|
.gif)
|
|
Sensitivity
requirements for the point of care diagnosis of Chlamydia
trachomatis and Neisseria gonorrhoeae in women.
Vickerman P,
Watts C, Alary M, Mabey D, Peeling RW.
Sexually
Transmitted Infections
2003;79:363-68
|
| by |
Margaret Sullivan
Pepe, PhD
University of Washington and
Fred Hutchinson Cancer Research Center
1100 Fairview Avenue N., M2-B500
Seattle, WA 98109-1024 |
This paper provides a rigorous
mathematical formula for the sensitivity required of a POC test so that
its use averts as many infections as the gold standard test. The formula
incorporates the many factors influencing STI detection, transmission and
treatment. However, there are several implicit assumptions made about the
context in which the POC test is evaluated. These assumptions deserve to
be highlighted.
- It is assumed that the current
standard of practice is to wait for the results of a gold standard
test before initiating treatment. The model does not consider that
treatment could be initiated because of symptoms apparent at the
initial visit. Neither does it allow for the fact that high risk
individuals who are unlikely to return for test results might be
treated without waiting for the results of the gold standard test.
Although either of these events invalidate the current mathematical
derivations, in the future the model could possibly be extended to
incorporate them.
- It is assumed that the false positive
rate of the POC test is less than or equal to that of the gold
standard. The authors therefore do not consider relative impacts of
false positive results because it is assumed that false positives will
not be increased with the POC test. However if the POC test is even
slightly less specific than the gold standard then increases in over
treatment and the social consequences of wrong STI diagnosis would
need to weigh into the decision to use the POC test.
- It is assumed that test results have
no impact on subsequent behavior related to STI. However, a false
negative test result provides false reassurance. Potentially this
could lead to more risky sexual practices (like reduced condom use).
Or, the subject might be less likely to seek testing in the future. If
either of these factors plays a role in practice, then using a test
with low sensitivity could have more negative impact than is
considered in this paper.
- It is assumed that women need to
return to the clinic for results of the gold standard test and
treatment. However, in developed countries access to test results and
treatment could be made more readily available without the burden of a
return visit and thereby increase the effective return rate. Making
the more accurate gold standard test more effective in this way is
also worth considering.
This paper investigates one important
aspect of evaluating if a POC test with suboptimal sensitivity may be
useful relative to the gold standard test for detecting STI. However, the
take home message is not that low sensitivity POC tests are generally
adequate. They may have utility in certain settings and must be evaluated
from various points of view relative to other tests and strategies
available
|
