Hormonal contraception and depression: another Pill scandal?

Publication of the article by Skovlund et al. ‘Association of hormonal contraception with depression’ [1] has produced a strong media reaction in several countries, with recent headlines like those in the German magazines Gala ‘Finally proven: the Pill causes depression’ and Brigitte ‘Danish scientists show the Pill increases the risk of depression in young women by 80%’. Although the article’s authors refer to association and not causality in the title and in the conclusion, and indicate that ‘further studies are warranted to examine depression as a potential adverse effect of hormonal contraception’, social and other media have seized on the study as proof that the Pill causes depression. After the media storms surrounding thromboembolic risk and sexual dysfunction, we now seem to have another public health issue linked to hormonal contraceptives. No doubt we will soon read about the money we could save if women stopped taking the Pill, by reducing the consumption of antidepressants.

First, we should ask the most important question: does this study prove that combined hormonal contraceptives cause depression? The answer, for several reasons, is no.

• Depression is a multifactorial disease in which genes, environment, life events, distress and hormones interact in a complicated, individual way. This study did not have the design or the information to control for these possible confounders.

• The diagnosis of depression is even more difficult than the diagnosis of thrombosis (itself quite difficult). The study took as an outcome not the diagnosis of depression but treatment with antidepressants, as a substitute for diagnosis. Another study outcome was hospitalisation due to depression. This is a methodological problem. It may well be that women who take the Pill have more connection with health care services and are more likely to report symptoms that lead to a prescription. It may well be that pre-existing depressive disorder is diagnosed at the time when the woman is using a combined oral contraceptive (COC) containing a specific progestogen (see below).

• The study’s ‘sensitivity analysis’ was insufficient to exclude psychosocial changes that might have contributed to the depressed mood of individual women.

• No information was given about the women who stopped taking COCs because of mood changes and what the outcome was. This is, however, what happens in real life and it is important to know whether the reported side effect disappeared after stopping or changing the contraceptive.

• In terms of absolute risk, the importance of the increase in risk seems much less impressive. Among women not taking hormonal birth control, 1.7% took antidepressants and 0.28% received a diagnosis of depression at a psychiatric hospital. By comparison, 2.2% of women who started birth control began taking antidepressants afterwards and 0.3% were diagnosed with depression at a hospital. Basically, about 0.5% of women who began hormonal contraception developed depression who might not have developed it otherwise.

• There are several results that raise concerns about biological plausibility. Why should teenagers using the patch or the ring have a statistically higher risk of depression than those using oral preparations containing the same steroids? It is hard to believe that the small pharmacokinetic and pharmacodynamic differences would account for this difference.

Apart from these limitations the publication suffers lacked a more critical discussion of other studies that reached different or conflicting results. For instance, a study by Keyes et al. [2] indicated a protective effect of hormonal contraception with respect to affective disorders. A Swedish observational study published in this journal found a positive association between the use of progestogen-only contraception and antidepressants, particularly among teenagers.[3,4] Regarding combined hormonal contraception, teenage users were more likely, but older women less likely, to be prescribed antidepressants compared with non-users. Another study, not mentioned in the Skovlund et al. article, may contribute to understanding the complexity.[5] This study included women under the age of 40 years suffering from major depression: 223 used combined hormonal contraceptives, 58 used progestogen-only preparations and 948 did not use hormonal contraceptives. The women who used combined hormonal contraceptives were significantly less depressed than the women who did not. Users also showed higher physical fitness and less comorbidity with compulsive disorders.

The second question is: does this study show that low or depressed mood in some women may be related to the progestogenic component of hormonal contraception? The answer is a cautious yes. This is something we already know not only from clinical experience but also from studies on the impact of hormones on mood and affect.

There is a huge body of literature about the impact of steroid hormones on the most important neurotransmitters in the brain. The interaction is complicated and depends not only on the presence or level of hormones but also on the hormone receptors in the brain. The vulnerability of women to endogenous steroid hormones has been very well studied in research aimed at trying to understand the hormonal mechanisms involved in premenstrual syndrome (PMS) and premenstrual dysphoric disorder (PMDD). These studies show that in vulnerable women the use of progestogens may increase the symptoms of depressed mood. There seems to be a difference between progestogens: antiandrogenic or neutral progestogens show a more favourable action compared with androgenic progestogens. The final action in the individual patient is, however, complicated by the fact that the estrogen component has a mood-elevating effect while at the same time reducing circulating levels of free testosterone, which may have a mood-lowering effect. We know from studies of the progestogen drospirenone that this COC is able to treat the psychiatric affective disorder PMDD apparently as effectively as the gold standard treatment with selective serotonin reuptake inhibitors.[6]

Other studies have shown the complexity and sometimes contradictory action of synthetic steroids on the brain: Oinonen and Mazmanian [7] found that, compared with non-users, oral contraceptive (OC) users experienced less variability in affect across the entire menstrual cycle, and less negative affect during menstruation (i.e., during withdrawal bleeding).

In women with negative mood and affect change related to OC use, potential mediators of the relationship between OCs and mood or affect have been identified: a history of depression, psychiatric symptoms, dysmenorrhoea and premenstrual mood symptoms prior to OC use; a history of pregnancy-related mood symptoms; a family history of OC-related mood complaints; being in the postpartum period; and age. Furthermore, a lower ratio of progestogen to estrogen is associated with more negative mood changes in women with a history of premenstrual mood symptoms; a higher progestogen to estrogen ratio is associated with increased negative mood effects in women without such a history; monophasic OCs have a greater stabilising effect on mood compared with triphasic OCs.

In an important study, Rapkin et al. [8] showed that in healthy women without underlying mood or anxiety disorder the use of a low-dose OC did not result in adverse psychological symptoms despite a significant reduction in neuroactive steroids, indicating that individual vulnerability to steroid action in the brain, as we know from PMS and PMDD, is a precondition for adverse effects and that in healthy women fluctuations of neurosteroids are well tolerated.

In summary, we do see women who react with depressed mood when taking certain hormonal contraceptives (probably related to the particular effect of the specific progestogen on receptors in the limbic system). Interestingly enough, the same women may feel better by changing progestogen. The Danish study [1] thus confirms what the European Society of Contraception and Reproductive Health have always advocated: that contraceptive counselling and care should be tailored to the individual.

• Carry out a biopsychosocial assessment of the woman, including her mental condition and environmental distress, as part of her psychosocial and cultural profile.

• Provide a choice of contraceptive methods, including the use of different progestogens with differential actions on different parts of the body such as the reproductive system, the cardiovascular system, the skin and hair, and last but not the least the brain.

• During use, ask proactively about side effects, including questions about sexual function and affective symptoms, and explore factors contributing to any complaints that may be due to the contraceptive used or to other causes.

• Discuss changing the contraceptive method or changing the type of hormonal contraceptive. Ask not only about bleeding or physical symptoms but also about mood and sexuality.

Johannes Bitzer

Editor-in-Chief

The European Journal of Contraception and Reproductive Health Care

[email protected]