C.1.What is the true death toll of AIDS in rural Malawi? Population estimates on the distribution of causes of death from the Karonga continuous registration system (crs)
By A Jahn, F Mwaungulu, V Msiska, A Crampin, L Kachiwanda, S Kasimba, N McGrath, P Fine,
Objectives and Scope: To estimate causes of death in an area of Karonga District where an estimated 15% of the adult population are HIV positive.
Methodology: Malawi’s first demographic surveillance system began in 2002 in the south of Karonga District covering a population of 30,000. Verbal autopsy (VA) interviews are conducted on all deaths and available health records reviewed. Three medical officers independently review each history including any available HIV test results and assign a cause of death. Discrepantly coded cases are discussed and consensus reached if possible.
Results: By December 2004, 494 deaths had been observed in 45,000 person-years (crude death rate: 11 per 100). Recent HIV test results were available for 50 deceased (10%) and for 38 mothers of children dying before age 10 (24%). There was an excess of young adult deaths from age 20. A cause of death could be assigned to 88% of deaths and 161 deaths (32%) were attributable to AIDS. Twelve percent (19/159) of deaths in children aged <10 and 66% (118/179) of deaths in people 20-49 were identified as AIDS-related. 50% of AIDS-deaths occurred in health facilities. Analyses of non-AIDS related causes of death will also be presented.
Conclusions:AIDS is the main adult killer in this population but is also responsible for a significant number of childhood deaths. Health facility-based death statistics would account for only 50% of the burden of AIDS deaths. The systematic review of health documents and supplementation with circumstantial information plus the availability of recent HIV test results is likely to have improved sensitivity and specificity of the VA tool in the data presented.
Recommendations:The CRS is a valuable tool in measuring cause specific mortality in Karonga. Additional sample vital registration systems would be needed in Malawi to obtain nationally representative estimates.