TO ALL, Please Read

This was posted by someone else on the forum. I posted it to Dr. HHH a while ago and he didnt delete it so it must have made some sense to him.

The question of windows period post-exposure is one of the most common ones on these boards.  I've spent some time doing investigation.  I am a scientist by trade, although not a biologist nor medical doctor.  

The first thing that struck me is the lack of recent studies to reaffirm the current guidelines of 13 weeks. Especially as tests have improved over time.  I've found only one well-written study and it is this one (published: 2000)
http://www.aidsonline.com/pt/re/aids/fulltext.00002030-200010200-00014.htm;jsessionid=GpvJP3Q2p2xvCGsVJg2yylpMDpM5nV62XVcl0zWMchp922jmlCKF!-756477024!181195629!8091!-1

What they did is used a series of blood draws from people that would turn out to be HIV infected, and ran them against a set of HIV tests.  They used modern tests of the time (1997) as well as data from tests in 1987.  A few interesting observations:

1) Of people they could pinpoint to a single exposure, they all showed symptoms in four weeks.  87% showed symptoms in first 15 days.  
2) The antibody tests in 1987 were a LOT less sensitive.  The most sensitive 1987 antibody test detected all positive samples by ~26 days after first *symptoms* appeared.  And the average 1987 antibody test was closer to ~36 days after first symptoms.  
If you add 15 (time for first symptoms) +36 this puts you to about 51 days post-exposure before there was high confidence that you could detect antibodies.  This probably explains the 13 week guidance.
3) The antibody tests in 1997 were much more sensitive!  Note, even first generation tests were drastically improved:
"By the 1987 Abbott first-generation test, employing cell-lysate antigens, 99% of seroconversions were estimated to occur within 48/78 (early/late) days. The improved 1997 Abbott first-generation test detected seroconversion significantly earlier with 99% estimated within 16/20 days."
The first generation tests removed 32/58 days from the window.  That's about a 5-8 week reduction!
4) The worst 1997 test detected antibodies at about 20 days from first symptoms, and most did so at ~14 days.
If you add 15 + 14 = 29.  Thus at about 29 days post-exposure most of the modern tests (even 1st generation) will have pretty high confidence.  If you factor in the person who had first symptoms at 28 days: 28 + 14 = 42 days (6 weeks).

The upshot is that 13 weeks is still the holy grail.  You never got fired for buying IBM (back in the day), likewise no one can dispute a 13 week window recommendation.  But there does seem to be strong evidence that 4-6 weeks is exceptionally encouraging with modern tests.  

[alienshadow]

i agree with this research

[liondabou]

Who conducted this research paper? I will read it right away.

[liondabou]

Ok, now I have read it. I happen to be Swedish myself and I can confirm that these researchers are the top experts on HIV in my country and that the study is definitely a legit medical study performed by the most respectable medical instute (Karolinska) in Sweden.

I hope we can see more scientific research like this in the future rather than the guessing and copy-paste information available in many places throughout the internet!

[Pecha1230]

Yup, but its nice to see that someone else is saying that there is great comfort in a negative test at 4-6 weeks.

[Redhead442]

Thanks for posting this!  

Regarding Item # 3 of the original post...When you say that the Abbott first-generation test detected seroconversion within 16/20 days, is that 16/20 days after possible exposure, or after onset of possible symptoms?  And how does an Abbott first-generation test compare to a standard antibody test?

[alienshadow]

yesy today are very sensitive then whay first generation test were still cdc recommends 3 months yes most who have positives do not have to wait this long for a positive result average time is 22 days to show a positive

[liondabou]

As I understand it it means days after the onset of symptoms, i.e. the test turned positive in infected individuals at the latest 20 days after the onset of symptoms. The latest onset of symptoms in the study was after 28 days (most of them around 15 days however) which would mean that seroconversion would at the LATEST occur about 48 days after exposure which is barely 7 weeks. This would suggest that 3 months testing gives a very significant safety margin.

[Pecha1230]

see

[liondabou]

Also remember this study is based on 1997 tests. Today is 2008 and one can assume that the test are probably even more accurate now after 10 years. At least they wouldn't be less accurate.

[Pecha1230]

Thats why Dr. HHH always says any test after six weeks is pretty much conclusive unless the risk was very high like unprotected anal sex

[Oral_Worry]

My buddy, so you are saying that a single six weeks test after low risk exposure is good enough? My buddy, I am happy after reading your research. You have done some good work.

[Pecha1230]

a 6 week test is a good result thus far

[Redhead442]

Can someone please provide more information on what an Abbott first-generation test is, compared to a typical antibody test?