TRUVADA: Why are we ALL not taking this???!!!!

[Westheangelino]

HUGE NEWS. Not often a study stops giving placebos mid study.

http://www.aidsmap.com/page/2917367/

[trish]

trish what are the risks bottoming without protection

First let me say that I'm not the person to ask. I have absolutely no training in any field related medicine or epidemiology. Like everyone else I read, I Google, I listen and I think.

This is a pertinent question. According to the CDC bottoming is the second riskiest behavior in regards to the transmission of HIV. (The riskiest activity is blood transfusion with from HIV infected person). Using the CDC numbers, the risk of receptive anal sex (apparently scientific etiquette for bottoming) is 12 to 13 times the risk of insertive anal sex. http://www.cdc.gov/hiv/policies/law/risk.html

The risk of HIV transmission for the unprotected bottom (according to the above CDC link) is 138 transmissions out of 10 000 exposures.

Now I’m going to engage in some risky behavior. I’m going to try to compute the probability of transmission given a single unprotected act of receptive anal with an infected partner. Here is the reasoning: Let p denote the probability we seek. Then the probability that single incident doesn’t result in transmission is (1-p). So the probability that there will be no transmission if there are two such incidents is (1-p)(1-p). The probability that there will be no transmission over a run of three such encounters is (1-p)(1-p)(1-p), and so on. So the probability of no transmissions during a run of N such encounters is (1-p)^N. So the probability that the bottom is infected sometime during a run of N such encounters is 1-(1-p)^N. Now according to the CDC numbers (if I interpret them correctly), the probability of infection sometime during a run of 10 000 exposures is 138/10000. So 1-(1-p)^(10000) = 138/10000 = 0.0138. This is an equation which can be solved for p. Solving one finds p = 1-(1-0.013^(1/10000) = .0000013896.

You can now personalize the model to your situation. For example: Suppose you have unprotected receptive anal sex with an infected partner once or twice a day every day, say on average 550 times a year; i.e. 550 exposures each year. Remember p = 0.0000013896. So the probability that you get infected before the year is out is 1-(1-p)^500 = 0.00069. The probability of transmission before the decade is out is 1-(1-p)^(10*500) = 0.0069. The probability of transmission during the next N years is 1-(1-p)^(N*500). The probability of getting infected over thirty years is 0.02. So there’s a 98% chance of no HIV infection over a thirty year period. That may sound like a risk you’re willing to take, but just keep in mind that 2 out of every hundred of you (all participating in 500 unprotected receptive encounters per year for thirty years) will acquire HIV (not to mention hepatitis, gonorrhea and other STDs).

If this is correct (and please -anyone- inform me of any errors you detect in the reasoning) I am forced to change the way I’ve been modeling Travada use. More on that in a later post.

I think the moral here that one should not bottom without protection (unless -perhaps- you’re monogamous, your partner is trustworthy and both of you are free of STDs).

[trish]

I entered this conversation back on page 21 in an attempt to answer the question, What is the probability of avoiding an HIV infection over a given period of time, if you’re having sex with an infected partner and using a given preventative strategy. I was looking for the probability escaping HIV infection per exposure which could then be easily employed to answer such a question. Wes, Broncofan and myself quickly decided that the numbers the lay-literature calls “effectiveness” was not that probability. So what was it? No one knew. The lay-literature was not helpful, though I should’ve been more persistent in my search. I interpreted “99% effective” in my posts prior to #372 to be the probability of avoiding infection over a year of exposure. Wes objected to my charts and conclusions but he could never put his finger on the source my error. He kept saying “Truvada is 99% effective” but he couldn’t tell me what that meant.

In answering MrBest’s most recent post I ran across these two links:

http://www.cdc.gov/hiv/policies/law/risk.html

http://www.cdc.gov/hiv/basics/prep.html

Both are CDC webpages. The first provides a table of clearly defined risks for an assortment of sexual activities. For example: The risk of unprotected receptive anal intercourse with an infected partner is ranked at 138 per 10 000 exposures. I explained in my last post (#372) how this determines the probability of a given person acquiring HIV through one unprotected anal receptive exposure to be equal to 0.0000013896.

The second link explains, “In several studies of PrEP, the risk of getting HIV infection was much lower—up to 92% lower—for those who took the medicines consistently than for those who didn’t take the medicines.”

I found these two pieces of information together to be quite enlightening. 92% of the 138 is about 127. The number that is 92% lower than 138 is therefore 138-127 = 11. So when using a PrEP (like Truvada) the risk of transmission through receptive anal sex is 11 out of 10000 exposures. (Notice this is approximately the risk of unprotected oral transmission according to the CDC).

We can now repeat the reasoning in post #372 to find the probability, when using a PrEP, of transmission through one anal receptive exposure. Call this probability p. It’s equal to p = 1-(1-11/10000)^(1/10000) = 0.00000011.

Given this value of p we can compute the probability of transmission during a run of N exposures as 1-(1-p)^N.

For example. Suppose you have anal receptive sex with your HIV infected partner on average of 550 times a year (one to two times a day). Then the probability of transmission within the next thirty years is 1-(1-p)^(30*550) = 0.0018. If there are 100 people exactly like you, then all are expected to be HIV free at the end of the three decade run. This is a much happier prediction than that derived from my prior understanding.

[trish]

Errata for Post #372
The 550 suffered a typo and mutated to 500 in post#372. The probability of transmission before the end of the next 30 thirty is still approximately 0.02 (the mutation error got taken up in the rounding off to the second decimal place).

[Westheangelino]

^ Trish, you're still only half right, but I'm glad you seem to be seeing the light.

Your conclusions are correct, but your reasoning is flawed. Your mixing risk of exposure with no medication with risk while on medication. That's like grouping together people who are vaccinated against polio and those who aren't.

[trish]

Not sure where you think I'm mixing the two. The CDC's 92% means that when using a PrEP the risk is 92% lower than the risk would be without using any protection. Hence one has to take 92% of the risk without protection and complement it to find the risk with PrEP. (This is explained in the second link under the section How Well Does PrEP Work?) Perhaps this is where you think I'm mixing the two erroneously, though at the moment it looks right to me.

[trish]

Errata for Post #373
In post 373 I made the parenthetical remark, "Notice this is approximately the risk of unprotected oral transmission according to the CDC."

However of risk of 11 transmissions per 10000 exposures is equivalent to the risk of unprotected insertive anal. So the correct statement would be: the use of PrEP can reduce the risk of transmission during receptive anal to the risk equivalent to that of unprotected insertive anal.

I misremembered the table and didn't double check before posting. Sorry.

[Westheangelino]

Can't wait til this is common practice

[trish]

I give up. The reported numbers in this field are all over the place. Consider these two links

http://www.cdc.gov/hiv/policies/law/risk.html

http://www.aidsmap.com/Estimated-ris.../page/1324038/

The first is a CDC webpage for the dissemination of information to the lay public. The second is a NAM page.

According to the CDC
"The estimated per act probability of acquiring HIV from an infected source by...receptive anal intercourse is...138 risks out of 10000 exposures"

According to NAM
The HIV transmission risk per exposure by...receptive anal sex amongst gay men, partner HIV positive is...82%.


What am I missing here? Why are these numbers no commensurate? Are they not both measuring the same thing, namely the risk of HIV transmission per exposure via unprotected receptive anal intercourse with an infected partner? According to the CDC the risk is 138 chances of transmission out of 10000 exposures. Even a naive division demonstrates that's way under 2%. According to NAM that same risk is 82%. Help?

[lifeisfiction]

I give up. The reported numbers in this field are all over the place. Consider these two links

http://www.cdc.gov/hiv/policies/law/risk.html

http://www.aidsmap.com/Estimated-ris.../page/1324038/

The first is a CDC webpage for the dissemination of information to the lay public. The second is a NAM page.

According to the CDC
"The estimated per act probability of acquiring HIV from an infected source by...receptive anal intercourse is...138 risks out of 10000 exposures"

According to NAM
The HIV transmission risk per exposure by...receptive anal sex amongst gay men, partner HIV positive is...82%.


What am I missing here? Why are these numbers no commensurate? Are they not both measuring the same thing, namely the risk of HIV transmission per exposure via unprotected receptive anal intercourse with an infected partner? According to the CDC the risk is 138 chances of transmission out of 10000 exposures. Even a naive division demonstrates that's way under 2%. According to NAM that same risk is 82%. Help?

Its how the disease is contracted. Just because a person has HIV, it doesn't mean they can always transmit the virus. Its gets so of weird and such so you have to look at the factors. Secondly there is the what should happen and what actually happens and those two numbers are different. Its easier to look at it that when you engage in risky sex you put yourself at high risk. That's all you have to know. So the question I ask is how much do you want to gamble with your health?