ENUMERATING AND MONITORING VULNERABLE SUB-POPULATIONS

Project Summary/AbstractCrude mortality has been used by the public health community as a cornerstone for defining a public healthemergency. Typically mortality in emergencies is monitored by surveillance processes that are incomplete orsurveys which are often done poorly and ignored where that is politically expedient. This project will attempt toestablish two different techniques, both simpler than surveys, and less likely to be dismissed by critics. 1) Theprocess of multiple source triangulation, or establishing the sensitivity of IDP and death tallies will draw uponmethods used widely in TB programs and the monitoring of homeless populations in large cities, but is rarelyused to assess mortality. This process of constructing a summary of all known deaths or IDP's, collecting anindependent sample, and examining the overlap between the two listings can be rapid, and can potentiallytranscend the biases typical in death and registration records. The project will select places to enumerate andmeasure mortality in at least one population displaced by an acute disaster, and at least one population that ispersecuted or associated with significant stigma'. 2) A statistically representative community-basedsurveillance network was established by RHA with the Ministry of Health in Fizi Health Zone, DRC. Communityhealth workers recorded birth, death, and migration events. Preliminary analyses suggest that the surveillancesystem detected 93% of births and 87% of deaths, with a specificity >99% for both measures, and positivepredictive values of 98% (births) and 91% (deaths). This project will attempt to see if a similar system can beestablished in the city of Goma, to monitor IDPs in an urban setting. While not applicable to all crises, IDPs inNairobi and Jordan and other large urban centers where IDPs remain stable over months or years couldpotentially be monitored with this approach. Once established, collection of other desired information from astatistically representative sample could easily be added into the surveillance process.To maximize policy impact, the later years of the project will involve extensive coordination with CDC, OFDA,and other donors in an attempt to establish procedures for mortality measurement in the locations and withpartners of priority to donors. Without a widespread and credible capacity to measure central outcomes likeincome, malnutrition, and mortality, funding of humanitarian emergencies will continue to be based more onpolitical interests than on public health need, and successful life-saving relief operations will remainindistinguishable from ineffective efforts. By adding two new simple and inexpensive options to the toolbox ofthe humanitarian community, we can help develop a community focused on outcomes instead of processes.