Abstract
Universal Risk Factors for Violence? Evidence from Low- and Middle-Income Countries
We know least about the causes of violence where it matters most. The highest rates of serious violence are found in Latin America and Sub-Saharan Africa, but nearly all major studies of the developmental origins of violence have taken place in a small number of high-income countries in Europe, North America, and Australasia. To reduce global violence by 50% in the next 30 years, major investment will be needed to identify causes of violence where rates are highest. Two large-scale birth cohort studies in the city of Pelotas, Brazil provide a unique opportunity to investigate the development of violence, and identify risk and protective factors in a middle-income context with a high rate of violence. Over 5,000 children in each study have been followed since their births in 1982 and 1993. Official crime records have been collected for both cohorts and their parents. Self-reported crime data was collected at age 18 in the 1993 cohort that can be directly compared with a well-matched British study (ALSPAC), including some 14,000 children born in the early 1990's in the Bristol city area. We have compared associations between health and socio-demographic risk factors in the perinatal period and violence at age 18 years between the 1993 Pelotas birth cohort and ALSPAC. Many of the same risk factors were important in both contexts; for example conduct problems were predicted in both Brazil and Britain by the following: unplanned pregnancy, maternal smoking in pregnancy, teenage mother, low maternal education, single mother, and low family income. However, several risk factor associations were significantly weaker in Brazil than in Britain and, strikingly, almost no risk factors were predictive of male youth violence in Brazil. Developmental origins of violence may vary across social context. Violence research must become truly global to produce an evidence base for effective action in every region.