The differences in advice between me and Dr. Bob are minor.  I look at thebody.com fairly regularly and 95% of our advice is more or less identical and all issues.

The minor difference in advice about timing of testing is one of perspetive.  Dr. Bob is primarily a clinician and I am primarily a public health expert.  He deals with patients, I deal with populations, which means I deal with odds and probabilities.  The individual person wants '100%' assurance (which never exists in medicine and biology).  Similar differences exist between, say, the Massachusetts dept of health (6 weeks is plenty) and CDC (3 months), primarily because Mass health has my public health perspective, whereas a federal agency such as CDC tends to have regulatory perspective that fosters a conservative stance.

On top of this, there simply are no precise data on seronversion window time.  In order to know the time with precision, you would have to study more than 1000 persons exposed at a precise, known moment in time to the exclusion of all other possible exposures, then test them all every 1-2 days for 3+ months.  That research will never happen. Thus, the data are imprecise and will remain that way; 10 years from now we won't know the true seroconversion limits any better than we do now.

Now take your case.  Assume a 10% chance your partner had HIV.  Unprotected receptive anal sex results in transmission 1% of the time. Now assume a 1% chance of transmission despite use of the condom (maybe it broke and your partner didn't know it).  These estimates would make your chance of infection 0.1 x 0.01 0.01 = 0.00001.  That is 1 chance in 100,000.  Now add the fact that a 6 week (40 days) HIV antibody test is 95% sensitive, that is it misses 5 cases in 100.  That sounds high, doesn't it?  'Holy ****, there is a 1 in 20 chance it missed an infection!'  But in fact, since even before the test your odds of having HIV were only 1 in 100,000, the your negative result at 40 days actually means the chance you have HIV is 0.00001 x 0.05 = 0.0000005.

That, my friend, is 5 chances in 10 million, i.e. 1 in 2 million.  If you live in the US, the chance you will die of a lightning strike has been calculated at 1 in 27,000.  In other words, the likelihood you will be killed by lightning someday is 740 higher than the chance you have HIV.

As a clinician, Dr. Bob (and CDC) tend to focus on the 95% senstitivity of the test at say 4-6 weeks, and say 'to reduce that to almost zero, be tested at 3 months'.  As a public health physician, I say that 1 in 2 million are plenty good odds which, for all practical purposes, means there is no chance you have HIV.  But if you want to drop those odds to 1 in 20 million, it's fine with me if you get retested.

Your symptoms mean nothing, by the way.  And my final advice is that you definitely should learn from this experience.  Most important, always ask your partners whether they have HIV, and don't have sex (even intended safe sex) for those who are infected or who are evasive in their answers.

Good luck--  HHH, MD

YGM:  Of course it is obvious when a condom breaks.  An uncovered penis isn't hard to spot.  I already answered your question about follow-up testing:  "...for all practical purpose...there is no chance you have HIV. But if you want to drop those odds to 1 in 20 million, it's fine with me if you get retested."  

Others:  The unusually thorough answer was only because once in a while, it probably is useful to go through the explanation (again) about interpreting test results.  And also because I have never directly addressed the reasons for the difference betwen Dr. Frascino's advice and mine about testing times.

HHH, MD

Ok I'm a female and I had unprotected sex with this guy on the 15th. It is now the 22nd. He didn't ejaculate inside me, he pulled out. After his ejaculation he went back in and when he was about to orgasm he pulled out again but there was no semen. I think he had an orgasm without ejaculation. I've read that can happen somewhere. Anyway I checked myself and nothing came out so I'm pretty sure thats what happened. I'm just real worried. I haven't shown any symptoms except a few headaches which alarms me because I never get them and a slight cough. I've asked him his status and he's reassured me that I have nothing to worry about, but those are just words I guess. I just want to know, how likely Is it that I've contracted the virus and how soon can I test to  be sure ??

Hello YGM,

I don't know how long you and your "long term boyfriend" were broken up for, but if it was any length of time, do you really want to take the chance of having unprotected sex with him.  Why not just use condoms and practice safe sex for a while then both go get tested together before practicing unprotected sex.  I don't understand why gay guys (I am not discriminating against gay guys, I am one myself) have such a problem with using condoms.  

Another thing to keep in mind, even in a "long term" relationship people cheat, especially young gay men.  

As for asking for Dr. HHH's opinion about continuing to have unprotected sex with your boyfriend, he is here to give expert advice on HIV, not to give consent to having unprotected sex.

You were obviously quite scared after your encounter with Mr internet.  You should have learned from that experience.  

Be safe and good luck,

Ryan  

Hi Doctor

Just a quick follow up, my long-term partner and I are getting back together, he has been tested out to six months, I have been tested out to 40 days, the one experience I had above (protected anal unprotected oral) is the only exposure I have had in the "window period".  

Based on this can we continue to have sex unprotected as we had before, we have had lengthly discussions about this and have equally shared our concerns about commitment and disease.  Based on your advice I am comfortable doing this, but wanted to get your opinion.

Thanks

Hi,

I've just put some figures from other posts into Dr H's calculations as I wanted to see what I got with my 5-week negative test. If I keep Dr H's number but put in the figure for female-to-male transmission as 1 in 2000 (0.0005):

10% chance that partner has HIV: 10% = 0.1
Female-to-male transmission risk: 1 in 2000 = 0.0005
1% chance of transmission: 0.01
Accuracy of anti-body test at 4-6 weeks: 95% = 0.05

Chance of infection = 0.1 * 0.0005 * 0.01 * 0.05 = 0.000000025 = 1 in 40,000,000.

If the anti-body test is ony 90% accurate at 5 weeks (just guessing as I don't know the accuracy at 5 weeks), the figures become:

0.1 * 0.0005 * 0.01 * 0.1 = 0.00000005 = 1 in 20,000,000.

(Assuming my arithmetic is ok - but if there's a glaring error please correct it).

For a list of odds of more likely things that might happen to us, a site I just found is http://funny2.com/odds.htm - a short extract is:

Odds of fatally slipping in bath or shower: 2,232 to 1
Chance of dying from any kind of injury during the next year: 1 in 1,820
Chance of dying from a car accident: 1 in 18,585
Chance of dying from any kind of fall: 1 in 20,666
Chance of dying from food poisoning: 1 in 3,000,000
Chance of having a stroke: 1 in 6
Chance of dying from heart disease: 1 in 3
Chance of American man developing cancer in his lifetime: 1 in 2
Chance of an American woman developing cancer in her lifetime: 1 in 3
Chance of getting prostate cancer: 1 in 6
Chance of getting breast cancer: 1 in 9
Chance of getting colon / rectal cancer: 1 in 26

Ok - so they are only odds and someone has to be the poor soul who is the statistic, but it does give a wider perspective to the risk.

Bert

The statistical difference between 40 and 42 days is negligible.
If the incident was protected - as it seems to have been - there really wasn't any need for testing at all. You behaved responsibly!

If you remain concerned, however, test again in a couple of weeks. The result will not change:)

That was a thorough answer, thank you!  Does generation of Elisa test play a factor in the 95% accuracy for 4-6 weeks?  From talking with home access for example with  them using 2nd generation, they say it is just as good as 3rd gen. after 6 weeks.

Thanks for the information, I found it helpful.

Given my situation I am not sure what you felt I did wrong, I did ask my partner his status before, he claimed he was negative, I asked him after he said he had a test 10 days after our meeting and it was also negative.

Secondly, I did use a condom for the intercourse, I've just never experienced a breakage before so I don't know if I would know had it happened, is this typically something someone would know?

If you feel a test at three months is necessariy, I will definately follow up, but honestly if you feel that given the current situation I am in the clear I am comfortable with that, how do you feel given the circumstances?  (Are the indivduals in Mass. correct in saying I'm not conclusive given I was two days short of 42 days??)

Thanks for all your help,

This has been a very thorough answer, and has answered a lot of my questions regarding why you and Dr. Bob differ on window period. Thank you for the answer!