As has been pointed out many times before, if Knox's recollection of this exchange is accurate (and of course there are plenty of reasons why it possibly might not be accurate), then the prison doctor absolutely would not have been following WHO protocol in respect of these 2-part assays which were used in 2007.
The very reason why 2-part tests were used at that time (before accurate and cost-effective 1-part assays were widely available) was this: the first of those tests (the ELISA test) was used to test for a true negative - i.e. if the ELISA test came back negative, then HIV was virtually guaranteed not to be present (= an extremely low false-negative ratio). The patient could at that point be reassured that they were not an HIV carrier, and no further testing would be required.
BUT
If the ELISA test came back positive, this in and of itself was extremely far from indicating that the patient was carrying HIV. In fact, if the patient did not identify in a high-risk group (e.g. promiscuous gay men or intravenous drug users who shared needles), an ELISA positive was still significantly more likely to end up with a negative diagnosis for HIV than a positive diagnosis.
And that's where the second test came in: the Western Blot test. This test took much longer to produce results than the ELISA test, and cost a lot more money. It also had a high false-negative ratio for HIV. But since false negatives had already effectively been ruled out by a positive ELISA test, if the subsequent Western Blot test came back negative, it was virtually certain to be a true negative, and therefore a positive ELISA followed by a negative WB would act as a confirmatory negative for HIV.
And conversely, the WB assay had a very low false-positive ratio (again, the exact opposite of ELISA). Therefore a positive ELISA followed by a positive WB served as a confirmatory positive for HIV.
What does all of that mean in respect of someone like Knox? Well, the most important part is the bolded sentence above. Which is precisely why WTO guidelines around these 2-part HIV assays clearly stated that if the patient's first ELISA test comes back positive, the patient (in all cases, but especially in the case of patients who were not identifying in any high-risk groups) should receive that information in an extremely carefully-framed manner. The standard practice would be for the physician to say something like: "We are going to need to carry out a second, confirmatory test on your sample; the first test indicated that there's at least a possibility that you might carry HIV, but it's nothing more than that at this stage and it's only by carrying out the second test that we will know properly." And for people identifying in low-risk groups it would have been appropriate to add: "For someone with your lifestyle, even a positive first test means it's still actually a good deal more likely that you do not carry HIV than that you do carry it."
It doesn't take a genius to figure out why it would have been ethically proper to be extremely careful about the way in which a positive ELISA test was revealed to a patient - particularly to one with a low-risk lifestyle......
