Article Text



Statistics from

It is thought that by August 2004 more than a million people in Darfur, Sudan had been displaced from their homes by the fighting and about 188 000 had fled to Chad. In February 2003 the emergence of two anti-government rebel groups (the Sudan Liberation Army and the Justice and Equality Movement) led to pro-government reprisals by the Janjaweed militia and units of the Sudanese army. Guesses about the number of people killed reached 30–50 thousand by July 2004. Workers with Médecins Sans Frontières (MSF) have reported data collected more systematically (

, see also Comment, ibid: 1290–1).

Non-governmental organisations were at first prevented from working in Darfur but MSF began to set up sites for internally displaced people in December 2004. They report mortality data collected retrospectively by surveys among 215 400 people at four sites in Western Darfur between April and June 2004. Heads of households were asked about deaths within a given recall period using a calendar of Islamic months and locally memorable events. At three of the sites mortality before arrival had been 5.9, 9.5, and 7.3 deaths per 10 000 people per day. About 70–90% of these deaths had been caused by violence. The victims were mostly adult men but women and children were also killed. The corresponding mortality rates in children under 5 years old were 2.8, 2.1, and 1.5 per 10 000 people per day. The proportion of deaths due to violence ranged from 0 to 35% among children under 15 years, from 36% to 91% among men, and from 0% to 47% among women. After arrival at the camps mortality rates fell five to eight fold but remained at emergency levels. The Janjaweed militia still attacked, however, and violence still accounted for between 7% and 21% of deaths. Among adults in the camps there was a dearth of men, with male-to-female ratios of between 0.61 and 0.83. There was a high proportion of men reported as dead or absent.

Novel methods of data recording and analysis have been used in this study. The collection of accurate data must be attempted during humanitarian emergencies so that effective interventions can be designed and, where there has been violence, so that those responsible can be held to account.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.