| Home | E-Submission | Sitemap | Contact Us |  
Search
Clin Exp Emerg Med Search

CLOSE

Clin Exp Emerg Med > Volume 10(4); 2023 > Article
Hwang, Jeong, and Jo: Relationships between trauma death, disability, and geographic factors: a systematic review

Abstract

Objective

Trauma is a global health problem. The causes of trauma-related deaths are diverse and may depend in part on socioeconomic and geographical factors; however, there have been few studies addressing such relationships. The aim of this study was to investigate the relationships between trauma and geographical factors in order to support policy recommendations to reduce trauma-related deaths and disability.

Methods

In accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we searched international and Korean databases to retrieve relevant literature published between 2000 and 2020.

Results

Thirty-two studies showed a positive relationship between the outcomes of major trauma and geographical factors. The study investigated regional factors including economic factors such as size of urban areas, gross domestic product, and poverty rate, as well as hospital parameters, such as presence of trauma centers and number of hospital beds. There was a tendency toward higher mortality rates in rural and low-income areas, and most of the studies reported that the presence of trauma centers reduced trauma-related mortality rates.

Conclusion

Our study showed that geographic factors influence trauma outcomes. The findings suggest geographical considerations be included in care plans to reduce death and disability caused by trauma.

INTRODUCTION

Trauma is a global health problem [1], causing a reported 58 million global deaths every year, which represents about 10% of all deaths [2]. Since trauma was recognized as a critical cause of death and disability in patients, multiple studies of trauma-related mortality and disability have been published [3]. According to an annual report summarizing the causes of death published by Statistics Korea, 304,948 Koreans died from trauma in 2020. Among these, 26,442 (8.7%) died of external causes such as suicide, traffic accidents, and falls [4]. When expressed as incidence per 100,000 people, 25.7 died from suicide, 7.7 from traffic accidents, and 5.2 from falls. Worldwide, trauma is one of the leading causes of death, especially in the working age group (15 to 44 years), with many of the fatalities resulting from traffic accidents and self-harm. Socioeconomic losses due to trauma account for 60% to 62% of the total socioeconomic costs, indicating trauma to be associated with a high socioeconomic burden [5].
In the past, trauma was perceived to be a greater issue among low-income people with poor medical access. The US Department of Health and Human Service Centers for Disease Control and Prevention reported that deaths from trauma are not only a matter of concern in low-income counties, but also in high-income countries [6]. Today, it is believed that trauma can happen to anyone but is often preventable: as such, there are ongoing efforts to reduce trauma cases [1] that include measurement of preventable trauma death rates. Preventable traumatic deaths are defined as trauma-related deaths that would have been avoided if trauma patients had received clinically “correct” diagnoses and optimal medical care [7]. There is a large body of research related to preventable traumatic death rates in Korea [7–10], reporting rates of 40.5%, 39.6%, and 39.6% of trauma-related deaths in 1997−1998, 2003−2004, and 2009−2010, respectively. Researchers have predicted that if the medical infrastructure in Korea was equivalent to that of advanced nations, the preventable death rates would fall to less than 20% [8–11]. However, the preventable death rates may vary by region within a given country. Even if a country has well-established trauma care systems and improves medical quality at a national level, the quality and quantity of care may vary between regions, especially with respect to prehospital and in-hospital systems, medical resources, and level of medical centers [7,12]. The accessibility of medical services and service quality may significantly affect an individual’s health [13]. The causes of trauma-related deaths may also vary regionally due to socioeconomic and geographical factors [3,14]. Many advanced nations have reported such a correlation between geographical factors and incidence of trauma [15,16]. There have been few Korean studies addressing the socioeconomic and geographical factors that influence trauma since most domestic studies have focused on trauma patterns, pattern analyses, and clinical outcomes related to trauma in certain body parts [17].
Based on the background above, the aim of the study was to investigate the relationships between trauma and geographical factors in order to support policies to reduce trauma-related deaths and disability. We performed a systematic review of published studies.

METHODS

The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) guidelines were used for this study [18]. To ensure objectivity, two researchers independently reviewed the retrieved articles based on the PRISMA flowchart. The two researchers reached a consensus through discussions when they disagreed with the assessment. A risk-of-bias assessment was conducted for the articles included in the final set of references.

Reference literature

The research was conducted with literature published in Korean databases and several international databases from January 1, 2000 to December 31, 2020: RISS (Research Information Sharing Service), KISS (Korean Studies Information Service System), KMbase (Korean Medical Database), KoreaMed, KNbase (Korean Nursing Database), National Assembly Library of Korea, PubMed, MEDLINE (via Ovid), Cochrane Library, ClinicalKey, CINAHL (Cumulative Index to Nursing and Allied Health Literature), and EBSCOhost, with the last date of search June 27, 2021.
Databases were selected based on systematic literature reviews. The search terms (both in Korean and English) were as follows: (1) “urbanization” AND “trauma” AND “outcome” OR “injury and wounds” AND “outcome” OR “trauma” AND “Glasgow Outcome Scale (GOS)” OR “injury and wound“ AND “GOS”; (2) “geographic” AND “trauma” AND “outcome” OR “injury and wounds” AND “outcome” OR “trauma” AND “GOS” OR “injury and wound” AND “GOS”; (3) “community” AND “trauma” AND “outcome” OR “injury and wounds” AND “outcome” OR “trauma” AND “GOS” OR “injury and wound” AND “GOS.”

Literature selection

The selection of publications for inclusion was conducted in two phases. In the first phase, the researchers reviewed the title and abstract of each of the initially selected articles and then excluded case studies and papers that did not consider geographic and sociodemographic factors. In the second phase, the articles selected in the first phase were reassessed based on the following factors: (1) whether dependent variables were related to trauma patients; (2) whether geographic factors were considered independent variables and if the study clarified, what causes influenced the patient’s diagnosis of trauma; (3) whether the papers were published in an academic journal and analyzed correlations between trauma and regional factors; and (4) whether the papers met the purpose of this study.

Risk-of-bias assessment

A risk-of-bias assessment was conducted for the 32 studies selected for inclusion. The risk-of-bias assessment may also be called a quality assessment or critical appraisal of the included studies. The risk-of-bias assessment is needed to avoid any wrong conclusion due to heterogeneity, publication bias, etc. during a systematic literature review [19]. We used a checklist (QualSyst) developed by the Alberta Heritage Foundation for Medical Research to assess the risk of bias in the included quantitative studies. QualSyst is a verified systematic review tool and very practical and useful for various study designs [20,21]. The QualSyst checklist includes 14 items, but only 11 were used for this study [22].

RESULTS

Literature selection and risk-of-bias assessment

A total of 170,425 articles was searched, of which 81,264 were selected after excluding duplicate publications. After the first phase of selection, 141 articles remained. Of the 141 articles, 109 were excluded based on the criteria for the second phase of literature selection. The reasons for exclusion were as follows: (1) 12 articles dealt with dependent variables but were not related to trauma; (2) 86 articles did not consider geographical variables; (3) five articles did not assess the outcomes of trauma; and (4) six articles did not meet the purpose of this study. If an article failed for more than two reasons, the first factor we considered was whether the outcome was due to injury (in order of (1), (2) and (3)). Finally, we reviewed the remaining 32 articles published from January 1, 2010 to December 31, 2020 (Fig. 1).
We assessed risk of bias in the 32 studies based on the 11 items of the QualSyst checklist and scored them 0 (no), 1 (partial), or 2 (yes). The two reviewers assessed the 32 studies and compared the results. When disagreements occurred, consensus was reached through discussions involving a third author (Table 1).
QualSyst recommends exclusion of articles that do not meet the minimum threshold of 0.55 from a systematic review [23]. Some articles did not meet some criteria, but they were also reviewed since the total score of each of them was more than 0.99 (Supplementary Table 1) [16,2454].

Structure of the study

We found that 27 of the 32 studies (84.4%) [16,24,25,2831,34,35,3754] were published after 2010, indicating that geographic factors influencing major trauma have recently become a priority for study. Of the 32 studies, 18 studies [16,24,25,27,29,30,32,3943,45,4750,52] selected study subjects of all ages, eight studies [26,34,35,38,44,46,51,53] were adults, one [54] studied people in their 50s or older, and five [28,31,33,36,37] studied children and adolescents.

Data sources and methods of analysis

Table 2 summarizes the general characteristics of the 32 studies [16,2454]. Of the 32 studies, one [26] used data obtained from insurance companies, while all other used representative regional and national data; specifically, nine [16,31,34,4042,46,48,51] used national data and one [32] used cross-national data, and the remainder [2430,33,3539,4345,47,49,50,5254] used regional data. Of the 32 articles, there was one systematic review [38] and one scoping review [25]. Nineteen studies (63.3%) [16,24,26,28,30,31,36,37,40,41,4350,52] used data from the United States, four studies (13.3%) [34,39,42,51] from Asia, three studies (10.0%) [29,33,53] from Oceania, three studies (10.0%) [32,35,54] from Europe, and one study (3.3%) [27] from Africa.
The studies were assigned to one of five categories for assessment: 16 [27,29,30,33,3537,40,4345,4953] in individual, four [16,39,41,42] in regional, 10 [24,26,28,31,32,34,4648,54] in individual and regional, one [38] in systematic review, and one [25] in scoping review. The sample population sizes ranged from 614 to 8,673,213 subjects. Among the 32 articles, 23 [24,2631,3337,40,4348,5052,54] analyzed more than 1,000 subjects.
While some of the 32 studies used more than one data analysis method, the main method in six studies [24,27,34,43,51,54] was Poisson regression, 10 [26,28,29,32,35,36,40,44,46,49] used logistic regression, three [48,52,53] used chi-square test, and five [16,41,42,45,50] used geospatial analytics. The remaining studies [25,30,31,33,3739,47] used other methods such as multivariate or multiple regression.

Geographic factors as independent variables and use of dependent variables

To classify the papers, we conducted two rounds of categorization: one considering the possibility of multiple category duplication and another with the aim of avoiding duplication. To avoid duplication, three authors discussed and selected the important variables and classified them into the corresponding categories geographic factors as independent variables were divided into three categories. First, we classified the studies focusing on the connections between social structures and geographic factors as “socioeconomic.” Second, we categorized the studies focusing on medical resources as “medical-resource–related.” Finally, we categorized the studies focusing on environments where trauma is likely to occur as “environmental.” The results of the analyses showed that, with the exception of systematic and scoping review articles, five studies [30,37,39,44,51] were classified as socioeconomic, six [16,36,45,46,48,50] as medical-resource–related, and 19 [24,2629,3135,4043,47,49,5254] as environmental (Table 3).
In another round where we considered the possibility of duplication (Table 4) [16,24,2637,3954], we divided the socioeconomic category into three subcategories: geodemographic, economic, and individual. Five studies [24,39,41,47,49] analyzed geodemographic factors such as population density; seven [34,39,42, 43,46,51,53] analyzed economic factors such as gross domestic product, resident tax, house income, type of insurance, middle-class household income, poverty rate, and unemployment rate; and 11 [16,30,33,36,37,44,45,49,51,53,54] analyzed individual factors such as race, level of education, and per capita income.
The medical-resource–related category was divided into two subcategories: prehospital and in-hospital. Six [24,32,35,4446] addressed prehospital factors such as distance from the site to a trauma center, time from scene to medical institution, type of transfer vehicle, and the numbers of helicopters and helipads; while 12 [16,28,32,35,36,40,4346,48,50] analyzed in-hospital factors such as presence of a trauma center, level of trauma center, and number of beds.
The environmental category was also divided into three subcategories: road environment, geospatial factors, and legal factors. Among the 27 studies, one [39] addressed road characteristics of speed limit, type of road, road conditions, and roadway congestion index; 14 [26,27,29,3335,4044,5254] analyzed geospatial factors such as urbanization and area of the city, and one [31] addressed legal factors such as regulations on seat belts and safety seats, red-light and speed-camera policies, speed limits, and laws applicable to driving under the influence.

Outcomes of trauma

We found that six types of dependent variable (mortality, disability, severity, survival and discharge, death and impairment, and work disability) were used in the 32 studies included in the analyses. Among the 32 studies, 22 [16,25,2831,34,35,37,3945,48,49,5154] identified incidences of death and survival rate as dependent variables, and the rest used disability rate [27], death and impairment rate [28,38,47], severity of trauma [32,36], survival and discharge rates [24,33,46,50], and work disability rates [26]. Dependent variables were identified through medical records in 28 studies (91%) [16,24,2837,3954], through a survey in one study (3%) [27], through a large insurance company in one study (3%) [26], and the remaining two studies were a systematic literature review [38] and a scoping review [25].
More specifically, Wang et al. [36] referred to medical records to obtain trauma severity scores. Brown et al. [46] analyzed survival and discharge rates obtained from medical records. Moshiro et al. [27] used duration of impairment as reported through a survey. Young et al. [26] identified work disability through insurance records. Finally, Newnam et al. [38] conducted a systematic review for analyzing death and impairment. The above findings indicate that medical records are being widely used to assess the causes of major trauma (Table 5).

Correlations between outcomes of major trauma and geographic factors

We investigated the correlations between the outcomes and geographic factors and divided the subjects into four groups: all ages, adults, children, and middle-aged to elderly (Table 6). After excluding systematic and scoping review articles, the remaining 30 studies [16,24,2637,3954] showed a positive relationship between the outcomes of major trauma and geographic factors. The 19 studies [24,2629,3135,4043,47,49,5254] analyzing environmental factors also showed such positive relationships, as did five studies [30,37,39,44,51] analyzing socioeconomic factors and six studies [16,36,45,46,48,50] assessing medical resource factors.
Among the studies that examined factors affecting all age groups, environmental factors (12 studies [24,25,27,29,32,4043,47,49,52]), socioeconomic factors (two studies [30,39]), and medical resources (four studies [16,45,48,50]) were the main variables identified. Studies that compared trauma outcomes in rural or urban environments showed that rural areas had higher mortality rates than urban areas, and studies that compared regional economic status found that low-income regions had higher mortality rates than high-income regions [29,39,51]. Bhutiani et al. [49] reported that traffic fatality rates increased as population density decreased.
With regard to medical resource factors, the better was accessibility to trauma centers, the lower were the death rates. Both death and disability rates are lower if patients can be transferred to hospitals with helicopters.
Five studies [26,34,35,51,53] focused on geographic factors affecting adult major trauma. Of these, four [26,34,35,53] addressed environmental factors, while one [51] addressed socioeconomic factors. In the four articles, three [26,34,53] showed higher mortality rates in rural area than in urban ones but an opposite result in impairment rate, with Dinh et al. [35] reporting opposite result regarding death. The articles analyzing socioeconomic factors showed higher incidence rates of trauma in low-income regions.
Among the five studies focusing on geographic factors influencing pediatric trauma, three [28,31,33] analyzed environmental factors, one [37] analyzed socioeconomic factors, and one [36] analyzed medical resource factors. The three articles assessing environmental factors showed that the risk of pediatric trauma was greater in rural areas and mortality was greater in regions where safety regulations were not established. One article [54] analyzed the outcomes of major geriatric trauma and showed that the female mortality rate increased in all regions except the capital.
Although the findings of the 32 studies [16,2454] varied depending on factors such as region, accessibility to trauma centers, injury severity, and traffic congestion, they generally indicated that trauma outcomes differed according to the characteristic of the area in which the patients lived. For example, pedestrian fatality rates tended to be lower in high-income areas [49]. Also, trauma death rates were greater in low-income areas than other areas [44,49]. Additionally, the mortality rates for injuries and head trauma were greater in rural areas than in urban areas, and prehospital trauma mortality was higher in both urban and rural areas than in metropolises [25,41,43].

DISCUSSION

This study used a systematic literature review to retrieve articles published in Korea and foreign countries that analyzed the correlations between disability or death in major trauma and geographical factors. The purpose of this study was to establish and strengthen policies on major trauma.
In this study, medical resources, including emergency medical systems and trauma care, were critical regional factors that need to be well-established to ensure timely and effective medical care. Having a well-established emergency medical system, as well as adequate trauma care facilities and sufficient beds and personnel, can significantly impact outcomes of emergency situations and foster prompt and effective care for patients with severe injuries. Therefore, ensuring adequate medical resources is crucial for the well-being of a region’s population. Studies in the United States have shown that, after improving the emergency medical system, the preventable death rate was reduced [5557]. To that end, it is necessary for people to perceive trauma as a regional or national matter rather than an individual matter. When assigning emergency medical resources, few studies have recommended considering regional characteristics to decrease mortality due to major trauma and to establish integrated trauma care systems from injury to discharge [58,59].
In previous studies, it has been suggested that in order to reduce the disparities in mortality rates that are attributed to socioeconomic factors such as gender, race, and region, it is necessary to establish prevention policies and programs tailored to high-risk areas and populations at the national and regional levels [30,37,39,52]. Policies based on injury prevention are needed, as shown by the effectiveness of policy and enforcement interventions such as speed limits, traffic lights, speed-camera enforcement to reduce injury and mortality rates due to car accidents, and installation of smoke alarms to reduce fire-related incidents [31,33].
It appears that the impacts of medical resources and environmental factors on trauma outcomes are not attributed to a single factor, but rather a complex interplay of multiple factors. In order for such strategies to be realized, several factors should be considered. First, more domestic trauma case studies should be conducted, as there have been too few conducted in Korea to fully understand the risk factors for major trauma. Among the 32 studies [16,2454] assessed herein, only two [34,51] analyzed domestic cases. Second, case studies of impairment should be also conducted, as there have been significantly fewer domestic studies addressing impairment due to trauma than those addressing death due to trauma, In addition, rather than using objective indicators, the domestic case studies employed relatively subjective indicators, including surveys, number of work absences after accidents, and availability of hospice wards. Finally, articles published in other countries have used various geographic variables such as time of transfer, type of vehicle for transfer, number of beds, and variables related to trauma centers. In future studies, Korea should consider incorporating similar geographic variables.
This systematic review has a few limitations. First, we focused only on geographic variables and excluded other variables. Second, we categorized geographic factors into socioeconomic, environmental, and medical-resource–related categories, and articles analyzing other topics were not considered. Despite these limitations, this study is meaningful as it presents the trends in trauma care, as determined through variables, analytical methods, and research outcomes.
In conclusion, although further research is needed, our study showed that geographic factors, including socioeconomic, medical resource, and environmental factors, such as population density, region (e.g., metropolis, rural), income levels of the area, and presence of trauma centers, appear to influence trauma outcomes. In order to reduce the incidence of death and disability due to trauma, it is important to establish appropriate care plans that take into account geographical factors. One approach could be to develop regional trauma care systems that address specific treatment difficulties in different areas.

SUPPLEMENTARY MATERIAL

Supplementary materials are available from https://doi.org/10.15441/ceem.23.009.

Supplementary Table 1.

. Result of assess of study using QualSyst (Alberta Heritage Foundation for Medical Research)
ceem-23-009-supplementary-Table-1.pdf

NOTES

ETHICS STATEMENT
Not applicable.
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
FUNDING
None.
AUTHOR CONTRIBUTIONS
Conceptualization: BH, TJ; Data curation: BH, TJ; Formal analysis: BH, JJ; Methodology: BH, TJ; Writing–original draft: BH; Writing–review & editing: TW, JJ. All authors read and approved the final manuscript.

REFERENCES

1. Lee EJ, Lee JS, Kim Y, et al. Patterns of unintentional domestic injuries in Korea. J Prev Med Public Health 2010; 43:84-92.
crossref pmid
2. Peden M, McGee K, Sharma G. the injury chart book: a graphical overview of the global burden of injuries. World Health Organization; 2002.

3. Hong TH, Lee SH, Kim HW, Jung MJ, Lee JG. Patterns of in-hospital mortality and causes of death in blunt poly-trauma patients. J Acute Care Surg 2014; 4:67-72.
crossref
4. Statistics Korea. Causes of death statistics in 2020. Statistics Korea; 2021.

5. Park K, Lee JS, Kim Y, Kim YI, Kim J. The socioeconomic cost of injuries in South Korea. J Prev Med Public Health 2009; 42:5-11.
crossref pmid
6. Borse N, Sleet DA. CDC Childhood Injury Report: patterns of unintentional injuries among 0- to 19-year olds in the United States, 2000-2006. Fam Community Health 2009; 32:189.
pmid
7. Moon S, Lee SH, Ryoo HW, et al. Preventable trauma death rate in Daegu, South Korea. Clin Exp Emerg Med 2015; 2:236-43.
crossref pmid pmc pdf
8. Jung KY, Kim JS, Kim Y. Problems in trauma care and preventable deaths. J Korean Soc Emerg Med 2001; 12:45-56.

9. Kim Y, Jung KY, Cho KH, et al. Preventable trauma deaths rates and management errors in emergency medical system in Korea. J Korean Soc Emerg Med 2006; 17:385-94.

10. Kim H, Jung KY, Kim SP, et al. Changes in preventable death rates and traumatic care systems in Korea. J Korean Soc Emerg Med 2012; 23:189-97.

11. Lee KH. Optimal trauma care system in Korea. J Korean Med Assoc 2013; 56:748-50.
crossref
12. Cho HM. Proposal for stabilization of regional trauma centers in Korea. J Korean Med Assoc 2016; 59:931-7.
crossref pdf
13. Brown AF, Ettner SL, Piette J, et al. Socioeconomic position and health among persons with diabetes mellitus: a conceptual framework and review of the literature. Epidemiol Rev 2004; 26:63-77.
crossref pmid
14. Di Bartolomeo S, Sanson G, Michelutto V, et al. Epidemiology of major injury in the population of Friuli Venezia Giulia-Italy. Injury 2004; 35:391-400.
crossref pmid
15. van Beeck EF, Mackenbach JP, Looman CW, Kunst AE. Determinants of traffic accident mortality in The Netherlands: a geographical analysis. Int J Epidemiol 1991; 20:698-706.
crossref pmid
16. Brown JB, Rosengart MR, Billiar TR, Peitzman AB, Sperry JL. Geographic distribution of trauma centers and injury-related mortality in the United States. J Trauma Acute Care Surg 2016; 80:42-50.
crossref pmid pmc
17. Choi YE, Lee KS. Multilevel analysis on factors influencing death and transfer in inpatient with severe injury. Health Policy Manag 2013; 23:233-43.
crossref
18. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med 2009; 151:264-9.
crossref pmid
19. Kim SY. Evidence-based medicine and systematic review. Qual Improv Health Care 2006; 12:35-40.

20. Phillips F. Adolescents living with a parent with advanced cancer: a review of the literature. Psychooncology 2014; 23:1323-39.
crossref pmid pdf
21. Kim WJ, Shin YJ, Kim SY, Kim JD. Analysis of dental unmet needs medical research trends and influence factors: using structural literature review. J Korean Soc Dent Hyg 2020; 20:243-56.

22. Kwon RA, Shin SS, Shin YJ. A systematic review of studies on association between drinking behavior and alcohol availability. Health Soc Welf Rev 2017; 37:543-67.
crossref
23. Kmet LM, Cook LS, Lee RC. Standard quality assessment criteria for evaluating primary research papers from a variety of fields. Alberta Heritage Foundation for Medical Research (AHFMR); 2004.

24. Minei JP, Schmicker RH, Kerby JD, et al. Severe traumatic injury: regional variation in incidence and outcome. Ann Surg 2010; 252:149-57.
pmid
25. Keeves J, Ekegren CL, Beck B, Gabbe BJ. The relationship between geographic location and outcomes following injury: a scoping review. Injury 2019; 50:1826-38.
crossref pmid
26. Young AE, Wasiak R, Webster BS, Shayne RG. Urban-rural differences in work disability after an occupational injury. Scand J Work Environ Health 2008; 34:158-64.
crossref pmid
27. Moshiro C, Heuch I, Astrom AN, Setel P, Kvale G. Effect of recall on estimation of non-fatal injury rates: a community based study in Tanzania. Inj Prev 2005; 11:48-52.
crossref pmid pmc
28. Snyder CW, Chandler NM, Litz CN, Pracht EE, Danielson PD, Ciesla DJ. Immature patients in a mature system: regional analysis of Florida’s pediatric trauma system. J Trauma Acute Care Surg 2017; 83:711-5.
pmid
29. Fatovich DM, Phillips M, Langford SA, Jacobs IG. A comparison of metropolitan vs rural major trauma in Western Australia. Resuscitation 2011; 82:886-90.
crossref pmid
30. Zandy M, Zhang LR, Kao D, et al. Area-based socioeconomic disparities in mortality due to unintentional injury and youth suicide in British Columbia, 2009-2013. Health Promot Chronic Dis Prev Can 2019; 39:35-44.
crossref pmid pmc
31. Wolf LL, Chowdhury R, Tweed J, et al. Factors associated with pediatric mortality from motor vehicle crashes in the United States: a state-based analysis. J Pediatr 2017; 187:295-302.
crossref pmid pmc
32. Gomez de Segura Nieva JL, Boncompte MM, Sucunza AE, Louis CL, Segui-Gomez M, Otano TB. Comparison of mortality due to severe multiple trauma in two comprehensive models of emergency care: Atlantic Pyrenees (France) and Navarra (Spain). J Emerg Med 2009; 37:189-200.
crossref pmid
33. Poulos RG, Hayen A, Chong SS, Finch CF. Geographic mapping as a tool for identifying communities at high risk of fire and burn injuries in children. Burns 2009; 35:417-24.
crossref pmid
34. Kim KH, Noh MS, Ha ID. A study using HGLM on regional difference of the dead due to injuries. J Korean Data Inf Sci Soc 2011; 22:137-48.

35. Dinh MM, Curtis K, Mitchell RJ, et al. Major trauma mortality in rural and metropolitan NSW, 2009-2014: a retrospective analysis of trauma registry data. Med J Aust 2016; 205:403-7.
crossref pmid pdf
36. Wang NE, Saynina O, Kuntz-Duriseti K, Mahlow P, Wise PH. Variability in pediatric utilization of trauma facilities in California: 1999 to 2005. Ann Emerg Med 2008; 52:607-15.
crossref pmid
37. Irizarry CR, Hardigan PC, Kenney MG, et al. Prevalence and ethnic/racial disparities in the distribution of pediatric injuries in South Florida: implications for the development of community prevention programs. Inj Epidemiol 2017; 4:12.
crossref pmid pmc pdf
38. Newnam S, Collie A, Vogel AP, Keleher H. The impacts of injury at the individual, community and societal levels: a systematic meta-review. Public Health 2014; 128:587-618.
crossref pmid
39. Razzaghi A, Soori H, Kavousi A, Abadi A, Khosravi A. Factors with the highest impact on road traffic deaths in Iran; an ecological study. Arch Acad Emerg Med 2019; 7:38.
pmid pmc
40. Jarman MP, Castillo RC, Carlini AR, Kodadek LM, Haider AH. Rural risk: geographic disparities in trauma mortality. Surgery 2016; 160:1551-9.
crossref pmid
41. Brown JB, Kheng M, Carney NA, Rubiano AM, Puyana JC. Geographical disparity and traumatic brain injury in America: rural areas suffer poorer outcomes. J Neurosci Rural Pract 2019; 10:10-5.
crossref pmid pmc
42. Pu H, Li B, Luo D, et al. Impact of urbanization factors on mortality due to unintentional injuries using panel data regression model and spatial-temporal analysis. Environ Sci Pollut Res Int 2020; 27:2945-54.
crossref pmid pdf
43. Jarman MP, Hashmi Z, Zerhouni Y, et al. Quantifying geographic barriers to trauma care: urban-rural variation in prehospital mortality. J Trauma Acute Care Surg 2019; 87:173-80.
crossref pmid
44. Jarman MP, Haut ER, Curriero FC, Castillo RC. Mapping areas with concentrated risk of trauma mortality: a first step toward mitigating geographic and socioeconomic disparities in trauma. J Trauma Acute Care Surg 2018; 85:54-61.
crossref pmid pmc
45. Wandling M, Behrens J, Hsia R, Crandall M. Geographic disparities in access to urban trauma care: defining the problem and identifying a solution for gunshot wound victims in Chicago. Am J Surg 2016; 212:587-91.
crossref pmid pmc
46. Brown JB, Gestring ML, Stassen NA, et al. Geographic variation in outcome benefits of helicopter transport for trauma in the United States: a retrospective cohort study. Ann Surg 2016; 263:406-12.
pmid
47. Beaulieu E, Smith J, Zheng A, Pike I. The geographic and demographic distribution of residential fires, related injuries, and deaths in four Canadian provinces. Can J Public Health 2020; 111:107-16.
crossref pmid pmc pdf
48. Elkbuli A, Dowd B, Flores R, McKenney M. The impact of geographic distribution on trauma center outcomes: do center outcomes vary by region? J Surg Res 2020; 252:107-15.
crossref pmid
49. Bhutiani N, Miller KR, Benns MV, et al. Correlating geographic location with incidence of motor vehicle-induced pedestrian injury. Am Surg 2018; 84:1049-53.
crossref pmid pdf
50. Ciesla DJ, Pracht EE, Cha JY, Langland-Orban B. Geographic distribution of severely injured patients: implications for trauma system development. J Trauma Acute Care Surg 2012; 73:618-24.
pmid
51. Lee J, Lee WY, Noh M, Khang YH. Does a geographical context of deprivation affect differences in injury mortality? A multilevel analysis in South Korean adults residing in metropolitan cities. J Epidemiol Community Health 2014; 68:457-65.
crossref pmid
52. Liu Q, Zhang L, Li J, et al. The gap in injury mortality rates between urban and rural residents of Hubei Province, China. BMC Public Health 2012; 12:180.
crossref pmid pmc pdf
53. Chen HY, Senserrick T, Martiniuk AL, et al. Fatal crash trends for Australian young drivers 1997-2007: geographic and socioeconomic differentials. J Safety Res 2010; 41:123-8.
crossref pmid
54. Klop C, van Staa TP, Cooper C, Harvey NC, de Vries F. The epidemiology of mortality after fracture in England: variation by age, sex, time, geographic location, and ethnicity. Osteoporos Int 2017; 28:161-8.
crossref pmid pmc pdf
55. West JG, Trunkey DD, Lim RC. Systems of trauma care: a study of two counties. Arch Surg 1979; 114:455-60.
crossref pmid
56. Guss DA, Meyer FT, Neuman TS, et al. The impact of a regionalized trauma system on trauma care in San Diego County. Ann Emerg Med 1989; 18:1141-5.
crossref pmid
57. Shackford SR, Mackersie RC, Hoyt DB, et al. Impact of a trauma system on outcome of severely injured patients. Arch Surg 1987; 122:523-7.
crossref pmid
58. Kim WN. Political direction for establishing a patient centered emergency medical services system [Internet]. HIRA OAK Repository; 2020 [cited 2023 Mar 2]. Available from: https://repository.hira.or.kr/handle/2019.oak/2235.

59. Lee KH. Establishment of regional emergency medical services system [Internet]. HIRA OAK Repository; 2020 [cited 2023 Mar 27]. Available from: https://repository.hira.or.kr/handle/2019.oak/2236.

Fig. 1.
Study flowchart. RISS, Research Information Sharing Service; KISS, Korean Studies Information Service System; KMbase, Korean Medical Database; KNbase, Korean Nursing Database; CINAHL, Cumulative Index to Nursing and Allied Health Literature.
ceem-23-009f1.jpg
Table 1.
QualSyst (Alberta Heritage Foundation for Medical Research) for quality assessment in the study (n=32)
Criteria Score
2 (Yes) 1 (Partial) 0 (No) Not applicable
Are questions/objectives sufficiently described? 24 (75.0) 7 (21.9) 1 (3.1) 0 (0)
Is study design evident and appropriate? 11 (34.4) 14 (43.8) 7 (21.9) 0 (0)
Are methods of subject/comparison group selections or source of information/input variables described and appropriate? 22 (68.8) 9 (28.1) 1 (3.1) 0 (0)
Are subject (and comparison group, if applicable) characteristics sufficiently described? 18 (56.3) 9 (28.1) 3 (9.4) 2 (6.3)
Are the outcomes and (if applicable) exposed measure(s) well defined and robust to measurement/misclassification bias? Are means of the assessment reported? 17 (53.1) 10 (31.3) 5 (15.6) 0 (0)
Are sample sizes appropriate? 28 (87.5) 4 (12.5) 0 (0) 0 (0)
Are analytic methods described/justified appropriately? 23 (71.9) 9 (28.1) 0 (0) 0 (0)
Are some estimates of variables reported for the main results? 11 (34.4) 3 (9.4) 17 (53.1) 1 (3.1)
Are confound variables well-controlled? 17 (53.1) 6 (18.8) 8 (25.0) 1 (3.1)
Are results reported in detail? 24 (75.0) 7 (21.9) 1 (3.1) 0 (0)
Does the conclusion correspond to the result? 29 (90.6) 2 (6.3) 1 (3.1) 0 (0)
Table 2.
General characteristics of cited studies
Study Age (yr) Category No. of subjects Country Database Statistical analysis
Moshiro et al. [27] (2005) All Individual 15,223 Tanzania Adult Morbidity and Mortality Project Poisson regression
Fatovich et al. [29] (2011) All Individual 3,333 Australia Royal Flying Doctor Service and Trauma Registries Weighted logistic regression
Zandy et al. [30] (2019) All Individual 10,445 Canada British Columbia Mortality data Age-standardized mortality rates
Poulos et al. [33] (2009) < 15 Individual 2,981 Australia New South Wales hospitals Spatial autocorrelation (Moran I statistic)
Dinh et al. [35] (2016) > 16 Individual 11,423 Australia New South Wales Trauma Registry Logistic regression, nonparametric GAM
Wang et al. [36] (2008) 1–14 Individual 65,434 USA California Office of Statewide Health Planning and Development Public Patient Discharge Database Logistic regression
Irizarry et al. [37] (2017) < 22 Individual 1,610 USA South Florida trauma center Cluster analysis
Jarman et al. [40] (2016) All Individual 8,673,213 USA 2009–2010 Nationwide Emergency Department Sample Multiple logistic regression
Jarman et al. [43] (2019) All Individual 1,108,211 USA CDC WONDER General linear model, Poisson regression
Jarman et al. [44] (2018) > 18 Individual 16,082 USA 2015 Maryland Adult Trauma Registry Geographically weighted regression, logistic regression
Wandling et al. [45] (2016) All Individual 11,744 USA Illinois State Trauma Registry Geospatial analysis
Bhutiani et al. [49] (2018) All Individual 615 USA University of Louisville Trauma Registry Multiple logistic regression
Ciesla et al. [50] (2012) All Individual 14,653 USA 2009 Florida Agency for Health Care Administration database Geospatial analysis using ArcGIS (Esri)
Lee et al. [51] (2014) > 35 Individual 36,242 Korea Statistics Korea (cause-of-death statistics, population and housing census) Mixture of Poisson regression model
Liu et al. [52] (2012) All Individual 9,714 China Disease Surveillance Point System Chi-square test
Chen et al. [53] (2010) 17–25 Individual 644 Australia New South Wales State records Chi-square test
Brown et al. [16] (2016) All Region (state) 48 States USA CDC Geospatial analysis
Razzaghi et al. [39] (2019) All Region (city) 31 Cities Iran Ministry of Health and Medical Education Multivariate regression
Brown et al. [41] (2019) All Region (state) 50 States USA CDC Geospatial analysis, linear regression
Pu et al. [42] (2020) All Region (prefecture) 161 Prefectures China Disease Surveillance Point System Spatial autocorrelation, linear regression
Minei et al. [24] (2010) All Individual and region 5,857 (8 States) USA ROC Epistry-Trauma Poisson regression
Young et al. [26] (2008) Adults Individual and region 11,576 (9 States) USA Administrative records of a large insurance company Logistic regression
Snyder et al. [28] (2017) < 15 Individual and region 34,816 (7 Provinces) USA Florida Agency for Health Care Administration Logistic regression
Wolf et al. [31] (2017) < 15 Individual and region 18,116 (51 States) USA Fatality Analysis Reporting System Multiple Regression
Gomez de Segura Nieva et al. [32] (2009) All Individual and region 614 (2 Nations) France, Spain Navarra and Atlantic Pyrenees district emergency center database Multinomial logistic regression
Kim et al. [34] (2011) > 18 Individual and region 21,868 (229 Cities, provinces, or districts) Korea Statistics Korea (cause-of-death statistics) Poisson HGLM
Brown et al. [46] (2016) > 15 Individual and region 193,629 (Northeast and southwest regions) USA National Trauma Database Logistic regression
Beaulieu et al. [47] (2020) All Individual and region 145,252 (4 Cities) Canada National Fire Information Database Simple linear regression
Elkbuli et al. [48] (2020) All Individual and region 317,500 (4 Regions) USA National Trauma Database Chi-square test, t-test
Klop et al. [54] (2017) > 50 Individual and region 82,375 (10 Provinces) UK Clinical Practice Research Datalink Poisson regression
Keeves et al. [25] (2019) All Other 47 Articles - Articles Scoping review
Newnam et al. [38] (2014) > 17 Other 96 Articles - Articles Systematic review

GAM, generalized additive models; CDC, Centers for Disease Control and Prevention; WONDER, Wide-ranging Online Data for Epidemiologic Research; HGLM, hierarchical generalized linear models.

Table 3.
Category and details of independent variables of 30 studies
Independent variable category Detail of independent variable Studya)
Socioeconomic Social deprivation, material deprivation, insurance, racial, ethnic, household income, residential, etc. 5 Studies [30,37,39,44,51]
Medical resource Presence of trauma center, helicopter transport 6 Studies [16,36,45,46,48,50]
Environmental Geographic regions, rural areas, red light camera legislation, population density, etc. 19 Studies [24,2629,3135,4043,47,49,5254]

a) Two studies were systematic review [38] and scoping review [25].

Table 4.
Detail of categorization for independent variables and their use of 30 studies
Variable Study
[16] [24] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54]
Socioeconomic
 Geodemographic
 Economic
 Individual
Medical resource
 Prehospital
 In-hospital
Environmental
 Road environment
 Geospatial
 Legal

Excluding systematic review [38] and scoping review [25].

Table 5.
Dependent variables and data source used for the 32 studies
Type of dependent variable Data source Detail of dependent variable
Mortality Medical records Death or survival
Scoping review
Disability Survey Duration of disability
Severity Medical records Injury Severity Score > 16
Survival and discharge Medical records Survival and discharge
Death and impairment Medical records Death and impairment (hospice)
Systematic review
Work disability Administrative records No. of absences
Table 6.
Geographical impact on the outcomes of trauma
Study Age Comparison Independent variable category Explanation and details of outcome of articles Representative outcome Dependent variable category
Moshiro et al. [27] (2005) All ages Urban vs. rural Environmental Lower impairment rates in urban areas during the recovery period*** Disability
Fatovich et al. [29] (2011) All ages Urban vs. rural Environmental Higher risk of death in rural areas* OR, 2.60 (95% CI, 1.05 to 6.53) Mortality
Zandy et al. [30] (2019) All ages Regions Socioeconomic Higher risk of death in low-income areas* 903.2 vs. 183.9 Standardized mortality rate per 100,000 for death Mortality
Gomez de Segura Nieva et al. [32] (2009) All ages Nations (French vs. Spain) Environmental Multivariates including age, sex, severity of trauma were corrected and the result showed the mortality rate at Atlantic Pyrenees (France) was 79% higher than Navarra (Spain) No significant differences in death Mortality
Brown et al. [16] (2016) All ages Regions Medical resource Fatality rate and nearest neighbor ratios were correlated* 56.9 (IQR, 46.5 to 58.9) vs. Mortality
Clustered states had a lower median injury fatality rate compared to dispersed states* 64.9 (IQR, 52.5 to 77.1)
Razzaghi et al. [39] (2019) All ages Urban vs. rural Socioeconomic A unit (1 million Iranian Rial) increase in the GDP of the province, the number of deaths decreased by as much as 0.0014** Range, 17.48 to 43.48 Mortality
When the population density increases one unit, the number of people died in traffic accidents rises 30***
Jarman et al. [40] (2016) All ages Urban vs. rural Environmental The probability of death in trauma was 14% higher in rural area*** OR, 1.34 (95% CI, 1.16 to 1.54) Mortality
Increased odds of death for rural residents were observed at level I (OR, 1.20), level II (OR, 1.34), and level IV and nontrauma centers (OR, 1.23) ***
Brown et al. [41] (2019) All ages Metropolitan vs. nonmetropolitan Environmental Among unintentional injuries, the head-injury death rate is 23% higher in rural areas than in urban areas, the injury by violence is 16% higher*** 12.78 vs. 9.81 in fatality rate per 100,000 Mortality
Pu et al. [42] (2020) All ages Regions Environmental As environmental urbanization rate increased, 7.3 increase in the mortality rate of traffic accident** t, –2.970766 Mortality
Jarman et al. [43] (2019) All ages Regions (by scale) Environmental Small cities and rural areas had higher prehospital trauma mortality rates than large cities*** IRR, 1.14 (95% CI, 1.05 to 1.23) Mortality
Jarman et al. [44] (2018) Adult Regions (by income) Socioeconomic The trauma mortality rate in low-income cities is higher than other *** OR, 1.98 (95% CI, 1.56 to 2.51) Mortality
Brown et al. [46] (2016) Adult Regions (east, west, south, north) Medical resource HT was associated with an increased odds of in-hospital survival (survival and discharge rates were 1.5 and 1.2 times higher, respectively, when patients are transferred by helicopters)*** AOR, 1.48 (95% CI, 1.44 to 1.52) Mortality
The survival rate was higher in the southern area than other areas*** OR, 1.29 (95% CI, 1.18 to 1.41)
Severity of injury, helicopters, and trauma centers, accessibility to trauma centers, and traffic congestion affected the transferability of helicopters*
Beaulieu et al. [47] (2020) All ages Regions Environmental The death rate per 1,000 residential fire incidents was higher in suburban areas than urban areas NA Mortality
Elkbuli et al. [48] (2020) All ages Regions (east, west, south, north) Medical resource Even controlled each level I to III trauma centers, there was difference mortality rate in each among areas NA Mortality
Bhutiani et al. [49] (2018) All ages Metropolitan vs. nonmetropolitan Environmental Pedestrian accidents were concentrated in cities OR, 0.67 (95% CI, 0.52 to 0.86) Mortality
Higher median household income*** and higher population density* were associated with decreased likelihood of death following pedestrian vs. motor vehicle accident OR, 0.97 (95% CI, 0.94 to 0.99)
Ciesla et al. [50] (2012) All ages Hospitals Medical resource Even though 93% of patients live in areas where trauma centers are located, the treatment rate vary by regions from 13% to 58% and the rate is relatively low NA Severely injured patients dis- charge
Trauma centers discharged 52% of major trauma patients
Liu et al. [52] (2012) All ages Urban vs. rural Environmental The trauma mortality rate was 2.4 higher in rural areas than urban areas** CRR, 1.9 (95% CI, 1.8 to 2.0) Mortality
Wandling et al. [45] (2016) All ages Regions Medical resource It takes more than 30 min to transfer gunshot wounds adult patients to trauma centers from the south of Chicago (pediatric trauma centers are in the southeast) NA Mortality
Minei et al. [24] (2010) All ages Regions Environmental There were regional differences in measures related to injury, such as incidence and survival rate* NA Survival ranged
Keeves et al. [25] (2019) All ages Urban vs. rural Environmental Both the hospital and prehospital mortality rates are higher in rural areas Scoping review Mortality
Young et al. [26] (2008) Adults Urban vs. rural Environmental The work disability rate was higher in urban areas than rural areas*** OR, 0.78 (95% CI, 0.72 to 0.84) Disability
Kim et al. [34] (2011) Adults Regions Environmental Lower mortality from transport accidents and suicides in high residence tax per person areas*** β, –0.141 (95% CI, –0.220 to –0.062) Mortality
Dinh et al. [35] (2016) Adults Urban vs. rural Environmental Inpatient mortality for those injured in metropolitan locations was 14.7% in 2009 and 16.1% in 2014* Mortality
In rural locations, there was a statistically significant decline in in-hospital mortality over the study period, from 12.1% in 2009 to 8.7% in 2014**
Lee et al. [51] (2014) Adults Regions Socioeconomic In areas where the poverty index is the lowest, traffic accidents, falling down, and suicide are 1.3**, 1.6***, and 1.1* times higher, respectively RR, 1.34 (1.05 to 1.73) Mortality
RR, 1.63 (1.20 to 2.20)
Chen et al. [53] (2010) Adults Urban vs. rural Environmental Traffic accident fatality rates in cities decrease by 5% every year χ2, 119.95 Mortality
**Traffic accidents were increased in rural areas.
It led by over speed limit, fatigue, drunk driving, and no seat belts**
Newnam et al. [38] (2014) Adults - - To seek the impact of injury to individual, community, societal level Systematic review Injury
Total 78 articles were included in the final set
The results showed that the vast majority of injury outcomes literature identified in this meta-review (83%) focused on the impacts of injury on the individual level
Snyder et al. [28] (2017) Children Regions Environmental The mortality rate varies from region to region* OR, 2.0 (95% CI, 1.6 to 2.6) Mortality
The treatment rate outside a region is different
Medical service fees are higher if patients get treated outside the region they resides in*
Wolf et al. [31] (2017) Children Regions Environmental Percentage of crashes on rural road* β, 0.05 Mortality
States without red light camera policies had higher mortality rates than the states with such policies***
Poulos et al. [33] (2009) Children Regions Environmental Children in rural areas are at a higher risk of burn injury than their counterparts in urban areas* RR > 1.2, PP > 0.8 (using Bayesian methods) Morbidity
Wang et al. [36] (2008) Children Existence of trauma center Medical resource Higher accessibility of children with moderate or major trauma to trauma centers*** OR, 3.95 (95% CI, 3.43 to 4.54) Hospitalized
Irizarry et al. [37] (2017) Children Regions Socioeconomic African American children have a higher GSW risk and mortality rate and most of the GSW cases for African American children occur in areas where they live** Hispanic children: RR, 0.37 (95% CI, 1.18 to 0.79) Prevalence, mortality
White children: RR, 0.08 (95% CI, 0.04 to 0.16)
Klop et al. [54] (2017) Middle-aged to elderly Regions Environmental Most cities, other than London, show higher mortality rates of fracture among women* RR, 1.37 (95% CI, 1.18 to 1.58) Mortality

OR, odds ratio; CI, confidence interval; IQR, interquartile range; GDP, gross domestic product; IRR, incidence rate ratios; AOR, adjusted odds ratio; NA, not available; CRR, crude rate ratio; RR, relative risk; PP, posterior probability; GSW, gunshot wound.

* P<0.05.

** P<0.01.

*** P<0.001.

No significance.

Editorial Office
The Korean Society of Emergency Medicine
101-3104, Brownstone Seoul, 464 Cheongpa-ro, Jung-gu, Seoul 04510, Korea
TEL: +82-31-709-0918   E-mail: office@ceemjournal.org
About |  Browse Articles |  Current Issue |  For Authors and Reviewers
Copyright © by The Korean Society of Emergency Medicine.                 Developed in M2PI