HIV Prevention and Testing > Am I Infected?

oral transmission

(1/3) > >>


I have read many answers on this issue and I believe there are some misconceptions in the way one can read results on serodiscordant couples that need addressing.

When a statistical analysis is performed, say of transmission frequency, the result is expressed as a confidence interval for a given probability of being right (eg 95%). So what these 3 studies prove is that 0 is one of the bounds of the confidence interval, ie that the null hypothesis "oral transmission of HIV never happens" is not disproved.

What these studies cannot prove is the hypothesis "oral transmission of HIV never happens"

In fact here we are dealing with statistical analysis of rare events. Longitudinal studies are not good at estimating the probability of rare events. Take another example : follow 1000 people for 10 years. If not one dies from a thunderbold, you cannot conclude from these longitudinal studies that death from thunderbolds do not happen. It would be invidious to assert that there is only unscientific, anecdotal evidence that cannot be accepted.

What we do instead is that we examine people that have died from thunderbolts and come with an extremely low frequency over the population as a whole of dying that way. And we can do the same for oral transmission of HIV. I feel that aidsmeds should at this point accept that and state that it does happen, if very rarely.

There are further issues to address. If oral transmission of HIV is only possible when virus load is extremely high, say between the third and sixth week after infection, then almost all observations made in the studies would be irrelevant. We're dealing with couples in which the positive partner has a viral charge of less than 500 000 copies, in all or almost all observation points, presumably, so the hypothesis "oral transmission of HIV happens with a 5% (or 1% or 10%) likelihood  when ejaculates contains more than 500 000 copies" has hardly been tested in these longitudinal studies.

What I think is warranted is the following : oral transmission of HIV happens very rarely. 3 studies of serodiscordant couples have shown no case of transmission. However there are isolated documented cases. It is likely that the very low odds of transmission increase when highly contaminant semen is swallowed (which reached the highest concentration in the second month of infection of the insertive partner) and when the receiving partner has cuts or sores in the mouth (even from invisible cuts that come from flossing or aggressive brushing)

Andy Velez:
Anytime you have sex with someone other than yourself there is potentially risk involved.

We don't say transmission cannot happen orally. What we say is that thus far from the evidence in over 25 years of the epidemic, only a minicule number have occured and even those are questionable because of other factors involved. So, could transmission happen orally? Yes. Does it happen? The evidence is that it doesn't.

You're entitled to add your caveat, and each person has to make a decision as to what they're comfortable doing in terms of actual risk v theoretical.

In your quest for a more exact answer I would suggest you have to be careful you do not unduly alarm readers who are in many cases already in an anxious state over something that wasn't really risky to begin with.

The bottom line is to be well educated about the issues and then make informed choices.


I just feel that some aidsmeds answers on oral transmission are ever so slightly misleading. They are based on dismissing all reported cases as unproven, unreliable, or unscientific. The Lancet published last year a case study on transmission via a bloody fight: rare things happen, and isolated cases are just that.

Saying that there is no evidence is not true. You can say that there is evidence outside of longitudinal studies, but that aidsmeds disputes it.

I understand the risk of unduly alarming readers about transmission routes that are very low odds. But one should not on that basis distort the record and say there are no odds. One has to balance the risk of dispensing information which would lead to some being contaminated based on wrong assurances given by aidsmeds. One case would be one too many.

That's why I think aidsmeds should just say that odds are very very small and leave it at that. These longitudinal studies have been excessively relied upon and do not really address the issue of contamination by insertive partners in the period where there are more contaminant by several orders of magnitude. For people who can give several blowjobs in a month to unknown, highly sexually active partners in backrooms, the risk is small but not nil.


Andy Velez:
Point taken. And we don't say there isn't any risk in giving oral.

But I don't really want to go back and forth on this.

I know we're committed to being very careful about what we say here. If you check through some of the threads you will again and again see mentions of the need to decide about one's personal comfort level in terms of deciding about giving oral m2m. 

The problem with all of it is telling the truth. How do we know the person involved in that fight wasn't of high risk group and is ashamed to admit it? We don't. Is it possible he got it from high risk sex? Yup. could he have gotten it from that fight? Yup. Do we realy know the truth? Nope.

I think the object is live your life safe, but live it.


[0] Message Index

[#] Next page

Go to full version