TY - JOUR
AU1 - USN, Mark Hammett, MC
AU2 - USN, Lisa Pearse, MC
AU3 - USN, Neal Naito, MC
AU4 - PhD, Dorraine Watts,
AU5 - MPH, Tomoko Hooper, MD
AB - ABSTRACT Objective: We examined common factors in vehicular drowning deaths that might lead to changes in equipment or training. Methods: Drowning deaths among service members deployed to Operations Iraqi Freedom or Enduring Freedom, 2003 to 2005, were ascertained using the Armed Forces Medical Examiner Tracking System database. Cases were linked to Army and Navy safety and investigative files. Results: Fifty-two cases of vehicular drowning deaths were identified. These occurred mostly at night, were almost always the result of a rollover, and were most frequently associated with a high-mobility medium-weight vehicle. Seat belts were rarely worn, but the majority of injuries should not have been severe enough to affect egress from the vehicle. These drowning deaths seldom occurred while engaged with the enemy and were rarely associated with bad road conditions or bad weather. Conclusions: Effective preventive strategies might focus on training and equipment to reduce rollover events and on the expeditious extrication of victims. Introduction Drowning is a relatively infrequent cause of death during combat operations. In the 15 years of hostilities in Vietnam, there were 1,207 documented drowning or suffocation deaths, which represented 2.1% of the total number killed during that war.1 Preliminary data from the Armed Forces Institute of Pathology (AFIP) indicated that 71 documented cases of drowning occurred among U.S. service members deployed in support of Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF) between 2003 and 2005. This represents 3% of the total number of deaths during that time period2,3 and is quite noteworthy since these countries are characterized as desert nations. The large majority of these deaths were the result of motor vehicle immersion (in roadside canals or ditches) and the proportion of fatalities resulting from vehicular crashes occurring in Iraq and Afghanistan is much higher than has been reported in civilian settings.4 There is very little in the published literature regarding the epidemiology of drowning fatalities related to motor vehicle crashes. A study of vehicular drowning deaths during Hurricane Floyd in 1999 reported a 73% fatality rate when total vehicle submersion occurred.5 No drowning deaths occurred when the vehicle was not completely submerged. No descriptions of safety equipment in use or nature or severity of injuries were included in this case-control study.5 A case series published in 1990 characterized deaths resulting from motor vehicle immersion occurring in Sacramento, California between 1974 and 1985.6 One of the important variables explored in this study was injury severity based on the autopsy report. The authors graded the amount of trauma that the victims sustained in the accident and concluded that almost all of the drowning victims sustained very little or no traumatic injury.6 There is no available literature on drowning fatalities related to motor vehicle crashes in the combat environment. Given the numbers of potentially preventable deaths due to motor vehicle immersion during the current conflict in Iraq and Afghanistan, we conducted an investigation of the epidemiological factors associated with this type of deployment-related death. Our primary objective was to characterize common factors in these drowning deaths that might lend themselves to changes in equipment, training, or safety procedures to decrease these tragic events in the future. Methods We conducted a descriptive epidemiological study of all drowning deaths occurring in OIF and OEF during the period 2003 to 2005. All active duty service member deaths are captured in the Department of Defense Medical Mortality Registry maintained at the AFIP, Office of the Medical Examiner. All deaths occurring in theater undergo autopsy by the Office of the Armed Forces Medical Examiner. Cases were identified by a keyword search of the Armed Forces Medical Examiner Tracking System database. A search was performed using the word “drowning” as the cause of death and yielded 71 cases of drowning during the time period 2003 to 2005. Although hostilities commenced in Afghanistan in 2001, there were no drowning deaths recorded in the database until 2003. Fifty-two of these cases were associated with motor vehicle accidents. The primary investigator (M. Hammett) reviewed the postmortem examination report for each case and cross-referenced them with mishap reports at the Army Combat Readiness Center (Fort Rucker, Alabama), the Navy Safety Center (Norfolk Naval Station, Virginia), and the First Marine Division Safety Office (Camp Pendleton, California), as well as files maintained by the Army Criminal Investigation Division (Washington, DC) and the Navy Criminal Investigative Service (Washington, DC) to examine the circumstances associated with the deaths. Four deaths (three separate accidents) did not have a mishap report performed on the case and some data variables in some of the cases were not present in any of the report sources. Missing data are excluded from the statistical analyses reported in this study. The key variables used to characterize the drowning deaths due to a motor vehicle crash in a combat zone included the following: age, location in the vehicle, seat belt use, vehicle type, whether a rollover had occurred, time of day, road and weather conditions, number of passengers in the vehicle, whether there were survivors, and the length of time the victim was in the water before being extracted. Additional variables of interest unique to the combat environment were whether the crash occurred during enemy engagement and whether night vision devices were in use. In addition, we assessed whether the victim might have sustained an incapacitating injury, which might have prevented egress from the vehicle following the crash. Two different methods were employed to categorize potentially incapacitating injuries. Both methods are based on the Abbreviated Injury Scale and Injury Severity Score (ISS) system used by investigators in the Sacramento, California study.6 In this study, trauma victims below the age of 49 with an ISS score of 5 or less are reported to have a 0% mortality rate.7,8 For our study, the primary investigator (M. Hammett) reviewed autopsy reports to grade traumatic injuries using the Abbreviated Injury Scale and ISS system. An ISS >5 was set as the threshold for categorizing injuries as severe enough to have potentially been a contributing factor in the drowning death. The second method used to impute incapacitating injury was based on a more liberal definition of any type of postmortem documentation of bruising about the head, as evidence that the victim may have been stunned or semiconscious as a result of the crash. One case that did not meet either definition of incapacitating injury was one in which the mishap report noted that the victim's hand was trapped under the vehicle, which physically prevented him from getting out of the water. The postmortem examination confirmed such an injury to the serviceman's hand and this case was grouped with other incapacitating injuries. Descriptive statistics were calculated, and tests of association were performed using SPSS version 12.0 (Chicago, Illinois). We assessed the association between time of day and having crash survivors. We used a two-tailed Fisher's exact test and calculated confidence intervals (CIs) at the 95% level. The study protocol was reviewed and approved by the Institutional Review Boards at the AFIP and the Uniformed Services University of the Health Sciences. Results Fifty-two drowning deaths involving motor vehicle immersion were identified for our study. The 19 excluded drowning cases involved the following: aircraft accidents, recreational swimming, falling from a boat during riverine operations, traversing waterways during combat operations, or rescue attempts (three drownings during attempted rescue following vehicular crashes that are part of this study). The 52 deaths in our case series involved 29 separate events. Two deaths from one accident occurred in Afghanistan and the remainder of the deaths occurred in Iraq. The type of vehicle involved in the majority of the deaths (31 of 51 (61%)) was the high-mobility medium-weight vehicle (HMMWV), which was not an unexpected finding since the HMMWV is the most frequently used vehicle in both theaters (Fig. 1). From the available data, we could not address whether HMMWVs were involved in a greater than expected proportion of motor vehicle crashes. The remainder of the deaths were associated with heavy armored/tracked vehicles, a bulldozer, and a 7-ton truck (Table I). Rollover accidents accounted for 89.6% of the deaths (n = 48) (Fig. 1). TABLE I ACCIDENT EVENTS AND NUMBERS OF DROWNING DEATHS DURING OIF AND OEF, 2003–2005, BY VEHICLE TYPE   Vehicle Type  Deaths  Valid Percent  Accident Events  Valid Percent  Valid  HMWWV  31     60.8  18    64.2    Heavy armor vehicles  18     35.2    8    28.6    Bulldozer    1      2.0    1     3.6    7-ton truck    1      2.0    1     3.6    Total  51  100.0  28  100.0  Missing  System    1      1    Total    52    29      Vehicle Type  Deaths  Valid Percent  Accident Events  Valid Percent  Valid  HMWWV  31     60.8  18    64.2    Heavy armor vehicles  18     35.2    8    28.6    Bulldozer    1      2.0    1     3.6    7-ton truck    1      2.0    1     3.6    Total  51  100.0  28  100.0  Missing  System    1      1    Total    52    29    View Large TABLE I ACCIDENT EVENTS AND NUMBERS OF DROWNING DEATHS DURING OIF AND OEF, 2003–2005, BY VEHICLE TYPE   Vehicle Type  Deaths  Valid Percent  Accident Events  Valid Percent  Valid  HMWWV  31     60.8  18    64.2    Heavy armor vehicles  18     35.2    8    28.6    Bulldozer    1      2.0    1     3.6    7-ton truck    1      2.0    1     3.6    Total  51  100.0  28  100.0  Missing  System    1      1    Total    52    29      Vehicle Type  Deaths  Valid Percent  Accident Events  Valid Percent  Valid  HMWWV  31     60.8  18    64.2    Heavy armor vehicles  18     35.2    8    28.6    Bulldozer    1      2.0    1     3.6    7-ton truck    1      2.0    1     3.6    Total  51  100.0  28  100.0  Missing  System    1      1    Total    52    29    View Large Fig. 1 View largeDownload slide The scene of a fatal vehicular drowning accident that depicts many of the typical features found in the study. Pictured is a HMMWV that is inverted in the shallow water of a roadside canal. Fig. 1 View largeDownload slide The scene of a fatal vehicular drowning accident that depicts many of the typical features found in the study. Pictured is a HMMWV that is inverted in the shallow water of a roadside canal. The age of the fatalities ranged from 19 to 40 years; the median age (24 years) reflects the demographics of those who are deployed in support of combat operations. Thirteen events (44.8% of 28) resulted in multiple fatalities; 8 events resulted in no survivors (30.8% of 27). In the incidents where the total number of passengers was known (n = 25), 60% involved either three or four passengers. The driver or front passenger (n = 47) accounted for 56.8% of the fatalities. Seat belts were not used by 95.8% of the fatalities (n = 48). Among the 46 fatalities in which the time of the accident was known, 33 (72%) occurred between dusk and dawn. When vehicle immersion occurred during darkness, there were no survivors in 35% of events. However, the difference in survivability between accidents occurring in the dark versus during daylight, although suggesting reduced crash survivability in darkness (risk ratio = 0.74; 95% CI: 0.49–1.13), did not reach statistical significance (Table II). Night vision devices were in use in only 7 of 26 events (26.9%). The amount of time that victims spent underwater before rescue was known in 25 of the 52 fatalities and ranged between 5 and 180 minutes for these casualties. TABLE II ACCIDENT EVENTS AND CRASH SURVIVABILITY BY LIGHT CONDITION, OIF AND OEF, 2003–2005   Survivors: Yes  Survivors: No  Total  Darkness  11  6  17  Daylight    7  1    8  Total  18  7  25    Survivors: Yes  Survivors: No  Total  Darkness  11  6  17  Daylight    7  1    8  Total  18  7  25  Risk ratio = 0.74 (95% CI: 0.49–1.13). View Large TABLE II ACCIDENT EVENTS AND CRASH SURVIVABILITY BY LIGHT CONDITION, OIF AND OEF, 2003–2005   Survivors: Yes  Survivors: No  Total  Darkness  11  6  17  Daylight    7  1    8  Total  18  7  25    Survivors: Yes  Survivors: No  Total  Darkness  11  6  17  Daylight    7  1    8  Total  18  7  25  Risk ratio = 0.74 (95% CI: 0.49–1.13). View Large Using the ISS score of 5 as a threshold for categorizing injuries as incapacitating, only 9 service members (17.3%) suffered injury severe enough to potentially preclude them from exiting the vehicle before drowning. Using the more liberal method of categorization for incapacitating injuries (evidence of some degree of trauma to the head), 27 of 52 drowning victims (52%) had no injuries that would seem to have precluded them from being able to exit the vehicle. Finally, poor weather or bad road conditions did not seem to be a contributing factor in the motor vehicle crashes. Clear or cloudy skies were noted in 93%, and dry roads were documented in 63%, of the cases in which complete information was available (n = 46). Discussion Although our findings should be considered preliminary, it is noteworthy that most of the victims did not appear to have sustained injuries severe enough (based on ISS score of 5) to interfere with egress from the vehicle following the crash. However, autopsy reports did provide evidence of some degree of trauma to the head in 48% of the cases. Whether the degree of head trauma was sufficient to affect vehicle egress is not clear. Another important finding from our analysis is that there were survivors in motor vehicle crashes that resulted in fatalities. It is important to further investigate the circumstances for survivability in these events. We found no statistically significant association between survivability and whether the crash occurred during daylight or under conditions of darkness. The Department of Defense has already taken steps to mitigate the mortality rate due to motor vehicle crashes in theater by instituting training and equipment changes, particularly in the HMMWV (D. Wren, unpublished data, Daily PM TV “HMMWV Enhancements” Status, U.S. Army Combat Readiness Center, Fort Rucker, Alabama). Modifications to the HMMWV include the following: improved restraint system (retrofitting beginning February 2006); improved gunner restraint system (retrofitting beginning December 2005); changes to operating procedures when on a road near water; water egress procedure training; and battle lock modification (both single movement combat lock system inside the vehicle and the battle lock opening wrench located on the vehicle exterior for rescuers to utilize in the event of emergency). The goal of these changes is to promote primary and secondary crash prevention mechanisms to reduce motor vehicle-associated injury and death, including drowning due to vehicle immersion. As a primary prevention measure, the Army Combat Readiness Center is studying the use of an inclinometer in vehicles, to increase driver awareness regarding the vehicle's roll angle along dangerous stretches of road to prevent a rollover accident (D. Wren, unpublished data). In our study, 96% of the fatalities occurred among those not wearing seat belts. Improvement in the HMMWV seat restraint system involves changing to a true three-point restraint system. This system is designed to allow for more movement during routine operations, added comfort and compatibility with the loaded flak vest worn in combat, easy removal, and effective crash restraint. If the new system is successfully implemented and well accepted by combat personnel, then the increased seat belt use may potentially reduce the percentage of incapacitating injuries in the event of a crash. Changing and implementing training procedures to make water egress a drilled response is a means of secondary prevention to make vehicle occupants aware of the hazard and to train them in proper egress procedures when a vehicle becomes submerged. Egress from a HMMWV that is upside down in the water is a difficult problem to overcome. Time is of the essence, and the first time to think about how to escape from a vehicle after a crash should not be when it is upside down and underwater. Those few seconds of improved reaction time may be all that is required to get a door open before it becomes inoperable. This is the same stance taken by the Navy and Marine Corps when they instituted a training program for helicopter egress using the “Helo-dunker.” This training was to make service personnel aware of the risks of helicopter crashes over water and to give them some survival skills in the event of such a crash. The Army has a basic procedure for water egress in the HMMWV and the Stryker vehicle.9 There is also a training video that covers vehicle rollover procedures in the HMMWV.10,11 Like the “Helo-dunker” training, it serves to familiarize personnel with the hazards that they face, and there is a large body of recent research that confirms the fact that people who visualize a task and practice the task do much better than those who do not.12,–14 The difference between successfully exiting a sinking vehicle and being trapped inside to drown might only be a few seconds, and egress training under conditions mimicking reality may help prevent a tragic outcome. In February 2005, the Deputy Secretary of Defense's Military Assistant asked, “Can we prevent vehicle occupant drowning by using Helicopter Emergency Egress Devices (HEED) as an auxiliary air supply to give occupants more time to escape from the vehicle?” The Army and Marine Corps responded to this inquiry, considering expense of procurement, maintenance, and training as well as the risk of mortality due to vehicle immersion. Alternate recommendations were presented, but a final decision has not yet been made on the use of the HEED (J. Seibert, unpublished data). The goal of this case series was to characterize common factors to generate testable hypotheses and to inform potential prevention strategies; some issues emerged that are worth further investigation. Clearly, some of the occupants involved in motor vehicle crashes escaped from the immersed vehicles. Mishap investigations have not focused on how survivors successfully exited the vehicles. A comprehensive study of the factors associated with survivability in this type of motor vehicle crash might be helpful in identifying prevention strategies. Another factor that emerges from the mishap reports is that these events are often witnessed. In some cases, witnesses become casualties in their heroic efforts to save their comrades. Our study showed that the majority of victims are under the water for over 5 minutes, even when rescuers are on the scene very close to the time of the vehicular mishap. Evidence contained in the majority of the mishap reports suggests that a victim's attempt to egress the vehicle may be the most difficult problem to overcome. This finding may lend itself to enhanced training on exit strategies following a vehicle crash and water immersion, as well as strategies for first responders. Training first responders on how to quickly extricate victims from a combat vehicle, particularly a HMWWV, might prove to be the most effective method of saving lives. Some of the standard tools aboard convoy vehicles might be used in the effort, such as a jack handle. The cost and benefit of adding a few simple tools to the standard equipment in a combat vehicle, to facilitate rescue attempts, needs to be assessed. Perhaps a re-examination of the use of the HEED system (or its replacement) might also be helpful in reducing vehicular drowning deaths. If a single HEED bottle were part of the standard equipment in each vehicle, rescuers in a convoy might have those critical extra minutes to help trapped passengers exit an immersed vehicle. The distribution, maintenance, and training costs would have to be weighed against the potential benefits of such a program. Finally, it was noted that several rescuers died in the act of trying to save their comrades. These tragedies may have been prevented by simple safety precautions for rescuers, such as having a rope tied around them. This is another issue that might be considered as part of extrication training in the future. Summary A high proportion of the drowning deaths in OEF/OIF have involved vehicle immersion. These events occurred mostly at night, the majority involving a HMMWV, and were almost always the result of a rollover. Seatbelts were rarely used; most injuries were apparently not severe enough to preclude egress from the vehicle; and events seldom occurred during enemy engagement. Equipment and training changes may serve to make survival more likely in the future. The Army has already implemented some changes to mitigate vehicular injuries and deaths. A subsequent study, using data from this analysis as a baseline, may help to establish the effectiveness of these changes. We recommend that an assessment of factors related to survivability in vehicular mishaps involving water immersion receive more emphasis and be included in mishap reports. Additionally, egress training, egress drills, and preconvoy briefings that include water hazard awareness and rollover prevention procedures may increase survivability in the event of a crash. Extrication training and drills for rescuers, to include rescuer safety precautions, as well as the HEED system, might also be effective means of reducing drowning deaths. Finally, supplementing standard equipment for convoy operations, to include rope and simple tools to pry open jammed doors, may represent a cost-effective method to not only improve crash survivability for those trapped in vehicles, but also to reduce the risk of rescuers becoming victims. Safety and training experts could consider all these strategies to potentially save the lives of American military personnel serving in combat operations. Acknowledgments We thank the following individuals for technical assistance or facilitating data access: CAPT Craig Mallak at the AFIP, Mr. Don Wren, Mr. Alan Longshore, Mr. Joseph McFadden at Army Combat Readiness Center; CDR Edward Hobbs, Mr. Gene Feierstein at Naval Safety Center; Mr. Paul O'Jala at 1st Marine Division Headquarters; Special Agent Jeanmarie Sentell at Naval Criminal Investigation Service; and Mrs. Cara Olsen at Uniformed Services University Biostatistical Consulting Center. References 1. Hull T Statistical information about casualties of the Vietnam conflict, 2005.  Available at http://archives.gov/research/vietnam-war/casualty-statistics.html; accessed March 28, 2006. 2. Center DMD Global War on Terrorism-Operation Iraqi Freedom casualties by month , 2006. Available at http://siadapp.dior.whs.mil/personnel/CASUALTY/OIF-Total-by-month.pdf; accessed June 12, 2006. 3. Center DMD Global War on Terrorism-Operation Enduring Freedom casualties by month , 2006. Available at http://siadapp.dior.whs.mil/personnel/CASUALTY/oefmonth.pdf; accessed June 12, 2006. 4. Smith G, Langley J Drowning surveillance: how well do E. codes identify submersion fatalities. Inj Prev  1998; 4: 135– 98. Google Scholar CrossRef Search ADS PubMed  5. Yale JD, Cole TB, Garrison HG, Runyan CW, Ruback JK Motor vehicle-related drowning deaths associated with inland flooding after hurricane Floyd: a field investigation. Traffic Inj Prev  2003; 4: 279– 84. Google Scholar CrossRef Search ADS PubMed  6. Wintemute GJ, Kraus JF, Teret SP, Wright MA Death resulting from motor vehicle immersions: the nature of the injuries, personal and environmental contributing factors, and potential interventions. Am J Public Health  1990; 80: 1068– 70. Google Scholar CrossRef Search ADS PubMed  7. Baker SP, O'Neill B, Haddon W, Long WB The Injury Severity Score: a method for describing patients with multiple injuries and evaluating emergency care. J Trauma  1974; 14: 187– 96. Google Scholar CrossRef Search ADS PubMed  8. Copes WS, Champion HR, Bain LW Progress in characterising anatomic injury. J Trauma  1990; 30: 1200– 7. Google Scholar CrossRef Search ADS PubMed  9. Army Combat Readiness Center, Fort Rucker, AL: Water Egress HMMWV Uparmored Rollover Task and Performance Measures. VolGTA 55-03-031 , 2005. 10. Army Combat Readiness Center, Fort Rucker, AL: Letters from War: Tactical Movements [DVD].  2005. 11. Army Combat Readiness Center, Fort Rucker, AL: Letters from War: Uparmored HMMWV Rollovers [DVD]. 2006.  12. Brouziyne M, Molinaro C Mental imagery combined with physical practice of approach shots for golf beginners. Percept Mot Skills  2005; 101: 203– 11. Google Scholar CrossRef Search ADS PubMed  13. Hayashi S, Shimura K, Kasai T Rapid plastic changes of human primary motor cortex with repetitive motor practice and transcranial magnetic stimulation. Percept Mot Skills  2005; 101: 575– 86. Google Scholar CrossRef Search ADS PubMed  14. Meiran N, Daichman A Advance task preparation reduces task error rate in the cuing task-switching paradigm. Mem Cognit  2005; 33: 1272– 88. Google Scholar CrossRef Search ADS PubMed  Reprint & Copyright © Association of Military Surgeons of the U.S.
TI - Drowning Deaths of U.S. Service Personnel Associated with Motor Vehicle Accidents Occurring in Operation Iraqi Freedom and Operation Enduring Freedom, 2003–2005
JF - Military Medicine
DO - 10.7205/MILMED.172.8.875
DA - 2007-08-01
UR - https://www.deepdyve.com/lp/oxford-university-press/drowning-deaths-of-u-s-service-personnel-associated-with-motor-vehicle-mYpymE9hvx
SP - 875
EP - 878
VL - 172
IS - 8
DP - DeepDyve
ER -