Skip to main content

Table 1 Agreement between the four validation methods in determining the causes of children’s deaths

From: Verbal autopsy as a tool for identifying children dying of sickle cell disease: a validation study conducted in Kilifi district, Kenya

Methods

AC1for all CODs, % (SE) (95% CI)

AC1for COD due to SCD, % (SE) (95% CI)

PCVA coder 1 versus PCVA coder 2 (n = 610)

54.6 (2.0) (50.7 to 58.5)

97.4 (0.6) (95.9 to 98.7)

InterVA-4 model versus PCVA consensus (n = 610)

50.6 (5.0) (48.7 to 57.2)

94.6 (0.9) (95.0 to 99.0)

InterVA-4 model versus paediatric ward COD (n = 134)

42.5 (3.6) (35.4.0 to 49.6)

96.9 (1.5) (95.0 to 99.0)

PCVA consensus versus pediatric ward COD (n = 134)

50.0 (5.1) (40.0 to 60.0)

95.5 (1.8) (92.0 to 99.0)

PCVA consensus versus laboratory evidence (n = 93)

–

87.5 (4.4) (78.9 to 96.1.)

  1. AC1, Agreement coefficient of Gwet; CI, confidence interval; COD, cause of death; SCD, sickle cell disease. SE, Standard error; VA, verbal autopsy.
Back to article page

BMC Medicine

ISSN: 1741-7015

Contact us