Apart from being a US based forum, the link alderick84 provides also shows EXACTLY why we stick to the 12-week conclusive result despite the DUO's extreme accuracy and an 8 week negative.
=====================================================
Dr HHH:
"Dr. Bob is simply saying that when the risk of HIV infection is particularly high, it is sometimes wise to have an extra test at the lastest possible time. WE DO THE SAME THING HERE. Even though we almost always say 6-8 weeks is plenty of time, when someone is at particularly high risk, WE ADVISE TESTING AT 3 MONTHS. "
"Now assume a known HIV infected partner has unprotected anal sex. The receptive partner has a 1% chance of catching HIV, or 1 in 100. If a test at 4 weeks is negative, his chance of being infected drops to 1 in 1,000. A lot better than 1 in 100, but obviously not good enough. Same test, done at the same time after exposure, but a very different level of reassurance. This person clearly needs another test or two. Even at 6-8 weeks, when the test picks up nearly all (let's say 99%) of infections, the chance he has HIV is still 1 in 10,000. Better still, but good enough? Maybe not. THEREFORE DR HOOK OR I WOULD RECOMMEND A 3 MONTH TEST."
Please follow the link alderick84 has kindly provided, which also redirects you to other links, and you will see the quotes above from Dr Handsdsfield.
http://www.medhelp.org/posts/HIV---Prevention/what-is-the-window-period-for-hiv/show/1704700