South African National Cause-of-Death Validation

Report 1 | Methodology and Description of a National Sample of Verbal Autopsies

Executive Summary

South Africa has a well-established Civil Registration and Vital Statistics System (CRVS) with a high proportion of deaths being registered. The quality of the cause of death statistics that arise from the death notification forms completed by medical doctors, however, is sub-optimal, making it difficult to obtain a reliable cause of death profile or monitor trends in the number of deaths from specific causes. In addition, there is extensive underreporting of HIV as an underlying cause of death. Although estimates based on demographic models of the epidemic suggest that 17% of the deaths in that 2016 were due to HIV, the country’s vital statistics report indicated that only 4.8% of all the registered deaths were due to HIV.

The South Africa (SA) National Cause of Death Validation Project (NCODVP) was implemented by the South African Medical Research Council (SAMRC) and partners to conduct a validation of CRVS cause-of-death information by linking CRVS data to data obtained from medical records, forensic pathology records, and verbal autopsy interviews for a national sample of deaths. The main purpose of the study was to compare the registered underlying cause of death indicated on the CRVS medical notification form with the highest level of information collected in the study (forensic pathology record followed by medical records and verbal autopsy) so that correction factors could be estimated. Additionally, the study aimed to compare the medical cause of death identified from the different sources to assess their performance in identifying cause of death. The study protocol was reviewed and approved by the SAMRC Ethics Committee and the United States Centers for Disease Control and Prevention (CDC)Centers for Global Health Associate Director for Science. Support was obtained from the National Department of Health and permissions were obtained from each provincial Department of Health and health facility included in the study.

View the complete South African National Cause-of-Death Validation Report

Report on the quality of Physician Reviewed Verbal Autopsies cause-of death data using ANACONDA for the South African National Cause-of-Death Validation Project

The Analysis of Causes of (National) Deaths for Action, or ANACONDA,1 software tool used in this analysis was developed by the Melbourne School of Population and Global Health at the University of Melbourne, with the support of the Swiss Tropical and Public Health Institute at the University of Basel. ANACONDA is specifically designed to be applied to large datasets such as those from civil registries, or any other sources that routinely collect and generate cause of death data coded to the International Standard Classification of Diseases and Related Health Problems, 10th Revision (ICD-10), classified by age and sex.

We have used ANACONDA to assess the quality of the cause of death data from the South African National Cause-of-Death (SA NCOD) Validation Project based on the physician reviewed verbal autopsy interviews. Several steps of ANACONDA, related to assessing the completeness of death registration, have been omitted as they were not relevant for these data.

View the complete Report on the quality of Physician Reviewed Verbal Autopsies cause-of death data using ANACONDA for the South African National Cause-of-Death Validation Project