Citation

Snow RW, Basto de Azevedo I, Forster D, Mwankuyse S, Bomu G, Kassiga G, Nyamawi C, Teuscher T, Marsh K. Int. J. Epidemiol. 1993; 22(4): 677-683.

Affiliation

Kenyan Medical Research Institute Coastal Unit, Kilifi, Kenya.

Copyright

(Copyright © 1993, International Epidemiological Association, Publisher Oxford University Press)

DOI

unavailable

PMID

8225743

Abstract

Verbal autopsies (VA) are widely used by population and health scientists to determine individual causes of death in areas where most deaths occur at home and well-documented clinical data on cause of death are usually unavailable. VA interviews are based upon key symptoms and signs recalled by relatives of the deceased. In order to assess the reliability of the technique, the accuracy with which mothers and normal guardians recognize and recalled specific symptoms and clinical signs over time was assessed in the cases of 491 children who died on the pediatric wards of 2 district hospitals in Ifakara, Tanzania, and Kilifi, Kenya. The bereaved were interviewed 3 days to 24 months after child death. Recall after 1 month was similar to recall after 6 months for most signs and symptoms except neonatal tetanus for which false positives reported by mothers increased with time after death. Kwashiorkor, measles, trauma, generalized convulsions, and neonatal tetanus were reported with a high degree of accuracy. Symptoms and signs commonly used to describe malaria, respiratory tract and diarrhea- related deaths, however, were reported by mothers to have been present during terminal illness in 43% of cases where the features were absent. Finally, recall abilities differed between the 2 communities studied.

Language: en