Mental health

Antidepressants and stroke

“Postmenopausal women who take antidepressants may be increasing their chance of suffering a stroke and dying prematurely,” reported the Daily Mail . It said a six-year study found a 45% increase in risk of strokes for women who used antidepressants compared to women who did not use them.

As the newspaper also reported, the absolute increase in risk of stroke (the number of women who might be affected) was small, equating to an increase of about 13 additional women in every 10,000 (0.43% of women on antidepressants compared to 0.3% of women not on them). In addition, depression itself is a known risk factor for stroke, so it is not clear how much of the increase was due to depression rather than the drugs.

Overall, this increase in risk was small and may not be attributable solely to the drugs themselves. Taking any medication involves weighing up the pros and cons of taking the drug compared to the prospects of leaving the disease untreated. As the British Heart Foundation said, “it is important to weigh up any small increase in the risk of stroke with the benefits of treating depression".

Where did the story come from?

This research was carried out by Dr Jordan W Smoller from the Department of Psychiatry at the Massachusetts General Hospital in Boston along with other investigators collectively known as the Women’s Health Initiative (WHI) Investigators. The WHI study is funded by the National Heart, Lung, and Blood Institute, National Institutes of Health, US Department of Health and Human Services.

This study was published in the peer-reviewed medical journal Archives of Internal Medicine.

What kind of research was this?

This was an analysis of data from a long-running cohort study. The researchers note that antidepressants are among the most widely prescribed medications and their effects on heart disease, the risk of stroke and death is unclear.

Here they investigated whether there is any association between antidepressant use and the risk of heart disease, stroke or death, comparing both older drugs used to treat depression called tricyclic antidepressants (TCAs) and the newer, more commonly used drugs, selective serotonin reuptake inhibitors (SSRIs).

The researchers also say that depression is an independent risk factor for cardiovascular morbidity and mortality, meaning that the disease itself can increase the risk of heart disease and strokes.

What did the research involve?

The WHI study involved 161,608 postmenopausal women aged 50 to 79 years in a series of overlapping clinical trials and a prospective cohort study. The participants were enrolled between 1993 and 1998 using mass mailings. As the research was investigating the risk factors for several chronic diseases, it excluded women who already had certain diseases (including depression) or were already taking antidepressants.

A first follow-up visit revealed that about 5,500 of these women had started or were currently using antidepressants. These women were then followed for about six years on average (maximum 10.8 years) to see if they developed heart disease, stroke or died. They were not asked about depression again.

The characteristics of women on antidepressants were compared to those of women taking no antidepressants. Accepted statistical techniques were used to assess the link between these factors and heart disease, stroke and death from these or any other cause.

Several adjustments were made to the analysis to reduce the possibility that other characteristics of the women, depression for example, affected the link between antidepressant use and disease or death (confounding). The researchers admit that this may not have been completely effective and that, technically, there may have been some ‘residual confounding’.

What were the basic results?

The researchers say that antidepressant use was not associated with coronary heart disease. However, with SSRI use there was an increased risk of stroke (hazard ratio [HR] 1.45, 95% confidence interval [CI] 1.08 to1.97) and death from any cause (HR 1.32, 95% CI 1.10 to 1.59). The excess risk of stroke with SSRIs was largely for haemorrhagic stroke (HR 2.12, 95% CI 1.10 to 4.07). This is one particular type of stroke and any increase in risk for the other type, ischaemic stroke, was not statistically significant.

The annual rate of stroke for women not taking an antidepressant was 2.99 per 1,000 women per year compared to 4.16 per 1,000 women per year for women using SSRIs. (The risk quoted here comes from all women in participating trials, whereas the risk quoted by some newspapers is from the observational studies only).

There were no significant differences between SSRI and TCA use in risk of any outcomes.

How did the researchers interpret the results?

The researchers conclude that for postmenopausal women:

  • There were no significant differences between SSRI and TCA use in risk of coronary heart disease, stroke or mortality
  • Antidepressants were not associated with risk of coronary heart disease
  • Tricyclic antidepressants and SSRIs may have been associated with increased risk of mortality, and SSRIs with increased risk of haemorrhagic and fatal stroke
  • The absolute risk of these events was low

They conclude that “these findings must be weighed against quality of life and established risks of cardiovascular disease and mortality associated with untreated depression”.


This study has collected and pooled a large amount of data from several studies of postmenopausal women. In the full journal article, the researchers are cautious in their interpretations of their results, discussing the issue of residual confounding and other limitations in four pages of comments.

The study has some limitations:

  • There was no difference in risk between the two different types of antidepressants (SSRIs compared to TCAs) despite the drugs working in different ways. This raises the possibility that depression could account for part of the excess risk, rather than the drugs that are used to treat it.
  • The women were only assessed once for their antidepressant use at the start of the follow-up period. It is possible that some women started antidepressants after this initial assessment, and these women would still be considered to be ‘non-users’. This would have reduced the effect.
  • The diagnosis of depression used at the first follow-up visit may not be as accurate as a clinical diagnosis of depression. This may have led to an inaccurate estimate of the rates of depression.
  • This study included only postmenopausal women. The results may not apply to younger women or to men.

The researchers comment that this is the largest study yet to look at this association and may be the closest researchers can get to an estimate of the effect of antidepressant use on these outcomes.

Because of the limitations noted above, this study has not conclusively proved that antidepressants, rather than depression, are the cause of the small increase in risk of stroke.

NHS Attribution