
in interpreting military casualty data and their derived statistics.
THE PROBLEM OF DEFINITIONS
Even the term “casualty” must be approached with cau-
tion when reviewing military medical data. “Casualty” in
customary military usage means active duty personnel lost to
the theater of operations for medical reasons.4The term
therefore includes illness and noncombat injuries as well
as combat injuries. For this discussion, we focus on battle
injuries sustained in combat, i.e. during hostile engage-
ment with a military enemy. However, even using this
definition, sub-groups of casualties may be included or
excluded from a given set of summary statistics, depending
on the definitions in use at the time, with important effects
both on the results and the inferences that are made from
these results, when compared with other data sets.5–7
Beebe and DeBakey, in their review of World War II
combat casualties, wrote: “The proportion of deaths among
all men hit is fundamental. . . although perhaps greatest in-
terest attaches to the proportion of the wounded (excluding
those designated as killed) who die of their wounds.”8In this
statement, the authors contrast the overall concept of all men
hit, to three groups: killed, wounded, and, as a sub-set of
Submitted for publication November 18, 2005.
Accepted for publication December 15, 2005.
Copyright © 2006 by Lippincott Williams & Wilkins, Inc.
From the US Army Institute of Surgical Research, Fort Sam Houston,
Texas (L.G.S., J.B.H. C.W.); and Uniformed Services University of the
Health Sciences, Bethesda, Maryland (H.R.C., R.F.B.).
Disclaimer: The opinions or assertions expressed herein are the private views
of the authors and are not to be construed as official or as reflecting the views of the
United States Army or the Department of Defense.
Address for Reprints: COL John B. Holcomb, MD, US Army Institute of
Surgical Research, 3400 Rawley E. Chambers Avenue, Fort Sam Houston, TX
78234; email: John.Holcomb@amedd.army.mil.
The Journal of TRAUMA?Injury, Infection, and Critical Care
Volume 60 • Number 2
Report Documentation Page
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it
does not display a currently valid OMB control number.
1. REPORT DATE
2. REPORT TYPE
3. DATES COVERED
00-00-2005 to 00-00-2005
4. TITLE AND SUBTITLE
Understanding Combat Casualty Care Statistics
5a. CONTRACT NUMBER
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
U.S. Army Institute of Surgical Research (USAISR),3400 Rawley E.
Chambers Avenue,Fort Sam Houston ,TX,78234-6315
8. PERFORMING ORGANIZATION
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)
10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF:
17. LIMITATION OF
19a. NAME OF
c. THIS PAGE
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
wounded, those “who die of their wounds.” In the discussion
of definitions that follows, it will be useful to keep in mind to
which of these groups the modern terms refer.
A key term used to define combat-injured casualties is
the number of wounded in action (WIA) and is the sum of
1. Died of Wounds (DOW, vide infra)
2. Those admitted to a medical treatment facility (MTF)
3. Returned to duty within 72 hours (defines minor
Conventionally, the subgroup of surviving WIAs who return
to duty within 72 hours, the RTD, is excluded from denom-
inators when proportional statistics are presented. This is
significant because this group traditionally represents about
50% of all wounded in action, and in the current conflict
represents 51% of all wounded.
The number and classification of wounded and deaths
from combat is classically used to provide insights into the
lethality of the battle, the effectiveness of the systems of care
and evacuation, and focus attention on required areas of
research. The following definitions, taken from Bellamy,4
standardize the numbers to allow a reasonable retrospective
comparison between conflicts.
Case Fatality Rate (CFR)
CFR refers to the fraction of an exposed group—all those
wounded in action including all those who die (at any level),
expressed as a percent.
CFR ?KIA ? DOW
KIA ? WIA? 100
This summary statistic provides a measure of the overall
lethality of the battlefield in those who receive combat wounds.
It includes the RTDs that are excluded in the denominator of
DOW and killed in action (KIA) rates defined below. However,
this statistic has been used both with and without the RTD
population, creating a major source of confusion when compar-
medical planning for reasons discussed below. The CFR is not a
total mortality rate that would describe all deaths relative to the
entire deployed population at risk.
Killed in Action (KIA)
KIA refers to the number of combat deaths that occur
before reaching an MTF (battalion aid station, forward sur-
gical, combat support and higher levels of hospital care),
expressed as a percent of the Wounded in Action minus the
Deaths before MTF
KIA ? ?WIA ? RTD?? 100
This statistic provides a measure of (1) the lethality of the
weapons (82% of KIAs are near-instant deaths from nonsur-
vivable injuries that result from the massive destructive na-
ture of military weapons); (2) the effectiveness of point-of-
wounding and medic care; and (3) the availability of
evacuation from the tactical setting.
Died of Wounds (DOW)
DOW is the number of all deaths that occur after reach-
ing an MTF, expressed as a percentage of total wounded
minus the RTDs.
%DOW ?Died after reaching MTF
?WIA ? RTD?
This statistic provides a measure of the effectiveness of
the MTF care and perhaps also of the appropriateness of field
triage, initial care, optimal evacuation routes and application
of a coordinated trauma systems approach in mature combat
settings. Deaths that occur at anytime after admission to an
MTF are included in this category.
It is important to note that the above two statistics, %KIA
and %DOW, have different denominators. The latter does not
include deaths before reaching a medical treatment facility
(or those who are dead on arrival at an MTF). This focuses
%DOW as a measure of MTF care. However, both denomi-
nators use the same definition of a battle injury: the first two
subgroups of WIA. The main difference is that the KIAs are
excluded from the DOW calculations. The %KIA and
%DOW cannot be summed to obtain a case fatality rate.
Over the past century, the %KIA has consistently re-
mained between 20 and 25%. The %DOW dropped signifi-
cantly toward the latter half of World War II when improved
evacuation, anesthesia, antibiotics, blood transfusion, and
surgical techniques all coalesced to bring the %DOW to less
than 5%, where it has stayed for the latter half of the 20th
Numerators and Denominators
As noted above, the inclusion or exclusion of the num-
bers of lightly wounded from the denominators of calcula-
tions of proportional mortality can have huge effects. The
Surgeon General’s 1981 revised report on the Vietnam War
(covering 1965–1974), summarized by Bellamy,4shows KIA
for the Army as 27,129, DOW as 3,529, wounded requiring
hospital care but surviving as 96,924 and RTD (using the
definitions current at the time) as 44,858.4Similar casualty
data for the Marines are 11,152; 1,454; 51,399; and 37,234
respectively.5Using the formula shown above, these data
result in a KIA rate of 20.0% when the RTD are excluded
from the denominator. For World War II, it has been less easy
to distinguish the cohort of RTD, but their existence is
recognized.8,9These raw data are summarized in Table 1.
What is included in the numerator can be a source of
confusion. The most striking example of this is World War II.
The overall case fatality rate for World War II using the data
available on the DIOR Website is 30%, however, the case
The Journal of TRAUMA?Injury, Infection, and Critical Care
fatality rate for the individual services in World War II
calculated from the same website is 49% for Navy, 26% for
Army (including both Air Corps and ground troops), and 22%
for Marines. Air Corps mortality for World War II is not
given on the DIOR Website but has been calculated by Beebe
and DeBakey as roughly 66%.8This high rate, like that for
sailors, is clearly associated with the environments (air,
ocean) in which battle is joined, i.e. larger numbers go into
the numerator, and small numbers in the denominator. This
same source shows a case fatality rate for what is variously
described as “infantry” or “ground troops” as ?23%. This is
less than the overall number, 30% used by Gawande,3and is
the more legitimate comparison with mortality for Vietnam,
Iraq, and Afghanistan.
The Problem of Samples
Another critical problem in battle casualty epidemiology
is that of nonrepresentative samples of casualties of variable
sizes being represented as theater-wide experiences. Beebe
and DeBakey, writing about World War II, expressed this
problem very clearly:
“At this writing , as was true throughout the combat
period, one must perforce rely upon a multitude of source-
materials of varying excellence, often without assurance as to
their comparability or even essential accuracy. . . . ”8
For Vietnam, the final compilations appear relatively com-
plete and have been reviewed and revised officially10and
data sets of well-defined samples have been compiled and
analyzed.11,12The Wound Data and Munitions Effectiveness
Team (WDMET) database from the Vietnam War is arguably
the most detailed source of information to date on weapons
and wounding on a sample of approximately 4% of the total
Vietnam casualties between 1965 and 1969. This initiative
provides a model for field data retrieval; five field teams
recorded the most complete, largest and detailed sample of
modern combat injuries. However, early reporting on the
medical consequences of both Vietnam and the current con-
flict have relied on reporting of data from individual medical
units, with little or no outcome data available from the follow
on levels of care and are necessarily skewed. In some of these
essentially anecdotal data sets, the surgeons involved clearly
identify this problem in the course of their reports,13–19but in
others, there is little recognition of its existence.
Time- and unit-specific sampling of casualty and out-
come data, however, planned and identified as such, is
strongly encouraged. At its best, this work provides details,
institutional memory, a scholarly foundation for combat ca-
sualty care, and generates hypotheses that can be tested on
appropriate data sets. Furthermore, clinical outcomes can be
reasonably expected to improve over the course of a conflict
as surgeons and clinical teams trained in noncombat situa-
tions gain experience. This is documented by the steady
decrease in Vietnam DOW rates from 6.1 to 2.4% between
1965 and 1971.10,20,21The time course, cause and dynamics
of such improvements are less apt to be identified in end-of-
conflict summary statistics.
The raw battle casualty data from the current United
States military engagement in Afghanistan and Iraq available
on the DIOR Website as of November 30, 2004, yielded a
case fatality rate of 10%, and the conclusion published in the
New England Journal of Medicine was that mortality had
improved significantly over time.3What is not obvious, how-
ever, is that this analysis used data from Vietnam that ex-
cluded the RTD from the denominator, and data for Afghan-
istan and Iraq that did not. The Defense Link Website, which
provides military casualty data for the current conflict, does
distinguish between the RTD and those more seriously
wounded.1By combining data from both websites, it is pos-
sible to adjust the Afghanistan/Iraq data to more accurately
equate with the denominator provided by the DIOR site for
Vietnam.1,2Table 2 displays the summary data available
Table 1 U.S. Military Combat Casualties, Afghanistan
and Iraq, October, 2001–October 2005, Data from the
Department of Defense1,2,4,5
Admitted & Evacuated‡
KIA, Killed in Action; DOW, Died of Wounds; RTD, Returned to
Duty in 72 hours; WIA, Wounded in Action (WIA ? RTD ? Evacu-
ated ? DOW).
* Does not include air combat wounded.
†Does not include 653 MIA or air combat wounded.
‡Admitted and Evacuated ? Not RTD in 72 hrs.
Table 2 U.S. Military Combat Casualties, Afghanistan and Iraq, October 2001–October 2005, Data from the
Department of Defense1,2
WIA RTD Evacuated DOWKIA
KIA, Killed in Action; DOW, Died of Wounds; RTD, Returned to Duty in 72 hours; WIA, Wounded in Action (WIA ? RTD ? Evacuated ?
DOW); Evacuated, Not RTD in 72 hours.
Combat Casualty Care Statistics
Volume 60 • Number 2
from the two Websites for the major categories of interest for
Iraq and Afghanistan.
Table 3 shows KIA, DOW, and CFR rates for three
conflicts using the most comparable numerator and denomi-
nator figures for each (i.e. ground troops only and the ability
to distinguish RTD) and using the definitions referred to
The case fatality rate (CFR) progressively decreased
over the conflicts WWII ? Vietnam ? Iraq and Afghanistan;
p ? 0.0001 between conflicts). A similar pattern was noted in
%KIA (WWII ? Vietnam ? Iraq and Afghanistan; p ?
0.0001 between conflicts). Understandably a different pattern
is seen for %DOW. There was an increase in %DOW during
the most recent conflict (Iraq and Afghanistan ? WWII ?
Vietnam; p ? 0.004 between conflicts). Interestingly, both
DOW and KIA are higher in Afghanistan than Iraq (p ?
Data were analyzed using SAS version 8.1 (SAS Institute
Inc., Cary, NC). To compare among and between conflicts for
the categorical variable Live/Die Chi-square tests were used.
A Bonferroni adjustment was used for multiple comparisons
and significance level is set at 0.05.
In the present conflict, now entering its fourth year,
case fatality rates (Table 3) for combat injury among
United States military personnel in Afghanistan and Iraq is
indeed roughly half that of Vietnam and one-third that of
World War II, (p ? 0.01). It is not unreasonable to judge
that some of this reduction may be a result of widespread
use of improved body armor, because chest wounds are
relatively decreased in preliminary data when compared
with previous conflicts.20Particularly for the reduction in
%KIA, (p ? 0.001), additional contributing factors may
include the successful transition of products from the 10
year DoD research program on improved hemorrhage con-
trol and increased focus on prehospital Tactical Combat
Casualty Care training,22coupled with rapid evacuation.
Some degree of reciprocity between KIA and DOW rates
is expected6,21as many of the more severly injured casu-
alties who in the past would have died before reaching
MTF care (KIA), now die after rapid evacuation to MTFs,
changing their classification to DOW. The observed in-
crease (p ? 0.01) in DOW rates would likely be higher if
not for the improvements in surgical management utilizing
damage control techniques, improved ICU care, earlier
recognition of abdominal compartment syndrome, liberal
use of fresh whole blood and recombinant factor VIIa
(rFVIIa), among other new techniques, and institution of a
theater-wide trauma systems approach. Interestingly, the
calculation of DOW for Afghanistan reveals a rate of 6.7%
while in Iraq it is 4.7 %, (p ? 0.05), while the KIA rate is
18.7 in Afghanistan and in Iraq it is 13.5% (p ? 0.05).
Only by using common definitions and consistent equa-
tions can these comparative rates be determined. The cause
of the differences between theaters is unclear. Smaller
numbers overall, different application of DOW, KIA and
dead on arrival definitions, wounding at altitude, much
longer evacuation distances, different applications of body
armor and different injury mechanisms are all probably
important variables. However, these and other hypotheses
cannot be tested until wound severity data are compiled in
a fashion that permits appropriate case-control compari-
sons. Taken together, these and other changes in practice
implemented on the current battlefield have resulted in a
statistically and clinically significant decrease in the the-
ater wide, four year CFR compared to previous conflicts,
(p ? 0.001).
In both WWII and Vietnam, of those Soldiers who died,
88% were KIA and 12% DOW.4,23Because of the significant
decrease in the KIA rate in the current war, a greater per-
centage of patients are dying after reaching a MTF. In Iraq
and Afghanistan of those who die, 23% are DOW and 77%
KIA. Though the CFR rate has decreased, the near doubling
of those patients now dying at the MTF’s emphasizes the
need to focus resources and research to aid these casualties.
Thoughtful review of KIA, DOW, and CFR rates for
combat trauma are important for optimal medical planning,
training, research, and resource allocation. The need to bring
combat casualty epidemiology to a civilian standard requires
utilization of both technology and organization that are rou-
tinelyutilized in theUnited
community.24,26Thanks to efforts by the Deputy Assistant
Secretary of Defense for Health Affairs and the Surgeons
General of each of the armed services, raw data appropriate
for this effort are now being collected in three separate
Table 3 Comparison of Proportional Statistics for Battle Casualties, U.S. Military Ground Troops, World War II,
Comparisons between WWII, Vietnam, and Total Iraq/Afghanistan, a,b,c, ? 0.05.
Comparison between Iraq and Afghanistan * p ? 0.05.
% KIA ? 100 ? KIA/(WIA ? RTD) ? KIA; % DOW ? 100 ? DOW/(WIA ? RTD); CFR ? 100 ? (KIA ? DOW)/(WIA ? KIA).
The Journal of TRAUMA?Injury, Infection, and Critical Care
databases developed by the United States Army Center for Download full-text
AMEDD Strategic Studies in conjunction with the United
States Army Institute of Surgical Research, the Armed Forces
Institute of Pathology, and the Navy/Marine Corps Naval
Health Research Center. Standard operational definitions are
in use for the cataloging and analysis of this complex infor-
mation. Injury severity data are recorded, scored, and ana-
lyzed by methods that both meet trauma-community stan-
dards and are appropriate to meet the unique aspects of battle
injuries. If these efforts are successful, the current war will be
the first in history from which detailed concurrent analyses of
the epidemiology, nature, and severity of injuries, care pro-
vided, and patient outcomes can be used to guide research,
training, and resource allocation for improved combat casu-
1. Casualty reports [database on the Internet]: US Department of
Defense. Available from: http://www.defenselink.mil/.
2. Directorate for information operations and reports [database on the
Internet]: Department of Defense. Available from: http://
3.Gawande A. Casualties of war: military care for the wounded from
Iraq and Afghanistan. N Engl J Med. 2004;351:2471–2475.
4. Bellamy RF. Combat Trauma Overview. Textbook of Military
Medicine: Anesthesia and Pre-Operative Care of the Combat
Casualty. Washington, DC: Department of the Army, Office of the
Surgeon General, Borden Institute; 1994:1–42.
5.Bellamy RF. Why is Marine combat mortality less than that of the
Army? Mil Med. 2000;165:362–367.
6. Henderson JV. The importance of operational definitions in design
of a combat casualty information system. J Med Syst. 1983;7:413–
7.Reister FA. Battle Casualties and Medical Statistics: US Army
Experience in Korea. Washington, DC: Department of the Army;
8.Beebe G, DeBakey ME. Battle Casualties: Incidence, Mortality and
Logistic Considerations. Springfield, IL: Charles C Thomas; 1952.
9. Reister FA. Medical Statistics in World War II. Washington, DC:
Department of the Army, Office of the Surgeon General; 1975.
10.Battle injury, active duty Army personnel, Vietnam origin, 1961–
1979. US Army Patient Administration Systems and Biostatistics
Activity. Fort Sam Houston, TX: AMEDD Center and School,
Department of the Army; 28 June 1984 (unpublished data).
11. Wound Data and Munitions Effectiveness Team. The WDMET Study.
Original data from the Uniformed Services University of the Health
Sciences, Bethesda, MD 20814–4799; summary volumes available
from: Defense Documentation Center, Cameron Station, Alexandria
VA 22304–6145; 1970.
Bellamy RF, Zajtchuk R. Assessing the effectiveness of conventional
weapons. In: Bellamy RF, Zajtchuk R, eds. Textbook of Military
Medicine: Conventional Warfare: Ballistic, Blast, and Burn Injuries.
Washington, DC: Department of the Army, Office of the Surgeon
General, Borden Institute; 1991:53–82.
Brass A. Medicine over there. JAMA. 1970;213:1473–1475.
Jaffe G. Army’s New Combat Helmet Might Have Fatal Flaw. Wall
Street Journal, August 27, 2004.
Jones EL, Peters AF, Gasior RM. Early management of battle
casualties in Vietnam. An analysis of 1,011 consecutive cases treated
at a mobile army surgical hospital. Arch Surg. 1968;97:1–15.
Levitsky S, James PM, Anderson RW, Hardaway RM, 3rd. Vascular
trauma in Vietnam battle casualties: an analysis of 55 consecutive
cases. Ann Surg. 1968;168:831–836.
Mabry RL, Holcomb JB, Baker AM, et al. United States Army
Rangers in Somalia: an analysis of combat casualties on an urban
battlefield. J Trauma. 2000;49:515–528.
Chambers LW, Rhee P, Baker BC, et al. Intitial experience of US
Marine Corps forward resuscitative surgical system during Operation
Iraqi Freedom. Arch Surg.2005;140:26–32.
Stevens RA, Bohman HR, Baker BC, Chambers LW. The US
Navy’s forward resuscitative surgery system during Operation Iraqi
Freedom. Mil Med. 2005;170:297–301.
Champion HR, Bellamy RF, Roberts CP, Leppaniemi A. A profile
of combat injury. J Trauma. 2003;54(Suppl):S13–19.
Whelan TJ. Surgery in Vietnam (correspondence). JAMA. 1971;
McSwain M, Frame S, Salomone J. Military Medicine. Prehospital
Advanced Life Support, 5th Edition. St Louis: Mosby; 2003.
Bellamy RF. The causes of death in conventional land warfare:
implications or combat casualty care research. Mil Med. 1984;
Champion HR, Copes WS, Sacco WJ, et al. The Major Trauma
Outcome Study: establishing national norms for trauma care.
J Trauma. 1990;30:1356–1365.
Branas CC, MacKenzie EJ, Williams JC, et al. Access to trauma
centers in the United States. JAMA. 2005;293:2626–2633.
Hoyt DB, Holcomb J, Abraham E, Atkins J, Sopko G. Working
Group on Trauma Research. Working Group on Trauma Research
Program summary report: National Heart Lung Blood Institute
(NHLBI), National Institute of General Medical Sciences (NIGMS),
and National Institute of Neurological Disorders and Stroke
(NINDS) of the National Institutes of Health (NIH), and the
Department of Defense (DOD). J Trauma. 2004;57:410–415.
Combat Casualty Care Statistics
Volume 60 • Number 2