hivtest {lcca}R Documentation

HIV test dataset

Description

This dataset concerns the diagnostic accuracy of tests for HIV infection. Results of four different tests are reported for 428 high-risk patients. Yang and Becker (1997) applied latent-class analysis to these data to estimate the sensitivity and specificity of these tests in the absence of a gold standard.

Usage

hivtest

Format

a data frame with 9 rows and 5 variables:

A

test result from radioimmunoassay (RIA) utilizing recombinant agl21 (1=negative, 2=positive)

B

test result from RIA utilizing purified HIV p24 (1=negative, 2=positive)

C

test result from RIA utilizing purified HIV gpl20 (1=negative, 2=positive)

D

test result from enzyme-linked immunosorbent assay (1=negative, 2=positive)

COUNT

frequency; number of patients exhibiting the given pattern of results

Source

Yang, I and Becker, M.P. (1997) Latent variable modeling of diagnostic accuracy. Biometrics, 53, 948-958.

References

For example analyses of this dataset using functions in the LCCA package, see the manual LCCA Package for R, Version 1 in the subdirectory doc.


[Package lcca version 2.0.0 Index]