Dear.Colleagues,

     I devote much time to train my students and residents about the VITAL importance of "proper estimation of Pretest Probability" before requesting any test for diagnosing or screening.  The US Preventive Services Task Force (USPSTF) strongly recommends that clinicians screen all people aged 15 to 65 years for HIV infection, according to a draft recommendation statement posted online November 20 (http://www.uspreventiveservicestaskforce.org/draftrec.htm).  The USPSTF is offering the public an opportunity to comment on this draft recommendation until December 17 and will consider all public comments when writing its final recommendation.  I have learned from this very useful paper ( http://psi.sagepub.com/content/8/2/53.short ) how not considering the "pretest probability for having HIV infection" may result in the TRAGEDY of false positive.  I am pasting the following text from the paper that addresses this issue ( Please see Figure 8 in attached file.  In recent years I have talked with WHO experts about the False positive results of HIV testing when the pretest probability is very low) :

The Illusion of Certainty

Physicians need to inform patients that even the best tests are not perfect and that every test result therefore needs to be interpreted with care or the test needs to be repeated. Some test  results are more threatening than others and need to be handled particularly carefully. One terrifying example is a positive HIV test result. At a conference on AIDS held in 1987, former Senator Lawton Chiles of Florida reported that of 22 blood donors in  Florida who had been notified that they had tested positive withthe ELISA test, 7 committed suicide. A medical text that documented this tragedy years later informed the reader that ‘‘even if the results of both AIDS tests, the ELISA and WB [Western blot], are positive, the chances are only 50-50 that the individual is infected’’ (Stine, 1999, p. 367). This holds for people with low risk behavior, such as blood donors. Indeed, the test (consisting of one or two ELISA tests and a Western Blot test, performed on a  single blood sample) has an extremely high sensitivity of about
99.9% and specificity of about 99.99% (numbers vary, because various criteria have been used that maximize specificity at the expense of sensitivity, or vice versa). Nonetheless, due to a very  low base rate in the order of 1 in 10,000 among heterosexual men with low-risk behavior, the chance of infection can be as low as 50% when a man tests positive in screening. This striking result
becomes clearer after these percentages are translated into natural frequencies: Out of every 10,000 men, it is expected that one will be infected and will test positive with high probability; out of the other, non infected men, it is expected that one will also test positive (the complement to the specificity of 99.99%).  Thus, two test positive, and one of these is infected (Fig. 8).AIDS counselors need to properly inform everyone who takes the test.

Best regards, Mohammad

Mohammad Zakaria Pezeshki, M.D.
Associate Professor
Program for Estimation of Pretest Probability
Department of Community Medicine,
Tabriz Medical School, Golgasht Avenue, Tabriz, Iran,
Tel: 0098 411  336 46 73
Fax: 0098 411 336 46 68