When the birth control pill debuted more than 50 years ago, women wanted to know: Is it safe? There wasn’t much evidence to answer that question, but women embraced the Pill as a revolutionary improvement in contraception.
Today, millions of women around the world use hormonal contraceptives that have expanded beyond the Pill to patches, implants, injections and uterine devices. Decades of research support their safety, and serious but very rare side effects such as blood clots are finally much better understood. But other areas of research lag, and we still don’t know as much as we’d like about how these medications affect women’s mental health.
So when a study came out linking hormonal birth control and depression, the headlines went wild. The stories made for good clicks, but not so great science reporting. Insufficient skepticism about a single study makes it easy to imply birth control definitely causes depression when the study shows nothing of the sort.
The study, which was conducted in Denmark and published in JAMA Psychiatry, analyzed 14 years’ worth of health data for more than 1 million women from national healthcare systems and databases not available in most other countries. It also measured depression two ways: diagnosis at a psychiatric hospital, which would be quite severe depression, or filling a prescription for antidepressants. Across the whole study, 2 percent of all women ages 15 to 34, were diagnosed with depression at a hospital and 13 percent began taking antidepressants.
Several news stories reported an 80 percent increase in risk of depression in some groups of women, but few noted that was relative risk, which is an expression of proportional increase. An 80 percent relative risk does not mean that 80 percent of women taking hormonal birth control develop depression. It means that if 10 women not taking hormonal birth control develop depression, then 18 women on the birth control will develop depression.
In this study, that 80 percent increase in relative risk referred specifically to the likelihood that those ages 15 to 19 taking combined oral contraceptives — the pill containing both progestin and estrogen — would begin taking antidepressants after going on birth control.
However, looking at absolute risks conveys a less dire interpretation. Among women not taking hormonal birth control, 1.7 percent took antidepressants and 0.28 percent received a depression diagnosis at a psychiatric hospital. By comparison, 2.2 percent of women who started birth control began taking antidepressants afterward, and 0.3 percent were diagnosed with depression at a hospital. Basically, about 0.5 percent of women who began hormonal contraception developed depression who might not have otherwise.
“Therefore for an individual woman, even one using a method of hormonal contraception, the overall probability of experiencing one of these outcomes in this study was still fairly low, particularly for diagnosis of depression,” explains Chelsea Polis, a senior research scientist at the Guttmacher Institute.
Rates of those filling antidepressant prescriptions were higher for other forms of hormonal birth control: 4.1 percent for the patch and 3.2 percent for the vaginal ring in the first year, for example. Psychiatric depression diagnoses occurred in 0.7 percent of patch users and 0.6 percent of vaginal ring users.
Even in terms of relative risk, though, many of the risk increases were modest: Among all women taking the combined pill, the increased risk was 10 percent for depression diagnosis and 20 percent for using antidepressants after statistically adjusting for women’s age, educational level, weight and history of endometriosis or polycystic ovary syndrome, all factors that could influence depression risk.
Other increased risks ranged from 20 to 70 percent for all women, depending on contraception type. The largest increases — up to triple the likelihood of starting antidepressants — occurred among teens using the ring or patch.
Another thing to consider is that these numbers represent correlations — two things occurring at the same time that may or may not be linked.
“Depression is common. Contraception use is common. So both of those things are commonly going to occur together,” explains Jeffrey Jensen, a professor of reproductive & developmental sciences and director of the Women’s Health Research Unit at Oregon Health & Science University in Portland.
The study used several methods to reduce the possibility that other things could be causing depression. Jensen pointed out that women who are more likely to take hormones for contraception would probably also be more likely to take antidepressants for depression, but the authors did a separate analysis to compare women to themselves before and after beginning contraception and still found a depression risk.
None of this means that birth control does not cause depression, but it doesn’t mean it does, either. Since a half percent of millions of women taking birth control adds up, it’s important to know whether such a large number of women could be more susceptible to depression, which can be a very serious illness, after starting hormonal contraception.
But it’s complicated, and subtle.
For example, the study also found that depression risk peaked six months after women began using contraceptives, but then decreased to the point that women using hormonal contraception for four years actually had lower rates of depression than those not taking it. Though this likely resulted from many women with depression stopping their birth control, those findings match up with a previous large study finding a protective effect against depression with hormonal contraception. But the studies were done differently: The earlier study included only sexually active women, unlike the new study.
“Unfortunately, the analysis did not provide information on the frequency of depression diagnoses or antidepressant use among women using nonhormonal methods of contraception, such as copper IUDs,” says Polis. “Such a comparison would help to clarify whether the associations were related to other factors common to women choosing to use contraception, rather than being specifically related to the hormonal content of certain contraceptive methods.”
For example, those who become sexually active in adolescence have a higher risk for depression and anxiety, as previous research has shown. Even among women in their 20s and 30s, the decision to begin hormonal contraception may accompany various other circumstances in their lives that could potentially increase the risk of depression or anxiety — not a stretch when the study identifies just a half percent of women with the increased risk.
But the possibility that sexual initiation might come with mental health risks baffled study co-author Lidegaard. “Sexual relationships are a good experience for the majority of women, so I cannot see why women would get depressed by starting sexual relationships,” Lidegaard said in an interview. He pointed out that women not in relationships may experience loneliness, a risk factor for depression, but when asked about single women having sex, he said he believes “the majority feel more happy by realizing how wonderful sexual experiences can be. Why should women get depressed from that?”
What this study does do is suggest that women may respond differently to hormones and medical treatments.
“Doctors should perhaps be more careful when they prescribe hormonal contraception to young women and get a history of previous depression first,” Lidegaard says.
There’s no question that women are going to respond differently to hormonal contraception. But those differences are not well understood.
“We’re all very, very different, and we’re moving in the direction of precision medicine,” adds Catherine Monk, an associate professor of psychiatry and director of research at the Women’s Program at Columbia University Medical Center. “There are some women who are just much more sensitive to these hormone changes.”
The higher risk of depression found among teenage girls, both Monk and Lidegaard pointed out, might be biological, since teens may be more sensitive to the hormonal changes happening during puberty.
“You need to know yourself and be really informed,” Monk says. “If I were of an age and reading this article right now, I would want to be thinking, ‘Who am I? Am I sometimes who has moodiness around my periods?’ ” She notes the importance of each woman considering these findings in conjunction with what she knows about her own body and her own circumstances.
Philosophical differences about medicine may also influence how people interpret these findings in the absence of studies showing causation. Jensen, for example, pointed out that women in developed countries no longer see other women dying during childbirth, illegal, unsafe abortions or other devastating health effects of unplanned pregnancy, and have come to places less value the effectiveness of hormonal birth control.
“Women are more skeptical of using hormonal therapy than ever before,” Jensen says. “It’s a tragedy of the riches. If you really want to be depressed, have an unintended pregnancy.”
Toward the other end of the spectrum, Monk believes we may have gone too far in using hormones to control contraception. “Getting away from barrier methods of contraception is getting away from our bodies,” she says, and she would like to see a much larger range of options for contraception for men and women.
The fact that dissent rages over this issue points out a larger question about women’s health research.
“Understanding women’s health has been neglected, and there’s not enough research into understanding our hormones,” Monk says. “This is partially a story about women’s health research and how we need more of it.”
Tara Haelle is a freelance science and multimedia journalist who specializes in writing about women’s health, prenatal health, pediatrics, mental health, vaccines, infectious disease, and other medical research. She is a Forbes contributor whose work also appears in NPR, Scientific American, Slate, Politico, Medscape, Everyday Health and elsewhere. This article was previously published on NPR.