Earlier this month, the surprising findings of some new research were presented at a conference. At the virtual European Congress of Psychiatry, Elena Toffol and her team from the University of Helsinki in Finland reported that they had found that attempted suicide rates were lower in women who used hormonal contraception compared to those who didn’t. In fact, the latter group were almost 40 percent more likely to attempt suicide than the former, they reported. 

These findings (which have not yet been peer-reviewed) might be the opposite of what you’ve heard—or experienced: Doesn’t hormonal birth control have a reputation for exacerbating mental illness? Your confusion would be forgiven. Perhaps you recall headlines from 2017, when a Danish study found that hormonal contraception was linked with an increase in attempted suicides. 

This giant contradiction is but one of many in the years of research that has tried to answer the question of whether hormonal birth control causes psychological side effects—and the jury is still out. In September 2016, The New York Times published an article with the headline “Contraceptives Tied to Depression Risk.” Six months later, the same publication came out with a piece headlined “Birth Control Causes Depression? Not So Fast.” 

Oral contraceptives, which first came on the market more than 60 years ago, are astonishingly popular. Over 100 million women worldwide are estimated to be current users. The pill, as the medication is known, comes in two forms: a progesterone-only version and a combined estrogen and progesterone version. Both contain synthetic hormones designed to stop or reduce ovulation—the release of the egg from the ovary. 

But the decision to use hormonal birth control is not always borne out of a desire to remain unpregnant. The name is rather a misnomer; a more fitting designation would be “hormone medication, often used as birth control.” Hormonal contraception is prescribed for a veritable smorgasbord of conditions, including migraines, cystic acne, chronic menstrual pain, polycystic ovary syndrome (PCOS), and endometriosis. 

Fears about the pill’s psychological side effects fall into a growing trend that has emerged in recent years: a widespread distrust of hormonal contraception and wariness of its downsides, now that the satisfaction of hard-won victories for women’s self-determination have worn off. A flurry of books questioning how hormonal birth control negatively affects its users have been published in the past decade. Coming out as the top concern are mood changes, which are reported to be the number one reason women choose to go off the pill.

But we don’t yet have a clear answer on whether the link between the pill and mood is real. The biggest problem is that most studies to date have been cross-sectional in design, meaning they involve taking a group of women who are using the pill and comparing them to a group who aren’t using it. “It doesn’t take into account that women who tried the pill and had negative mood effects or sexuality negative effects would come off it,” says Cynthia Graham, a professor in sexual and reproductive health at the University of Southampton and editor-in-chief of The Journal of Sex Research. “That, to me, is one big reason why it’s difficult to answer the question.” This is called the survival bias, or healthy-user bias. 

Plus, studying whether a drug causes conditions like depression and suicidality are just plain tricky. Extraneous factors like socioeconomic background, relationship status, a family history of mental illness diagnose all muddy any conclusions drawn.

The most famous research on the topic is a Danish study from 2016, led by Charlotte Wessel Skovlund, which followed over 1 million women between the ages of 15 and 34 for over 14 years, using registry data. It reported that women aged between 15 and 19 taking oral contraceptives were about 80 percent more likely than women who weren’t to be diagnosed with depression or prescribed antidepressants. Participants who were taking progesterone-only contraception were twice as likely to be depressed, the study found. 

The study made international headlines, and women reported feeling vindicated and believed about their experiences with the medication. And it wasn’t just for people who took the pill—other long-term contraceptive methods like hormonal implants, patches, and intrauterine devices also appeared to have a negative effect on mood. But the study is not the hard-and-fast ending to the saga. 

“That was a well-done study for what’s possible with observational data,” says Ruben Arslan, a personality psychologist at the University of Leipzig in Germany. Note the term “observational”—the biggest drawback of the study is a classic tale of “correlation, not causation.” They found a link between contraception and depression, but not definitive proof that one was causing the other. The study authors did control for certain factors that could confound the results; they excluded women who had received a depression diagnosis prior to the study, and women who were pregnant or who had given birth in the last six months to control for postpartum depression. To screen for depression, the study used a depression diagnosis or antidepressant prescription as indicators, but many cases of depression go undiagnosed and untreated. 

As is true with all scientific queries, the best way to answer them is through a randomized placebo-controlled trial—the gold standard of research. Instead of observing how the pill behaves out in the world—where all sorts of other factors could be at play—with this type of study you’d create two highly similar sets of participants, and give one group the pill and the other a placebo, without indicating who has received what. You can then be much more confident that any differences between the groups are caused by the pill and not some other factor.

Such trials take a lot of effort. Only a handful have been done on this subject, about six or seven in total, and they’re spread over the six or so decades the medication has been available, says Graham. The glaring gap is due to a number of factors: a lack of funding for women’s health research, fears that such research could be viewed as “pill-bashing,” and that many people in the field, Graham says, simply throw their hands up and say the matter is too difficult or too subjective—or too trivial—to study. “For me, that’s not a reason not to study it,” she says. 

“I’m a little uneasy about how much the Skovlund studies got covered, and how little coverage the randomized controlled trials got,” says Arslan. The most recent, a 2017 study from researchers in Sweden, took 340 women aged between 18 and 35 and gave one group the combined pill and the other a placebo, and followed them for three months. They found that while the pill caused a general decrease in quality of life, there was no increase in depressive symptoms. “This is some of the best evidence we have,” he says. “I think the reason they weren’t so big in the media is because the effects don’t lend themselves to some grand narrative.”

If either side proves to be correct, an injustice has been done. On one hand, people have been prescribed medication that can trigger or exacerbate mental illness with potentially insufficient warning or research into how damaging this could be. On the other, it means that women may have forgone effective contraception based on unfounded fears. The true answer probably falls somewhere in the middle. 

It’s clear that some women experience negative emotional side effects; the problem is we have no way of predicting who will, says Graham. “I think we have enough evidence, for sure, over all the years since the pill was first approved, that there are some women who react negatively in terms of their mood.” But what we don’t have, she says, is really good research studies that can say this is more likely for women who have a certain type of characteristic, or for a certain type of pill or a certain type of progesterone used. An ideal study for Graham would be a randomized controlled study that counted depression using both formal diagnoses of depression and a validated depression questionnaire. 

Arslan echoes the point that the experience varies from woman to woman. “A lot of women have more data that is relevant to their experience than any scientist can marshal—because we know the reality is heterogeneity,” he says. “To get an answer that’s useful as advice for individual women, I think that’s really hard.”