Clinical and Health Affairs
Epidemiology of Emergency Department-Treated Traumatic Brain Injury in Minnesota
By Heather Day, R.N., M.P.H., Jon Roesler, M.S., Anna Gaichas, M.S., and Mark Kinde, M.P.H.
Abstract
A number of population-based studies have been published on mortality and hospitalizations associated with traumatic brain injury (TBI). However, very little has been published on treatment of TBI in the emergency department (ED), despite the fact that most cases are seen in such settings. Minnesota is 1 of 2 states funded by the Centers for Disease Control and Prevention since 2001 to conduct surveillance of TBIs treated in EDs. A sample of medical records from 2002 and 2003 was reviewed to better understand the epidemiology of ED-treated TBI and identify risk factors for and interventions to prevent them. We found males, infants, adolescents and young adults, blacks, and residents of the 7-county Twin Cities metropolitan area were most likely to be seen in the ED for TBI. Sports and recreational activities were the leading causes of these injuries, followed by falls and motor vehicle crashes.
The Centers for Disease Control and Prevention (CDC) has estimated that approximately 80% of patients with traumatic brain injuries (TBIs) are treated in and released from hospital emergency departments (EDs).1 In 2003, the CDC published a report that stated that “mild TBI is most often treated in emergency departments or in nonhospital medical settings, or not treated at all.”2 Although TBI is recognized as a significant cause of mortality and nonfatal hospitalized morbidity, few population-based studies have been published on TBI cases treated in emergency departments.3-6 Since 2001, the CDC has provided funding to 2 states—Minnesota and South Carolina—to conduct state-level surveillance of TBIs treated in EDs.
In September 2005, the Minnesota Department of Health’s Injury and Violence Prevention Unit published its first report on TBIs treated in EDs in Minnesota.7 Department of Health investigators applied the CDC operational TBI case definition (Table 1) to outpatient electronic Universal Bill-1992 version (UB-92) data provided by the Minnesota Hospital Association. The UB-92 is primarily used for billing purposes and does not include clinical details (eg, length of post-traumatic amnesia) and demographic information (eg, race).
As of 2003, 117 (91%) of 128 facilities with 24-hour EDs were reporting data to the Minnesota Hospital Association. Using data from the Department of Health’s Health Care Cost Information System, which provides information about hospital utilization and services, we estimated that their reports represented approximately 94% of all ED visits. Findings included estimates of the annual incidence of the diagnosis “head injury, unspecified” and of all other TBI-related diagnoses treated in EDs. Rates were calculated by gender, age group, county of residence, leading causes, season, and associated treatment charges by age group and year of treatment.
A random sample of cases was drawn for medical review from hospital discharge data (diagnostic code, age, gender, ZIP code, etc.) from 2002 and 2003 TBI cases treated in EDs in order to capture additional information and determine how many cases with TBI diagnostic codes actually had TBI clinical indicators (eg, skull fracture and altered mental status). Building on the previous study, we sought to better understand 1) the epidemiology of TBIs treated in Minnesota EDs and 2) the causes of TBI cases seen in EDs in order to design intervention activities to prevent them.
Methods
The CDC’s TBI operational case definition also was used for this study.8 In each of the 2 study years—2002 and 2003—a stratified random sample of 1,500 cases was selected for medical record review, yielding a total of 3,000 cases. One thousand had TBI diagnostic codes; 500 were coded as “head injury, unspecified.” This diagnosis is frequently assigned and is included in the CDC’s TBI operational case definition. However, previous analysis of 2001 data on TBIs treated in EDs showed it was not a good predictor of the true occurrence of TBI. Patients who died, who were admitted to the hospital or transferred to another facility, and who were residents of other states were excluded from the analysis, as were individuals with brain injuries caused by trauma-related anoxia and patients with brain dysfunction caused by congenital or degenerative disorders or birth trauma.
Abstraction forms and accompanying data dictionaries were developed using existing CDC central nervous system data submission standards as were TBI clinical indicators (skull fracture, intracranial lesion, neurological abnormalities, neuropsychological abnormalities, amnesia, level of consciousness, and physician diagnosis).7,8 A medical record review was conducted by 1 trained full-time nurse abstractor who had limited assistance from 3 trained contract abstractors. Cause and intent data were obtained using external cause-of-injury codes (E-codes) when available. For cases without E-codes, cause and intent of injury were assigned based on the medical record review.
We linked cases to UB-92 data based on medical record number, facility, patient’s date of birth, etc. Information was entered into a Microsoft Visual FoxPro database. Data quality checks were performed to look for nonsense values prior to the creation of a final dataset. Data analysis was conducted using SAS version 9.1 and Microsoft Excel 2000.
Data from the medical record review were used to determine the predictive value positive of the operational case definition (using the TBI clinical indicators) and to calculate the number of true positive sampled ED-treated TBI cases for the 2-year period. Predictive value positive is the proportion of true-positive cases with TBI clinical indicators (such as intracranial lesion) among individuals who have a diagnostic code included in the CDC’s TBI operational case definition; it is expressed as a percentage. Each true-positive ED-treated TBI case was assigned a weight to represent a proportion of the number of ED-treated TBI cases in Minnesota. Denominator data were calculated using 2000 census data from the state demographer’s office.9 Annualized incidence rates per 100,000 population by year, age-adjusted to the year 2000 U.S. population standard, were calculated using weighted proportions along with accompanying 95% confidence intervals.10,11
Findings
A total of 2,737 ED records were reviewed, of which 1,854 were true-positives based on the presence of one or more TBI clinical indicators (predictive value positive, 49%). We were unable to review 212 medical records, and 51 cases were excluded because of hospitalization or death. Based on weights assigned to true-positive cases, an estimated 9,300 ED-treated TBI incident cases occurred annually among Minnesota residents during 2002 and 2003.
Overall age-adjusted annualized rates by gender, cause, intent, location, race, and ethnicity are presented in Table 2. Rates differed by location, and subsequent analyses found that ED-treated TBI rates in the 7-county Twin Cities metro area (Anoka, Carver, Dakota, Hennepin, Ramsey, Scott, and Washington counties) by cause and intent were significantly higher than in greater Minnesota. Sports and recreational activities were the primary cause of ED-treated TBI among adolescents ages 15 to 19 years, with the rate being 454.7/100,000 population for males and 141.6/100,000 for females.
The sports and recreational activities most frequently associated with TBIs were biking, hockey, football, snowboarding, baseball/softball, playing on playground equipment, basketball and soccer (tie), horseback riding, and ice skating. The ranking differed by location. In the metro area, the 6 sports and recreational activities most often connected with TBIs were biking, snowboarding, hockey, football, soccer, and baseball/softball; in greater Minnesota, they were biking, football, hockey, baseball/softball, snowboarding, and riding all-terrain vehicles (ATVs). All-terrain vehicle- and snowmobile-related injuries were twice as frequent among greater Minnesota residents, while biking- and snowboard-related injuries were more than 1 1/2 times more common among metro-area residents.
Fall-related TBIs were most common among children younger than 5 years of age, with a slight increase again occurring among adults 80 years and older. Motor vehicle-related ED-treated TBIs were relatively constant across the age span, with a small peak occurring among 15- to -29-year-olds (Figure).
Looking at age group and gender, the overall rate for males (227.6/100,000) was higher than that for females (146.2/100,000); this difference was most pronounced among 15- to 19-year-olds, where males were more than twice as likely to be treated for TBI in the ED than females (708.6/100,000 for males; 296.5/100,000 for females).
Information on race was found in the medical records of approximately 83% of the true-positive cases, and ethnicity was assigned for 93% of true-positive cases. Sports and recreation was the leading cause of ED-treated TBI for whites, assaultive injuries for American Indians/Alaska Natives, and falls for blacks, Hispanics/Latinos, and Asian/Pacific Islander/Native Hawaiians.
Conclusion
This study was subject to some limitations that may have biased the results. First, coverage of the surveillance was not quite complete, with 117 of the 128 Minnesota hospitals with 24-hour EDs reporting data to the Minnesota Hospital Association during 2002 and 2003. Second, 212 (7%) records from the original sample were unavailable for review. Finally, it was not possible to consistently find some data elements (eg, race, alcohol use, use of personal protective equipment) during the review.
Our findings confirm that TBIs treated in the ED accounted for approximately 63% of all TBI cases reported in Minnesota during the study period—considerably less than the CDC’s previous estimate. To the best of our knowledge, this is the first study in the United States to report a rate for ED-treated TBI using only true-positive cases and using all possible TBI codes, including “head injury, unspecified.” Despite the limitations of the surveillance system, striking differences in the incidence of ED-treated TBI were seen by gender, age group, cause, intent, and location. Males ages 15 to 19 years were found to be at particular risk. Our findings underscore the importance of a recommendation in the American Medical Association’s Guidelines of Adolescent Preventive Services that all adolescents should receive health guidance annually to promote the reduction of injuries.13 One way they can reduce the potential for injury is by wearing seat belts when driving or riding in a car.
The increased incidence of TBI treated in EDs among residents of the 7-county metro area bears further investigation. Data for Minnesota residents who sought treatment for their TBI in facilities outside of Minnesota were not available. Previous analysis by the Department of Health estimated that approximately 480 cases (<3%) with TBI diagnostic codes are lost to “border leakage” each year.7 At this time, the Department of Health does not receive information about urgent care or clinic visits; therefore, it is not possible to assess whether more residents in greater Minnesota are going to a clinic rather than the hospital ED for treatment of their traumatic brain injuries or if they are less likely to seek treatment for such injuries at all.
Sports- and recreation-related ED-treated TBI was especially common among children and adolescents. This finding presents a unique challenge, given the emphasis to increase physical activity among children in order to prevent obesity. How do we keep youth safe while they engage in sports and recreational activities? Helmets are recommended for 6 of 10 leading activities shown to result in TBIs: biking, hockey, football, baseball/softball (batting and catching), snowboarding, and horseback riding. In addition, helmets have been demonstrated to reduce the risk of head injury while skateboarding, in-line skating, riding a scooter, riding an ATV, snowmobiling, and skiing.14-18
Consistent helmet use among children and youths during these activities is an ongoing challenge. One recently published randomized controlled trial at Children’s Hospitals and Clinics of Minnesota found that giving young patients a free helmet in the pediatric ED appears to be an effective hospital-based strategy for increasing helmet use.19 Similarly, health care providers need to be knowledgeable about return-to-play recommendations for participants in competitive activities. This is especially important given that children and adolescents take longer to recover from concussions than adults, have a greater chance of sustaining a subsequent concussion during recovery, and are at risk for second-impact syndrome—a rare, but often fatal, complication of concussion.20-23
There are a number of ways to prevent falls in children age 5 and younger and in older adults. Several of these strategies are included within the Minnesota Department of Health’s Home Safety Checklist (www.health.state.mn.us/divs/fh/mch/fhv/safety.html). The CDC also has helpful information on its Web site about fall prevention for seniors (www.cdc.gov/ncipc/duip/spotlite/falltips.htm#3) that focuses on strengthening through exercise, modifying the home, having a health care provider examine medication use, and having regular vision exams. Ways to prevent motor vehicle- and traffic-related ED-treated TBI include seat belt use, proper installation and use of car seats, use of booster seats for children ages 4 to 8 years, and not driving when under the influence of alcohol or drugs, or while drowsy or distracted (eg, when using a cell phone). Injuries caused by assault have complex etiologies, and interventions that have proved to reduce these injuries are few. It is notable, however, that public health nursing visits for mothers of high-risk infants decrease both child maltreatment and youth violence.24
Minnesota anticipates continued CDC funding for its ED-treated TBI surveillance efforts through 2010. This will allow for further refinement of surveillance methodology and evaluation of trends by age group, gender, cause, intent, etc., along with the identification of cause-specific risk factors such as type of activity (eg, hockey practice versus a league game). Based on recommendations in the CDC’s report to Congress on mild TBI in the United States, future surveillance activities will also help determine the level of severity (ie, mild versus moderate/severe) through analysis of clinical data such as neuroimaging results, Glasgow Coma Scale (GCS) score, length of time unconscious, and length of amnesia. Moreover, it is hoped that data collection will improve over time for information that has been difficult to obtain during medical record reviews such as race, use of personal protective equipment, and alcohol use.
Although the proportion of TBIs that are treated in EDs is lower than previous estimates, ED-treated TBI still represents the majority of medically treated TBI cases. It is important for physicians to screen for and recognize TBI in the ED because some patients, often referred to as the “miserable minority,” will experience untoward outcomes such as persistent headache, vision problems, and learning and memory difficulties.25 Appropriate recognition and documentation of TBI could help in later treatment.
Our results underscore both the frequency and significance of ED-treated TBI and identify several groups that would particularly benefit from guidance on how to prevent it. It is our hope that future ED-treated TBI surveillance activities will garner additional information on severity and about modifiable risk factors that will help Minnesota physicians and health care providers recognize the problems associated with TBI earlier and prevent such injuries from happening in the first place. MM
This study was supported by cooperative agreement No. U17/CCU519419-04 from the National Center for Injury Prevention and Control, CDC. Its contents are solely the responsibility of the authors and do not necessarily represent official views of the CDC. The work was also supported in part by the Minnesota Department of Health TBI /SCI Registry.
The authors wish to thank the CDC for the opportunity to conduct surveillance of ED-TBI, and specifically Joe Russel, Angela Marr, M.P.H., and Victor Coronado, M.D., from the CDC NCIPC for their advice and encouragement. We also wish to thank Sara Westberg for her tireless and meticulous review of medical records for this manuscript, Curtis Fraser and Mark Phillips for their information technology expertise, Dan Vesey for his careful and timely data entry, Maureen Holmes for her thorough review of the manuscript, and the Minnesota Trauma Data Bank Advisory Committee for its ongoing feedback.
Heather Day is an epidemiologist, Jon Roesler is the epidemiologist supervisor, and Anna Gaichas is a research analyst with the Minnesota Department of Health’s Injury and Violence Prevention Unit. Mark Kinde is an epidemiologist with and supervisor of the unit.
References
1. Langlois JA, Rutland-Brown W, and Thomas KE. Traumatic Brain Injury in the United States: Emergency Department Visits, Hospitalizations, and Deaths. Atlanta, Centers for Disease Control and Prevention, National Center for Injury Prevention and Control; 2004.
2. National Center for Injury Prevention and Control. Report to Congress on Mild Traumatic Brain Injury in the United States: Steps to Prevent a Serious Public Health Problem. 2003. Atlanta (GA): Centers for Disease Control and Prevention.
3. Minnesota Department of Health. Nonfatal Hospitalized Traumatic Brain Injury, Minnesota, 1998-2003. St. Paul, Minn. Department of Health Injury and Violence Prevention Unit; 2005.
4. Langlois JA, Kegler SR, Butler JA, et al. Traumatic brain injury-related hospital discharges: results from a 14-state surveillance system, 1997. MMWR Surveill Summ. 2003;52(SS04):1-18.
5. Thurman DJ, Adedkoya N, Coronado VG, Roesler J. Epidemiological studies of traumatic brain injury in the United States. Brain Injury Source. 2002;6:12-7.
6. Thomas C, Butler J, Davies M, Johnson R. State Injury Indicators Report, Second Edition—1999 Data. Atlanta, Centers for Disease Control and Prevention, National Center for Injury and Violence Control; 2004.
7. Minnesota Department of Health. Emergency Department-Treated Traumatic Brain Injury, Minnesota 1998-2003. St. Paul, Minn. Department of Health Injury and Violence Prevention Unit; 2005.
8. Marr A, Coronado V. Central Nervous System Injury Surveillance Data Submission Standards. Atlanta, Center for Disease Control and Prevention, Center for Injury Prevention and Control; 2004.
9. Minnesota Department of Administration, Office of Geographic and Demographic Analysis, Land Management Information Center, 2000 Census SF1 and SF3: report and mapping menu: http://www.lmic.state.mn.us/datanetweb/php/census2000/c2000.html. Accessed March 1, 2006.
10. Minnesota Department of Health. Report of Injury, Version 6 Data Dictionary. St. Paul, Minn. Minnesota Department of Health Injury and Violence Prevention Unit; 7-1-2003.
11. Klein RJ, Schoenborn CA. Age adjustment using the 2000 projected U.S. population 2000. Healthy People Statistical Notes. 2001; Jan (20); 1-9.
12. Vyrostek SB, Annest JL, Ryan GW. Surveillance for fatal and nonfatal injuries—United States, 2001. MMWR Surveill Summ. 2004;53(SS07):1-57.
13. Montalto NJ. Implementing the guidelines for adolescent preventive services. Am Fam Physician. 1998;57(9):2181-90.
14. Schieber RA, Branche-Dorsey CM, Ryan GW. Comparison of in-line skating injuries with rollerskating and skateboarding injuries. JAMA. 1994;271(23):1856-8.
15. Committee on Injury and Poison Prevention, American Academy of Pediatrics. Skateboard and scooter injuries. Pediatrics. 2002;109(3):542-3.
16. Murphy N, Yanchar NL. Yet more pediatric injuries associated with all-terrain vehicles: should kids be using them? J Trauma. 2004;56(6):1185-90.
17. Nayci A, Stavlo PL, Zarroug AE, Zietlow SP, Moir CR, Rodeberg DA. Snowmobile injuries in children and adolescents. Mayo Clin Proc. 2006;81(1):39-44.
18. Sulheim S, Holme I, Ekeland A, Bahr R. Helmet use and risk of head injuries in alpine skiers and snowboarders. JAMA. 2006;295(8):919-24.
19. Wu BC, Oakes JM. A randomized controlled trial of sport helmet interventions in a pediatric emergency department. Pediatr Emerg Care. 2005;21(11):730-5.
20. Field M, Collins MW, Lovell MR, Maroon J. Does age play a role in recovery from sports-related concussion? A comparison of high school and collegiate athletes. J Pediatr. 2003;142(5):546-53.
21. Collins MW, Iverson GL, Lovell MR, McKeag DB, Norwig J, Maroon J. On-field predictors of neuropsychological and symptom deficit following sports-related concussion. Clin J Sport Med. 2003;13(4):222-9.
22. Collins MW, Hawn KL. The clinical management of sports concussion. Curr Sports Med Rep. 2002;1(1):12-22.
23. Solomos NJ. Management guidelines for sport-related concussions. Am Fam Physician. 2002;65(12):2435-6
24. Olds DL, Kitzman H, Cole R, et al. Effects of nurse home-visiting on maternal life course and child development: age 6 follow-up results of a randomized trial. Pediatrics. 2004;114(6):1550-9
25. Bohnen N, Jolles J, Twijnstra A. Neuropsychological deficits in patients with persistent symptoms six months after mild head injury. Neurosurgery. 1992;30:692-5.
26. Bazarian JJ, Wong T, Harris M, Leahy N, Mookerjee S, Dombovy M. Epidemiology and predictors of post-concussive syndrome after minor head injury in an emergency population. Brain Inj. 1999;13:173-89.
27. Ruff R. Two decades of advances in understanding of mild traumatic brain injury. J Head Trauma Rehabil. 2005;20:5-18.