"You can die of a broken heart, study indicates," The Guardian reports. The study found that people who lost a partner – especially if the death was unexpected – had an increased risk of developing an irregular heartbeat up to a year after the death.
The study specifically looked at a type of heart condition called atrial fibrillation that causes an irregular and often abnormally fast heart rate. People with atrial fibrillation are at increased risk of having a stroke and, less commonly, heart failure.
The study found people whose cohabiting partner or spouse died had an increased risk of getting atrial fibrillation in the first month after the death; this is estimated to be around 41% higher than average. The risk was highest in the second week after the bereavement (90% higher than average) and gradually declined to almost the same as someone who had not been bereaved by the end of the first year.
We don't know for certain that bereavement directly caused the increased risk of atrial fibrillation. Also, despite the headlines, we don't know whether anyone in the study died. Although atrial fibrillation can raise the risk of having a stroke or getting heart failure, it is not usually life-threatening and can be treated.
Anyone who notices a sudden change in heartbeat, or a consistently fast or irregular heartbeat, should seek medical attention.
Where did the story come from?
The study was carried out by researchers from Aarhus University in Denmark and was funded by a grant from the Lundbeck Foundation. Lundbeck is a pharmaceutical company which makes cardiovascular drugs.
The UK media was unable to resist the cliché about dying from a broken heart (which, to be fair, neither could we), even though the study did not include any data on deaths from atrial fibrillation after bereavement.
The leading charity for atrial fibrillation, the AF Association, reports that: "AF is also associated with a slightly increased risk of death although this is a very small risk and generally AF is not considered a life threatening disease in its own right."
However, beyond the headlines, most reported the results accurately.
Although most news sources included the relative risk figure of 41% increased risk, they did not carry the absolute percentages of people with atrial fibrillation who'd had a bereavement, which only showed a small increase in absolute risk.
What kind of research was this?
This was a population-based case control study. Researchers wanted to see if people with atrial fibrillation were more likely than people without to have experienced the recent death of a partner. Case control studies, especially big ones like this, are useful ways of identifying differences between people with and without a condition. However, they can't prove cause and effect.
What did the research involve?
Researchers identified everyone in Denmark who had been treated in hospital for a first episode of atrial fibrillation from 1995 to 2014. They then chose 10 "control" subjects for each person with atrial fibrillation, to compare them. They recorded whether each person had experienced the death of a cohabiting partner, and how recently.
After adjusting their figures to take account of confounding factors that could affect the results, they calculated the chances of having atrial fibrillation in the days, weeks and months after the death of a partner, compared to people who'd not had a bereavement.
The figures came from national Danish registries of health and civil status, which give high-quality information about people's hospital treatment, prescriptions, cohabitant status (i.e. whether they lived with a partner) and deaths.
The researchers controlled for people's age, sex, education level, civil status and whether they'd previously had cardiovascular disease, diabetes, or were taking cardiovascular medicines.
Researchers carried out additional analyses to check for the influence of these potential confounding factors. They calculated the overall chances of having atrial fibrillation for bereaved versus non-bereaved people, then calculated the chances according to time since bereavement. They also looked at whether unexpected deaths were more likely to be linked to atrial fibrillation, by calculating the likelihood of death one month before it happened, using health data.
What were the basic results?
The study included 88,612 people with atrial fibrillation, 19.72% of whom had experienced the bereavement of a partner. Of the 886,120 controls without atrial fibrillation, 19.07% had lost a partner. This translated into a 41% higher relative risk of atrial fibrillation in the month after the bereavement (odds ratio [OR] 1.41, 95% confidence interval [CI] 1.17 to 1.7).
The increased risk was highest 8 to 14 days after the death, and gradually declined until a year afterwards, at which point the risk was "close to that of the non-bereaved population". Risk was higher among younger people (under 60) and among those where death was unexpected. Where deaths were likely due to ill health, there was no increased risk of atrial fibrillation in the partner after the death.
How did the researchers interpret the results?
The researchers said that in their study: "The severely stressful life event of losing a partner was associated with a transiently increased risk of atrial fibrillation, which lasted for about one year. The elevated risk was especially high for those who were young and those who lost a relatively healthy partner."
They say "residual confounding cannot be ruled out" because they were unable to include lifestyle factors, physical activities and family history of atrial fibrillation. However, they say they believe the risk of any of these factors affecting the results is small, "as we cannot think of any possible confounder that could cause a transiently increased risk of atrial fibrillation shortly after bereavement".
They discuss two possible ways in which bereavement could trigger atrial fibrillation. They suggest that acute stress could influence heart rate directly via the sympathetic nervous system, and promote the release of inflammatory substances called cytokines. They also say that factors such as drinking more alcohol, not sleeping, eating a poor diet and doing less physical activity could be a cause both of AF, and a direct response to bereavement.
The study found that people were more likely to have AF for the first time in the weeks immediately after a bereavement, but that this raised risk does not last. Despite the headlines, this does not mean that anyone who's had a bereavement is at immediate risk of "dying of a broken heart". Atrial fibrillation is treatable and not usually life-threatening.
Because this was an observational study, we cannot rule out the possibility that factors such as family history of atrial fibrillation or lifestyle factors could have affected the results, although the researcher's conclusion that this is a small possibility seems reasonable. The size of the study, and its use of big national databases, adds to its value.
It's not surprising to learn that people may be at increased risk of ill health immediately after the death of a partner, which is one of the most distressing events we are likely to face. The study gives us more reason to look after people who have been bereaved, as their health is vulnerable during this time.
Anyone with symptoms of atrial fibrillation, such as a very fast or irregular heartbeat, should see their GP. Anyone with chest pain and symptoms of a heart attack, such as pain travelling from your chest to your arms, jaw or neck, and shortness of breath, should call an ambulance and take an aspirin while they are waiting for it to arrive.