Everyone talks about how the 98% prevention rate is over the course of a year. How does this relate to a per use scenario and why do they use the by year rate vs the per use. I feel like the per use would be much lower.
In: 0
The effectiveness per use would have to be much higher than the per year rate. Let’s say you have sex 50 times in a year. If condoms work 99.95% (fails 1 time out of every 2000 uses) of the time, they would have a 97.5% effectiveness rate for the year. The math is: .9995^50 =.975
It’s more intuitive to me if you lower the effectiveness rate to 50% per use. Odds of success in 2 uses would be .5 x .5 = .25. Odds of success in 3 uses would be .5^3 =.125. Odds of making it through a years worth of 50 uses approaches 0.
Per use has to be higher than 98% in that case, because that’s just how math works. If the per use percentage were lower, say 95%, it would be statistically impossible to hit a 98% success rate in trials over the course of a year.
If something is 98% effective over a year of use, depending on how many the average use is, the effectiveness would likely be above 99%, and likely super close to 100%.
Since condom tests are done with couples who use condoms regularly over the course of a year, let’s just throw out a number and say that they have sex once a week. 52 uses per year, and let’s assume that gets us to a 98% effectiveness rate. This means that the effectiveness per use is about 99.96%. And if we were to start with a higher usage rate per year (more than once weekly, which is perfectly likely), we’d be even higher. Used properly, condoms have an awesome success rate. But *used properly* is the key here!
The 98% prevention rate is measured as ‘how many people get pregnant, when exclusively using condoms to prevent pregnancy per year’. So 2 out of 100 people get pregnant per year despite the correct use of condoms.
(When accounting for user error, that 98% rate drops to a real life rate of approximately 85%.)
It would be pretty difficult to break it down to a ‘per use’ statistic, because you’d have to find out how often all the people had sex overall. Then compare that to the number of pregnancies. The resulting rate would be very, very close to 100%.
Let’s say (for the sake of nice numbers) each of those 100 people had sex 10 times in a year (or less than once a month), always using a condom perfectly. So 1000 sexual encounters in that year lead to 2 pregnancies
Problem: Did the condom fail once and the person got pregnant immediately? Or did multiple condom failures occur and only one time they got pregnant? Maybe more condom failures after they were already pregnant and didn’t know yet or still used them because of STIs?
In a ‘per year’ setting those questions are unimportant. In a ‘per use’ setting we’d have to answer them in a meaningful way. But let’s just claim 1 pregnancy = 1 failure and continue with the math:
998 out of 1000 uses did not lead to pregnancies, so 99.8% prevention rate ‘per use’.
So we run into the next issue: Most people will read 99.8% effective and think that it’s basically 100%, so nothing could ever go wrong.
Latest Answers