Validating Verbal Autopsy Methods for Death Causes
Author Information
Author(s): Bauni Evasius, Ndila Carolyne, Mochamah George, Nyutu Gideon, Matata Lena, Ondieki Charles, Mambo Barbara, Mutinda Maureen, Tsofa Benjamin, Maitha Eric, Etyang Anthony, Williams Thomas N
Primary Institution: KEMRI-Wellcome Trust Research Programme
Hypothesis
Can verbal autopsy methods accurately determine causes of death compared to hospital records?
Conclusion
Both the InterVA and PCVA methods effectively identified the top five causes of death in a rural Kenyan community.
Supporting Evidence
- The InterVA model and PCVA identified similar top causes of death.
- Kappa values indicated moderate agreement between methods and hospital records.
- The study validated the WHO verbal autopsy tool against hospital cause of death.
Takeaway
This study shows that asking families about how someone died can help figure out the main reasons people die, even when doctors can't check directly.
Methodology
Verbal autopsy interviews were conducted for 145 adult deaths, and data were analyzed using kappa statistics and ROC curves.
Potential Biases
Potential biases may arise from the reliance on verbal accounts and the absence of comprehensive medical records for all deaths.
Limitations
The study had a small sample size and lacked postmortem examinations, which could affect the accuracy of cause of death determinations.
Participant Demographics
The mean age at death was 55 years, with 56% males and 44% females.
Statistical Information
P-Value
0.32 for InterVA, 0.52 for PCVA
Confidence Interval
95% CI: 0.37, 0.48
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website