Well, yes, but….

The first legally approved HIV self testing kit has gone sale in the UK.

The BioSURE HIV Self Test will enable people to test themselves when and where they like, with a 99.7 per cent accuracy rate.

Don’t we want to know both the false positive and false negative rates?

15 comments on “Well, yes, but….

  1. As somebody who has failed an HIV test thanks to an incompetent Australian lab, and thererfore read up considerably on the subject, false negatives are almost unheard of. If you have a positive – real or false – you always do another test to confirm, and then further tests. The HIV test as we know it is really about proving the absence of antibodies, not the presence.

  2. So if you’ve got a .1% false positive rate, that’s about 60,000 false positives if you do the entire population, right?

  3. .2%, specificity is the false positive rate. This (and cost) is why we don’t screen the whole population for HIV.

    On the other hand, if you’re self testing then the false positive just means you go to the doctor and get a better test, so apart from the stress and the period of abstinence while the results come back it’s not nearly as bad as a false negative.

  4. MatthewL: ‘This (and cost) is why we don’t screen the whole population for HIV.’

    Also a little matter like consent. You do understand that we aren’t all cattle or possessions of your beloved state, surely?

  5. .2%, specificity is the false positive rate. This (and cost) is why we don’t screen the whole population for HIV.

    Exactly: HIV is pretty difficult to catch, and despite the scare stories most of the population of the UK is simply not exposed. If you’re using needles for drugs, having unprotected gay sex with random men, or using prostitutes from/in Africa then you’re exposed. Otherwise, you’re not at risk, in which case they test you for chlamydia on the reasonable grounds that any straight sexual partner who might have HIV will almost certainly have that.

    On the other hand, if you’re self testing then the false positive just means you go to the doctor and get a better test, so apart from the stress and the period of abstinence while the results come back it’s not nearly as bad as a false negative.

    Yup, that’s exactly what happens. They tell you to wait three months, and then you say “fuck that” and get another test within 24 hours. The period of abstinence is easy enough, the stress no fun at all. Really, no fun at all.

  6. Matthew L – “.2%, specificity is the false positive rate. This (and cost) is why we don’t screen the whole population for HIV. ”

    And, of course, because the Gay lobby is against any measure that might actually work to restrict the spread of HIV. They would prefer to blame Reagan.

  7. .2%, specificity is the false positive rate

    Actually, a result of “inconclusive” is more common than a false positive. I don’t know if the former is included in that .2%. An inconclusive result is that of the 5 separate tests for HIV antibodies, some come back positive, some negative. This is common when somebody has just caught HIV and the different antibodies are still forming (and during this period you are very infectious). So even though the odds of me being in that .2% were pretty slim, the chances of me having contracted HIV in the 3 months prior to the test were precisely zero (unless there has been a species leap from human sexual contact to internet pr0n).

    What frightened me was the doctor called me up and said:

    “Don’t worry, this happens with some people, who have a certain blood characteristic and we just test again and if we get the same result we conclude it’s negative. Have you had an HIV test before?”

    “Yes, about 20 in all.”

    “Did you ever have this problem before?”

    “No”

    “Oh”

    Thanks, doc.

  8. False negatives are impossible to know. If A gets a negative in January but tests positive in June, how do we know whether the test in January was false? It may have been correct, but infection occurred after, or the test taken too early to say.

    This is why screening is really a waste of time for any disease because a negative result has a short shelf life, and really we should screen for everything once a week.

    False positives are not readily apparent. False positive breast cancer is often discovered after surgery when the removed tissue is examined for carcinoma cells.

    Screening is expensive, causes many people stress because of uncertainty due to repeat testing, often invasive next time round, unnecessary treatment, and false positives result in delayed reporting of symptoms because the individual has ‘proof’ it cannot be that particular disease.

    The actual number of lives ‘saved’ is far short of the number of lives it is claimed, and grossly overestimates cost benefit claimed by those who promote screening… those being the ones to gain most one way or another.

  9. Don’t be an idiot, Julia, a screening program doesn’t mean everyone is rounded up and herded through a testing shed.

  10. Matthew L – “Don’t be an idiot, Julia, a screening program doesn’t mean everyone is rounded up and herded through a testing shed.”

    Although the New York Times is perfectly happy when Cuba does it. They published an interesting article a few years back on how Cuba uses its street committees to report anyone who is unusually sexually active. They are then compulsorily tested. And jailed if they test positive.

    Cuba does not have much of an HIV problem.

    But what was interesting is how strongly they approved of this.

  11. Here’s where Bayes’ theorem comes into play. If a disease or infection is rare (which HIV still is in the West) then the low a priori probability of any given person being HIV+ blows up the false positive ratio. Let’s say 0.3% of people in the UK really are HIV+. What is the probability that a given a positive test result a person really is HIV+?

    P(HIV+|+) = P(+|HIV+)P(HIV+)/(P(+|HIV+) + P(+|HIV-)P(HIV-))

    = (0.999 × 0.003)/(0.999 × 0.003 + 0.002 × 0.997)

    = ~60%

    So if you test positive, there’s still a 40% chance you are not HIV+. The false negative i.e. P(HIV+|-) result, which has the most public health significance, is negligibly probable:

    P(-|HIV+)P(HIV+)/(P(-|HIV+) + P(-|HIV-)P(HIV-))

    = (0.001 × 0.003)/(0.001 × 0.003 + 0.998 × 0.997)

    = ~0.0003%

  12. John B – “Screening is expensive, causes many people stress because of uncertainty due to repeat testing, often invasive next time round, unnecessary treatment, and false positives result in delayed reporting of symptoms because the individual has ‘proof’ it cannot be that particular disease.”

    And yet the Social Justice Warriors insist that screening *must* be done for the right sort of diseases like breast cancer, but that it must *not* be done for the other sort of politically correct diseases like HIV.

    One might think that health is not the real issue.

    The simplest approach would be to insist that everyone coming to the UK gets a combined DNA/HIV test as a condition for their visa. If they test positive, they don’t get to come.

  13. What a good idea – anyone testing positive for DNA is kept out of the country.

    Not least because it wouldn’t affect this site much – most of the commentators are either ex-pats or bots programmed by some mad scientist in the bowels of UKIP House.

Leave a Reply

Name and email are required. Your email address will not be published.