Understanding the uneven distribution of the incidence of homicide in Latin America
1 Laboratorio de Ciencias Sociales, Universidad Central de Venezuela, Caracas, Venezuela.
2 Instituto Cisalva, Universidad del Valle, Cali, Colombia.
3 UNC – Injury Prevention Research Center, Chapel Hill, NC, USA.
4 Department of Epidemiology, University of North Carolina, Chapel Hill NC, USA
5 Pan-American Health Organization, Sustainable Development and Environmental Health Area, Washington DC, USA.
* Corresponding author. Laboratorio de Ciencias Sociales (LACSO), Universidad Central de Venezuela, Apartado Postal 47795, Caracas 1040-A, Venezuela. E-mail: bricenoleon{at}cantv.net
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
Background Throughout the final years of the twentieth century and into the beginning of the twenty-first, violence has been one of the main public health issues in Latin America, a region which has some of the highest mortality rates due to violence in the world. However, there seems to be an uneven geographical distribution of such instances.
Methods We reviewed epidemiological data on violence globally and in Latin America, and here, we discuss differences between the Latin American countries in the context of a sociological framework as well as from a public health perspective.
Results Our results indicate marked differences by country in terms of rates of violence. Countries such as Argentina, Chile, Costa Rica, and Uruguay, have low violence mortality rates; Peru, Nicaragua, Ecuador, Dominican Republic, Panama, and Paraguay have moderate rates, and Brazil, Mexico, Colombia, El Salvador, Honduras and Venezuela have high to extremely high mortality rates. Factors related to violence include social inequalities, lack of employment opportunities, urban segregation, a culture of masculinity, local drug markets, and the availability of firearms and widespread use of alcohol.
Conclusion The observed homicide variability between Latin American countries can be explained largely by differences in the countries’ social contexts and political models. In those countries where homicide rates are extremely high, governments should review their current policies and take preventive actions. Fortunately increasingly nowadays there are promising advancements in that direction.
Keywords Violence, public health, Latin America, prevention
Accepted 9 June 2008
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
Historically, until the 1970's interpersonal violence was studied more from the perspective of criminology, policing, and legal disciplines. While the social causes of violence had also been studied since the early twentieth century, such as in the French movement of the ‘Milieu Social’,1 its focus was more on individual and reactive perspectives with emphasis on approaches from psychiatry and the sociology of deviant behaviours applied to criminological and policing strategies. In the early 1960s, the first publication calling for the study of violence using epidemiologic methods was written by a Colombian epidemiologist, Professor Héctor Abad Gómez, who promoted a more widespread use of surveillance systems and epidemiologic methods to study violence in society.2 This publication paralleled the development of a public health approach to studying unintentional injuries which was largely developed by Dr William Haddon Jr.3 In subsequent years, the principles of public health were more strongly applied to studying violence and these efforts were consolidated in the three following decades.
Between the 1970s and 1980s, and paired with a steep rise in homicide rates in some Latin American countries, the study of violence grew by further focusing on social rather than individual problems. The public health approach strengthened the concept of prevention through the study of risk factors associated with violence, both from an individual perspective and as a population as a whole.4 From a sociological perspective, the high rates of violence made it impossible for social researchers to accept Durkheim's5 idea that delinquency and social transgressions exist in all societies as a norm. The sheer magnitude of the problem between the 1980s and 1990s contradicted this statement. With the development of the public health approach to studying violence came the involvement of new agencies in this field. In Latin America, the study of violence initially focused on the development of reliable information systems, at that time non-existent or fragmented, and of poor quality. In the 1990s the development of these systems and further research on violence was guided in part by collaborative work between city governments, academic institutions, and agencies such as the Pan-American Health Organization, the U.S. Center for Disease Control and Prevention, and later by the Inter-American Development Bank, World Bank and the World Health Organization. Efforts towards the development and improvement of violence and injury surveillance systems continue to date and have yielded technical guidelines6,7 as well as more reliable information, and allowed for the evaluation of preventive measures and selected policies aimed at curbing violence. The purpose of this paper is to review the instances of homicidal violence in Latin America in relation both to the global context and by making comparisons between countries of the region, as well as providing ideas that highlight the social aspects of violence and its prevention in a wider public health context.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
We reviewed several public databases that contain information about violent deaths in the world and, more specifically, in Latin America. Here, we present data by country where it was available, and by magnitude of violence-related mortality rates. Data are presented in comparison with other regions of the world as well as with Latin American countries. In addition, we cross-tabulated information on country-specific violence mortality rates with corresponding poverty rates, measures of inequality, and urban concentration of population.
We further present differences in violence-related mortality rates according to sex and age, where data were available. We then discuss these findings within a sociological framework and highlight selected successful interventions and evaluations of policies in different countries.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
Global violent death rates
According to the World Report on Violence and Health8 there were approximately 1.6 million violent deaths globally in 2000. These deaths include interpersonal violence, self-inflicted, and collective violence (war). Interpersonal violence, the majority of which were homicides, accounted for about a third of these deaths (rate 8.8 per 100,000 population) and were more than those caused by armed conflicts (rate 5.2 per 100,000 pop.). While the region of Sub-Saharan Africa reported higher rates than the region of the Americas (22.2 per 100,000 pop. vs. 19.3 per 100,000 pop.), once high income countries are eliminated and observations are focused more among middle- and low-income countries in the Americas (all Latin American countries), the rates due to homicide become the highest in the world (rate 27.5 per 100,000 pop.), over three times greater than those for the European Region, almost 4 times greater than rates reported by the Eastern Mediterranean Region, close to 5 times the rates of South-East Asia, and 8 times greater than homicide rates reported by the Western Pacific Region. Differences in magnitude of homicide rates also apply to comparisons between high-income and middle-and low-income countries, the latter about 3.5 times greater than the former (2.9 per 100,000 pop. vs. 10.1 per 100,000 pop.). This specific grouping implies differences beyond variation in income, because societies also display specific institutional, cultural, and behavioural characteristics that increase the overall risk of victimisation among their populations. Also noticeable in Latin America is the fact that homicide rates are much higher than suicide rates (27.7 per 100,000 pop. vs. 6.3 per 100,000 pop.), a reverse relationship from that which occurs in the rest of the world as a group (8.8 per 100,000 pop. vs. 14.5 per 100,000 pop. respectively).
Homicide in Latin America
Homicide rates in Latin America are not homogeneous. Remarkable differences exist between countries that have low rates of homicide (i.e. Uruguay, Costa Rica, and Chile) versus countries with very high rates of homicide (Colombia, El Salvador, and Venezuela) (Table 1).
|
The relationship between urban population, poverty levels, and the incidence of homicide is still unclear. Countries that report the highest rates of homicide tend to be countries with high proportions of urban population and high rates of poverty (Colombia, El Salvador, Honduras, Venezuela, and Brazil). Among those which are highly urbanized but with less poverty (Uruguay, Chile, Costa Rica) homicide rates tend to be lower. The high rates of urbanisation and poverty can also be associated with social inequalities commonly measured by the Gini Coefficient (Table 2.) Extremely high GINI coefficients appear to be associated with high homicide rates in Latin America and countries with the lowest GINI coefficients have low homicide rates. Regional comparisons between Latin America and other parts of the world confirm these differences. The Latin American region overall has, on average, the highest GINI coefficient in the world. A few countries outside of Latin America with similar coefficients (i.e. South Africa), tend to have similar levels of violence.
|
Characteristics of victims
Most victims of homicide in Latin America are young males of lower socio-economic levels. This pattern remains true, independent of the overall homicide rate of the country. As homicide rates increase, the disparity between sexes becomes more apparent as shown in Table 3.
|
Young populations are at particularly high risk of homicide, and young males between the ages of 15 to 24 years are most at risk. Furthermore, in Latin America most homicides are actually committed by young men. While the demographic structure of the population in Latin America is young, this predominant homicide pattern favouring younger age groups is also replicated in high-income countries where the population tends to be older. The fact that the highest proportion of the population in Latin America belongs to younger age groups further exacerbates the economic and social problems of violence and may well contribute to the high homicide rate. Furthermore. compared with countries with similar population structures outside Latin America, homicide rates are higher. Within the region, there are still varying proportions of deaths caused by homicide in young males (Table 4).
|
Temporal trends
Temporal trends indicate wide variations in homicide rates in Latin America. Strong efforts to control violence have been successful in reducing the magnitude of the problem. The development of multi-institutional efforts and the evaluation of strategies to control violence in large urban centres have shown benefits as depicted in Figures 1 and 2. These efforts have been mostly developed in Colombian cities (Bogotá and Medellín) and in São Paulo, Brazil, though local efforts in the latter have not affected the overall country rates.
|
|
Non-fatal interpersonal violence
In addition to the characteristics displayed by victims of fatal violence, other surveys have indicated that non-fatal violence affects all social strata but tends to be more prevalent among lower socio-economic levels.13 Studies of non-fatal violence in Latin America are scarcer overall but, as other literature indicates, involvement in non-fatal violence is associated with increased risks of exposure to fatal events.14,15
The costs to society
The overall effects of homicide on society far exceed individual consequences and pose considerable costs on the country's economy and its development. According to data from the United Nations Development Program (UNDP)16 homicidal violence has a direct impact on healthcare costs, other institutional costs, private insurance costs, and material losses, among other categories. The costs of violence as a proportion of the gross domestic product (GDP) vary by country. Estimates suggest that for Peru and Brazil it is 1.4% of the GDP, for Mexico 3.6%, for El Salvador 4.9%, for Colombia 6.4%, for Venezuela 6.6%, and for Guatemala 6.7%.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
A model for understanding violence in Latin America
To better approach and interpret violence, several models have been developed and are aimed at organising in some way the multiple factors that influence the occurrence of violence. A common model adopted by public health practitioners to understand violence is Bronfenbrenner's socio-ecological model originally proposed in the 1970s to understand human development and behaviour.17 Other models have been used focusing on social capital,18 economic models,19 criminological models,20 and social models.21 A formal sociological model22 is presented here and aims to establish relationships between different levels (similar to the socio-ecological model) but is defined as originating, promoting, and facilitating levels that incorporate material, situational, and cultural factors without focusing on the individual per se, but rather on the circumstances or context surrounding the person.
|
To better understand this sociological model, the originating level can be understood as the causal factors for violence which mostly refer to societal and cultural characteristics. The promoting factors are the second level, and are related to the material conditions in which people exist. At the lowest level, the facilitating factors contribute to the increased occurrence of violence or to its lethality, without being actual causes. The facilitating factors are imbedded within the level of promoting factors which are in turn imbedded within the larger, more macro originating factors, thus establishing an inclusive model that links all levels.
Factors from whence violence originates are mostly related to the overall social conditions of Latin American countries which are characterised by very high levels of inequality.23,24 Other prevailing characteristics are increases in poverty and youth unemployment. These factors add to the loss of traditional mechanisms of social control, in particular, those led by families and religious beliefs. Furthermore, young populations with higher expectations are provided with fewer opportunities to thrive in their social environment. These situations are more common in countries where inequality is greater and not in countries with higher overall levels of poverty such as Bolivia and Nicaragua. In areas where wealth and extreme poverty cohabit, violence tends to occur more frequently. Examples of this are found in urban areas of Brazil, Mexico, Venezuela, and Colombia where such conditions are further exacerbated by marked increases in poor urban populations.25
Unemployment among young people has also increased. In 1995 there were 7.2 million unemployed young people in Latin America. These numbers increased to 9.5 million in 2005. The youth unemployment rate is 16.6%; twice as much as the overall regional average. One in 2 of the total unemployed are young and the ratio of youth to adult unemployment rate is 2.8:1.26 This situation is worsened by the lack of orientation, mentoring, and institutional support aimed at addressing youth behaviour and establishing positive social norms.
Families have also been affected and now seem to have less influence on individuals. The percentage of women who did not work in two-parent families decreased from 46.2% in 1990 to 36.2% in 2002.27 This increased number of women going out to work has not been accompanied by compensatory social support mechanisms for families and children.
Religious institutions and, in particular, the Catholic Church have lost some of their social control, especially in urban settings. The process of secularisation in Latin America has been widespread and religious norms have been replaced by civil codes. Likewise, in this region, a large process of democratisation of expectations has occurred. This means that individuals, no matter of what origin, desire and strive towards owning similar material goods. Naturally, the capacity to satisfy these needs differs greatly between individuals who are wealthy compared to those who are poor. The latter 35% of youth who are under the poverty line (earning daily less than USD $2) have less capacity to acquire material things by legal means.26 These factors constitute important causes of violence in Latin American societies.
Factors that promote the occurrence of violence can be grouped into four categories: the specific structure of the built environment in cities, the culture of masculinity, drug markets, and inefficient judicial systems that foster impunity. Violence in Latin America is concentrated in urban areas where, because of higher population density and unplanned or poorly planned development, zones of exclusion are quickly created. Within these zones characterised by poor connectivity to other urban areas, topographic difficulties, and poor access to social services, alternative mechanisms of social control through gangs or subversive groups develop easily.28–30
As previously shown, there are marked gender differences in involvement in violence. This can be explained partially by the culture of masculinity that forces men to confront others when challenged. The perception generally prevails that the avoidance of confrontation is more associated with females. Studies on youth violence in Venezuela show the importance for males of "earning" respect in front of others. A common mechanism to achieve this is through acts of violence.31,32
A third promoting factor is the drug market. In several urban areas in Brazil and Colombia, but also beyond Latin America, the search for new markets and competition for other markets has been linked to illegal drug commerce.33 In Latin America, these markets are also associated with illegal gun markets by weapon-producing countries, thus linking some of the promoting factors with the facilitating factors. A fourth factor, the judicial system, is frequently characterised by inefficient responses and a high degree of impunity. Data from the DESEPAZ surveillance system in Cali, Colombia34 collects adequate information on victims of homicide but lacks over 90% of vital information on perpetrators. Impunity has a further deleterious effect on violence as perpetrators know they are unlikely to be punished. These deficiencies are compounded by the fact that, even if the majority of perpetrators were caught by the police, there does not exist sufficient space or resources to keep them in penitentiaries. Overpopulation in these centres is widespread.
Finally, factors that facilitate violence are related to social norms that promote alcohol consumption and trends such as the carrying of firearms. Alcohol consumption and the presence of alcohol vendors has been linked to violence.35 Legislation has been passed in some urban settings aimed at restricting hours of alcohol consumption. These measures have been linked to a reduction in the incidence of violence following the passing of these laws.
The carrying of firearms has also been linked to increased fatal violence in Latin America. However, restrictions to carrying firearms in cities such as Cali and Bogotá, in periods with high rates of homicide, have proved effective.36 Firearms increase the lethality of violence. According to the Small Arms Survey, Latin America has the greatest proportion of firearm-related fatalities in the world.37
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
Homicide rates in Latin America are unevenly distributed. This uneven distribution is not only geographical, but also appears in different population sub-groups and can be explained by the existence of diverse social, political, and cultural contexts between and within countries. In those countries where homicide rates are extremely high, governments should review current policies and identify what is effective and what is not. This could contribute towards identifying effective preventive actions to reduce homicide. Fortunately, there are current initiatives in the region that have focused on changing facilitating, or promoting factors, and are aiming to change originating factors of violence.
|
Funding |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
Fondo Nacional de Ciencia, Tecnología e Investigación, FONACIT, Venezuela (to R.B.L); U. S. Centers for Disease Control and Prevention (R49/CCR402444 to A.V.)
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsFundingReferences |
|---|
1 Renneville M. L’Anthropologie du Criminel en France. Criminologie (1994) XXVII:185–209.
2 Abad H. La violencia necesita estudios epidemiológicos. Tribuna Médica (1962) 2:10–12.
3 Haddon W Jr. The changing approach to the epidemiology, prevention, and amelioration of trauma: the transition to approaches etiologically rather than descriptively based. Am J Public Health Nations Health (1968) 58:1431–38.[Web of Science][Medline]
4 Centers for Disease Control and Prevention. The Public Health Approach to Violence Prevention. Accessed 1 May, 2008. [World Wide Web] 2008. Available at: http://www.cdc.gov/ncipc/dvp/PublicHealthApproachTo_ViolencePrevention.htm.
5 Durkheim E. Les règles de la méthode sociologique (1968) 17 édn. Paris: Presses universitaires de France.
6 Centers for Disease Control and Prevention – C.D.C. Recommended Framework for Presenting Injury Mortality Data. MMWR Morb Mortal Wkly Rep (Recommendations and Reports) (1997) 46(RR–14):i–iv, 1–33.
7 Concha-Eastman A, Villaveces A. Guidelines for the Epidemiological Surveillance on Violence and Injuries (2001) Washington, DC: Pan American Health Organization/World Health Organization. Report No.: HCP/HCN/01.01.
8 Krug E, Dahlberg LL, Mercy JA, Zwi AB, Lozano R. World Report on Violence & Health (2002) Geneva: World Health Organization.
9 Pan American Health Organization. Health Conditions in the Americas: Basic Indicators (2006) 1st. Washington DC: Panamerican Health Association (PAHO).
10 Comisión Económica para América Latina y el Caribe (CEPAL). Estadísticas Sociales, Anuario Estadístico de América Latina y el Caribe 2007 (2008) Santiago de Chile: United Nations Organization.
11 Observatorio Centroamericano sobre Violencia (OCAVI). Tasas de homicidios dolosos en Centroamérica y República Dominicana. Accessed 1 May, 2008. [World Wide Web] 2007. Available at: http://www.ocavi.com/docs_files/file_384.pdf.
12 Programa Venezolano de Educación-Acción en Derechos Humanos. In: Derecho a la seguridad ciudadana (2006) Caracas: Programa Venezolano de Educación-Acción en Derechos Humanos (PROVEA).
13 Cruz JM. La Victimación por violencia urbana: niveles y factores asociados en ciudades de América Latina y España. Rev Panam Salud Publica (1999) 5:259–67.[Medline]
14 Dobrin A, Brusk JJ. The risk of offending on homicide victimization: a public health concern. Am J Health Behav (2003) 27:603–12.[Web of Science][Medline]
15 Gartner R. The victims of homicide: a temporal and cross-national comparison. Am Sociol Rev (1990) 55:92–106.[CrossRef][Web of Science]
16 Programa de Seguridad Ciudadana y Prevención de la Violencia El costo económico de la violencia em GuatemalaPNUD, Guatemala 2006. (2006) Ciudad de Guatemala: Programa de las Naciones Unidas para el Desarrollo (PNUD).
17 Bronfenbrenner U. The Ecology of Human Development: Experiments by Nature and Design (1979) Cambridge, MA: Harvard University Press.
18 Moser C, Shrader E. Crimen, violencia y pobreza urbana en América latina: Hacia un marco de referencia integrado (1998) Report No.: Working paper. Washington DC: The World Bank.
19 Rubio M. Crimen e impunidad: precisiones sobre la violencia (1999) 1a ed. Bogotá, Colombia: TM Editores, Universidad de Los Andes (CEDE).
20 Cerqueira D, Lobão W. Determinantes da criminalidade: arcabouços teóricos e resultados empíricos. Dados (2004) 47:233–69.
21 Concha-Eastman A. Urban violence in Latin America and the Caribbean: dimensions, explanations, actions. In: Citizens of Fear: Urban Violence in Latin America—Rotker S, ed. (2001) New Brunswick: Rutgers University Press.
22 Briceno-Leon R. Urban violence and public health in Latin America: a sociological explanatory framework. Cadernos de saude publica/Ministerio da Saude, Fundacao Oswaldo Cruz, Escola Nacional de Saude Publica (2005) 21:1629–48. discussion 49–64.
23 Briceno-Leon R. Violence in Venezuela: Oil rent and Political Crisis. Ciencia & Saúde Colectiva (2006) 11:315–25.
24 Inter-American Development Bank. Youth violence prevention: Technical Note 10 (1999) Washington DC: Inter-American Development Bank.
25 CEPAL. Panorama social de América Latina (2004) Santiago de Chile: CEPAL.
26 International Labour Office. Global Employment Trends for Youth (2006) Geneva: International Labour Office.
27 Centro Latinoamericano y Caribeño de demografía (CELADE). Transformaciones demográficas en América Latina y el caribe y conseucencias para las políticas públicas. In: Panorama Social de América Latina (2004) Santiago de Chile: Comisión Económica para América Latina y el Caribe (CEPAL).
28 Guerrero R, Concha-Eastman A. An Epidemiological approach for the prevention of urban violence: The case of Cali, Colombia. J Health Popul Dev Ctries (2001) 4.
29 Concha-Eastman A. Ten years of a successful violence reduction program in Bogota, Colombia. In: Preventing Violence from Global Perspectives to National Action—McVeigh C HK, Lushey C, Bellis M, eds. (2005) Liverpool: Centre for Public Health, Liverpool John Moores University.
30 Concha-Eastman A. The DESEPAZ Programme: Development, Security and Peace. In:. In: Ciudad y Violencias en América Latina—Concha Eastman A, Carrion F, Cobo G, eds. (1994) Quito: Programa Gestión Urbana y Alcaldía de Cali. 119–53.
31 Márquez PC. The Street is my Home: Youth and Violence in Caracas (1999) Stanford, CA: Stanford University Press.
32 Zubillaga V, Briceño-León R. Exclusión, masculinidad y respeto: Algunas claves para entender la violencia entre adolescentes en barrios. Nueva Sociedad (2001) 173:34–78.
33 National Institute of Justice. Youth Violence, Guns & Illicit Drug Markets (1996) Washington DC: US Department of Justice.
34 Instituto Cisalva. Sistema de Información sobre muertes violentas en Santiago de Cali (1993–2008) CA: Universidad del Valle.
35 Gruenewald PJ, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res (2006) 30:1184–93.[CrossRef][Web of Science][Medline]
36 Villaveces A, Cummings P, Espitia VE, Koepsell TD, McKnight B, Kellermann AL. Effect of a ban on carrying firearms on homicide rates in 2 Colombian cities. JAMA (2000) 283:1205–09.
37 Small Arms Survey. 2004: Rights at Risk (2004) 1st edn. Geneva: Switzerland: Oxford University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. A. Lotufo and I. M. Bensenor Income inequality and male homicide rates: Sao Paulo, Brazil, 1996-2007 Eur J Public Health, December 1, 2009; 19(6): 602 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G A. Nieto, F S. Mutis, R Mercer, M Bonati, and I Choonara Child health in Colombia Arch. Dis. Child., November 1, 2009; 94(11): 900 - 903. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. J. Cook, W. Cukier, and K. Krause The illicit firearms trade in North America JCriminology and Criminal Justice, August 1, 2009; 9(3): 265 - 286. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||