Town-level comparisons may be an effective alternative in comparing rural and urban differences: a look at accidental traumatic brain injuries in North Texas children

full article:

Introduction

Injuries are the leading cause of death for individuals under 45 years of age, and are a significant public health problem in the USA¹. Specifically, traumatic brain injury accounts for almost one-third of all injury deaths². This problem is even more pronounced when looking specifically at the pediatric population³, where traumatic brain injury is the leading cause of morbidity and mortality^4-15 . Further, a myriad of problems exist when looking specifically at accessing the necessary medical care to treat traumatic brain injuries in rural-area children.

Access to healthcare services in rural areas is a well known problem, and includes limited availability of health services through the remote location, the isolation of rural areas as injury sites, difficulties with transportation, extended travel distances to healthcare facilities, lack of health insurance, and additional access problems associated with lower socioeconomic status¹⁶. This becomes even more problematic when looking at the pediatric populations in rural areas. With isolation and access to care already problematic for rural areas, rural children may be more isolated than their adult counterparts when considering the level of care needed to treat pediatric traumatic injuries. In interviews conducted by Robertson in 2009, Emergency Medical Service (EMS) providers indicated that while time and distance are known problems plaguing rural areas, children sustaining traumatic brain injury are generally best served at high-level trauma centers specializing in pediatric care¹⁷. Additionally, pediatric traumatic brain injuries require direct transport to these high-level trauma centers, receiving little to no benefit from being treated at local lower-level hospitals. Where adults can be treated for similar injuries at these lower-level hospitals, such centers are generally not equipped or staffed to effectively manage severe brain trauma in children^18,19. While high-level pediatric trauma hospitals are better equipped to treat the severity and unique presentation of a pediatric injury, they are generally located in urban centers. As a result, fewer appropriately equipped hospitals are available to rural children who, because of this, may need to endure extended travel time to access an appropriate level of care.

While many health service studies examine differences at the rural and urban county levels, relatively few studies exist that compare resources and service provision according to town size. It stands to reason that larger towns will have better access to funds, are closer in proximity to the higher levels of health services found in the urban centers, and that individuals injured in such towns may benefit from these two factors. Henderson and Taylor posited that technological advances in healthcare delivery further divide the delivery of care between rural and urban areas due to resource differences at the town level²⁰.

This report is a follow-up to Robertson's 2009 study of severe traumatic brain injury in children admitted to the Children's Medical Center Dallas¹⁷, one of several level I trauma centers in North Texas, and the only trauma center in the region specializing in children during the time of the data collection. Data collected and analyzed at the county level for a previous study were re-analyzed here to examine injury severity and outcome between town sizes and determine if significant differences can be established at this finer geo-demographic level. Given that rural areas are more isolated, have limited funds, and limited resources, it is assumed that these problems are compounded at the town level. It is then hypothesized that comparisons between town sizes will yield more significant differences in injury severity and outcome than comparisons between county designations.

Methods

Following Institutional Review Board approval at both UT Southwestern Medical Center and the University of Texas at Dallas, data were collected on all patients seen at the Children's Medical Center Dallas for severe traumatic brain injury. The Children's Medical Center Dallas trauma census was utilized to collect data over a five-year period. Inclusion criteria were based on a previous study¹⁷, using ICD-9 codes 800, 801, 802, 804, 850-854, and 959.01. Medical charts were reviewed to gather data on patient age, sex, county of injury, designation of the county as either rural or urban, the Rural-Urban Commuting Area 2 (RUCA 2) code for the patient's residence, the type of injury as accidental or non-accidental, whether the patient had a localized head injury or if the injury affected other body systems, the Glasgow Coma Scale (GCS) at the scene of injury, GCS on admission, Injury Severity Score (ISS) on admission, the Trauma Score and Injury Severity Score (TRISS), Trauma Score, length of hospital stay (LOS), and outcome mortality (defined as whether the patient recovered to discharge from the hospital or died during the hospital stay). Patients were excluded if they did not have an accidental injury, or were identified as any form of abuse, assault, or injury through other purposeful means.

The GCS is a clinical measure of neurological functioning ranging from 3 to 15 and is based on eye movement, motor movement, and verbal appropriateness⁶. Severe traumatic brain injury is associated with GCS scores of less than 8^4,8. The ISS is another clinical measure of injury severity based on an examination of six different body systems^17,21, where higher ISS scores signal worse injuries. The TRISS is a third clinical measure predicting survivability of a particular injury^22,23, and Trauma Scores are the final clinical measure. Trauma Scores were collected and recorded by Trauma Services at Children's Medical Center Dallas. The calculation was reported previously¹⁷, but is a summed value of scores ranging from -1 to +2 on 6 additional physiological variables (patient size, airway, level of consciousness, blood pressure, fractures, and cutaneous examination). These values were added to a revised GCS score where points were assigned to a specific GCS range, where four points were assigned to a GCS range of 13-15, three were assigned to a GCS range of 9-12, two points were assigned to a GCS range of 6-8, one point assigned to a GCS range of 4 or 5, and no points assigned to a GCS of 3. Lower Trauma Scores are associated with worse injuries.

Urban-rural differences are often analyzed using county-level data, typically out of either convenience afforded by readily available and widely used Department of Agriculture Urban Rural Continuum codes²⁴, the unambiguous boundaries that identify its spatial characteristics, or the abundant supply of socio-economic data that can be accessed and used in an analysis of the types of differences of interest. A major confounding factor in the use of counties, however, is the frequent existence of urban centers within officially designated rural counties and rural fringe in otherwise urbanized counties. As an alternative to a county-level analysis employing problematic urban/rural designations, this analysis adopts an alternative geographic coding system, also a product of the Department of Agriculture, the RUCA2²⁵, a coding system that allows the identification and codification of towns according to size and activity patterns. Using the stratification methods previously employed by Robertson¹⁷, RUCA2 codes were broken down in the following manner to determine town sizes:

• Urban: 1, 1.2, 2.1, 3
• Large town: 4, 4.1, 4.2, 5, 5.2, 6, 6.1
• Small town: 7, 7.1, 7.3, 7.4, 8, 8.3, 9, 9.2
• Isolated town: 10, 10.2, 10.4, 10.5, 10.6.

Groups were then stratified by the size of the town into either small or large towns, and were compared for severity and outcome for accidental injuries. Urban and large towns were designated 'Larger Towns', and small and isolated towns were classified as 'Smaller Towns'. Data were analyzed using χ² for nominal data, and t-tests for continuous variables. All statistical analyses were conducted using SPSS v15.0 (SPSS; Chicago, IL, USA; www.spss.com) with p-values <0.05 considered statistically significant.

Results

In all, 444 patients of the original sample of 545 met the 'accidental injury' inclusion criterion for patients admitted to the Children's Medical Center Dallas with severe traumatic brain injuries. A significant majority of the injuries occurred in Texas, but patients injured in Oklahoma and Arkansas were also seen at the Children's Medical Center Dallas and were included in the analysis. Demographic information by city size is presented (Table 1). Smaller-town patients accounted for 8.6% (n = 38) of the study sample, while larger-town patients accounted for 91.4% (n = 406). The average age at presentation was 5.67 years, with smaller-town patients presenting almost a full year older than larger-town patients. No statistical differences between town sizes were found for age, the type of injury, sex, or discharge status.

Severity comparisons are shown for smaller and larger towns (Table 2) and rural and urban counties (Table 3). As hypothesized, comparisons between town sizes for GCS on admission, ISS, and LOS become even greater compared with observed differences between urban and rural counties; however, only LOS showed statistically significant differences at the 0.05 p-value for these variables. Conversely, Trauma Scores, GCS on scene and TRISS scores become less diverse when comparing town sizes, rather than for county type comparisons. Interestingly, while TRISS and LOS comparisons between county designations have p-values of 0.055, the p-values for these measures drop below the 0.05 significance level when drawing comparisons between town sizes. The ISS, however, are not statistically significant despite expanding from a difference of 1.72 between rural and urban counties to an almost 3 point difference in town-size comparisons.

Table 1: Demographic information



Table 2: Severity measures' comparison of small and large towns



Table 3: County severity measures according to rural and urban locationz¹⁷



Discussion

Access to healthcare in rural populations is a serious public health problem in the USA. The authors report that while 20% of the American population currently resides in a rural area, only 10% of physicians practice in these areas²⁶. Access to basic healthcare in rural areas is a problem, and several factors contribute to the actual use of these services, including convenience, cost, and time²⁷. Additionally, rural environments differ from urban environments not only in location and socioeconomic status, but also in the nature of work dynamics and the environmental conditions found in rural areas. Rural areas typically have more labor-intensive jobs that expose individuals to environmental and chemical hazards^28-30 , and these problems often extend to children, particularly concerning agricultural work^31-33 . It is estimated that 100 childhood deaths, and 23 000 childhood injuries are related to agriculture annually³³. Traumatic brain injury, accounts for almost 90% of all pediatric injury deaths⁶.

Rural areas also have the added problem of lower levels of funding and available resources compared to more urbanized areas. This results in limited funds to training EMS personnel, fewer resources to treat traumatic injuries, and a heavier reliance on lower-trained emergency medical staff and volunteer emergency workers^34,35. This is not to ignore the value of the contribution of volunteers in the rural health workforce, but as stated by Rogers et al³⁵, volunteers are by definition secondary EMS personnel. However, some authors dispute the benefit of being treated by a highly-trained paramedic due to delays in obtaining hospital care because of a paramedic's ability to treat emergencies and the subsequent increased time spent at the scene of the injury³⁶. Despite these arguments, the limited availability of funds and resources in view of the pressing healthcare needs in rural areas cannot be ignored.

Many existing studies compare rural and urban areas for injury severity, and many studies collect data on zip (postal) codes. However, few studies stratify the zip codes to draw direct comparisons between town sizes. One study compaing rural and urban differences collected zip code data for firearm deaths³⁷, but used county-level designations as the primary means of determining rurality. Additionally, some studies collected and compared zip codes directly for cases such as unintentional injuries³⁸, and violence associated with social deprivation³⁹. Likewise, the RUCA codes have been used in a number of studies, often broken down into the four classes previously mentioned (urban, large town, small town, and isolated town). These RUCA code classes have been examined for rural hospital outcomes¹⁶, nursing homes⁴⁰, EMS^41,42, and in comparing severity and outcome of pediatric traumatic brain injuries¹⁷.

Similarly, some authors have used RUCA codes to identify and compare differences between rural and urban areas in terms of seat-belt usage⁴³, outcome of injury in Medicare patients⁴⁴, quality of life studies⁴⁵, healthcare utilization⁴⁶, and mental health⁴⁷. Weeks et al went so far as to use RUCA codes to determine three levels of rurality: urban, suburban, and rural⁴⁵. Despite the use of RUCA codes to define rural and urban areas or to compare among zip code levels, an extensive review of the literature indicates that this is the first study using RUCA2 codes to compare the severity of injury outcome between small and large towns for traumatic brain injury.

One of the problems with this study is the selection bias from looking at only one hospital system, and the limited number of patients living in small and isolated towns in the North Texas area. Whereas Robertson found 18% of the study population was injured in a rural county¹⁷, this study finds only 9% of the study sample residing in small or isolated towns. While only having 9% of a study population fall into one comparison group leads to unequal groups, the key problem here may be related to the high number of urban counties in the Dallas area and the general size of the cities within those counties. For example, rural counties may contain large towns and urban cities, but urban counties rarely contain small or isolated towns. As Robertson noted, 59% of the patients residing in rural counties who suffered a severe traumatic brain injury lived in urban or large towns¹⁷. However, only 2% of the patients living in urban counties lived in rural or isolated towns, once again suggesting the problematic use of a simple urban-rural continuum. Additionally, the urban concentration of the Dallas area causes many of the towns anecdotally considered 'small' to be classified as larger towns, based solely on commutable distance to an urban center. More research using a larger sample size that encompasses more patients residing in small and isolated towns is necessary.

The urban makeup and spatial dimensions of the Dallas metropolitan area means that many of the patients from isolated or small towns must travel a substantial distance to high-level trauma centers should these emergencies arise. Additionally, because many local hospitals are not staffed or equipped to manage severe brain injuries in children, this equates to extended response times for EMS workers, longer traveling distances, and further isolation for the pediatric population¹⁷. As noted, while the severity of such injuries dictates that these patients are best served at specialized pediatric trauma centers, many of these patients may have difficulty reaching these hospitals, or may never reach them at all. Patients who need immediate emergency life-saving care, such as in the case of loss of airway or those who sustain cardiac arrest during transport, are immediately taken to the nearest hospital to preserve vital signs. Despite the injury or the ability to treat these injuries at lower-level trauma centers, these are two instances in which the patient's condition is best served at a the nearest medical center, as opposed to a Mobile Intensive Care Unit en route to a pediatric trauma center¹⁷.

The retrospective study design of this research does not allow for cause-and-effect analysis or a greater direction on establishing any public health programs geared toward injury prevention. Additionally, the retrospective nature of the study limits the availability to obtain certain information. Whereas Robertson drew comparisons between county of injury¹⁷, the specific zip code of injury occurrence could not be identified in this study. Consequently, questions can be raised regarding the geographic location of where the injury was actually sustained in relation to residence. Given the differences in service availability and resources afforded to different town sizes, the exact zip code of injury may be highly valuable as injuries may not always be sustained in the same zip code as the person's residence. Resource differences at the town level are key constraints in the need for town-level comparisons, and the inability to determine the exact location of injury may inadvertently affect the outcome.

Conclusion

The results of this study indicate that town-size comparisons may be an effective means to define demographic groups because of the similarity and continuity between small and isolated towns. Rural counties may include large and urban towns, while small and isolated towns are rarely found in urban counties. Injury severity comparisons at the county level may be inadvertently affected by the size of the town because of the resources available at the town level. Research on the severity of rural-county injuries and patient outcomes may show improved outcomes due to those injuries being sustained in urban towns located in rural counties. Likewise, data on urban-county injury severities may appear worse because of the injuries sustained in those smaller and less-equipped towns found within urban counties. However, a larger sample size is needed to better gauge the efficacy of these comparisons, and more research is needed to better understand the differences in the actual services available to the individual towns, and the nature of injuries in smaller towns and rural counties, particularly in terms of mechanism and cause of injury.

When it comes to improving service delivery and reducing heath disparities in rural and underserved areas, understanding the differences between towns and counties can help practitioners better create effective public health programs and interventions geared toward reducing the health problems that plague certain concentrated areas. Specifically, this differentiation in geographic designation is required because with respect to rural/urban designations, congruence is not assured between town and county experiences in types or causes of injuries. Additionally, the resources available to individual towns may not reflect the resources available at the county level, and understanding these differences is necessary to improve the delivery of healthcare services in these areas.

Acknowledgement

The Trauma Registry of Children's Medical Center Dallas, Dallas, Texas was the source of the data. This work was supported by a grant from the Perot Family Center for the Care of Brain and Nerve Injuries at Children's Medical Center Dallas.

References

1. Peek-Asa C, Zwerling C, Stallones L. Acute traumatic injuries in rural populations. American Journal of Public Health 2004; 94(10): 1689-1693.

2. Thurman D, Guerrero J. Trends in hospitalization associated with traumatic brain injury. JAMA 1999; 282(10): 954-957.

3. Pertidou E, Anastasiou A, Katsiardanis K, Dessypris N, Spyridopolous T, Trichopoulos D. A prospective population based study of childhood injuries: the Velestino town study. European Journal of Public Health 2005; 15(1): 9-14.

4. Calvert S, Miller H, Curran A, Hameed B, McCarter R, Edwards R et al. The King's Outcome Scale for Childhood Head Injury and injury severity and outcome measures in children with traumatic brain injury. Developmental Medicine & Child Neurology 2008; 50: 426-431.

5. Halldorsson J, Flekkoy K, Gudmundsson K, Arnkelsson G, Arnarson E. Urban-rural differences in pediatric traumatic head injuries: a prospective nationwide study. Neuropsychiatric Disease and Treatment 2007; 3(6): 935-941.

6. Atabaki S. Pediatric head injury. Pediatrics in Review 2007; 28(6): 215-224.

7. Da Dalt L, Marchi A, Laudizi L, Crichiutti G, Messi G, Pavanello L et al. Predictors of intracranial injuries in children after blunt head trauma. European Journal of Pediatrics 2006; 165: 142-148.

8. Ducrocq S, Meyer P, Orliaguet G, Blanot S, Laurent-Vannier A, Renier D et al. Epidemiology and early predictive factors of mortality and outcome in children with traumatic severe brain injury: experience of a French pediatric trauma center. Pediatric Critical Care Medicine 2006; 7(5): 461-467.

9. Keenan H, Bratton S. Epidemiology and outcomes of pediatric traumatic brain injury. Developmental Neuroscience 2006; 28: 256-263.

10. Ciurea A, Coman T, Ro?u L, Ciurea J, B?i?u S. Severe brain injuries in children. Acta Neurochirurgica 2005; 93(Suppl): 209-212.

11. Reid S, Roesler J, Gaichas A, Tsai A. The epidemiology of pediatric traumatic brain injury in Minnesota. Archives of Pediatric and Adolescent Medicine 2001; 155: 784-789.

12. Arnarson E, Halldorsson J. Head trauma among children in Reyjkavik. Acta Pædiatrica 1995; 84: 96-99.

13. Jennett B. Epidemiology of head injury. Archives of Diseases in Childhood 1998; 78: 403-406.

14. Brooke O. Delayed effects of head injuries in children. BMJ 1988; 296: 948.

15. Kraus J, Fife D, Cox P, Ramstein K, Conroy C. Incidence, severity, external causes of pediatric brain injury. American Journal of Diseases of Children 1986; 140: 687-693.

16. Bowman S, Zimmerman F, Sharar S, Baker M, Martin D. Rural trauma: is hospital designation associated with better hospital outcomes? Journal of Rural Health 2008; 24(3): 263-268.

17. Robertson B. Differences in service delivery and patient outcome between rural and urban areas: the case of traumatic brain injury at a level I pediatric trauma center in North Texas. Ann Arbor, MI. UMI/ProQuest, 2010.

18. Rogers F, Osler T, Shackford S, Martin F, Healey M, Pilcher D. Population-based study of hospital trauma care in a rural state without a formal trauma system. Journal of Trauma 2001; 50(3): 409-414.

19. Edge W, Kanter R, Weigle C, Walsh R. Reduction of morbidity in interhospital transport by specialized pediatric staff. Critical Care Medicine 1994; 22(7): 1186-1191.

20. Henderson J, Taylor B. Rural isolation and the availability of hospital services. Journal of Rural Studies 2003; 19: 363-372.

21. Trauma.Org. Injury severity score. (Online) no date. Available: http://www.trauma.org/archive/scores/iss.html (Accessed 15 June 2009).

22. Davis D, Serrano J, Vilke G, Sise M, Kennedy F, Eastman A et al. The predictive value of field versus arrival Glasgow Coma Score and TRISS calculations in moderate-to-severe traumatic brain injury. Journal of Trauma 2006; 60(5): 985-990.

23. Chawda M, Hildebrand F, Pape H, Giannoudis P. Predicting outcome after multiple trauma: which scoring system? Injury: International Journal of the Care of the Injured 2004; 35: 347-358.

24. US Department of Agriculture. Measuring rurality: rural-urban continuum codes. (Online) 2004. Available: http://www.ers.usda.gov/Briefing/Rurality/RuralUrbCon (Accessed 24 February 2010).

25. US Department of Agriculture. Measuring rurality: rural-urban commuting area codes. (Online) 2005. Available: http://www.ers.usda.gov/briefing/Rurality/RuralUrbanCommutingAreas/ (Accessed 24 February 2010).

26. Arcury T, Gesler W, Preisser J, Sherman J, Spencer J, Spencer J et al. The effects of geography and special behavior on healthcare utilization among residents of a rural region. Health Services Research 2005; 40(1): 135-155.

27. Kennedy V. Locational aspects of medical care-seeking in a rural population, with some implications for public policy research. Journal of Health Politics, Policy and Law 1980; 5(1): 142-151.

28. Ruhm C. Are recessions good for your health? Quarterly Journal of Economics 2000; 115(2): 617-650.

29. Findeis J, Snyder A, Jayaraman A. The well-being of U.S. farm worker employee benefits, public assistance, and long-term effects. Review of Agricultural Economics 2005; 27(3): 361-368.

30. Variyam J, Mishra A. The well-being of U.S. farm workers: a look at health. Review of Agricultural Economics 2005; 27(3): 369-376.

31. Zietlow S, Swanson J. Childhood farm injuries. The American Surgeon 1999; 65(7): 693-698.

32. Gerberich S, Gibson R, French L, Renier C, Lee T, Carr W et al. Injuries among children and youth in farm households: Regional Rural Injury Study I. Injury Prevention 2001; 7: 117-122.

33. Cherry D, Huggins B, Gilmore K. Children's health in the rural environment. Pediatric Clinics of North America 2007; 54: 121-133.

34. Rogers F, Osler T, Shackford S, Cohen M, Camp L, Lesage M. Study of the outcome of patients transferred to a Level I hospital after stabilization at an outlying hospital in a rural setting. Journal of Trauma 1999; 46(2): 328-333.

35. Rogers F, Shackford S, Osler T, Vane D, Davis J. Rural trauma: the challenge of the next decade. Journal of Trauma 1999; 47(4): 802-821.

36. Gonzalez R, Cummings G, Phelan H, Mulekar M, Rodning C. On-scene intravenous line insertion adversely impacts prehospital time in rural vehicular trauma. American Surgeon 2008; 74(11): 1083-1087.

37. Dresang L. Gun deaths in rural and urban settings: recommendations for prevention. Journal of the American Board of Family Practice 2001; 14(2): 107-115.

38. Tarrant County Community Health Report. Monitoring & Assessment Project. (Online) 2003. Available: http://www.tarrantcounty.com/ehealth/lib/ehealth/Injuries.pdf (Accessed 22 February 2010).

39. Gruenewald P, Freisthler B, Remer L, LaScala E, Treno A. Ecological models of alcohol outlets and violent assaults: crime potentials and geospatial analysis. Addiction 2006; 101: 666-667.

40. Phillips C, Hawes C, Williams M. Nursing Homes in Rural and Urban Areas, 2000. College Station, TX: Texas A&M University Health Science Center, School of Rural Public Health, Southwest Rural Health Research Center, 2003.

41. Rural Health Resource Center. An alternative approach to defining rural for the purposes of providing Emergency Medical Services (EMS). (Online) 2004. Available: http://www.ruralcenter.org/sites/default/files/EMS_Definition_Of_Rural_Complete_0.pdf (Accessed 16 November 2009).

42. Murdock T, Knapp J, Dowd M, Campbell J. Bridging the emergency medical services children information gap. Archives of Pediatric and Adolescent Medicine 1999; 153: 281-285

43. Chang B, Ebel B, Rivara F. Child passenger safety: potential impact of the Washington State booster seat law on childcare centers. Injury Prevention 2002; 8: 284-288.

44. Gorra A, Clark D, Mullins R, MDeLorenzo M. Regional variation in hospital mortality and 30-day mortality for injured Medicare patients. World Journal of Surgery 2008; 38: 954-959.

45. Weeks W , Kazis L, Shen Y, Cong Z, Ren X, Miller D et al. Differences in health-related quality of life in rural and urban veterans. American Journal of Public Health 2004; 94(10): 1762-1767.

46. Weeks W, Lee R, Wallace A, West A, Baglan J. Do older rural and urban veterans experience different rates of unplanned readmission to VA and non-VA hospitals? Journal of Rural Health 2009; 25(1): 62-69.

47. Uphold C, Rane D, Reid K, Tomar S. Mental health differences between rural and urban men living with HIV infection in various age groups. Journal of Community Health 2005; 30(5): 355-375.