OVER THE LAST few years, there has been a growing appreciation of the need for more information about complex humanitarian emergencies in order to develop an understanding of such events and and more effective reactions to them. This book is a result of the collaboration of four institutions, the U.S. National Academy of Sciences, the National Academy of Engineering, the Institute of Medicine and the National Research Council. A roundtable was developed from this collaboration, which led to a workshop in 1999; this book is drawn from papers presented at that workshop.
The introduction reviews the information on mortality in complex emergencies and related data issues. It provides an overview of the state of knowledge, the gaps that need attention, and aspects of the social and operational situation that affect data collection, interpretation and application. The case studies in this volume are from four different regions and examine four different kinds of crises, namely from Rwanda, Kosovo, North Korea and Cambodia. They seek to summarize what is known but they also illustrate the need for progress in the knowledge base used to address complex emergencies.
In the first case study, the flight of Rwandan refugees into Eastern Zaire (now the Democratic Republic of Congo) after the 1994 genocide is described. Mortality at various stages of the forced migration is discussed. Using a combination of surveillance systems and retrospective mortality surveys, mortality rates were estimated for the same refugee population at four different points in time and at four different geographical locations. The pattern that emerges is disturbing; by the final estimation only about 20 per cent of the refugee population remained and the rest were either dead or missing. The merits and drawbacks of both systems of mortality estimation are also addressed.
The same methods were used in the case study of Albanian Kosovar refugees from March to June 1999. One set of data was collected from surveillance systems that were operational in refugee areas in Albania and the former Yugoslav Republic of Macedonia during the refugee crisis. The second data set was collected in Kosovo in September 1999 after the majority of the refugees had returned home. The results were compared and it was found that the overall mortality in the Kosovo crisis was relatively low. The different nature of the populations and the fact that the crisis was in a more developed region raises several methodological issues about estimating mortality, such as the importance of chronic diseases, reproductive health and psycho-social trauma. In the third case study, indirect estimation techniques were used to estimate mortality rates in an isolated population suffering from famine in North Korea. By interviewing North Korean migrants who had crossed the border into China in search of food about their own household experiences and the experiences of a sibling, non-migrant household, they were able to estimate mortality rates. Although the sample is non-representative, the study gives an insight into what was happening inside North Korea. The final case study was not presented at the workshop but was commissioned afterwards. It describes a variety of data sources and techniques that can be used to estimate the total excess mortality during the Cambodian crisis of 1975 to 1979. Survey and census data are discussed but the focus is again on indirect estimation techniques, including demographic projection methods that seek to estimate total excess mortality and decomposition methods to obtain age- and cause-specific mortality.
The book ends with a reflection on the four case studies and the difficulty and necessity of collecting statistics in emergency situations is raised. Practitioners and researchers are reminded that each crisis is a unique event and must be understood not only on the basis of its similarities to other events but also on the basis of its specificity.