Community level risk factors for numbers of landmine victims in Chad and Thailand.

STUDY OBJECTIVE: To determine commonalities of landmine victim risk factors in two very different countries. DESIGN AND SETTING: Data on 249 communities in Chad and 530 in Thailand were collected during 2000-2001 as part of the Global Landmine Survey. Community level variables were analysed in a series of Poisson mixture models with number of landmine victims as the dependent variable. Models developed for each country were tested on the other to investigate similarities and robustness of identification of risk factors. MAIN RESULTS: Increased community level risk was associated with population size, closeness to another community with victims, emplacement in the previous two years, blocked water or pasture, and the proximity of unexploded ordnance or anti-tank mines. In Chad, risk factors tended to be more related to identifying communities that had crossed a threshold between near zero and moderate risk; Thailand, factors were more related to increases in victim rates. CONCLUSIONS: Current systems of collecting data on community characteristics and landmine victims can provide meaningful risk factor information. Remediation approaches that focus on blockage of important resources and areas of recent, high intensity conflicts may be the most beneficial in reducing the numbers of victims.

Uncertainty and information flows in humanitarian agencies.

Humanitarian agencies struggle with high uncertainty. The focus here is on uncertainty that grows out of the internal complexity of the agencies themselves, rather than simply emanating from their war-ridden environments. This is illustrated with a study of uncertainty management and information processing in two UN agencies working for the victims of the conflict in southern Sudan. Data from 1995 relief distributions are used to simulate aspects of uncertainty; paradoxically, uncertainty increased with growing success of the relief operation. The findings is in tune with observations from the same region that Jok (1996) reported in this journal. While uncertainty may remain difficult to reduce, co-operation among agencies and public confidence in their work provide functional equivalents for certainty.

Computer Simulation as a Means of Dialogue between Local Committees and Relief Agencies: A Case from Southern Sudan.

Computer simulations, using modest equipment - a portable computer with a familiar spreadsheet programme - can facilitate dialogue between local committees and relief agencies in the management of disasters. An example is given from the southern Sudan in which the outcomes of different relief strategies, some urged by the donors, were simulated. This helped the local committee to defend difficult choices and the donors to realise that the decision-making autonomy of the affected community should be respected.