Missing Research on NSAIDs and Contraception

Could disrupting prostaglandins at the right time bring safer, cheaper, more effective birth control to billions?

Does this make me look fat, or am I on birth control?

New research suggests hormonal contraceptives promote adipogenesis (e.g., triglyceride accumulation and/or pre-adipocyte proliferation) (“Molecular Assessment of Pro-Adipogenic Effects for Common-Use Contraceptives and their Mixtures,” Yu-Ting Tiffany Chiang and Christopher D Kassotis, Endocrinology). This joins a wide array of evidence suggesting standard contraception methods may be associated with weight gain — including progestin-only methods and copper IUDs.

We don’t know whether these effects are causal, or (if they are) what the mechanisms are, or how big the effects are. We just know that a lot of women perceive birth control hurts their metabolic health — as expressed in their waistlines.

A lot of women complain about other birth control side effects, like depression. And a lot of women discontinue birth control when they can: Around 30-60% of women get off the pill in 6-12 months, and around 50% ditch copper and progestin IUDs within the first year of insertion, with bleeding the most commonly cited concern.

This is not the behavior of a happy customer base. Reproductive-age women are a big market segment. There’s a lot of money to be made here. So where’s the research on better options? There are recent leads…

NSAIDs and Emergency Contraception

Some of the most exciting women’s health research in the past year was on non-steroidal anti-inflammatories (NSAIDs) and emergency contraception. In the Lancet last fall, Raymond Hang Wun Li et al published the results of a randomized trial comparing a usual emergency contraception (levonorgestrel, a progestin - synthetic progesterone) with or without the NSAID piroxicam. The results were impressive: Only one pregnancy of 418 subjects in the piroxicam group, versus seven of 418 in the usual levonorgestrel-only treatment group. This efficacy boost came from just a single 40 mg piroxicam pill.

Hormonal emergency contraception does not work when women have already ovulated. So it was really just the piroxicam and not the levonorgestrel working with it for those women. How did it work? Would it work by itself? What about NSAIDs as contraception?

Piroxicam probably prevented pregnancies by disrupting prostaglandins. Piroxi-what?

If you know what NSAIDs are but haven’t heard of piroxicam, you’re not alone. This one was an odd choice. It’s not as widely used as, say, ibuprofen, for a reason: According to Wikipedia, “Compared to other NSAIDs it is more prone to causing gastrointestinal disturbances and serious skin reactions.”

Bracketing that, prostaglandins are a group of hormone-like substances that derive from arachidonic acid. NSAIDs disrupt the arachidonic acid cascade implicated in inflammatory processes including those that synthesize prostaglandins. So NSAIDs decrease prostaglandin levels by cutting them off before they get made. And prostaglandins are implicated in ovulation and implantation.

Previous research has tended to look at NSAIDs as a group, and suggests NSAID use may substantially increase early miscarriage risk. Researchers think the mechanism is prostaglandin disruption inhibiting normal implantation. The effect is dose-responsive, and the risks are from NSAID use around conception.

Another, related set of risks are to ovulation itself, with more research suggesting NSAIDs can inhibit or delay ovulation.

This growing body of evidence suggests disrupting prostaglandins around the time of ovulation could bring safer, cheaper, more effective birth control to billions. It may also be more logical than other approaches, because it minimizes the possibility for contraception to do more than it’s supposed to do — to change women’s bodies and brains in a bunch of other, unintended and often undesired ways…

Minimizing Intervention Exposure

For some people, it’s intuitive that we want to minimize exposure to contraceptive interventions that can change bodies and brains. So intuitive that it means women should do most of the work of contraception. Because, of the sexes, women pose less conception risk.

At least, that’s how a young tech millionaire once mansplained the gendered division of contraceptive labor to me: Men are fertile all the time, and women are only fertile for a small fraction of the month, so it makes sense for women to manipulate their fertility in order to avoid those relatively few fertile periods. Fewer fertile days. So less risk to mitigate. So less manipulation required. Leaving men out of it, in this story, is only fair.

This explanation does not survive contact with the present realities of available contraceptive options. With the exception of mechanical methods (predominantly guy-controlled condoms and withdrawal), the most common contraceptive methods (predominantly gal-controlled hormone pills and implanted devices like IUDs) work all day, every day, constantly manipulating female reproductive systems that are already designed to be infertile (in the sense of lacking the possibility of conception from sex) the vast majority of the time.

Despite his ignorance of how current reproductive technology actually works, Tech Bro may have been onto something. Why don’t we have contraception that lets women block the relatively small window of conception possibility, instead of manipulating the reproductive system — and other systems along with it — all the time? Chemical barrier methods that use what we know to give people what they want — reliable contraception without net mortality risks or big quality of life costs?

Perverse Incentives

The jaded answer is that it could hurt pharmaceutical companies’ bottom lines if this panned out. If well-timed ibuprofen prevented pregnancy reliably — say, 200 mg, 2x/day, every three days from cycle days 5 through 20 to account for the wackiness of hidden ovulation — it would cost less than $10/month. The effective dose and/or duration* could turn out to be half that, slashing the price. Compare that to the cost of the pill — which, depending on health insurance coverage and formulation, may run about double or more.

(*It might be harder to use this method right than other methods. One might refer to this as user error, “noncompliance,” or “nonadherence.” It’s fine to say this method may risk higher such error rates than we’d expect to see with other methods. Or that there are women this method wouldn’t suit, because they’d have to take pills every three days for a few weeks — instead of taking one every day or not at all with other contraceptive options. But it’s not ok to say we shouldn’t figure out if this method works, just because it wouldn’t be for everyone.)

Could disrupting prostaglandins at the right time bring safer, cheaper, more effective birth control to billions, and be as easy as taking some ibuprofen? We don’t know, and may not find out, because social and political forces shape science. Pharmaceutical companies have monetary incentives to offer consumers daily pills instead of monthly ones. They’re not the only ones with perverse incentives…

Scientists have long had perverse incentives to prioritize contraceptive efficacy rates over women’s lives in trying to find the effective recipe and dose, with the so-called low-dose progestin-only pill a relative latecomer. Instead of starting low and going up to find effective hormonal contraception, researchers started high, killing some women with preventable cardiovascular events and breast cancers. Knowingly or not, they did this so that they could get better results, publish them, and serve the monetary interests of the pharmaceutical companies supporting their careers.

This risky research also reflected the social and political priorities of powerful networks: States wanted more effective contraception at their disposal at the population level, fast. State and corporate interests beat out ordinary women’s voices in designing and implementing research that shaped (and sometimes took) their lives.

It doesn’t require a conspiracy of ill-intentioned collaborators for scientists to structure their research on the status quo’s terms. That’s just how normal science works, since we’re all a product and part of the psychosocial ecosystems in which we exist.

So there seems to be missing research on NSAIDs and contraception. PubMed searches for relevant terms including ibuprofen and contraception don’t turn up relevant results. Perverse incentives could explain why this research seems to be missing.

What’s the bigger picture in terms of known risks and benefits?

Risks and Benefits of Contraceptive Options: It’s Complicated

It’s hard to say. But for a lot of women, finding something better than current contraceptive options is more than a matter of side effects. As serious as metabolic and mental health harms can be (think cardiovascular disease and crippling depression), the possible harms of typical contraceptive options extend farther. Hormonal contraception with estrogen substantially increases life-threatening cardiovascular risks. Hormonal contraception (whatever the route or formulation) increases breast cancer risks. IUDs risk potentially fertility-threatening injuries and infections. The stakes could not be higher for women’s health.

And then there arethe potential quality of life gains — which can translate into relationship, reproductive, and professional success or failure — for women suffering common side effects like crippling depression (hormones), chronic infections (both), and heavy bleeding (IUDs). If you’re not one of those women (or someone who cares about them), and you just want to count the bodies, then working out the net mortality effects of the pill alone is still an exercise in spotting mistakes in the underlying medical literature. There’s a lot of incomplete information and crappy trade-offs where there should be a simple solution to a common problem. It’s enough to give you a headache, for which you might just take some ibuprofen and call it a day.

Anyway, NSAIDs like ibuprofen carry their own set of possible risks and benefits. The big risks have to do with possible GI bleeding, and seem low in general. The big benefits have to do with possible lower cancer risks.

We don’t know what we don’t know, but most people have taken NSAIDs for pain, and common side effects do not include the complaints (like depression and weight gain) that get women off the pill or getting their IUDs out in droves. One might remark that the bleeding risk resonates with the bleeding that turns a lot of women off IUDs — and worry that NSAIDs could cause such bleeding, too.

But the opposite seems to be the case. The reason is that IUDs ramp up prostaglandins, and disrupting prostaglandin synthesis with ibuprofen is a well-established method for preventing and treating painful, heavy IUD periods. Even for women without IUDs, NSAIDs are commonly recommended for period problems for that reason.

Knowledge Is Power

The future, for some people, is now. Common contraceptive options like the pill and IUDs offer safe, cheap, reliable options for birth control on-demand, net improving women’s health and quality of life.

Some of us, however, are not impressed by the status quo. Odds are, the half of women (give or take) who discontinue their contraception within a year of trying it, would like to have some more options. The literature has had leads for future research on this for a long time. There are several, but NSAIDs are my favorite. Recent research confirmed their potential efficacy in the emergency contraception context. When will someone say “Screw the incentives,” and follow up with randomized trials to see what works?

As usual, I think women could and should do this research themselves using a free online citizen science platform (that doesn’t yet seem to exist). Women are pretty clear that better contraceptive options should be on the research agenda; they are voting with their bodies when they discontinue the usual suspects. NSAIDs like ibuprofen are over-the-counter, safe and well-tolerated enough that they’re already in common use, and would seem to have better net risk profiles than hormonal birth control — if they worked reliably for this purpose. (Otherwise, bets are off; pregnancy and birth are risky.)

Knowledge is power, and the ordinary people most affected by ordinary health practices should have it — and be part of making it, where it’s missing. That said, it’s not clear how one would go about discovering what works here…

Research Is Hard

We would want to compare a known-effective birth control method with the unknown one; what would that look like? Like with any other idealized (target) trial, we want to start by thinking about minimizing risk of bias. But what if different groups of women tend to select different birth control methods, making the groups systematically different? That could introduce unmeasured confounding. One possible way to address this would be to compare women who are randomly assigned to stick with their current contraceptive method, with women who are randomly assigned to try a new one.

But most of those methods may also have other health effects that women may not want, beyond just preventing pregnancy (e.g., ramping up prostaglandin production in ways that cause a lot of bleeding — or undermining metabolic or mental health). It might minimize risk of bias from those other possible effects to instead compare ibuprofen with less medically invasive contraceptive methods like condoms, withdrawal, and the rhythm method that don’t carry those risks — but tend to be less reliable than the other available options. That is, if the pill works in part by hurting women’s metabolic health, then maybe we don’t want to ask how its contraceptive efficacy compares to ibuprofen’s. Maybe we just want to know what works without hurting women’s health.

Relatedly, time could also matter in terms of both side effects and efficacy. If it does, then it’s not an apples-to-apples comparison to pit women’s current contraceptive methods (continuation) against a new option. So ideally, I think it would make sense to recruit women who want to change their current contraceptive method for this target trial. Recruiting women who want to try less medically invasive options would further protect against risks of bias from possible health risks of other options.

Ethically, it also makes sense to compare these options, in the sense that they may all have higher failure rates than other birth control (a lot depends on subgroups). It might even make sense to explicitly recruit women who would be ok with getting pregnant, even though they’re not planning on it. Although pregnancy intentionality is complex and could introduce bias into trials, at least equalizing it across study arms would account for that risk — as well as mitigating the significance of the risk of unintended pregnancy in the experimental group.