- Email: [email protected]
Acute healthcare utilization by children after motor vehicle crashes
Accident Analysis and Prevention 36 (2004) 507–511
Acute healthcare utilization by children after motor vehicle crashes Flaura K. Winston a,e,∗ , Michael R. Elliott c , Irene G. Chen e , Edith M. Simpson d,e , Dennis R. Durbin b,c,e a
Division of General Pediatrics, Department of Pediatrics, School of Medicine, The University of Pennsylvania, PA 19104, USA b Division of Emergency Medicine, Department of Pediatrics, School of Medicine, The University of Pennsylvania, c/o The Children’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., Philadelphia, PA 19104, USA c The Center for Clinical Epidemiology and Biostatistics, The University of Pennsylvania, Blockley Hall, Room 818, 423 Guardian Drive, Philadelphia, PA 19104, USA d School of Nursing, The University of Pennsylvania, Room 419 NEB, 420 Guardian Drive, Philadelphia, PA 19104, USA e TraumaLink: The Pediatric Interdisciplinary Injury Control Research Center, The Children’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., 3535 TraumaLink 10th Floor, Philadelphia, PA 19104, USA Received 18 December 2002; received in revised form 29 January 2003; accepted 17 March 2003
Abstract This study, describing the overall patterns of acute healthcare resource utilization by child crash victims (age 15 years and younger), was conducted between 28 July 1999 and 30 November 2000 as part of an on-going large-scale, child-specific crash surveillance system, Partners for Child Passenger Safety: insurance claims from 15 states and the District of Columbia function as the source of subjects, with telephone survey and on-site crash investigations serving as the primary sources of data. A probability sample of 4862 eligible crashes with 7368 child occupants formed the study sample. Our results suggest that for every 1000 children involved in crashes, 3 are hospitalized; 108 are treated and released from an emergency department (ED); 48 are evaluated in a physician’s office, urgent care center, or other facility; and 841 receive no care at all. Comprehensive surveillance systems for motor vehicle crashes must capture children treated in physicians’ offices, emergency departments, and other healthcare facilities in order to provide accurate estimates of the impact on the health care system related to motor vehicle trauma. © 2003 Elsevier Ltd. All rights reserved. Keywords: Healthcare utilization; Health service research; Trauma; Injury; Motor vehicle crashes; Children
1. Introduction Over 1.5 million children are in crashes each year in the United States (Durbin et al., 2001). Estimates of the burden of these injuries to children are based primarily on the number of children hospitalized following a crash (Agran et al., 2001; Osmond et al., 2002). In particular, the majority of child passenger trauma studies have focused on seriously injured children treated in trauma centers (Brown et al., 2001; Holmes et al., 2002; Scheidler et al., 2000; Silber et al., 2001). These studies have been invaluable in advancing our understanding of injury mechanisms and the development of trauma center protocols. However, they provide little insight on the care received by the vast majority of children who are not treated at a trauma center following a crash. Therefore, little is known about the overall patterns of acute healthcare resource utilization by child crash victims. ∗ Corresponding author. Tel.: +1-215-590-3118; fax: +1-215-590-5425. E-mail address: [email protected] (F.K. Winston).
0001-4575/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0001-4575(03)00056-3
An estimate of the impact of motor vehicle crashes on children’s use of acute care resources requires an understanding of healthcare utilization by children across the spectrum of crash and injury severity. These data would be beneficial in planning for the allocation of acute care resources for child crash victims. Thus, the purpose of this study was to provide crash exposure-based estimates of acute health care utilization by children involved in motor vehicle crashes and to describe patterns of this care. 2. Methods 2.1. Study population and data collection Data were collected over a 16-month period from 28 July 1999 to 30 November 2000. A description of the study methods has been published previously (Durbin et al., 2001). The project consists of a large-scale, child-specific crash surveillance system, including a cross-sectional population-based
508
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
sample survey regarding children in passenger vehicle crashes.1 Vehicles qualifying for inclusion were model year 1990 or newer, and involved in a crash with at least one child occupant under 16 years of age. Qualifying crashes were limited to those that occurred in fifteen states and the District of Columbia, representing three large regions of the United States (east: NY, NJ, PA, DE, MD, VA, WV, NC, DC; midwest: OH, MI, IN, IL; west: CA, NV, AZ). A probability sample of 4862 eligible crashes with 7368 child occupants formed the study sample. Vehicles containing children who received medical treatment following the crash were over-sampled so that the majority of injured children would be selected while maintaining the representativeness of the overall population. Survey data were obtained on all child occupants of selected vehicles. Consent was obtained from 76% of eligible policyholders to participate in the survey; of these, 19% were sampled for interview and an estimated 81% of these were successfully interviewed. The full interview involved a 30 min telephone survey with the driver of the vehicle and parent(s) of the involved children. The median length of time between the date of the crash and the completion of the interview was 6 days. Comparing respondents with the eligible population showed virtually no differences with respect to region of residence, vehicle fleet makeup, or age of child.
Farm Insurance Company (State Farm) to The Children’s Hospital of Philadelphia and The University of Pennsylvania, for the conduct of the telephone survey, and for the conduct of the crash investigation. The study protocol was reviewed and approved by the Institutional Review Boards of both The Children’s Hospital of Philadelphia and School of Medicine, The University of Pennsylvania.
2.2. Variable definitions
All age groups were well represented in the study sample (29%: 0–3 years, 33%: 4–8 years, 23%: 9–12 years, and 16%: 13–15 years) as were boys (49%) and girls (51%). Passenger cars were the most common vehicle type (49% of all vehicles) and in the majority of crashes (68%), the vehicle could be driven from the scene. Nearly all children (97%) in this insured population were reported as restrained in some way (see Table 1). Among all children in crashes, only 13% were assessed at the scene by emergency medical services personnel; even among those in non-drivable vehicles, only 30% were assessed. Of those assessed, 53% had subsequent emergency department care or hospitalization, 5% received care at a physician’s office or other care facility, while 42% received no further care after scene assessment. Among the 87% of children who were not assessed at the scene by EMS personnel, 10% had subsequent medical care: put another way, 54% of those receiving medical care were not assessed on scene. Among all children in crashes, the vast majority (84%) received no medical care other than EMS assessment. As depicted in Fig. 1 in which weighted number and percentages shown, of those receiving care (16% of the total sample), 30% were not treated in hospital emergency department or inpatient settings: in other words, a quarter of them were seen only in physician’s offices while 5% were receiving medical care only in urgent care center or chiropractor. Of note, among the children who first sought care other than at an ED, 11% were eventually treated in an ED.
Survey questions regarding injuries to children were designed to provide responses that were classified by body region and severity based on The Abbreviated Injury Scale (AIS) score (The Abbreviated Injury Scale, 1990) and have been previously validated for their ability to distinguish AIS 2+ from less severe injuries (Durbin et al., 1999; Seidel and Henderson, 1991). For the purposes of this study, a consequential injury was defined as one likely to require medical attention: any injury with an AIS score of 2 or greater (concussions and more serious brain injuries, all internal organ injuries, spinal cord injuries, and extremity fractures) or facial lacerations. Minor injuries were defined as all other lacerations, contusions and abrasions. Medical care received was categorized as first care received (in an emergency department (ED), physician’s office, chiropractor’s office, urgent care center, or care not otherwise specified) and highest level of care received (hospitalized higher than emergency department only higher than either physician’s office, chiropractor’s office, urgent care center, or care not otherwise specified). In addition, information on whether emergency medical services (EMS) personnel attended the scene of the crash was obtained. Separate verbal consent was obtained from eligible participants for the transfer of claim information from State 1 The sample is drawn through, and thus restricted to, State Farm-insured vehicles (State Farm Insurance Company, Bloomington, IL).
2.3. Data analysis The primary purpose of this analysis was to describe the pattern of care received by children in passenger vehicle crashes. Case weights equal to the inverse of the probability of selection were used to account for the over-sampling of crashes in which an injury to a child likely occurred. To adjust inference to account for the disproportional sampling of subjects by medical treatment and clustering of subjects by vehicle, analytic methods were used to account for sampling weights, sampling strata, and sampling units. In particular, 95% confidence intervals were calculated using SAS-callable SUDAAN® : software for the statistical analysis of correlated data, Version 7.5 (Research Triangle Institute, Research Triangle Park, NC, 1999). 3. Results
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511 Table 1 Descriptive statistics on the study sample Variable name
Unweighted number of children in sample (%)
Gender Male Female
3511 (49.2) 3856 (50.8)
Age of child (years) 0–3 4–8 9–12 13–15
1681 2311 1882 1487
(28.6) (32.7) (22.7) (15.9)
Vehicle type Passenger car Passenger van (minivan) Sport utility vehicle Pickup truck Large van
4154 1647 964 391 210
(49.2) (25.5) (15.7) (5.5) (4.1)
Crash severity Drivable from scene Non-drivable from scene
2866 (67.8) 4322 (32.2)
Restraint status Restrained Unrestrained
6873 (96.7) 494 (3.3)
Values shown in the parentheses are the weighted % of children in sample.
Approximately 2% of children in crashes who received care required hospitalization, with 39% of these hospitalizations including admission to an intensive care unit. The majority of children who received care received it on the day of the crash; however this differed by the type
509
of care received. Ninety-one percent of children treated in an ED received the care the same day while only 28% of children treated in physicians’ offices received the care the same day (P < 0.001). Overall, 13% of children received minor injuries while 1.6% of the children in motor vehicle crashes sustained a consequential injury. The majority of consequential injuries were to the head (58%) and face (22%), while 4% involved the chest, 9% the abdomen, 7% the neck, spine, or back, and 17% upper or lower extremities. The majority (75%) of children who were hospitalized for their injuries suffered consequential injuries. However, only 7.6% of children who received care in an ED had sustained a consequential injury while 1.8% of children who received care in a physician’s office without further care in an ED had sustained a consequential injury. Of those reporting consequential injuries, 14% did not obtain care until at least 1 day after the crash. Table 2 summarizes our estimates with 95% confidence intervals for acute care resource utilization for children involved in motor vehicle crashes. Our results suggest that for every 1000 children involved in crashes, 3 are hospitalized; 108 are treated and released from an emergency department; 48 are evaluated in a physician’s office, urgent care center, or other facility; and 841 receive no care at all. 4. Discussion This study describes current patterns of acute healthcare resource utilization among children in motor vehicle crashes. A small percentage of children who are in motor vehicle
Fig. 1. Distribution of sites of care.
510
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
Table 2 Distribution of care for a hypothetical cohort of 1000 children in crashes Parameter
Estimate (per 1000 children in crashes)
95% confidence interval for estimate
EMS assessment? Yes No
126 874
113–140 860–887
Care received? Yes No
159 841
146–172 828–854
Site of first care Physician’s office ED Other
105 44 10
95–114 36–51 6–15
Site of highest level of care Physician’s office 39 ED 108 Inpatient 3 Other 9
33–46 98–132 2–3 5–13
crashes sustain injuries serious enough to require hospitalization. Therefore, comprehensive surveillance systems for motor vehicle crashes must capture children treated in physicians’ offices, urgent care centers and emergency departments in order to provide accurate estimates of the impact on the health care system related to motor vehicle trauma. Our study finds that more than half of children medically treated for crash injury (54%) are not assessed at the crash scene; one-fourth (27%) of the medically treated patients never encountered either the hospital-based or field emergency medical services systems. This raises questions about relying on EMS run reports, emergency department, hospital, and other medical records for crash surveillance, as is done in the NHTSA CODES project (National Highway Traffic Safety Administration, 2003). The patterns of care reflect the importance of training in child trauma care across the emergency care system, beginning with pre-hospital care. Our results suggest that EMS personnel are often the only source of care received by children. Over half of children assessed by EMS personnel had no subsequent evaluation and care. Previous publications have highlighted the unique needs of children in EMS Systems (Ludwig and Selbst, 1990; Seidel and Henderson, 1991). Our data provide further insight on the importance of pediatric training for EMS personnel, highlighting the fact that these personnel often deliver what turns out to be the only care a child receives following a crash. While the majority of children who received care were treated in emergency departments and hospitals, 30% received care in physician’s offices, chiropractic offices, and urgent care centers. This distribution may reflect the recent decline in rates of hospitalization following injury for children (Durbin et al., 2000) and the increase in injury-related visits to primary care offices in the United States (Hambidge et al., 2002). While most injury-related visits to primary care are secondary to sports injuries, our data suggest that primary care physicians are also evaluating a significant num-
ber of children following a crash. This suggests that primary care physicians and staff should be trained in the proper assessment and stabilization of children at risk for injuries to multiple body regions following a crash. Current guidelines exist for equipping primary care offices for the emergency care of children (Seidel and Knapp, 2000). It is important to consider the majority of children in crashes who do not receive any medical care. Our data suggested that the level of care received was related to the likelihood of having a clinically significant physical injury. Recent research has demonstrated that children and their parents can suffer psychological consequences, including posttraumatic stress disorder, after traffic crashes (or other causes of injury/violence) regardless of whether the child is injured (de Vries et al., 1999). As these symptoms often go largely unrecognized, the high incidence of crashes involving children indicates the importance for primary care pediatricians and school personnel to inquire about crash exposure for children who are exhibiting changes in behavior. Beyond the data presented, we are not able to comment on the appropriateness of the care received by children. Given the wide range of care received following a crash, this is an area that should be targeted for further research. 4.1. Study limitations Most of the data for this study were obtained by parent/driver report and may be subject to recall bias. Previous studies have suggested that the accuracy of recall for injuries and care received varies inversely with the length of the recall period with optimal recall within 1 month of the injury (Harel et al., 1994; Landen and Hendricks, 1995; Zwerling et al., 1995; Durbin, 2001). On average, data for this study were collected within the first week following the crash. Therefore, it is likely that parents accurately recalled patterns of care. The sample on which these treatment pattern descriptions are based is a population of insured drivers with newer model year vehicles. Recent data indicate that approximately two-thirds of US children are covered by private health insurance, with the remainder either uninsured (15%) or covered by public sources (19%) (McCormick et al., 2001). Patterns of care vary by health insurance coverage with non-privately insured children more reliant on the emergency department for their care (McCormick et al., 2001). Because health insurance coverage may be greater on average among families with auto insurance, our results may generally underestimate acute care resource utilization by children following crashes. In addition, the study was conducted in English and, as such, the results may not be generalizable to the non-English-speaking population.
5. Conclusions A large number of children are involved in motor vehicle crashes each year, the majority of whom do not receive care
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
following the crash. In order to assess the full impact of motor vehicle crashes on children’s health, one must account for children being cared for across the emergency medical services spectrum, including prehospital care, as well as care in physician’s offices, emergency departments and urgent care centers. This raises questions about the reliability of passenger vehicle injury surveillance that relies only on EMS intake. Further research is needed to assess the appropriateness of care received in all of these settings, as well as the potential morbidity—both physical and psychological— incurred by children following crashes in order to minimize the impact of crashes on children’s health.
Acknowledgements The authors would like to thank State Farm Insurance Companies for their financial support of this work through the Partners for Child Passenger Safety Project. In addition, the authors would like to thank the many dedicated claim representatives and personnel from State Farm, the Research Team on the Partners Project and at TraumaLink who devoted countless hours to this study, and the parents who generously agreed to participate in the study. References Agran, P.F., Winn, D., Anderson, C., et al., 2001. Rates of pediatric and adolescent injuries by year of age. Pediatrics 108 (3), E45. Brown, R.L., Brunn, M.A., Garcia, V.F., 2001. Cervical spine injuries in children: a review of 103 patients treated consecutively at a level 1 pediatric trauma center. J. Pediatr. Surg. 36 (8), 1107–1114. de Vries, A.P., Kassam-Adams, N., Cnaan, A., et al., 1999. Looking beyond the physical injury: posttraumatic stress disorder in children and parents after pediatric traffic injury. Pediatrics 104 (6), 1293– 1299. Durbin, D.R., Winston, F.K., Applegate, S.M., Moll, E.K., Holmes, J.H., 1999. Development and validation of ISAS/PR: a new injury severity assessment survey. Arch. Pediatr. Adol. Med. 153 (4), 404–408. Durbin, D.R., Schwarz, D.F., Localio, A.R., et al., 2000. Trends in
511
incidence of pediatric injury hospitalizations in Pennsylvania. Am. J. Public Health 90 (11), 1782–1784. Durbin, D.R., Bhatia, E., Holmes, J.H., et al., 2001. Partners for Child Passenger Safety: a unique child-specific crash surveillance system. Accid. Anal. Prev. 33 (3), 407–412. Hambidge, S.J., Davidson, A.J., Gonzales, R., et al., 2002. Epidemiology of pediatric injury-related primary care office visits in the United States. Pediatrics 109 (4), 559–565. Harel, Y., Overpeck, M.D., Jones, D.H., et al., 1994. The effects of recall on estimating annual nonfatal injury rates for children and adolescents. Am. J. Public Health 84 (4), 599–605. Holmes, J.F., Sokolove, P.E., Brant, W.E., et al., 2002. A clinical decision rule for identifying children with thoracic injuries after blunt torso trauma. Ann. Emerg. Med. 39 (5), 492–499. Landen, D.D., Hendricks, S., 1995. Effect of recall on reporting of at-work injuries. Public Health Rep. 110 (3), 350–354. Ludwig, S., Selbst, S., 1990. A child-oriented emergency medical services system. Curr. Probl. Pediatr. 20 (3), 109–158. McCormick, M.C., Weinick, R.M., Elixhauser, A., et al., 2001. Annual report on access to and utilization of health care for children and youth in the United States—2000. Ambul. Pediatr. 1 (1), 3–15. National Highway Traffic Safety Administration, 2003. The CODES Model. National Highway Traffic Safety Administration, US Department of Transportation, Washington, DC, http://www.nhtsa.dot. gov/people/perform/trafrecords/pages/codes/codes.htm (accessed 13 May 2003). Osmond, M.H., Brennan-Barnes, M., Shephard, A.L., 2002. A 4-year review of severe pediatric trauma in eastern Ontario: a descriptive analysis. J. Trauma 52 (1), 8–12. Scheidler, M.G., Shultz, B.L., Schall, L., et al., 2000. Risk factors and predictors of mortality in children after ejection from motor vehicle crashes. J. Trauma 49 (5), 864–868. Seidel, J.S., Henderson, D.P., 1991. Emergency Medical Services for Children: A Report to the Nation. National Center for Education in Maternal and Child Health, Washington, DC. Seidel, J.S., Knapp, J.F., 2000. Pediatric emergencies in the office, hospital, and community: organizing systems of care. Pediatrics 106 (2 Pt. 1), 337–338. Silber, J.S., Flynn, J.M., Koffler, K.M., et al., 2001. Analysis of the cause, classification, and associated injuries of 166 consecutive pediatric pelvic fractures. J. Pediatr. Orthop. 21 (4), 446–450. The Abbreviated Injury Scale, 1990. Association for the Advancement of Automotive Medicine, Des Plaines, IL. Zwerling, C., Sprince, N.L., Wallace, R.B., et al., 1995. Effect of recall period on the reporting of occupational injuries among older workers in the Health and Retirement Study. Am. J. Ind. Med. 28 (5), 583–590.
Acute healthcare utilization by children after motor vehicle crashes Flaura K. Winston a,e,∗ , Michael R. Elliott c , Irene G. Chen e , Edith M. Simpson d,e , Dennis R. Durbin b,c,e a
Division of General Pediatrics, Department of Pediatrics, School of Medicine, The University of Pennsylvania, PA 19104, USA b Division of Emergency Medicine, Department of Pediatrics, School of Medicine, The University of Pennsylvania, c/o The Children’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., Philadelphia, PA 19104, USA c The Center for Clinical Epidemiology and Biostatistics, The University of Pennsylvania, Blockley Hall, Room 818, 423 Guardian Drive, Philadelphia, PA 19104, USA d School of Nursing, The University of Pennsylvania, Room 419 NEB, 420 Guardian Drive, Philadelphia, PA 19104, USA e TraumaLink: The Pediatric Interdisciplinary Injury Control Research Center, The Children’s Hospital of Philadelphia, 34th Street and Civic Center Blvd., 3535 TraumaLink 10th Floor, Philadelphia, PA 19104, USA Received 18 December 2002; received in revised form 29 January 2003; accepted 17 March 2003
Abstract This study, describing the overall patterns of acute healthcare resource utilization by child crash victims (age 15 years and younger), was conducted between 28 July 1999 and 30 November 2000 as part of an on-going large-scale, child-specific crash surveillance system, Partners for Child Passenger Safety: insurance claims from 15 states and the District of Columbia function as the source of subjects, with telephone survey and on-site crash investigations serving as the primary sources of data. A probability sample of 4862 eligible crashes with 7368 child occupants formed the study sample. Our results suggest that for every 1000 children involved in crashes, 3 are hospitalized; 108 are treated and released from an emergency department (ED); 48 are evaluated in a physician’s office, urgent care center, or other facility; and 841 receive no care at all. Comprehensive surveillance systems for motor vehicle crashes must capture children treated in physicians’ offices, emergency departments, and other healthcare facilities in order to provide accurate estimates of the impact on the health care system related to motor vehicle trauma. © 2003 Elsevier Ltd. All rights reserved. Keywords: Healthcare utilization; Health service research; Trauma; Injury; Motor vehicle crashes; Children
1. Introduction Over 1.5 million children are in crashes each year in the United States (Durbin et al., 2001). Estimates of the burden of these injuries to children are based primarily on the number of children hospitalized following a crash (Agran et al., 2001; Osmond et al., 2002). In particular, the majority of child passenger trauma studies have focused on seriously injured children treated in trauma centers (Brown et al., 2001; Holmes et al., 2002; Scheidler et al., 2000; Silber et al., 2001). These studies have been invaluable in advancing our understanding of injury mechanisms and the development of trauma center protocols. However, they provide little insight on the care received by the vast majority of children who are not treated at a trauma center following a crash. Therefore, little is known about the overall patterns of acute healthcare resource utilization by child crash victims. ∗ Corresponding author. Tel.: +1-215-590-3118; fax: +1-215-590-5425. E-mail address: [email protected] (F.K. Winston).
0001-4575/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0001-4575(03)00056-3
An estimate of the impact of motor vehicle crashes on children’s use of acute care resources requires an understanding of healthcare utilization by children across the spectrum of crash and injury severity. These data would be beneficial in planning for the allocation of acute care resources for child crash victims. Thus, the purpose of this study was to provide crash exposure-based estimates of acute health care utilization by children involved in motor vehicle crashes and to describe patterns of this care. 2. Methods 2.1. Study population and data collection Data were collected over a 16-month period from 28 July 1999 to 30 November 2000. A description of the study methods has been published previously (Durbin et al., 2001). The project consists of a large-scale, child-specific crash surveillance system, including a cross-sectional population-based
508
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
sample survey regarding children in passenger vehicle crashes.1 Vehicles qualifying for inclusion were model year 1990 or newer, and involved in a crash with at least one child occupant under 16 years of age. Qualifying crashes were limited to those that occurred in fifteen states and the District of Columbia, representing three large regions of the United States (east: NY, NJ, PA, DE, MD, VA, WV, NC, DC; midwest: OH, MI, IN, IL; west: CA, NV, AZ). A probability sample of 4862 eligible crashes with 7368 child occupants formed the study sample. Vehicles containing children who received medical treatment following the crash were over-sampled so that the majority of injured children would be selected while maintaining the representativeness of the overall population. Survey data were obtained on all child occupants of selected vehicles. Consent was obtained from 76% of eligible policyholders to participate in the survey; of these, 19% were sampled for interview and an estimated 81% of these were successfully interviewed. The full interview involved a 30 min telephone survey with the driver of the vehicle and parent(s) of the involved children. The median length of time between the date of the crash and the completion of the interview was 6 days. Comparing respondents with the eligible population showed virtually no differences with respect to region of residence, vehicle fleet makeup, or age of child.
Farm Insurance Company (State Farm) to The Children’s Hospital of Philadelphia and The University of Pennsylvania, for the conduct of the telephone survey, and for the conduct of the crash investigation. The study protocol was reviewed and approved by the Institutional Review Boards of both The Children’s Hospital of Philadelphia and School of Medicine, The University of Pennsylvania.
2.2. Variable definitions
All age groups were well represented in the study sample (29%: 0–3 years, 33%: 4–8 years, 23%: 9–12 years, and 16%: 13–15 years) as were boys (49%) and girls (51%). Passenger cars were the most common vehicle type (49% of all vehicles) and in the majority of crashes (68%), the vehicle could be driven from the scene. Nearly all children (97%) in this insured population were reported as restrained in some way (see Table 1). Among all children in crashes, only 13% were assessed at the scene by emergency medical services personnel; even among those in non-drivable vehicles, only 30% were assessed. Of those assessed, 53% had subsequent emergency department care or hospitalization, 5% received care at a physician’s office or other care facility, while 42% received no further care after scene assessment. Among the 87% of children who were not assessed at the scene by EMS personnel, 10% had subsequent medical care: put another way, 54% of those receiving medical care were not assessed on scene. Among all children in crashes, the vast majority (84%) received no medical care other than EMS assessment. As depicted in Fig. 1 in which weighted number and percentages shown, of those receiving care (16% of the total sample), 30% were not treated in hospital emergency department or inpatient settings: in other words, a quarter of them were seen only in physician’s offices while 5% were receiving medical care only in urgent care center or chiropractor. Of note, among the children who first sought care other than at an ED, 11% were eventually treated in an ED.
Survey questions regarding injuries to children were designed to provide responses that were classified by body region and severity based on The Abbreviated Injury Scale (AIS) score (The Abbreviated Injury Scale, 1990) and have been previously validated for their ability to distinguish AIS 2+ from less severe injuries (Durbin et al., 1999; Seidel and Henderson, 1991). For the purposes of this study, a consequential injury was defined as one likely to require medical attention: any injury with an AIS score of 2 or greater (concussions and more serious brain injuries, all internal organ injuries, spinal cord injuries, and extremity fractures) or facial lacerations. Minor injuries were defined as all other lacerations, contusions and abrasions. Medical care received was categorized as first care received (in an emergency department (ED), physician’s office, chiropractor’s office, urgent care center, or care not otherwise specified) and highest level of care received (hospitalized higher than emergency department only higher than either physician’s office, chiropractor’s office, urgent care center, or care not otherwise specified). In addition, information on whether emergency medical services (EMS) personnel attended the scene of the crash was obtained. Separate verbal consent was obtained from eligible participants for the transfer of claim information from State 1 The sample is drawn through, and thus restricted to, State Farm-insured vehicles (State Farm Insurance Company, Bloomington, IL).
2.3. Data analysis The primary purpose of this analysis was to describe the pattern of care received by children in passenger vehicle crashes. Case weights equal to the inverse of the probability of selection were used to account for the over-sampling of crashes in which an injury to a child likely occurred. To adjust inference to account for the disproportional sampling of subjects by medical treatment and clustering of subjects by vehicle, analytic methods were used to account for sampling weights, sampling strata, and sampling units. In particular, 95% confidence intervals were calculated using SAS-callable SUDAAN® : software for the statistical analysis of correlated data, Version 7.5 (Research Triangle Institute, Research Triangle Park, NC, 1999). 3. Results
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511 Table 1 Descriptive statistics on the study sample Variable name
Unweighted number of children in sample (%)
Gender Male Female
3511 (49.2) 3856 (50.8)
Age of child (years) 0–3 4–8 9–12 13–15
1681 2311 1882 1487
(28.6) (32.7) (22.7) (15.9)
Vehicle type Passenger car Passenger van (minivan) Sport utility vehicle Pickup truck Large van
4154 1647 964 391 210
(49.2) (25.5) (15.7) (5.5) (4.1)
Crash severity Drivable from scene Non-drivable from scene
2866 (67.8) 4322 (32.2)
Restraint status Restrained Unrestrained
6873 (96.7) 494 (3.3)
Values shown in the parentheses are the weighted % of children in sample.
Approximately 2% of children in crashes who received care required hospitalization, with 39% of these hospitalizations including admission to an intensive care unit. The majority of children who received care received it on the day of the crash; however this differed by the type
509
of care received. Ninety-one percent of children treated in an ED received the care the same day while only 28% of children treated in physicians’ offices received the care the same day (P < 0.001). Overall, 13% of children received minor injuries while 1.6% of the children in motor vehicle crashes sustained a consequential injury. The majority of consequential injuries were to the head (58%) and face (22%), while 4% involved the chest, 9% the abdomen, 7% the neck, spine, or back, and 17% upper or lower extremities. The majority (75%) of children who were hospitalized for their injuries suffered consequential injuries. However, only 7.6% of children who received care in an ED had sustained a consequential injury while 1.8% of children who received care in a physician’s office without further care in an ED had sustained a consequential injury. Of those reporting consequential injuries, 14% did not obtain care until at least 1 day after the crash. Table 2 summarizes our estimates with 95% confidence intervals for acute care resource utilization for children involved in motor vehicle crashes. Our results suggest that for every 1000 children involved in crashes, 3 are hospitalized; 108 are treated and released from an emergency department; 48 are evaluated in a physician’s office, urgent care center, or other facility; and 841 receive no care at all. 4. Discussion This study describes current patterns of acute healthcare resource utilization among children in motor vehicle crashes. A small percentage of children who are in motor vehicle
Fig. 1. Distribution of sites of care.
510
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
Table 2 Distribution of care for a hypothetical cohort of 1000 children in crashes Parameter
Estimate (per 1000 children in crashes)
95% confidence interval for estimate
EMS assessment? Yes No
126 874
113–140 860–887
Care received? Yes No
159 841
146–172 828–854
Site of first care Physician’s office ED Other
105 44 10
95–114 36–51 6–15
Site of highest level of care Physician’s office 39 ED 108 Inpatient 3 Other 9
33–46 98–132 2–3 5–13
crashes sustain injuries serious enough to require hospitalization. Therefore, comprehensive surveillance systems for motor vehicle crashes must capture children treated in physicians’ offices, urgent care centers and emergency departments in order to provide accurate estimates of the impact on the health care system related to motor vehicle trauma. Our study finds that more than half of children medically treated for crash injury (54%) are not assessed at the crash scene; one-fourth (27%) of the medically treated patients never encountered either the hospital-based or field emergency medical services systems. This raises questions about relying on EMS run reports, emergency department, hospital, and other medical records for crash surveillance, as is done in the NHTSA CODES project (National Highway Traffic Safety Administration, 2003). The patterns of care reflect the importance of training in child trauma care across the emergency care system, beginning with pre-hospital care. Our results suggest that EMS personnel are often the only source of care received by children. Over half of children assessed by EMS personnel had no subsequent evaluation and care. Previous publications have highlighted the unique needs of children in EMS Systems (Ludwig and Selbst, 1990; Seidel and Henderson, 1991). Our data provide further insight on the importance of pediatric training for EMS personnel, highlighting the fact that these personnel often deliver what turns out to be the only care a child receives following a crash. While the majority of children who received care were treated in emergency departments and hospitals, 30% received care in physician’s offices, chiropractic offices, and urgent care centers. This distribution may reflect the recent decline in rates of hospitalization following injury for children (Durbin et al., 2000) and the increase in injury-related visits to primary care offices in the United States (Hambidge et al., 2002). While most injury-related visits to primary care are secondary to sports injuries, our data suggest that primary care physicians are also evaluating a significant num-
ber of children following a crash. This suggests that primary care physicians and staff should be trained in the proper assessment and stabilization of children at risk for injuries to multiple body regions following a crash. Current guidelines exist for equipping primary care offices for the emergency care of children (Seidel and Knapp, 2000). It is important to consider the majority of children in crashes who do not receive any medical care. Our data suggested that the level of care received was related to the likelihood of having a clinically significant physical injury. Recent research has demonstrated that children and their parents can suffer psychological consequences, including posttraumatic stress disorder, after traffic crashes (or other causes of injury/violence) regardless of whether the child is injured (de Vries et al., 1999). As these symptoms often go largely unrecognized, the high incidence of crashes involving children indicates the importance for primary care pediatricians and school personnel to inquire about crash exposure for children who are exhibiting changes in behavior. Beyond the data presented, we are not able to comment on the appropriateness of the care received by children. Given the wide range of care received following a crash, this is an area that should be targeted for further research. 4.1. Study limitations Most of the data for this study were obtained by parent/driver report and may be subject to recall bias. Previous studies have suggested that the accuracy of recall for injuries and care received varies inversely with the length of the recall period with optimal recall within 1 month of the injury (Harel et al., 1994; Landen and Hendricks, 1995; Zwerling et al., 1995; Durbin, 2001). On average, data for this study were collected within the first week following the crash. Therefore, it is likely that parents accurately recalled patterns of care. The sample on which these treatment pattern descriptions are based is a population of insured drivers with newer model year vehicles. Recent data indicate that approximately two-thirds of US children are covered by private health insurance, with the remainder either uninsured (15%) or covered by public sources (19%) (McCormick et al., 2001). Patterns of care vary by health insurance coverage with non-privately insured children more reliant on the emergency department for their care (McCormick et al., 2001). Because health insurance coverage may be greater on average among families with auto insurance, our results may generally underestimate acute care resource utilization by children following crashes. In addition, the study was conducted in English and, as such, the results may not be generalizable to the non-English-speaking population.
5. Conclusions A large number of children are involved in motor vehicle crashes each year, the majority of whom do not receive care
F.K. Winston et al. / Accident Analysis and Prevention 36 (2004) 507–511
following the crash. In order to assess the full impact of motor vehicle crashes on children’s health, one must account for children being cared for across the emergency medical services spectrum, including prehospital care, as well as care in physician’s offices, emergency departments and urgent care centers. This raises questions about the reliability of passenger vehicle injury surveillance that relies only on EMS intake. Further research is needed to assess the appropriateness of care received in all of these settings, as well as the potential morbidity—both physical and psychological— incurred by children following crashes in order to minimize the impact of crashes on children’s health.
Acknowledgements The authors would like to thank State Farm Insurance Companies for their financial support of this work through the Partners for Child Passenger Safety Project. In addition, the authors would like to thank the many dedicated claim representatives and personnel from State Farm, the Research Team on the Partners Project and at TraumaLink who devoted countless hours to this study, and the parents who generously agreed to participate in the study. References Agran, P.F., Winn, D., Anderson, C., et al., 2001. Rates of pediatric and adolescent injuries by year of age. Pediatrics 108 (3), E45. Brown, R.L., Brunn, M.A., Garcia, V.F., 2001. Cervical spine injuries in children: a review of 103 patients treated consecutively at a level 1 pediatric trauma center. J. Pediatr. Surg. 36 (8), 1107–1114. de Vries, A.P., Kassam-Adams, N., Cnaan, A., et al., 1999. Looking beyond the physical injury: posttraumatic stress disorder in children and parents after pediatric traffic injury. Pediatrics 104 (6), 1293– 1299. Durbin, D.R., Winston, F.K., Applegate, S.M., Moll, E.K., Holmes, J.H., 1999. Development and validation of ISAS/PR: a new injury severity assessment survey. Arch. Pediatr. Adol. Med. 153 (4), 404–408. Durbin, D.R., Schwarz, D.F., Localio, A.R., et al., 2000. Trends in
511
incidence of pediatric injury hospitalizations in Pennsylvania. Am. J. Public Health 90 (11), 1782–1784. Durbin, D.R., Bhatia, E., Holmes, J.H., et al., 2001. Partners for Child Passenger Safety: a unique child-specific crash surveillance system. Accid. Anal. Prev. 33 (3), 407–412. Hambidge, S.J., Davidson, A.J., Gonzales, R., et al., 2002. Epidemiology of pediatric injury-related primary care office visits in the United States. Pediatrics 109 (4), 559–565. Harel, Y., Overpeck, M.D., Jones, D.H., et al., 1994. The effects of recall on estimating annual nonfatal injury rates for children and adolescents. Am. J. Public Health 84 (4), 599–605. Holmes, J.F., Sokolove, P.E., Brant, W.E., et al., 2002. A clinical decision rule for identifying children with thoracic injuries after blunt torso trauma. Ann. Emerg. Med. 39 (5), 492–499. Landen, D.D., Hendricks, S., 1995. Effect of recall on reporting of at-work injuries. Public Health Rep. 110 (3), 350–354. Ludwig, S., Selbst, S., 1990. A child-oriented emergency medical services system. Curr. Probl. Pediatr. 20 (3), 109–158. McCormick, M.C., Weinick, R.M., Elixhauser, A., et al., 2001. Annual report on access to and utilization of health care for children and youth in the United States—2000. Ambul. Pediatr. 1 (1), 3–15. National Highway Traffic Safety Administration, 2003. The CODES Model. National Highway Traffic Safety Administration, US Department of Transportation, Washington, DC, http://www.nhtsa.dot. gov/people/perform/trafrecords/pages/codes/codes.htm (accessed 13 May 2003). Osmond, M.H., Brennan-Barnes, M., Shephard, A.L., 2002. A 4-year review of severe pediatric trauma in eastern Ontario: a descriptive analysis. J. Trauma 52 (1), 8–12. Scheidler, M.G., Shultz, B.L., Schall, L., et al., 2000. Risk factors and predictors of mortality in children after ejection from motor vehicle crashes. J. Trauma 49 (5), 864–868. Seidel, J.S., Henderson, D.P., 1991. Emergency Medical Services for Children: A Report to the Nation. National Center for Education in Maternal and Child Health, Washington, DC. Seidel, J.S., Knapp, J.F., 2000. Pediatric emergencies in the office, hospital, and community: organizing systems of care. Pediatrics 106 (2 Pt. 1), 337–338. Silber, J.S., Flynn, J.M., Koffler, K.M., et al., 2001. Analysis of the cause, classification, and associated injuries of 166 consecutive pediatric pelvic fractures. J. Pediatr. Orthop. 21 (4), 446–450. The Abbreviated Injury Scale, 1990. Association for the Advancement of Automotive Medicine, Des Plaines, IL. Zwerling, C., Sprince, N.L., Wallace, R.B., et al., 1995. Effect of recall period on the reporting of occupational injuries among older workers in the Health and Retirement Study. Am. J. Ind. Med. 28 (5), 583–590.