Hivemind math advice requested: someone on another website is claiming that "studies show" that a woman who's been on oral contraceptives for 3 years increases her cancer risk by 1000%. Which is pretty obviously nonsense--I remember just enough of How To Lie With Statistics to feel pretty confident in calling it nonsense, but I'm kind of a numbers moron myself and can't figure out how to quantify the idiocy.

Can anyone who's mathier than me take a crack at it? I know the ACS current lifetime risk estimate for the average USian woman is 1 in 8; and the Guttmacher Institute says that the US has approximately 43 million reproductive age, sexually active women, of whom about 10.7 million are on the Pill, most for at least 3 years over their lifetimes, if not a whole lot longer. I'm pretty sure all those numbers add up to "1000% is a completely crack-addled junk stat you just pulled out of your ass," but I'm not sure why, and I'm getting a math-anxiety stomachache just thinking about it. Help a sister out, anyone?

Um, of WHAT cancer? Any cancer? Cause being on the pill reduces your risk of several cancers, so that seems...unlikely. I mean, even if it was "increases 100%" (which would be doubling your risk).

I think a 100% increase would mean twice the risk and extrapolating would mean that the risk would be 10 in 8. My guess is that this figure comes from a very narrow and carefully cherry-picked metric.

I think a 100% increase would mean twice the risk and extrapolating would mean that the risk would be 10 in 8. My guess is that this figure comes from a very narrow and carefully cherry-picked metric.

Okay, yeah, that was my first thought. It's just that my second thought was that there's no such thing as ten out of eight women, so clearly I'd done something wrong.

Did I mention the math anxiety? Even when I stumble on the right answer, I immediately proceed to talk myself out of it.

I'm horrific at political debates because I want to make sure my numbers are accurate and I don't want to dismiss other claims out-of-hand. This means I have trouble with anyone who pulls figures out of their ass or is basing their info on BS information.

I don't think you can extrapolate from the numbers JZ gave, because those are numbers for all women...which means that women who do or don't take the pill are included in it. And it depends on how LONG you take the pill also.

See the below link--decreases ovarian and endometrial risk, may increase breast cancer risk but after 10 years off (which is MOST women, since most women aren't diagnosed until they're old) it's the same, and increases liver cancer for white women but not others. May increase cervical, but really mostly that risk is about HPV.

[link]

The actual answer is that there is no or very little increased risk for breast cancer. There may be a slightly higher breast cancer risk while a woman is actually taking oral contraceptives, but that goes away completely 10 years after she stops taking them. Some studies show none at all. There's more correlation with shift work than oral contraceptives.

Contraceptive use decreases the risk of ovarian cancer and increases the rate of cervical cancer, but the latter may be related to more sexual partners.

eta: Or what Meara said.

My guess is that this figure comes from a very narrow and carefully cherry-picked metric.

Also, this? I'm starting to think that the figure just comes from a bunch of out-and-out lies. Just for sick, bitter fun, last week I started going through one pro-life org's list of scary scary cites for peer-reviewed articles that prove the cancer link, but probably 3/4 of the list was pure bullshit. Real articles, sure enough, but real articles that state right in the conclusions (and sometimes in the abstract, in bold), "THESE NUMBERS FAIL TO DEMONSTRATE ANY INCREASED RISK."

I still don't know why I'm so shocked, except that I had been so prepared to find a bunch of cherry-picking that it had never crossed my mind that they might jump straight to naked lying.

Yeah, it's ok to lie if it's for Jesus.

And yet, isn't "Thou shalt not bear false witness" one of the Big Ten?