Activities
The openVA Team focuses on improving computer-based approaches to cause ascertainment using verbal autopsy data. This includes:
- developing new statistical/computational algorithms for cause classification,
- improving the symptom-cause information available for algorithmic cause assignment,
- supporting work on global standards for the VA instrument and symptom-cause information,
- developing machine learning/artificial intelligence approaches to cause classification,
- building training data resources for machine learning/artificial intelligence cause classification, and
- supporting work to bring pathology-based information into automated cause classification procedures using information from minimally-invasive tissue sampling.
Verbal Autopsy
Verbal autopsy (VA) is an old and well-established approach to ascertain cause of death when it is not feasible or practical to conduct medical certification or full autopsies. After a death is identified, a specially trained fieldworker interviews the caregivers (usually family members) of the decedent. A typical VA interview includes a set of structured questions with categorical/quantitative responses and a narrative account that records the 'story' of the death from the respondent's point of view as unstructured text. The resulting data are interpreted by various means to assign causes to the death.
Assigning Causes of Death Using Verbal Autopsy Data
Physician Assignment
The most common practice has been to have clinically trained, experienced physicians read the interviews and determine causes. To address the fact that physicians frequently do not agree on causes, VA interviews are often read by two physicians, and sometimes three, and the final causes are determined through a consensus mechanism. This implicitly acknowledges two of the challenges inherent to VA:
- without either clinical data or an autopsy it is difficult to be either specific or certain about the cause of death, and
- each physician has unique training and experience and is therefore biased in various ways when assigning causes of death.
The first is a fundamental limitation of VA resulting from the fact that VA data contain comparatively less information than clinical records and autopsy. A clear consequence of the second is that traditional physician-assigned VA causes of death are biased and not strictly comparable across groups of physicians.
Computer Assignment
An alternative to physician review is the use of an algorithmic method that processes the categorical/quantitative responses in VA interviews to identify causes of death. The algorithmic approach has three important advantages:
- physicians are free to spend their time caring for patients,
- VAs can be coded very quickly without having to wait for the always-lengthy physician review process, and
- physician-associated bias is removed from the process so that cause assignments are reproducible and comparable.