The answer two your first two questions is no:  there is no "secondary" window, nor has there ever been a reported case of HIV in which the antigen-antibody (AgAb, 4th generation) test was negative beyond 45 days. (We round that off to 6 weeks, for convenience; and there is no apparent difference in test performance between 42 and 45 days.) p24 antigen generally rises to detectable levels in 10-14 days; and it is the antibody itself that removes the antigen from the blood. Hence once the test is positive for Ag, Ab or both, the test remains positive for life. Typically the test is positive by 3-4 weeks (at least 95% by 4 weeks); 6 weeks provides a safety margin.

Calculation of ultimate test sensitivity is largely a statistical exercise. If, say, there are 400 patients in the database, and all 400 are positive by 6 weeks, statistically it remains theoretically possible that if a larger sample were available, there could be a rare miss. Stated in typical statistical terminology, for 400/400, the lower bound of the 95% confidence interval is around 99%, so there is no guarantee (statistically) that the true sensitivity is "only" 99%, and that is the figure the US FDA and other countries' equivalent agencies will cite and permit the manufacturer to say. (I'm using approximate numbers here, but you get the idea.) So the answer to your last question is a) common sense and b) the sorts of experience I mention above -- i.e. few if any experts have reported anyone with HIV whose AgAb test remained beyond 6 weeks after infection. (There is one exception:  if someone takes anti-HIV drug to prevent infection after the exposure (post exposure prophylaxis, PEP) and it doesn't work, the window period starts at the end of PEP, not at exposure. That is 6 weeks after the last dose, or ~10 weeks after exposure.)

Finally, consider a typical exposure. Let's say someone has sex with someone with a 1% chance of having HIV, and has an exposure that carries one chance in a thousand of transmitting the virus. (The average transmission rate, with unprotected vaginal sex with an infected partner not on treatment, is 1 in a thousand for the female partner.) Thus even before testing, her chance of being infected is 1 chance in 100,000. If she now has a test with "only" 99% sensitivity, it becomes 1 chance in 10 million, which obviously is zero for all practical purposes!

If you'd like to say more about an event that might have put you at risk, I'd be happy to help you evaluate the risk and interpretation of your test results. In the meantime, I hope these comments are helpful. Let me know if anything isn't clear.

HHH, MD

One additional fact that might help reassure you:  In the 15 years of this and our preceding forum at MedHelp.org, with thousands of questions from people worried about HIV exposure, nobody has yet reported they turned out to have HIV. You won't be the first! If and when it happens, undoubtedly it will be a typical, high risk exposure without atypical test results before testing positive.

Can you please give some reassurance that this is not a common occurrence? . 

What are the chances of someone seroconverting after a 14 week antibodies only test (I assume it was a 3rd gen test, they never told me)? (No pep, chemotherapy, immunosuppressive drugs, etc) 

If he was seroconverting during his 4th gen test would it still have been positive? 

He told me he saw a different woman 2 weeks before me. Say he caught hiv from her. That would make his negative 4th gen lab test 10 weeks after exposure, would that be even more accurate? Do tests become more accurate with more and more time from exposure?  I read somewhere that a negative 4th gen test passed 50 days has a 0% chance of a false negative is that true? 

Would a rapid test using blood from a vein find infection sooner? 

Thank you and I’m sorry for the extensive questions. 

The oral fluids test (Oraquick) is inherently less reliable than all HIV blood tests. it misses 2-4% of HIV infections even beyond 6 weeks, and is never appropriate to exclude HIV infection with 100% certainty. (It tests crevicular fluid, i.e. moisture produced between teeth and gums. HIV antibody is inherently far lower in crevicular fluid than blood, and thus a greater challenge for the technology of detection.) If you have had a negative lab-based blood test more than 6 weeks after your potential exposure, then for sure you do not have HIV. If you have not had such a test, do it now.

Your question about timing of the third generation HIV antibody tests is answered above:  they always are positive by 8 weeks, even though some authorities recommend waiting until 12 weeks for definitive results. There is no chance your partner is infected.

Threads normally are closed after 2 follow-up exchanges, and we're there. I'll leave this open so you can confirm you have had at least one lab-based AgAb (4th generation) blood test at least 6 weeks after the exposure (which is what I understood above). But that's all -- no more other questions about test timing etc. And before you do that, I suggest you go back and re-read all your questions and my replies above. And also stop searching the internet on this! Like most anxious persons, you are being drawn to information out of context that inflames your conviction you have HIV and missing all the reassuring stuff that also is there.

That concludes this thread. I hope it has been helpful. Do your best to accept this reasoned, science based evaluation and advice. Best wishes and stay safe.