Can the following statements be reconciled:

  1. Condoms form an essentially impermeable barrier to HIV
  2. Sexual activities other than penetration carry a very low risk of transmission
  3. The consistent use of condoms reduces the risk of HIV transmission to about a fifth of what it would otherwise have been (source: Am Fam Physician. 2004 Oct 1;70(7):1268-1269 as summarized at aafp.org/afp/2004/1001/p1268.html)

Obviously the effectiveness of condoms is going to depend on them being used correctly, but surely they are not used so incorrectly that the effectiveness drops as much as these statements imply - assuming all 3 statements are correct, is there a more plausible explanation?


closed as off-topic by David, anongoodnurse, AliceD Oct 4 '17 at 22:00

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "Personal medical questions and health advice are off-topic on Biology. We cannot safely answer questions for your specific situation and you should always consult a doctor for medical advice." – AliceD
If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ I'm voting to close this question as off-topic because it is not about biology but an attempt to start a discussion either on scientific logic or some hobby horse issue. $\endgroup$ – David Sep 27 '17 at 12:26
  • 2
    $\begingroup$ No, it is a genuine request for help understanding the numbers. The one fifth comes from Am Fam Physician. 2004 Oct 1;70(7):1268-1269 as summarized at aafp.org/afp/2004/1001/p1268.html. It is expressed as 80% effectiveness but I was hoping that the contrast with the other two statements would be clearer if I expressed it as a reduction in risk. The other statements can be found on any number of apparently reliable sites. Is the risk of transmission virtually nil, as the first and second statements imply, or is it 20% of whatever it would be if you were not using condoms? $\endgroup$ – Riedt Sep 27 '17 at 13:58
  • $\begingroup$ Then please put this info in your question. $\endgroup$ – David Sep 27 '17 at 14:16
  • $\begingroup$ Please see skeptics.stackexchange.com/q/3167/24840 $\endgroup$ – anongoodnurse Sep 27 '17 at 21:16
  • $\begingroup$ Thanks for the reference anongoodnurse - I don't seem to be able to rate it as useful, I guess because I'm a newb. $\endgroup$ – Riedt Sep 28 '17 at 12:17

Note that those three statements are also approximately equivalent to pregnancy risk in the context of vaginal intercourse:

  1. Condoms form an essentially impermeable barrier to sperm
  2. Sexual activities other than penetration carry a very low risk of pregnancy
  3. The use of condoms has an annual failure rate around 15%, i.e. on average 15/100 couples using exclusively condoms will conceive in a year

There are dozens of sources that reconcile these issues, and it all comes down to the failure rate and misuse rate of condoms. Even the source you cite talks about this explicitly, so I'm not sure why you still have a question about it that isn't already answered there.

  • $\begingroup$ It's because the discrepancy seems too large to be explained that way. Failure of the condom itself is rare: thebody.com/content/art2436.html. Unintended pregnancies are overwhelmingly due to user failure, which is overwhelmingly non-use rather than incorrect use (ibid). The study cited previously excluded couples reporting less than 100% use. Given the pregnancy statistics, incorrect use is not a plausible explanation for a 20% rate of failure to protect (same whether per act or per year). Did the couples misreport consistency of use, or is there something else going on? $\endgroup$ – Riedt Sep 28 '17 at 8:12
  • $\begingroup$ The answer may be that the per year multiplier only approximates the per act multiplier when the risk involved in unprotected sex is very small. The probability of HIV transmission through unprotected sex, given a serodiscordant couple, seems to be much lower than the probability of conception, given a couple in the normal fertility range, so the same rate of failure per act may translate into a very different rate per year. $\endgroup$ – Riedt Sep 28 '17 at 10:15
  • $\begingroup$ Yes, the annual multiplier is almost the same for HIV transmission as for conception at about 0.2, which sounds as though it makes sense until you realize that it corresponds to a per act failure rate of about 0.2 in the case of HIV and about 0.1 in the case of conception (as best I can calculate it), and that the failure rate for conception includes cases where the couple didn't actually use a condom whereas - if we accept what the study participants reported - the failure rate for HIV does not. If condoms themselves are equally effective, the annual failure rates should differ significantly. $\endgroup$ – Riedt Sep 28 '17 at 10:55
  • $\begingroup$ @Riedt Oops, I read the paper you linked incorrectly, though it seems like you might have too; they are not taking about a 20% rate of failure but rather a 80% reduction. The actual rate of failure they are talking about is 1% per year. You also seem to be missing the very important statement in the abstract of the paper: Consistent use is defined as using a condom for all acts of penetrative vaginal intercourse. Because the studies used in this review did not report on whether condoms were used correctly for every act of intercourse, effectiveness and not efficacy is estimated. $\endgroup$ – Bryan Krause Sep 28 '17 at 15:46
  • $\begingroup$ No, I realize that. What I mean by 20% failure is 20% failure of protection, not 20% transmission. It's the same as an 80% reduction in risk. I have commented on your second point below. $\endgroup$ – Riedt Sep 29 '17 at 2:32

No, they are not reconcilable unless we assume either that serodiscordant couples are very much more likely to use condoms incorrectly than seronegative couples, or that the effectiveness of a condom in terms of protecting against HIV is very much more easily compromised by incorrect use than its effectiveness in terms of preventing conception. It is difficult to assess whether it is more plausible that participants in the studies analyzed in Am Fam Physician. 2004 Oct 1;70(7):1268-1269 reported that they had used condoms consistently (which was defined for the purposes of the analysis as meaning "always") when in fact they had only used them sometimes. This would make the third statement unreliable. While that is a possibility, it is wise to be extremely careful storing handling and using condoms (don't keep them in your wallet, for example) and to make sure that it is your own that are used. It is extremely unwise to assume that as long as you are using a condom for vaginal sex, you do not need to be concerned about your partner's HIV status.

  • $\begingroup$ I commented this in response to your comments on my own answer, but I'll add it here as well: Consistent use is defined as using a condom for all acts of penetrative vaginal intercourse. Because the studies used in this review did not report on whether condoms were used correctly for every act of intercourse, effectiveness and not efficacy is estimated. I'd also add that by posting this poorly thought-out answer in addition to your original question suggests that some users' suspicions that you are here to rant rather than ask a question is validated. $\endgroup$ – Bryan Krause Sep 28 '17 at 15:48
  • $\begingroup$ I am sorry to have given you that impression. It is not the case at all. I am at risk and am trying to quantify the risk and the extent to which it can be reduced. I am aware of the statement you quote in italics - in fact that is largely what gives rise to the discrepancy I have been trying to understand. Are you perhaps reading it as saying that the study does not report on whether condoms were used for every act? It does not mean that, it means that the couples reported using them for every act (meeting the definition) but we don't know whether they used them correctly. $\endgroup$ – Riedt Sep 29 '17 at 2:29
  • $\begingroup$ Hopefully it is clear that I am looking to explore this in the abstract, from an academic perspective. I am not asking for medical advice. I only mention my own circumstances because I don't know how else to allay the suspicions you mention. $\endgroup$ – Riedt Sep 29 '17 at 3:25

Not the answer you're looking for? Browse other questions tagged or ask your own question.