Introduction
Violence undermines people’s health and some groups are more affected by violence than others. Race and ethnicity are among the factors that distinguish population groups most at risk of violence, with ethnic minorities being disproportionately affected [
1‐
7]. Studies in several countries have shown that appreciable portions of the gaps in exposure to violence and its consequences between ethnic groups can be accounted for by family background, individual differences, and neighborhood factors [
8‐
10]. Ethnic differences associated with violence are also linked to sociocultural (such as parenting styles and masculinity norms) and immigration factors, such as social disintegration [
1,
11‐
13].
According to a report by the World Health Organization (WHO) Region for Europe, in comparison to other European countries, Israel ranked 9th in homicide among people aged 10–29 years at a rate of 4.30 per 100,000 population [
1]. A report produced by the Israel Ministry of Public Security in collaboration with the Central Bureau of Statistics (CBS) showed that there were approximately 620,000 annual incidents of violence in Israel between 2003 and 2010; with an upward trend between 2011 and 2012, particularly for serious violent offenses [
14]. In addition, disparities in violence-related injuries between ethnic groups in Israel have been reported [
15,
16]. The risk of being hospitalized for violence-related injuries was higher among non-Jewish Israeli residents compared with Jewish Israelis [
17].
The Israel society is composed of two main ethnic populations: Jews and Arabs, comprising 75.1 and 20.6%, respectively, according to the Israeli CBS data for the years 2008–2017 [
18]. Jews and Arabs are the two major distinct ethnic groups who may experience different risks of exposure, vulnerability and outcomes from injuries as a result of violence, because these populations differ in many ways, including socio-cultural differences [
17,
19‐
22]. Among the Jewish population, there were two major waves of immigration to Israel in the 1990’s. These immigrant groups included Ethiopian Jewish immigrants and immigrants from the former Soviet Union, 1.0 and 7.9%, respectively, of the general Israel population, according to the Israeli CBS, 2008–2017 [
23]. It is important to note that the proportion of Ethiopian and former Soviet Union immigrants presented above does not include descendants born in Israel. These population groups faced the challenges of reestablishing life in a new country [
24,
25], which may be associated with increased susceptibility to violence and related injuries. In comparison with Arab and Jewish born Israelis, studying violence-related injuries among these immigrant populations can help understand the potential impact of immigration on violence-related injuries. The Ethiopian community and Arabs are amongst the ethnic groups in Israeli society with the lowest socioeconomic status (SES). These two population groups are comprised of large families and a relatively young population. The majority of Arabs live in towns and villages where most of the residents are Arabs. While the majority of Ethiopian Israelis reside in Jewish cities, they often live in segregated neighborhoods with large numbers of Ethiopian-Israeli residents [
25‐
29]. These sociodemographic characteristics may be associated with differential injury risks from violence [
30‐
32]. It should be noted that every Israeli resident, regardless of ethnicity, gender or country of origin is entitled to health care services under the National Health Insurance Law. Equally important, fees are not prerequisite for receiving care [
33].
In this paper we sought to examine ethnic differences associated with violence-related injuries. In Israel, characteristics and circumstances of violence-related hospitalizations, hospital resource utilization and in-hospital mortality in relation to ethnicity and country of origin have not been well documented. The outcomes from this study provide evidence based data which can enable policy makers to focus on intervention programs for population at high risk.
Methods
This study was based on the Israeli National Trauma Registry (ITR) database of patients hospitalized for injuries due to violence between January 1, 2008 and December 31, 2017. The ITR is an extensive database of hospitalized trauma patients, providing a broad geographic and demographic coverage in the country [
15,
16]. All the six Level Ι Trauma Centers and 14 Level ΙΙ Trauma Centers participated in the ITR during the study period. There were five hospitals that did not participate in the ITR during the study period, representing only 5% of the total hospitalized trauma patients in Israel. Included in the ITR are injured trauma patients with an International Classification of Diseases, Ninth Revision (ICD-9-CM) diagnosis code of 800–959.9 who were hospitalized, died in the hospital (including deaths in the emergency department) or were transferred to or from another hospital for admission. The ITR does not include patients who died at the scene of the event, on the way to hospital or on arrival; who were discharged following treatment in the emergency department; or who were hospitalized 72 h or more after the event. Data reported in the registry are recorded by trained trauma registrars at each trauma center under the supervision of a trauma director and trauma coordinator. Electronic files are transferred daily to the National Center for Trauma and Emergency Medicine Research where quality assurance is carried out prior to the data being analyzed. Unclear or erroneous data are referred back to the trauma centers for clarification or completion. The data in the ITR are anonymous and there is no way to identify patients; the study received the approval of the Sheba Medical Center Institutional Review Board (IRB) (Number 5138–18 SMC). From this comprehensive database, this study focused on analyzing violence-related injuries, identified using codes for external cause (E-code) of injury ‘E960.0’-'E968–.9′. Self-inflicted injuries and injuries from terror attacks and war were not included. In order to avoid double counting for transferred patients, the data from their last hospitalization was used. Since the study focused only on Israeli residents, Non-Israeli Arab residents of East Jerusalem (311, 1.8%), foreign workers (544, 3.1%), tourists (141, 0.8%) and unidentified casualties (90, 0.5%) were excluded from the analysis. These groups were excluded on the assumption that hospitalization characteristics and outcomes at hospital discharge of non-citizens may differ from those of citizens primarily because of insurance coverage differences.
The data measured included demographic characteristics (age, gender, ethnicity, country of birth); injury characteristics (mechanism of violence, type of injury, injury severity, injured body region, traumatic brain injury); injury setting and timing of hospital arrival; hospital resource utilization (treatment in trauma resuscitation unit, undergoing surgery, admission to intensive care unit and length of hospital stay) and in-hospital mortality. Age in years was divided into five categories: 0–14, 15–24, 25–44, 45–64, 65+. The Injury Severity Score (ISS) was used to quantify injury severity, which was classified into two groups: 1–14 (mild and moderate injuries) and 16–75 (severe and critical injuries), derived from its four categories [
34,
35]. Mechanism of violence was categorized as: firearm, stabbing, assault with object, unarmed brawl and other. Type of injury was categorized as penetrating versus non-penetrating and place of injury was classified as injury event occurring on the street/road, at home, in public (which included commercial/leisure places, playing/tour places or sport centers) and other (which included residential institution, workplace, school, army base, farm land, sea/lake, air and others). Time of hospital arrival was defined as follows: weekday as daytime from 06:00 to 18:59 Sunday through Thursday; weeknight as night hours from 19:00 to 05:59 from Sunday19:00 to Thursday 05:59; and weekend as any time between Thursday at 19:00 and Sunday at 05:59. It is important to note that in Israel Sunday is a work day and Friday is not a work day for government offices. For this study’s purpose, time of patient presentation to the hospital could be taken as a proxy for time of injury occurrence, as the country is of a relatively small surface area [
36], and it is known for its advanced trauma system with short arrival times. Trauma centers are located throughout the country; enabling trauma casualties to receive rapid treatment in all parts of the country. Due to redistribution of ambulance dispatch centers and the use of geographic information systems and Global Positioning System, in recent year’s response times have been reduced [
15,
16,
37‐
40]. The Abbreviated Injury Scale (AIS) codes and scores of nine body regions (head, face, neck, thorax, abdomen, spine, upper extremity, lower extremity and external) were used for recognizing the injured body regions. Traumatic brain injury was defined as any recorded evidence of brain injury in accordance with the AIS 1990 Revision Manual [
41]. Number of injured body regions was categorized as single (when only one body region was injured) and multiple (when more than one region was injured). Treatment in trauma resuscitation unit (yes/no), admission to intensive care unit (yes/no), undergoing surgery (yes/ no), length of hospital stay (
> 7 days or < 7 days) and in-hospital mortality (yes/no) were coded as binary variables.
Analyses were performed using SAS statistical software version 9.4 (SAS Institute, Cary. NC, USA). Missing data for each variable entered into the analyses were less than 0.1%. Descriptive data were compared by χ2-test and multiple logistic regression models were developed to compare the mechanisms of violence as well as injury severity, injury type and requirement of admission to intensive care unit between the ethnic groups. A p-value of < 0.05 was considered statistically significant.
Discussion
The outcomes from this study provide evidence suggesting that ethnicity of violence-related casualties influence the mechanism, circumstances and severity of injury, and consequently hospital resource utilization.
The evidence shows that the mechanism of injury varies by ethnicity, IBE are at greater risk of hospitalization due to stab injuries, IBFSU for unarmed brawl injuries and AI for firearm injuries. Both AI and IBE are also at a greater risk of hospitalization due to injuries by assault with an object. These findings may be partially associated with the age and gender distributions in each population group. In comparison to IBFSU and AOI, IBE and AI males and young adults, aged 15–44 years, are at greater risk for violence-related hospitalizations. The younger age and male gender characteristics of IBE and AI casualties may contribute to their increased risk for injuries due to stabbings and fire arms as supported by our observation in adjusted analyses, and consistent with the available evidence reporting that violence by sharp objects and firearm is more prevalent in these segments of population [
42]. Attributed to masculinity and gender values, younger people and males are more apt to risky behaviors and thus may have greater chance of being involved in violence in general, and in more severe forms of violence in particular [
2,
3,
43‐
46].
In addition to injury mechanism, differences in place and time of event were observed. The street was a common place of injury for the immigrant groups, IBE and IBFSU. While among all groups the majority of hospitalizations occurred on the weekend or at night, the percentage was even greater among the Ethiopian immigrants. These outcomes may indicate that immigration has influential effects on where injury events occur. This information should be used by policy makers to develop interventions which focus on the peak times and places of violence-related injury events. It can also be likely that the injury setting characteristics may contribute to the observed differences in violence mechanism between the population groups, which were demonstrated in the adjusted analyses. The comparison population groups had a higher risk of hospitalization due to specific types of violence, even after controlling for possible confounding factors, implying that ethnicity is an independent factor.
It is possible that situational factors such as ease of access to weapons may play a role in determining violence and its consequences [
47], although this study has not investigated these aspects. There is evidence suggesting that practices of carrying and using knives in Israel are growing [
1,
21], since they are easily available. There is also evidence reporting that pupils from low socioeconomic status (SES) families have higher levels of weapon carrying [
48]. The availability of weapons and the act of carrying them are risk factors for violence. The high prevalence of carrying weapons, such as knives, and using them in Israel [
1,
21] can be speculated to be among the factors associated with increased rates of stab-related injuries, warranting the importance of intensifying control measures against access to and carrying of such weapons. The findings presented identify IBE as a distinct high risk group and their needs must be addressed. The availability of firearms is a major determinant of their use and influence on homicide rates [
1,
2,
46]. Surprisingly, even the possession of licensed firearms has been reported to affect the risk of violence-related injuries [
17]. People with firearm possession are more likely to endorse aggressive attitudes that increase their risk for retaliatory violence [
49]. In Israel, there are increasing media reports of illegal gun possession and armed crimes, particularly in Arab towns. It has been reported that firearms are easy to obtain, that is, both makeshift pistols which are being manufactured in workshops in Arab villages or in the Palestinian territories, and weapons which are often smuggled or stolen. There are also social media reports of increasing gun carrying to school by children in the Arab sector [
22]. Accordingly, all such factors may increase risk of firearm injuries among Arabs, which is supported by our finding. The international literature also has shown evidence that stab and firearm injuries are more common among ethnic minorities, similar to our findings of higher risks among the minority groups in Israel [
3‐
5,
7,
46].
The differences in injury type between the population groups can be attributed to their differences in the involved violence mechanism. For example, the frequency of a combination of firearm and stab injuries was highest among AI, which will explain for the greater rate of penetrating injuries in this population group. Israel has a lower rate of personal gun ownership, stricter gun control laws, and its policy discourages personal gun ownership [
50]. Our findings, however, may highlight the importance of intensifying regulatory and monitoring activities on firearms. In an effort to reduce firearm-related injuries, there may be an urgent need to enforce the illegal access and distribution of firearms, especially by theft or unlawful sales, in addition to enhancing efforts in restricting weapons in certain settings, for example, leisure facilities, public places and school premises, as such interventions are proven to be effective in reducing violent injuries [
42,
43]. In addition, due to the ethnic differences, formulating population-specific and socio-culturally appropriate violence prevention and intervention programs is crucial.
In comparison to the general Israeli population, the three minority groups studied, IBE, AI and IBFSU appear to be disproportionately affected by violence-related injuries, as demonstrated by the greater proportion of IBE, AI and IBFSU casualties from the overall violence casualties than their respective share in the general Israeli population; 3.2 times greater for IBE, 2.2 times greater for AI and 1.6 greater for IBFSU. These disparities can be speculated to be due to differential exposure to risk and protective factors. It is known that there is no single reason explaining why some populations are more vulnerable to injury in general, and specifically to different types of violence-related injuries. Nevertheless, these differences in violence injuries between the ethnic groups compared in our study may be, at least in part, be a reflection of the many socioeconomic differences. There is evidence that the SES of minority groups in Israel, in particular the Ethiopian community and Arabs, is lower than that of the other population groups and they often reside in disadvantaged neighborhoods. The Ethiopian community has the lowest SES amongst all population groups in the country [
25‐
28]. A large body of evidence shows an association between lower SES and an increased risk of sustaining violence-related injuries [
1,
2,
20,
30‐
32,
51‐
55]. Low income, low education level, unemployment, unskilled labor and job dissatisfaction, poverty, low social standing and integration difficulties may be speculated to be associated with behavioral changes, which may increase exposure and susceptibility to violence and worse outcomes. Socioeconomic disparities may be associated with various risky behaviors, including substance abuse and crime, which may increase risk of violence-related injuries [
6,
17,
30‐
32]. In addition, there may be low negotiation skills attributable to lower education level, and as a result unprecedented situations may end up with events that can lead to injury, in particular to the more severe forms of violence. Cultural practices and social values may play roles in experiencing inequalities in violence. Violence may be regarded as a practical option to conflict resolution [
42,
56]. Contextual factors in disadvantaged neighborhoods may create suitable situations to exposure and susceptibility to violence, as in such neighborhoods there may be high rates of crimes, illicit drugs dealing, gang membership and deviant peer groups [
8,
31,
57,
58].
In addition to changes in the socioeconomic and professional status, the immigration process, which includes obstacles, such as language barrier and cultural differences, may hinder integration of immigrants into the Israeli society and lead to violence-related behaviors, which may contribute to violence injuries among immigrants from Ethiopia and the former Soviet Union [
17,
59]. It is challenging for immigrants to adapt between their cultures of the past and fitting into the new culture, the Israeli society. These clashes often play a crucial role in violent behavior and consequently on injuries [
11,
60]. Furthermore, new immigrants and minority groups may experience discrimination leading to violence-related behaviors [
25,
61]. Due to language and cultural barriers, access to and utilization of preventive services among Ethiopian immigrants may also be lacking.
While there were significant differences in hospital resource utilization, as a function of injury severity, no difference in in-hospital mortality was found. This finding could be a result of appropriate trauma care to all population groups, without any discrimination. In addition, it shows the efficacy of the Israeli trauma care system and the national health insurance law, which provides equal quality of care for all Israeli residents regardless of ethnicity, gender or country of origin where fees are not prerequisite for receiving care [
33,
37,
38].
Limitations
First, the study population includes victims of violence, and not necessarily the perpetrators. Thus, we are unable to identify the victim-perpetrator relationship, such as stranger, friend, family member, or intimate partner, or other demographic characteristic.
Second, the study included only the first generation of immigrants from Ethiopia or former Soviet Union, that is, those born abroad. Although identifying the immigration effect on the next generation is important, the current trauma registry database does not provide the parents’ country of origin, which can be considered an important limitation.
A third limitation originates from the inclusion criteria of the ITR. The ITR does not include mortality prior to arriving at the medical center. In addition, casualties with minor injuries, who were not hospitalized, were not included in this study, which might have resulted in selection biased estimate. Since the majority of hospitalized trauma patients and almost all severely wounded patients are treated in the ITR participating hospitals, we can conclude that this study provides plentiful, representative and valuable information in understanding violence injury characteristics and outcomes in Israel.
Fourth, since the trauma registry does not include information on socio-economic position, its potential influence could not be investigated. We recommend future research to explore the contribution of socioeconomic position on violence-related injuries among different population groups.
Acknowledgements
The authors wish to thank the Israel Trauma Group for their responsibility for the data completing. The Trauma Group includes: H. Bahouth, A. Becker, A. Hadary, I. Jeroukhimov, M. Karawani, B. Kessel, Y. Klein, G. Lin, O. Merin, B. Miklush, Y. Mnouskin, A. Rivkind, G. Shaked, G. Sibak, D. Soffer, M. Stein, M. Wais, H. Pharan and I. Garbetzev. The authors would also like to thank Sharon Goldman for the English proof reading and corrections.