Cause of death: where there is no physician
Verbal autopsy diagnostic algorithms
As noted earlier, traditionally VA interviews are analysed by physicians to come up with a likely COD (PCVA). Given the workload on physicians, this approach leads to delays due to time constraints faced by physicians who are involved in day-to-day care of patients. Also, physicians do not always agree on the diagnosis of COD, so it is necessary to involve two or more physicians to determine a final, agreed cause. Therefore, inter-rater-reliability (or diagnostic consistency) among physicians reading the same VA interview could be low, and unstandarised across (and within) countries.
For VA to be used routinely and generate CSMFs in a timelier way, standardised, and cost-effective way, more efficient ways of determining the COD based on the interviews are needed. This has led to the development of automated algorithms. These methods have been shown to perform as well as, if not better than PCVA and have the benefit of consistency, extremely low-cost and reliability, as well as speed. Results can be available immediately following the interview.
The PHMRC questionnaire collects only the information required by the Tariff method that uses SmartVA Analyze.
The WHO 2016 VA questionnaire, asks a number of questions about symptoms in order to run any one or all of the three automated diagnostic methods currently available: Tariff, InterVA and InSilicoVA. This could lead to three different diagnoses of the most probable cause of death. Guidance to assist countries with the interpretation of multiple diagnoses of the most probable cause of death is being developed.
Desai N et al. (2014). Performance of four computer-coded verbal autopsy methods for cause of death assignment compared with physician coding on 24,000 deaths in low- and middle-income countries. BMC Medicine.
Institute for Health Metrics and Evaluation (2011). Verbal autopsy: innovations, applications and opportunities. Population Health Metrics.
Institute for Health Metrics and Evaluation. Verbal Autopsy Tools.
Leitao J et al (2014). Comparison of physician-certified verbal autopsy with computer-coded verbal autopsy for cause of death assignment in hospitalized patients in low- and middle-income countries: systematic review. BMC Medicine.
Lopez AD et al (2011). Verbal autopsy: Innovations, applications, opportunities—Improving cause of death measurement. Population Health Metrics.
Murray CJL et al (2011). Population Health Metrics Research Consortium Gold Standard Verbal Autopsy Validation Study: design, implementation, and development of analysis datasets. Population Health Metrics.
Serina P et al (2015). A shortened verbal autopsy instrument for use in routine mortality surveillance systems. BMC Medicine.
Serina P et al (2015). Improving performance of the Tariff Method for assigning causes of death to verbal autopsies. BMC Medicine.
Tyler H et al (2016). Probabilistic cause-of-death assignment using verbal autopsies. Journal of the American Statistical Association.