ArticlePLOS global public health2026
A comparison of verbal autopsy assignment methods to obtain adult cause-specific mortality in two longitudinal studies in Rakai and Kalungu districts of South Central, Uganda.
Article in PLOS global public health, 2026. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
9 authors.
Funding
Abstract
Verbal Autopsy (VA) data determines cause of death (CoD) in settings where certification is lacking, using structured interviews on symptoms and circumstances before death. Various methods are used to interpret this information and assign a probable cause of death. Debate exists over replacing physician reviews with automated methods. We compared physician reviews with two computer algorithms in assigning adult CoD. We used adult (≥15 years) VA data collected at two Health and Demographic Surveillance System (HDSS) sites in Uganda (Rakai and Kyamulibwa) collected between 2013 and 2021. CoD were categorized into six broad groups, and observed cause-specific mortality fractions (CSMF) were calculated. We evaluated the performance of physician reviews and two algorithms (InterVA-5 and InSilicoVA) based on four individual and population-level metrics (CSMF accuracy, percentage agreement, sensitivity, and Spearman's correlation coefficient). Data from Rakai compared physician reviews and algorithms, while Kyamulibwa data compared the two algorithms. A total of 1564 VA records from Rakai and Kyamulibwa were analysed. In Rakai, InSilicoVA showed higher CSMF for other communicable causes, excluding HIV/TB (males 24.1%, 95% CI:20.1-28.7; females 25.2%, 95% CI: 20.8-30.3) than Physician reviews (males 14.8%, 95% CI: 11.6-18.8; females 17.5%, 95% CI: 13.8-22.1) and InterVA-5 (males 11.1%, 95% CI: 8.3-14.8; females11.7%, 8.6-15.7). Non-communicable diseases CSMF was lowest with InSilicoVA (males 30.5%, 95% CI:26.1-35.4, and females 38.46%, 33.31-43.89) compared to the InterVA-5. The CSMF accuracy and percentage agreement demonstrated comparable performance between physician reviews and computer algorithms, with substantial agreement in identifying causes of death. InSilicoVA was more sensitive for infectious, pregnancy, and external causes, while InterVA-5 better identified non-communicable and HIV/TB-related deaths. Computer algorithms can complement physician review in resource-limited settings, but current VA tools rely on structured symptom data and exclude rich narrative information. Incorporating qualitative information in future algorithms may improve symptom-cause relationships, accuracy, and cause-of-death assignment.
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.