Family doctor

 There are some easy mistakes to make when considering the relationship between a medical treatment and side effects. The first is to mix up correlation and causation- just because something is seen more often within a certain group of people does not mean that a causal link is present. This is the problem with observational studies; they have no real way of controlling for the possible confounding factors that can independently influence outcomes. This is even more pronounced when the two exposures are common in the general population.

Proving direct causation between an exposure and an outcome that you can’t randomise people to receive (for ethical or practical reasons) is tricky. Instead, the most accurate picture we are able to get is through the synthesis of multiple observational studies, as well as considering the biological plausibility of the link, ie is there a realistic, scientific reason why a medication may cause the given side effect?

Trying to understand the relationship between hormonal contraception and depression demonstrates these difficulties well. Both hormonal contraception use and depression are very common occurrences and are readily influenced by other factors. We are reliant (for the most part) on observational data in this area because of the difficulties of randomising people to receive hormonal contraception.

A prospective trial published in JAMA Psychiatry has reignited our debate. The study found that hormonal contraception is associated with an increased chance of beginning antidepressants ¹. It concludes that depression is a real, potential side effect of hormonal contraception that is perhaps under recognised in current practice.

Is there a plausible reason for a connection?

It is well documented that the life-time prevalence of depression in women is double that seen in men following puberty ². Evidence suggests that depression is more common in women during times of hormonal fluctuation such as menopause (when levels of oestrogen decline) and during the perinatal period where the placenta makes additional hormones.

Studies have suggested that there is a close relationship between oestrogen and serotonin signalling pathways in the brain ³, with lower levels usually linked to increased rates of depression. Other sex steroid hormones such as progesterone have been linked to emotional and cognitive processing through increasing an inhibitory receptor in the brain (GABA) ⁴.

Public opinion has played perhaps the most significant role within this connection, and anecdotally (more so than any other form of evidence) progesterone contraceptives are linked to self-reported lower mood.

It is important to keep in mind that depression is the result of a complex interplay between a biologically-vulnerable individual and their environment. With known contributors including vulnerable personality styles, limited social support, daily increased life stress and stressful childhood experience, as well as a family history and previous history of depression.

Our brains are too complex for something as simple as an added hormone or low hormonal levels to be the direct and sole cause of an experience as varied and complex as depression. However, it is scientifically plausible that hormonal levels influence mood and in turn, may increase the rates of depression in certain individuals.

What are our potential confounding factors?

It is possible that hormonal contraception may increase the likelihood of depression through factors other than the effects of the hormone itself. These include the experience of being in an intimate relationship (which may add increased life stress), and having painful periods or endometriosis for which people require hormonal contraception for treatment. It could also be that people who are engaged with a doctor (evidenced by being prescribed hormonal contraception) are more likely to see a doctor regarding their low mood, and therefore to be prescribed antidepressants. It is important to keep these interactions in mind when interpreting the evidence as these factors make it tricky to separate correlation from causation.

So what does the evidence say?

It would be great if we had a concrete answer, but nothing about the body (particularly the brain) is straightforward. Our conclusions are largely limited by variations in the study methods as well as drug formulations, and the difficulty of randomising and placebo-controlling individuals to contraception.

There is evidence both for and against a relationship between hormonal contraception and depression. While most of the data has focused on a link to low mood, there is also evidence to suggest that some forms of contraception are mood-stabilising and can prevent the fluctuations in mood seen during normal, unaltered menstrual cycles ⁵.

One large review on oral contraceptive use and depression, found that the increased risk was most pronounced in women with a high risk of depression in the first place, such as having had prior depression, experiencing other psychiatric symptoms and pre-menstrual mood symptoms prior to the pill ⁶.

An observational study (part of a larger, Australian longitudinal study on Women’s Health) on young, Australian women found that after adjusting for potential confounders, there was no difference in depressive symptoms between those using oral contraception and those who weren’t ⁷. They also found that women using the pill for non contraceptive reasons (e.g minimising pain of endometriosis) were 1.32 times more likely to experience depressive symptoms than those using the pill for contraception. This highlights just how vulnerable these studies are too confounding and external bias.

We will now go on to look at the results of a new study that has been published this month, reigniting the debate over whether hormonal contraception can cause depression.

The Danish study

A recent study of over 1 million Danish women has found a statistically significant link between starting hormonal contraceptive treatment and subsequent antidepressant use.

The study was an observational, prospective study. This means that rather than taking one snapshot in time and measuring how many people on contraception were also taking antidepressants, the study set a time point (the year 2000) and measured the incidence of new antidepressant prescriptions following hormonal contraception use. One of the study’s great strengths is that it captured the entire Danish female population who had started contraception after the study began, including over 1 million women in the analysis.

It found that a wide variety of hormonal treatments including the combined oral contraceptive pill, the progesterone only ‘mini pill’, the depot (progesterone injection), implant (Jadelle in New Zealand), intrauterine device (Mirena), vaginal ring (infrequently used in New Zealand) and progesterone transdermal patch (also infrequently used) were all associated with an increased risk of antidepressant use. This was particularly pronounced in adolescents, and seemed to peak at 6 months of hormonal use.

The risk associated with each form of contraception was slightly different, and is presented below as a risk ratio (e.g 1.5 means you are 50% more likely to be started on antidepressants if taking the hormonal contraception):

Combined oral contraceptive pill: 1.23

Progesterone-only pill: 1.34

Progesterone implant (similar to the Jadelle): 2.1

Progesterone injection (also known as the depot): 2.7

Progesterone intrauterine system (similar to the Mirena): 1.4

Progesterone patch: 2.0

Vaginal ring: 1.6

Potential limitations with the research:

The study corrected for obesity, smoking, polycystic ovary syndrome, endometriosis and sexual intercourse in its analysis. The suggestion was that the beginning of a sexual relationship may in itself be related to depression but they found no added relationship. The study also suggested that many sexually active adolescents begin contraception with condom use and start on the oral contraceptive later, reducing the effects of this potential bias. There were no differences seen when the study adjusted for obesity, smoking, polycystic ovary syndrome and endometriosis status.

The study did find a link between low levels of education and the depot and implant use and suggested that perhaps that this could influence the data seen, with higher rates of poor education linked to increased chance of depression. The study also hypothesised that significantly increased risk of antidepressant use in these two groups could be dose-related, in that they receive a much higher dose of progesterone at once with the injection and implant.

When considering the biological plausibility of these results, keeping in mind the hypothesis that progesterone increases the risk of low mood, it is interesting to note that there is still a high risk of antidepressant use in those using the intrauterine device and ring. Theoretically, there is much less absorption of progesterone into the body if it is being applied straight into the site of action (the uterus and vagina). However, the study still found a significantly increased risk of antidepressant use. It would have been interesting if the study had included non-hormonal contraception use (Copper IUD) as a control to compare results.

So where to from here?

The Danish study has reignited the conversation about whether hormonal contraception use increases the risk of depression. While it is a comprehensive, detailed piece of research, it is inherently flawed by its observational nature and should not be considered the be-all and end-all of research in this area.

The decision to start hormonal contraception should still be treated in an individual context, and is a personal discussion to be had with your doctor. Women at a high risk of depression, either with a background history or strong mood symptoms during their period, may be best to try non hormonal methods of contraception to begin with such as the copper IUD. However, there is no clear answer yet, and it is likely that the anecdotal evidence will remain strong and dominate the debate, while the concrete, scientific evidence will take longer to form conclusions and provide us with answers.

References:

1. Skovlund C, Mørch L, Kessing L, Lidegaard Ø. Association of Hormonal Contraception With Depression. JAMA Psychiatry. 2016. doi:10.1001/jamapsychiatry.2016.2387.

2. Desai HD, Jann MW. Major depression in women: a review of the literature. J Am Pharm Assoc (Wash).2000; 40(4):525-537.

3. McEwen BS. Invited review: estrogens effects on the brain: multiple sites and molecular mechanisms.J Appl Physiol (1985). 2001;91(6):2785-2801.

4. Andréen L, Nyberg S, Turkmen S, van Wingen G, Fernández G, Bäckström T. Sex steroid induced negative mood may be explained by the paradoxical effect mediated by GABAA modulators. Psychoneuroendocrinology. 2009;34(8):1121-1132.

5. Oinonen KA, Mazmanian D. To what extent do oral contraceptives influence mood and affect? J Affect Disord 2002;70:229 – 40.

6. Kahn LS, Halbreich U. Oral contraceptives and mood. Expert Opinion

Pharmacotherapy. 2001;2:1367 – 82.

7. Duke JM, Sibbritt DW, Young AF. Is there an association between the use of oral contraception and depressive symptoms in young Australian women? Contraception. 2007;75(1):27-31.