An Analysis of Pediatric By Dennis Vane,
Frederick G. Shedd,
Trauma Deaths in Indiana
Jay L. Grosfeld,
Karen W. West,
Indianapolis, 6 From June 1986 to May 1968. there were 1.931 childhood deaths recorded in Indiana. Eight hundred six children (0 to 18 years old) died as a result of trauma (41.4% of all deaths). Seventy percent of all traumatic deaths occurred in boys. Blunt trauma accounted for 54% of deaths, asphyxia or drowning 26%. penetrating trauma 15%. electrocution 3%. and burns 1%. Sixty percent of deaths occurred in rural areas and 40% occurred in urban centers: however, state-wide demographics define the population as 70% urban. The percentage of deaths due to trauma within a given race was: hispanic 71%, Caucasian 42%. black 35%. and others 50%. However, when deaths occurring in infants less than 30 days of age were eliminated. the percentages changed: hispanics 70%. Caucasian 45%. black 45%. and others 50%. Traumatic deaths were 1.6 times as likely to occur during the months of June through October (n = 85 deaths/ma) as opposed to November through May In = 53 deaths/ma) [P -C .05). Mortality from burns was limited to children less than 5 years of age and penetrating trauma mortality was twice as likely to occur in children over 15 years (10% Y 20%). Fifty-two percent of all traumatic deaths in children occurred between 15 and 18 years of age. Major burns account for only 1% of traumatic deaths in this state. Asphyxia and drowning were more common in young children, and blunt traumatic deaths more common in older children. In 1988, the first state-wide accident awareness program was instituted. This, coupled with a number of safety programs initiated from the private and public sector, may in part be responsible for the fact that trauma is no longer responsible for majority of childhood mortalities in this state. Demographics indicate that deaths from trauma are more likely to occur in a rural community rather than in urban areas. In addition, the cause of traumatic death appears to be age-related. These data indicate that age-related and demographic-targeted trauma education programs may be of benefit in further reducing mortality. Q 1990 by W.B. Saunders Company. INDEX WORDS: Childhood trauma, mortality.
T
RAUMA IS the most frequent cause of death in American children between the ages of 1 and 15 years. The type of injury may differ considerably in large metropolitan coastal communities that are highly industrialized and smaller communities with a strong agricultural economy. The purpose of the study was to evaluate the cause of death related to trauma in a midwestern state with a balanced economy involving both heavy industrial and agricultural communities. MATERIALS
AND
METHODS
Death certificates issued in the state of Indiana from June 1, 1986 through May 30, 1988 were retrospectively reviewed. AI1 deaths occurring in children and adolescents (0 to 18 years of age) were
Journa/ of f’ediatric Surgery, Vol 25. No 9
(September),
Randall J. Franiak, Jeffery C. Ulrich,
and Frederick J. Rescorla
1990: pp 955-960
Indiana
analyzed. Mechanism of death, sex, age, race, region of occurrence, and time of occurrence were tabulated. All nontrauma-related deaths were excluded from the study. If information was lacking on the original death certificate, coroners’ reports were consulted. RESULTS
From June 1, 1986 through May 30, 1988, 1,93 1 children from birth to 18 years of age died in the state of Indiana. Eight hundred six died of traumatic injuries (41.4% of all deaths), which was the single most common cause of death in this population. Seventy percent of deaths due to trauma were in boys. Etiology of traumatic deaths predominantly centered around blunt injuries, which resulted in 54% of the mortality (432) (Fig 1). The majority of these were caused by motor vehicle accidents with the child as a pedestrian or on a bicycle (201 children). The remaining children who died or suffered as a result of blunt trauma were occupants or operators of motor vehicles (18 1 children) or the victims of falls and other causes (50 children). Penetrating trauma occurred in 15% of the population (125 cases), and was very much agerelated with 84 children (67%) 15 years of age or older. Asphyxia or drownings accounted for 211 deaths including 72 related to fires. This etiology was more common in younger children, with 87 of these deaths occurring in children less than 5 years of age. Only two deaths occurred in children over 13 years of age. Sixty-two deaths related to asphyxia occurred in situations where smoke detectors were not present (all cases involving children over 5) and in six infants, child neglect was documented. Electrocution caused 3% of the recorded deaths and major burns resulted in the remaining 1%. Region of injury occurrence played an important part in mortality. Sixty percent of deaths occurred in rural areas (480 deaths), although the state census lists
From the Section of Pediatric Surgery, Department of Surgery, Indiana University Medical Center, Kiwanis/Riley Trauma Life Center for Children, and the James Whitcomb Riley Hospital for Children. Indianapolis, IN. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Chicago, Illinois, October 21-23.1989. Address reprint requests to Jay L. Grosfeld, MD, Surgeon-inChiel. J. W. Riley Hospitalfor Children. 702 Barnhill Dr, Indianapolis, IN 46202-5200. Q 1990 by W. B. Saunders Company. 0022-3468/90/2509-0007$03.00/0
955
956
VANE ET AL
BLUNT
TRAU 55%
LECTROCUTION 3% PENETRATING 15%
ASPHYXIA
PERCENT
OR DROWNING 26%
OF CASES
Age was perhaps the most significant factor in pediatric trauma mortality. Although death rates were highest in the infant age group (720 deaths in children less than 1 year of age), the percent of traumatic deaths was only 3.5%. This number increased 20-fold to an 80% incidence at 18 years of age (Fig 3). This trend was true of all modes of trauma (ie, penetrating, blunt, electrocution, asphyxia) except major burns, where no mortality occurred in children over 5 years of age.
Fig 1. The causes of traumatic deaths in children remain unchanged, with blunt trauma accounting for 54% of the mortality.
70% of the childhood population as residing in urban areas. The remaining 40% of injuries occurred in an urban setting and could not be separated from those occurring in rural areas by cause. The incidence of mortality due to penetrating injuries, blunt trauma, asphyxia, and drownings was similar in both the rural and urban setting. Burn deaths occurred predominantly in the urban regions, but this was not significant due to the small number of cases encountered. Racial incidence of traumatic deaths were as follows: Hispanics 71%, Caucasians 42%, blacks 35%, and others 50%. However, when children less than 30 days of age were excluded, percentages changed to Hispanics 70%, Caucasians 45%, blacks 45%, and other 50%. Traumatic deaths were more common in Hispanics than other racial groups and this was significant to the P < .05 level. Deaths due to traumatic injuries in children were seasonal in nature (Fig 2). An injury resulting in mortality was 1.6 times as likely to occur in this state during the months of June through October (n = 85 deaths/ma) than November through May (n = 53 deaths/ma) (P < .05). No single cause of death was more commonly observed in a specific season however, with a relatively even distribution of causes noted over the entire year.
DISCUSSION
Trauma remains the leading cause of death in children from 1 to 18 years of age in the United States.’ This review of mortality in Indiana documents an overall incidence of deaths due to trauma of 41.4%. This number is somewhat lower than previously published studies that indicate that injuries cause approximately 47% of deaths in this age group.2 This is confirmed with a new incidence of 15.7 deaths per 100,000 per year, which is somewhat lower than the 18 per 100,000 per year previously reported for this state in 1983.3 Boys accounted for 70% of the deaths from trauma but represent only 51% of the population of the state.4 This increase in male mortality is constant regardless of age, mechanism of injury, race, or demographics in this study and has been similarly observed by others in previous reviews.5 Although no specific reason for this male predominance is postulated, it is of note that sports injuries, car accidents, and drownings all are more frequent in males.2 Suicide is the only mechanism of death with a female predominance in childhood. However, there were not enough suicide cases in this study to statistically confirm that observation.2 The fact that boys are more likely to be killed by trauma is not age-related, and subsequently, thoughts about hormonal drives remain speculative. However, it would seem reasonable that one could conclude that boys are more aggressive and adventur-
120,
100
I 100
80 %
#
80
D k
D 60
:
S
s 20 ;
40 -
60
z JAN
FEB
MAR
APR
MAY JUNEJULY
AUG
SEP
OCT
NOV
DEC
MONTH Fig 2. A higher mortality was noted during the summer months. However, the ceuses of death were similar to those observed in the other seasons of the year.
20
40 11
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 16
AGE (YRS) Fig 3. Age was a significant factor. Death due to trauma represented only 3.6% of all deaths in children less than 1 year of age but greater then 88% in those over 18 years.
PEDIATRIC TRAUMA
DEATHS IN INDIANA
ous and more likely to be injured. Whether this behavioral pattern is inborn or an acquired trait is debatable. As in all studies involving childhood trauma, blunt injuries are responsible for most deaths in children (Fig 1). In this series, accidents involving a motor vehicle resulted in 88% of the blunt trauma mortality, which in itself resulted in 54% of all the deaths caused by trauma. A recent study from the Johns Hopkins Injury Prevention Center lists death as an occupant of the vehicle as the major cause of mortality in children during the years 1980 to 1985.* This report finds that blunt trauma due to motor vehicle accident in which the injured child was a pedestrian, or is on a bicycle exceeds that of children injured as an occupant in the vehicle (201 v 181). This more recent data is more in line with reports from urban centers that describe pedestrian injuries as more common than occupant injuries. 6 The lower mortality for childhood car occupants may be related to the widespread use of seatbelts by children after a mandatory seat belt law was introduced in this state in 1987.7 Penetrating trauma accounted for approximately 15% of the deaths of children in the state. Due to many changes in the sociological strata of the country this number appears to be increasing dramatically.* Reports from large urban centers describe a 25% incidence of mortality in children from gunshot wounds, with a significant number of these cases related to intentional injury.’ This inner-city epidemic is not apparent in the state of Indiana although the actual number of penetrating injuries appear to be increasing. In Indiana the incidence of injury from all types of penetrating trauma appears to be equally distributed between urban and rural settings, with a very low incidence of intentional injury noted. This trend with equal distribution between rural and urban areas is present in other reports where urban and rural populations are mixed.” This observation is of concern because, in the past, investigators considered gunshot wounds an urban problem only. Asphyxia or drownings were more prevalent in younger children. Thirty-four percent of these deaths occurred in fire-related incidents. This number coupled with the deaths resulting from burns pushes the Indiana mortality rate from fires to 2.8 per 100,000 deaths per year, which is slightly above the national average.3 Only two fire-related deaths occurred in children over 13 years of age in this series and 57% of the deaths were in children under 4 (all burn deaths that occurred were in children under 5). All fire-related deaths in children over 5 years of age occurred in an environment without smoke alarms. This suggests that at least in the older age group the presence of smoke alarms is
957
an effective prevention method. It is of interest that deaths from fire-related causes in Indiana were not seasonal in nature, as suggested by other reports3,” The region in the state where the injury occurred was a very important factor affecting the mortality rate. Indiana population data indicate that 70% of the childhood population (under 18 years of age) resides in urban centers. However, this report documents that 60% of the trauma deaths (480 children) occurred in rural areas. Various causes of fatality were similar in both urban and rural populations, (eg, blunt trauma, penetrating trauma, etc). Race was not an important factor with regard to mortality. Although burn deaths were more common in the urban setting, the difference was not statistically significant. Some reports indicate that farm-related injuries are more prevalent than other types of injury in rural communities; however, we were not able to confirm these data in this study.‘* The present data showed that motor vehicular accidents with the child as pedestrian or on a bicycle far exceeded the number of farm injuries. This observation is similar to the report by Jagger et al.13 The single most important factor affecting an increased rural mortality is limited access to medical care.14 This is a major concern in many regions of the country and is a serious health care problem that must be addressed. The state of Indiana has a number of counties with no medical care available at all. The only Level I comprehensive trauma care centers are located in one area (the city of Indianapolis) (Fig 4). It is apparent that improved access to emergency care must be established before significant reduction in childhood mortality from injury can be accomplished. Race was not a significant factor in determining mortality statistics for trauma in this study. Black infant mortality in Indiana is relatively high in the newborn period and subsequently, when this population is excluded, mortality from trauma becomes more important in this group. Although other studies have indicated that the mortality from homicide is higher in blacks, this was not confirmed by this study.3*‘5-‘7The socioeconomic status of the victim is also reported to influence mortality in childhood trauma; however, we were also unable to document this in these data.‘*-*’ Data from this study clearly indicate that traumatic deaths in Indiana are seasonal in occurrence (Fig 2). Several other reports have documented similar variances with seasons. 6*‘3This study confirms an increase in childhood mortality during the months associated school vacation. However, it is of interest that penetrating injuries in larger cities have less seasonal variance than other types of injuries in other regions.6 Our study did not show a change in the seasonal variance for penetrating injuries, indicating that play activities,
958
VANE ET AL
Fig 4. This map of the counties in Indiana indicates that access to emergency medical care is a problem in rural areas of the state. m. Emergency facilities: a, general care facilities: q. comprehensive care facilities; 0. no facilities.
rather than intentional homicide (related to drug dealing, etc) is a more important consideration in this state. Age played an important role in the mortality in Indiana children. The incidence of mortality from
injury varied directly with the child’s age (Fig 4). This increase was present in all types of injury except fatal burns, which all occurred in children less than 5 years of age. This variance from burn fatalities has been previously described in a review in 1984.‘6*‘7 Age variation is present regardless of sex or mechanism of injury. Location of injury was also age related in this study. Several large reviews have documented an increased mortality in adolescents.2*‘0~‘6~17 However, a report from the state of Massachusetts indicates that pedestrian injuries were fairly constant throughout childhood, which differs significantly from these data.5 Accident-related death is one of the most important health care problems affecting our youth. The increased mortality observed in rural areas is related to limited access to adequate emergency medical care in these areas. However, developing adequate rural health care facilities is not the only answer. Even in urban areas where sophisticated medical facilities are available, accidental (and other causes of injury) are the leading cause of death in children. Data concerning the impact of age, patterns of injury, and differing mechanisms of injury on mortality indicate that modifications in educational programs relating to injury prevention are mandatory. Several directed programs have made significant inroads into changing the patterns of injury in some regions, although all have not been successfu1.21-23It is apparent that although access to medical care may be more of a priority in rural areas, fire prevention and warning systems may be better focused on parents rather than children. For this reason carefully directed safety education programs for specific populations at risk appear more cost-effective. In any case it is imperative that society recognize trauma as a major health problem for our youth and make prevention and other educational programs a high priority.
REFERENCES 1. Baker SP: Injuries: The neglected epidemic, Stone Lecture, 1985, American Trauma Society Meeting. J Trauma 27:343-348, 1987 2. Paulson JA: The epidemiology of injuries in adolescents. Pediatr Ann 1785-96, 1988 3. Wailer AE, Baker SP, Szocka A: Childhood injury deaths: National analysis and geographic variations. Am J Public Health 79:310-315.1989 4. 1987 Actuarial Population Estimate. Indianapolis, IN, State Board of Health of Indiana 5. Guyer B, Talbot AM, Pless IB: Pedestrian injuries to children and youth. Pediatr Clin North Am 32:163-174, 1985 6. Velcek FT, Weiss A, DiMaio D, et al: Traumatic death in urban children. J Pediatr Surg 12:375-384,1977 7. Indiana State Board of Health: Personal communication, September 1989
8. Ordog GJ, Wasserberger J, Schatz I, et al: Gunshot wounds in children under 10 years of age: A new epidemic, in Harris B, Coran AG (eds): Progress in Pediatric Trauma. Boston, MA, The Kiwanis Pediatric Trauma Institute, 1989 9. Davidson LL, Barlow B, Durkin MS, et al: Epidermiology of pediatric gunshot wounds in Northern Manhattan, in Harris B, Coran AG (eds): Progress in Pediatric Trauma. Boston, MA, The Kiwanis Pediatric Trauma Institute, 1989 10. Bergqvist D, Hedelin H, Lindblad B: Penetrating abdominal trauma. Acta Chir Stand 146:417-420, 1980 11. Mierley MC, Baker SP: Fatal house fires in an urban population. JAMA 249:1466-1468, 1983 12. Swanson JA, Sachs MI, Dahlgren KA, et al: Accidental farm injuries in children. Am J Dis Child 141:1276-1279, 1987 13. Jagger J, Levine J, Jane JA, et al: Epidemiologic features of
PEDIATRIC TRAUMA
head injury
959
DEATHS IN INDIANA
in a predominantly
rural
population.
J Trauma
24:40-
44.1984 14. Krob MJ, Cram AE, Vargish T, et al: Rural trauma care: A study of trauma care in a rural emergency medical services region. Ann Emerg Med 13:891-895, 1984 15. Joffe A: Adolescent suicide, injuries. Maryland Med J 37:955-958,
homicide, 1988
and
unintentional
16. Gulaid JA, Onwuachi-Saunders C, Sacks JJ, et al: Differences in death rates due to injury among blacks and whites, 1984. JAMA 261:214-216.1989 17. Fingerhut fatalities among 1988
LA, Kleinman young children.
JC, Malloy MH, et al: Injury Public Health Rep 103:399-405,
18. Schor
EL: Unintentional
injuries:
Patterns
within
families.
Am J Dis Child 141:1280-1284, 1987 19. Pearn JH: Secular trends in fatal and disabling child trauma. Aust Paediatr J 21:81-84, 1985 20. Spivak H, Prothrow-Stith D, Hausman AJ: Dying is no accident: Adolescents, violence, and intentional injury. Pediatr Clin North Am 35:1339-1347, 1988 21. Gallagher SS, Hunter P, Guyer B: A home injury prevention program for children. Pediatr Clin North Am 32:95-l 12. 1985 22. Pearn JH: Current controversies in child accident prevention. An analysis of some areas of dispute in the prevention of child trauma. Aust N Z J Med 15:782-787, 1985 23. Wheatley J. Cass DT: Traumatic deaths in children: The importance of prevention. Med J Aust 150~72-77. 1989
Discussion J. Templeton (Philadelphia, PA): In the abstract, you indicated that the traumatic deaths since 1988 are no longer the leading cause of mortality in children. D. Vane (response): That was traumatic deaths compared with all deaths. It didn’t compromise the majority of deaths of children because of our high infant mortality rate. J. Templeton (Philadelphia, PA): But you had a better than 40% mortality due to trauma in the early part of your survey. D. Vane (response): Right. J. Templeton (Philadelphia, PA): Have you had a real drop in a l-year time of education in the percentages of traumatic deaths? D. Vane (response): No. What we had was a real increase in the number of mortalities secondary to other causes. In other words, if you took trauma as the single cause of death, it was the single most common cause. However, if you related all deaths, trauma no longer exceeded all other deaths combined. J. Templeton (Philadelphia, PA): That is still surprising. The usual data shows that trauma causes four times as many deaths as cancer, for example, which is a typical high incidence. Is there an area in which you’ve seen an actual improvement in trauma mortality with prevention programs? D. Vane (responsej: We can’t say that. Our prevention program started in 1987, so we’re right in the middle of this. We didn’t see any change between the 2 years, 1986 and 1987. However, what we did see was an increase in the infant mortality rate in Indiana. So, our supposition is that the number of deaths from trauma stayed about the same. What changed apparently are the other mortality rates in the state. M. Matlak (Salt Lake City, UT): Did you look at
when and where these children died? Did the majority die within the first hour of injury as shown by Don Trunkey or are they making it to a definitive medical facility? D. Vane (response): We didn’t have data on all these children because it’s difficult to relate exactly where the deaths occurred. One of the problems that we have is that children who apparently died at the scene were not pronounced dead immediately but were pronounced in Indianapolis, which may have been 2 hours after the fact. What we were able to glean from the coroner’s report, and these are really soft data, is that apparently death occurred in the first 20 to 30 minutes except in the rural areas, where the deaths appeared to be occurring later as well. So, apparently, this postulates that about 20 minutes to reach good care is very important, but we don’t have the statistical data to prove that. S. Luck (Chicago, IL): I wonder if your study has taken into account the fact that Chicago, especially the Children’s Memorial Hospital, serves as a tertiary care center for Lake County in Indiana and, in fact, three of the northern counties, which you list on your map as having no tertiary care facilities. We very frequently get trauma patients from that part of Indiana. Frequently, they are less than an hour by helicopter and I certainly hope that we contribute to the better care of those patients. D. Vane (response): Certainly you contribute to the care of these patients. As you can see, the counties that surround Indianapolis have no comprehensive care facilities, as well. In point of fact, these are all the deaths that occurred in Indiana, ie, Indiana residents, this excludes patients that died in Indiana that were residing elsewhere, for example, tourists, etc.
VANE ET AL
960
R. Touloukian (New Haven, CT): Maybe help you out a little bit with what appears discrepancy with your data. It may well be that getting better reporting of infant mortality to for an apparent decrease in the number of deaths. I don’t think you want to imply that
we can to be a you are account trauma there’s
suddenly been a rather dramatic increase in infant deaths in your state. D. Vane (response): Actually, Indiana has drawn attention because of its infant mortality rate. We now have regional and federal interest because our infant mortality rate is so high.
Frederick G. Shedd,
Trauma Deaths in Indiana
Jay L. Grosfeld,
Karen W. West,
Indianapolis, 6 From June 1986 to May 1968. there were 1.931 childhood deaths recorded in Indiana. Eight hundred six children (0 to 18 years old) died as a result of trauma (41.4% of all deaths). Seventy percent of all traumatic deaths occurred in boys. Blunt trauma accounted for 54% of deaths, asphyxia or drowning 26%. penetrating trauma 15%. electrocution 3%. and burns 1%. Sixty percent of deaths occurred in rural areas and 40% occurred in urban centers: however, state-wide demographics define the population as 70% urban. The percentage of deaths due to trauma within a given race was: hispanic 71%, Caucasian 42%. black 35%. and others 50%. However, when deaths occurring in infants less than 30 days of age were eliminated. the percentages changed: hispanics 70%. Caucasian 45%. black 45%. and others 50%. Traumatic deaths were 1.6 times as likely to occur during the months of June through October (n = 85 deaths/ma) as opposed to November through May In = 53 deaths/ma) [P -C .05). Mortality from burns was limited to children less than 5 years of age and penetrating trauma mortality was twice as likely to occur in children over 15 years (10% Y 20%). Fifty-two percent of all traumatic deaths in children occurred between 15 and 18 years of age. Major burns account for only 1% of traumatic deaths in this state. Asphyxia and drowning were more common in young children, and blunt traumatic deaths more common in older children. In 1988, the first state-wide accident awareness program was instituted. This, coupled with a number of safety programs initiated from the private and public sector, may in part be responsible for the fact that trauma is no longer responsible for majority of childhood mortalities in this state. Demographics indicate that deaths from trauma are more likely to occur in a rural community rather than in urban areas. In addition, the cause of traumatic death appears to be age-related. These data indicate that age-related and demographic-targeted trauma education programs may be of benefit in further reducing mortality. Q 1990 by W.B. Saunders Company. INDEX WORDS: Childhood trauma, mortality.
T
RAUMA IS the most frequent cause of death in American children between the ages of 1 and 15 years. The type of injury may differ considerably in large metropolitan coastal communities that are highly industrialized and smaller communities with a strong agricultural economy. The purpose of the study was to evaluate the cause of death related to trauma in a midwestern state with a balanced economy involving both heavy industrial and agricultural communities. MATERIALS
AND
METHODS
Death certificates issued in the state of Indiana from June 1, 1986 through May 30, 1988 were retrospectively reviewed. AI1 deaths occurring in children and adolescents (0 to 18 years of age) were
Journa/ of f’ediatric Surgery, Vol 25. No 9
(September),
Randall J. Franiak, Jeffery C. Ulrich,
and Frederick J. Rescorla
1990: pp 955-960
Indiana
analyzed. Mechanism of death, sex, age, race, region of occurrence, and time of occurrence were tabulated. All nontrauma-related deaths were excluded from the study. If information was lacking on the original death certificate, coroners’ reports were consulted. RESULTS
From June 1, 1986 through May 30, 1988, 1,93 1 children from birth to 18 years of age died in the state of Indiana. Eight hundred six died of traumatic injuries (41.4% of all deaths), which was the single most common cause of death in this population. Seventy percent of deaths due to trauma were in boys. Etiology of traumatic deaths predominantly centered around blunt injuries, which resulted in 54% of the mortality (432) (Fig 1). The majority of these were caused by motor vehicle accidents with the child as a pedestrian or on a bicycle (201 children). The remaining children who died or suffered as a result of blunt trauma were occupants or operators of motor vehicles (18 1 children) or the victims of falls and other causes (50 children). Penetrating trauma occurred in 15% of the population (125 cases), and was very much agerelated with 84 children (67%) 15 years of age or older. Asphyxia or drownings accounted for 211 deaths including 72 related to fires. This etiology was more common in younger children, with 87 of these deaths occurring in children less than 5 years of age. Only two deaths occurred in children over 13 years of age. Sixty-two deaths related to asphyxia occurred in situations where smoke detectors were not present (all cases involving children over 5) and in six infants, child neglect was documented. Electrocution caused 3% of the recorded deaths and major burns resulted in the remaining 1%. Region of injury occurrence played an important part in mortality. Sixty percent of deaths occurred in rural areas (480 deaths), although the state census lists
From the Section of Pediatric Surgery, Department of Surgery, Indiana University Medical Center, Kiwanis/Riley Trauma Life Center for Children, and the James Whitcomb Riley Hospital for Children. Indianapolis, IN. Presented at the 38th Annual Meeting of the Surgical Section of the American Academy of Pediatrics, Chicago, Illinois, October 21-23.1989. Address reprint requests to Jay L. Grosfeld, MD, Surgeon-inChiel. J. W. Riley Hospitalfor Children. 702 Barnhill Dr, Indianapolis, IN 46202-5200. Q 1990 by W. B. Saunders Company. 0022-3468/90/2509-0007$03.00/0
955
956
VANE ET AL
BLUNT
TRAU 55%
LECTROCUTION 3% PENETRATING 15%
ASPHYXIA
PERCENT
OR DROWNING 26%
OF CASES
Age was perhaps the most significant factor in pediatric trauma mortality. Although death rates were highest in the infant age group (720 deaths in children less than 1 year of age), the percent of traumatic deaths was only 3.5%. This number increased 20-fold to an 80% incidence at 18 years of age (Fig 3). This trend was true of all modes of trauma (ie, penetrating, blunt, electrocution, asphyxia) except major burns, where no mortality occurred in children over 5 years of age.
Fig 1. The causes of traumatic deaths in children remain unchanged, with blunt trauma accounting for 54% of the mortality.
70% of the childhood population as residing in urban areas. The remaining 40% of injuries occurred in an urban setting and could not be separated from those occurring in rural areas by cause. The incidence of mortality due to penetrating injuries, blunt trauma, asphyxia, and drownings was similar in both the rural and urban setting. Burn deaths occurred predominantly in the urban regions, but this was not significant due to the small number of cases encountered. Racial incidence of traumatic deaths were as follows: Hispanics 71%, Caucasians 42%, blacks 35%, and others 50%. However, when children less than 30 days of age were excluded, percentages changed to Hispanics 70%, Caucasians 45%, blacks 45%, and other 50%. Traumatic deaths were more common in Hispanics than other racial groups and this was significant to the P < .05 level. Deaths due to traumatic injuries in children were seasonal in nature (Fig 2). An injury resulting in mortality was 1.6 times as likely to occur in this state during the months of June through October (n = 85 deaths/ma) than November through May (n = 53 deaths/ma) (P < .05). No single cause of death was more commonly observed in a specific season however, with a relatively even distribution of causes noted over the entire year.
DISCUSSION
Trauma remains the leading cause of death in children from 1 to 18 years of age in the United States.’ This review of mortality in Indiana documents an overall incidence of deaths due to trauma of 41.4%. This number is somewhat lower than previously published studies that indicate that injuries cause approximately 47% of deaths in this age group.2 This is confirmed with a new incidence of 15.7 deaths per 100,000 per year, which is somewhat lower than the 18 per 100,000 per year previously reported for this state in 1983.3 Boys accounted for 70% of the deaths from trauma but represent only 51% of the population of the state.4 This increase in male mortality is constant regardless of age, mechanism of injury, race, or demographics in this study and has been similarly observed by others in previous reviews.5 Although no specific reason for this male predominance is postulated, it is of note that sports injuries, car accidents, and drownings all are more frequent in males.2 Suicide is the only mechanism of death with a female predominance in childhood. However, there were not enough suicide cases in this study to statistically confirm that observation.2 The fact that boys are more likely to be killed by trauma is not age-related, and subsequently, thoughts about hormonal drives remain speculative. However, it would seem reasonable that one could conclude that boys are more aggressive and adventur-
120,
100
I 100
80 %
#
80
D k
D 60
:
S
s 20 ;
40 -
60
z JAN
FEB
MAR
APR
MAY JUNEJULY
AUG
SEP
OCT
NOV
DEC
MONTH Fig 2. A higher mortality was noted during the summer months. However, the ceuses of death were similar to those observed in the other seasons of the year.
20
40 11
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 16
AGE (YRS) Fig 3. Age was a significant factor. Death due to trauma represented only 3.6% of all deaths in children less than 1 year of age but greater then 88% in those over 18 years.
PEDIATRIC TRAUMA
DEATHS IN INDIANA
ous and more likely to be injured. Whether this behavioral pattern is inborn or an acquired trait is debatable. As in all studies involving childhood trauma, blunt injuries are responsible for most deaths in children (Fig 1). In this series, accidents involving a motor vehicle resulted in 88% of the blunt trauma mortality, which in itself resulted in 54% of all the deaths caused by trauma. A recent study from the Johns Hopkins Injury Prevention Center lists death as an occupant of the vehicle as the major cause of mortality in children during the years 1980 to 1985.* This report finds that blunt trauma due to motor vehicle accident in which the injured child was a pedestrian, or is on a bicycle exceeds that of children injured as an occupant in the vehicle (201 v 181). This more recent data is more in line with reports from urban centers that describe pedestrian injuries as more common than occupant injuries. 6 The lower mortality for childhood car occupants may be related to the widespread use of seatbelts by children after a mandatory seat belt law was introduced in this state in 1987.7 Penetrating trauma accounted for approximately 15% of the deaths of children in the state. Due to many changes in the sociological strata of the country this number appears to be increasing dramatically.* Reports from large urban centers describe a 25% incidence of mortality in children from gunshot wounds, with a significant number of these cases related to intentional injury.’ This inner-city epidemic is not apparent in the state of Indiana although the actual number of penetrating injuries appear to be increasing. In Indiana the incidence of injury from all types of penetrating trauma appears to be equally distributed between urban and rural settings, with a very low incidence of intentional injury noted. This trend with equal distribution between rural and urban areas is present in other reports where urban and rural populations are mixed.” This observation is of concern because, in the past, investigators considered gunshot wounds an urban problem only. Asphyxia or drownings were more prevalent in younger children. Thirty-four percent of these deaths occurred in fire-related incidents. This number coupled with the deaths resulting from burns pushes the Indiana mortality rate from fires to 2.8 per 100,000 deaths per year, which is slightly above the national average.3 Only two fire-related deaths occurred in children over 13 years of age in this series and 57% of the deaths were in children under 4 (all burn deaths that occurred were in children under 5). All fire-related deaths in children over 5 years of age occurred in an environment without smoke alarms. This suggests that at least in the older age group the presence of smoke alarms is
957
an effective prevention method. It is of interest that deaths from fire-related causes in Indiana were not seasonal in nature, as suggested by other reports3,” The region in the state where the injury occurred was a very important factor affecting the mortality rate. Indiana population data indicate that 70% of the childhood population (under 18 years of age) resides in urban centers. However, this report documents that 60% of the trauma deaths (480 children) occurred in rural areas. Various causes of fatality were similar in both urban and rural populations, (eg, blunt trauma, penetrating trauma, etc). Race was not an important factor with regard to mortality. Although burn deaths were more common in the urban setting, the difference was not statistically significant. Some reports indicate that farm-related injuries are more prevalent than other types of injury in rural communities; however, we were not able to confirm these data in this study.‘* The present data showed that motor vehicular accidents with the child as pedestrian or on a bicycle far exceeded the number of farm injuries. This observation is similar to the report by Jagger et al.13 The single most important factor affecting an increased rural mortality is limited access to medical care.14 This is a major concern in many regions of the country and is a serious health care problem that must be addressed. The state of Indiana has a number of counties with no medical care available at all. The only Level I comprehensive trauma care centers are located in one area (the city of Indianapolis) (Fig 4). It is apparent that improved access to emergency care must be established before significant reduction in childhood mortality from injury can be accomplished. Race was not a significant factor in determining mortality statistics for trauma in this study. Black infant mortality in Indiana is relatively high in the newborn period and subsequently, when this population is excluded, mortality from trauma becomes more important in this group. Although other studies have indicated that the mortality from homicide is higher in blacks, this was not confirmed by this study.3*‘5-‘7The socioeconomic status of the victim is also reported to influence mortality in childhood trauma; however, we were also unable to document this in these data.‘*-*’ Data from this study clearly indicate that traumatic deaths in Indiana are seasonal in occurrence (Fig 2). Several other reports have documented similar variances with seasons. 6*‘3This study confirms an increase in childhood mortality during the months associated school vacation. However, it is of interest that penetrating injuries in larger cities have less seasonal variance than other types of injuries in other regions.6 Our study did not show a change in the seasonal variance for penetrating injuries, indicating that play activities,
958
VANE ET AL
Fig 4. This map of the counties in Indiana indicates that access to emergency medical care is a problem in rural areas of the state. m. Emergency facilities: a, general care facilities: q. comprehensive care facilities; 0. no facilities.
rather than intentional homicide (related to drug dealing, etc) is a more important consideration in this state. Age played an important role in the mortality in Indiana children. The incidence of mortality from
injury varied directly with the child’s age (Fig 4). This increase was present in all types of injury except fatal burns, which all occurred in children less than 5 years of age. This variance from burn fatalities has been previously described in a review in 1984.‘6*‘7 Age variation is present regardless of sex or mechanism of injury. Location of injury was also age related in this study. Several large reviews have documented an increased mortality in adolescents.2*‘0~‘6~17 However, a report from the state of Massachusetts indicates that pedestrian injuries were fairly constant throughout childhood, which differs significantly from these data.5 Accident-related death is one of the most important health care problems affecting our youth. The increased mortality observed in rural areas is related to limited access to adequate emergency medical care in these areas. However, developing adequate rural health care facilities is not the only answer. Even in urban areas where sophisticated medical facilities are available, accidental (and other causes of injury) are the leading cause of death in children. Data concerning the impact of age, patterns of injury, and differing mechanisms of injury on mortality indicate that modifications in educational programs relating to injury prevention are mandatory. Several directed programs have made significant inroads into changing the patterns of injury in some regions, although all have not been successfu1.21-23It is apparent that although access to medical care may be more of a priority in rural areas, fire prevention and warning systems may be better focused on parents rather than children. For this reason carefully directed safety education programs for specific populations at risk appear more cost-effective. In any case it is imperative that society recognize trauma as a major health problem for our youth and make prevention and other educational programs a high priority.
REFERENCES 1. Baker SP: Injuries: The neglected epidemic, Stone Lecture, 1985, American Trauma Society Meeting. J Trauma 27:343-348, 1987 2. Paulson JA: The epidemiology of injuries in adolescents. Pediatr Ann 1785-96, 1988 3. Wailer AE, Baker SP, Szocka A: Childhood injury deaths: National analysis and geographic variations. Am J Public Health 79:310-315.1989 4. 1987 Actuarial Population Estimate. Indianapolis, IN, State Board of Health of Indiana 5. Guyer B, Talbot AM, Pless IB: Pedestrian injuries to children and youth. Pediatr Clin North Am 32:163-174, 1985 6. Velcek FT, Weiss A, DiMaio D, et al: Traumatic death in urban children. J Pediatr Surg 12:375-384,1977 7. Indiana State Board of Health: Personal communication, September 1989
8. Ordog GJ, Wasserberger J, Schatz I, et al: Gunshot wounds in children under 10 years of age: A new epidemic, in Harris B, Coran AG (eds): Progress in Pediatric Trauma. Boston, MA, The Kiwanis Pediatric Trauma Institute, 1989 9. Davidson LL, Barlow B, Durkin MS, et al: Epidermiology of pediatric gunshot wounds in Northern Manhattan, in Harris B, Coran AG (eds): Progress in Pediatric Trauma. Boston, MA, The Kiwanis Pediatric Trauma Institute, 1989 10. Bergqvist D, Hedelin H, Lindblad B: Penetrating abdominal trauma. Acta Chir Stand 146:417-420, 1980 11. Mierley MC, Baker SP: Fatal house fires in an urban population. JAMA 249:1466-1468, 1983 12. Swanson JA, Sachs MI, Dahlgren KA, et al: Accidental farm injuries in children. Am J Dis Child 141:1276-1279, 1987 13. Jagger J, Levine J, Jane JA, et al: Epidemiologic features of
PEDIATRIC TRAUMA
head injury
959
DEATHS IN INDIANA
in a predominantly
rural
population.
J Trauma
24:40-
44.1984 14. Krob MJ, Cram AE, Vargish T, et al: Rural trauma care: A study of trauma care in a rural emergency medical services region. Ann Emerg Med 13:891-895, 1984 15. Joffe A: Adolescent suicide, injuries. Maryland Med J 37:955-958,
homicide, 1988
and
unintentional
16. Gulaid JA, Onwuachi-Saunders C, Sacks JJ, et al: Differences in death rates due to injury among blacks and whites, 1984. JAMA 261:214-216.1989 17. Fingerhut fatalities among 1988
LA, Kleinman young children.
JC, Malloy MH, et al: Injury Public Health Rep 103:399-405,
18. Schor
EL: Unintentional
injuries:
Patterns
within
families.
Am J Dis Child 141:1280-1284, 1987 19. Pearn JH: Secular trends in fatal and disabling child trauma. Aust Paediatr J 21:81-84, 1985 20. Spivak H, Prothrow-Stith D, Hausman AJ: Dying is no accident: Adolescents, violence, and intentional injury. Pediatr Clin North Am 35:1339-1347, 1988 21. Gallagher SS, Hunter P, Guyer B: A home injury prevention program for children. Pediatr Clin North Am 32:95-l 12. 1985 22. Pearn JH: Current controversies in child accident prevention. An analysis of some areas of dispute in the prevention of child trauma. Aust N Z J Med 15:782-787, 1985 23. Wheatley J. Cass DT: Traumatic deaths in children: The importance of prevention. Med J Aust 150~72-77. 1989
Discussion J. Templeton (Philadelphia, PA): In the abstract, you indicated that the traumatic deaths since 1988 are no longer the leading cause of mortality in children. D. Vane (response): That was traumatic deaths compared with all deaths. It didn’t compromise the majority of deaths of children because of our high infant mortality rate. J. Templeton (Philadelphia, PA): But you had a better than 40% mortality due to trauma in the early part of your survey. D. Vane (response): Right. J. Templeton (Philadelphia, PA): Have you had a real drop in a l-year time of education in the percentages of traumatic deaths? D. Vane (response): No. What we had was a real increase in the number of mortalities secondary to other causes. In other words, if you took trauma as the single cause of death, it was the single most common cause. However, if you related all deaths, trauma no longer exceeded all other deaths combined. J. Templeton (Philadelphia, PA): That is still surprising. The usual data shows that trauma causes four times as many deaths as cancer, for example, which is a typical high incidence. Is there an area in which you’ve seen an actual improvement in trauma mortality with prevention programs? D. Vane (responsej: We can’t say that. Our prevention program started in 1987, so we’re right in the middle of this. We didn’t see any change between the 2 years, 1986 and 1987. However, what we did see was an increase in the infant mortality rate in Indiana. So, our supposition is that the number of deaths from trauma stayed about the same. What changed apparently are the other mortality rates in the state. M. Matlak (Salt Lake City, UT): Did you look at
when and where these children died? Did the majority die within the first hour of injury as shown by Don Trunkey or are they making it to a definitive medical facility? D. Vane (response): We didn’t have data on all these children because it’s difficult to relate exactly where the deaths occurred. One of the problems that we have is that children who apparently died at the scene were not pronounced dead immediately but were pronounced in Indianapolis, which may have been 2 hours after the fact. What we were able to glean from the coroner’s report, and these are really soft data, is that apparently death occurred in the first 20 to 30 minutes except in the rural areas, where the deaths appeared to be occurring later as well. So, apparently, this postulates that about 20 minutes to reach good care is very important, but we don’t have the statistical data to prove that. S. Luck (Chicago, IL): I wonder if your study has taken into account the fact that Chicago, especially the Children’s Memorial Hospital, serves as a tertiary care center for Lake County in Indiana and, in fact, three of the northern counties, which you list on your map as having no tertiary care facilities. We very frequently get trauma patients from that part of Indiana. Frequently, they are less than an hour by helicopter and I certainly hope that we contribute to the better care of those patients. D. Vane (response): Certainly you contribute to the care of these patients. As you can see, the counties that surround Indianapolis have no comprehensive care facilities, as well. In point of fact, these are all the deaths that occurred in Indiana, ie, Indiana residents, this excludes patients that died in Indiana that were residing elsewhere, for example, tourists, etc.
VANE ET AL
960
R. Touloukian (New Haven, CT): Maybe help you out a little bit with what appears discrepancy with your data. It may well be that getting better reporting of infant mortality to for an apparent decrease in the number of deaths. I don’t think you want to imply that
we can to be a you are account trauma there’s
suddenly been a rather dramatic increase in infant deaths in your state. D. Vane (response): Actually, Indiana has drawn attention because of its infant mortality rate. We now have regional and federal interest because our infant mortality rate is so high.
