Hormonal contraception and depression: outcomes and implications of a new Danish national database study

A study recently published in JAMA Psychiatry linking hormonal contraception with depression has provoked a lot of media interest and may raise questions from our patients. This summary provides an overview of the study and its implications so we can support informed decision-making in relation to contraceptive choices.

While an association has been found between hormonal contraception and the first use of an anti-depressant medication, particularly for young women, as well as with a diagnosis of depression it is important to understand that this type of observational study cannot prove causality. It is also important to put the risks into context by looking at the absolute numbers as well as the relative risks which in this case are small.

How was the study carried out?

This national Danish prospective observational cohort study gathered data over a 14-year time frame starting in 1995 from over one million women aged 15 to 34 years of age using national registry data.  Data was collected on redeemed prescriptions for hormonal contraception (combined pills, progestogen only pills, vaginal ring or patch and hormonal IUD) as well as redeemed prescriptions for first use of an anti-depressant and diagnosis of depression at a psychiatric hospital. Women who had a prior psychiatric history, had used anti-depressants and had a history of cancer, venous thrombosis or infertility were excluded.

Note that the progestogen only pill use analysis included the high dose 75mcg desogestrel POP which is unavailable in Australia (but available in New Zealand) and that both the contraceptive implant and depo injection were excluded from the main study outcomes as the authors considered it more likely that they would be used by institutionalised women and those with more severe psychiatric conditions or intellectual disability.

 The authors controlled for other factors which may be associated with a diagnosis of depression and use of anti-depressants including educational level, endometriosis and polycystic ovary syndrome (PCOS) and limited the analysis to women who had already experienced first intercourse (they speculated that initiation of a first sexual relationship may be associated with a higher risk for depression). The authors also took changes in anti-depressant prescribing patterns over time into account by controlling for calendar year. I am sure we can all think of other factors that may influence the diagnosis and treatment of depression including  a higher likelihood for women visiting a doctor for prescribed contraception, and the researchers acknowledged the limitations of the study in this article.

What did the study find?

The Risk Ratio (RR) and 95% CI for 1^st use of an anti-depressant/diagnosis of depression compared with non-users:

• Combined oral contraception:  1.23 (1.22-1.25); 1.1 (1.08-1.14)
• Progestogen only pill 1.34 (1.27 – 1.40); 1.2 (1.04-1.31)
• Vaginal ring 1.6 (1.55-1.69); 1.6 (1.45-1.77)
• LNG-IUD 1.4 (1.31-1.42); 1.4 (1.22 – 1.50)

The relative risks generally decreased with increasing age and adolescents aged 15-19 years had a RR of 1.8 and 2.2 for combined pills and progestogen only pills respectively.

If we look at the absolute risks for 1^st use of an anti-depressant, non-users of hormonal contraception had an incidence rate of 1.7 per 100 woman years which increased to 2.2 per 100 woman years in users. This equates to an additional one woman prescribed a 1^st anti-depressant in every 200 users.

What are the implications of the study?

This observational study provides important new evidence about depression and hormonal contraception but it does not provide evidence of causality which would require a randomised controlled trial. The study, while important to be aware of, does not change clinical practice and women should be reassured that if they are happy with their current contraceptive method there is no indication to stop or switch methods.

A history or development of depression is not a contraindication to the provision or ongoing use of any hormonal method of contraception (UKMEC 1 for all methods) but women should be advised about the possibility of side-effects including mood changes for all hormonal methods and that these cannot be predicted on an individual basis. Women can be advised to return for review and a possible change of formulation or method if mood symptoms occur and made aware that the copper IUD provides a highly effective non-hormonal alternative.

Reference: Wessel Skovlund C. et al. Association of hormonal contraception with depression. JAMA Psychiatry Published online September 28, 2016.