996
False Positive Rate of Rapid Oral Fluid HIV Tests Increases as Kits Near Expiration Date
Shelley Facente*1, T Dowling1, E Vittinghoff2, D Sykes3, and G Colfax1
1San Francisco Dept of Publ Hlth, CA, US; 2Univ of California, San Francisco, US; and 3California Dept of Publ Hlth, Ctr for Infectious Diseases, Sacramento, US
Background: Since the
launch of the OraQuick Advance Rapid HIV-1/2 Antibody Test, episodes of
excess false positive results with oral fluid specimens have been well
documented in several US cities. To date, these false positive clusters have
remained unexplained. Because a recent cluster of false positive test results
in San Francisco occurred with kits close to their expiration date, we
hypothesized that there was a greater probability of false positive results as
the test kit approached expiration.
Methods: Results were
analyzed for 20,904 tests with either an initial HIV– result (n = 20,828),
assumed to be true negative results, or a preliminary positive result that was
then negative on indirect immunofluorescence assay or Western blot (n = 76),
the set of false positives. For each test, we computed time to expiration as
the interval from the date the kit was used to the expiration date printed on
the test pouch, which were both systematically recorded for each kit. We first
computed specificity for kits with time to expiration from ≤1 to ≥6
months, with exact binomial confidence intervals. Logistic regression was then
used to estimate the independent association of time to expiration with false
positive results, adjusting for site and technician effects.
Results: For 1108 kits
used in the last month before expiration, specificity was 98.83% (95% exact binomial
confidence interval 98.00 to 99.37%); the upper bound is below the specificity
of 99.60% claimed by the manufacturer. In a logistic model adjusting for site
and technician effects, false positive results were significantly more common
for the 1108 kits with ≤1 month remaining to expiration than for 1251
kits with ≥6 months remaining to expiration (odds ratio 16.7, 95%CI 2.17
to 129.4, p = 0.007); adjusted specificity for those kits was 99.51%
(95%CI 99.07 to 99.75%).
Conclusions: Specificity of
the OraQuick Advance rapid HIV test used with oral fluid declined
significantly with ≤1 month remaining to expiration. The specificity of
kits used in the last month before expiration appears to leave little margin
for error from other sources.
|
