Can a loved one break your heart?

Behind the Headlines

Tuesday October 9 2007

Picture of a couple having a fight

Psychological or social stress is not good for you

“Being in a stormy relationship is bad for the heart,” reported the Daily Mail today. Other newspapers that covered the story included the Daily Express, which said that “Stress and anxiety caused by hostile partners or relationship breakdown can boost the chance of a heart attack or chest pain by 34%.”

The stories were based on a study that asked 9,000 civil servants about the “negative aspects” of their relationships and then followed them for more than 12 years to see if they experienced heart disease.

This well-designed study shows there is an association between “negative aspects” of relationships, such as arguments, and the risk of heart disease. However, although it has demonstrated that an association exists, this study does not show that negative aspects of relationships actually directly cause heart disease.

There may be unknown factors giving rise to this association, such as family history of heart disease, the quality of other personal relationships, or other personality factors.

Finally, the nature of the relationships the participants described at the outset could well have changed over the lengthy study period.

Where did the story come from?

Doctors Roberto De Vogli and colleagues from University College London carried out this research. The study was funded by the Medical Research Council, British Heart Foundation, Health and Safety Executive, and Department of Health in the UK, the National Institutes of Health in the US; and the John D. and Catherine T. MacArthur Foundation Research Networks on Successful Midlife Development and Socioeconomic Status and Health.The study was published in the peer-reviewed medical journal: Archives of Internal Medicine.

What kind of scientific study was this?

This study analysed results obtained in a prospective cohort study, called the Whitehall II study, which recruited 10,308 civil servants between 1985 and 1988, who were aged 35 to 55 years old and working in London.

For this study, the researchers selected people who had no history of heart disease at the time of their enrolment into the study. Of these, 9,011 people gave information on their close relationships either when first enrolled (1985-1988) or in the first period of the study (1989 to 1990).

Participants completed a questionnaire which asked about the negative aspects (such as arguments) of up to four of their close relationships, and whether they had received emotional and practical support in these relationships in the past year. This particular study looked only at data on the closest relationship that each participant reported and in about two thirds of cases, this was a spouse. Participants were separated into three groups based on their scoring on these questionnaires (lowest, middle, and high score groups).

The researchers then followed participants an average of 12 years up to 2004, to see who developed heart disease (heart attack or angina). They checked the NHS Central Registry death records to identify the participants who had died of a heart attack between 1990 and 2004. Participants who were still alive were contacted and asked if they had experienced chest pain or had a heart attack in this period. Medical records were checked to confirm any possible non-fatal heart attacks or angina; these people were also given ECGs or angiograms to help confirm diagnoses. Only those events that could be confirmed were included in the analyses.

Researchers then analysed the data to see if there was any difference in the rate of heart disease between the three groups. They initially adjusted these analyses for factors that might influence the risk of heart disease, including age, gender, employment grade, marital status, blood pressure, cholesterol level, obesity, diabetes, and social support. Further adjustments were made for depression, work stress, smoking, exercise, alcohol intake, and fruit and vegetable consumption.


What were the results of the study?


This evidence supports the belief that bad relationships increase the risk of certain diseases, and that reducing the risk of disease has to be set in the context of the individuals life and social environment.

Sir Muir Gray

About 7% of people developed heart disease (heart attack or angina) during follow-up. People who experienced the most negative aspects in their close relationship were 1.34 times more likely to experience a heart disease event than people who had the least negative aspects.

However, once they made further adjustments to these analyses for behavioural aspects, such as smoking, alcohol consumption, and exercise, and psychological factors such as depression, this increase in risk became a little smaller.

How much emotional or practical support a person had, made no difference to the risk of heart disease.

What interpretations did the researchers draw from these results?

The researchers concluded that “adverse close relationships may increase the risk of heart disease”.

What does the NHS Knowledge Service make of this study?

This was large and well-designed study. There are a few points to consider:

  • The “negative aspects” of relationships were reported by the individual only, and how they reported these negative aspects may have been affected by their personality. So if two people experienced similar events in their relationships, they might rate the negative aspects differently. This would suggest that it may be a person’s perception of events and their ability to cope with them, rather than the events themselves that may have an effect.
  • Relationships were assessed at one time point only, it is possible that the nature of the relationships might change over the years of follow up, or that the relationship assessed might end and another replace it. Similarly, potential confounding factors, such as smoking, were also only assessed at one time point. Changes in these factors over time may affect the results.
  • As with all studies of this type, although there appears to be an association between negative relationships and heart disease, we cannot say for certain that having more negative relationships causes heart disease, as there may be confounding factors playing a part in this association. The authors adjusted for known confounding factors, including age, gender, obesity, diabetes, high blood pressure, and high cholesterol, which does increase our confidence that we may be seeing a true association. A replication of this finding in another large study would further increase our confidence in this association.
  • These results were obtained in a very specific population: civil servants in office jobs. Therefore the results might not apply to the population as a whole.

However, these results do seem reliable, and highlights an area for future research. We cannot as yet say whether interventions aimed at improving close relationships would reduce risk of heart disease.

Sir Muir Gray adds...

I find it helpful to distinguish between strain - the external pressure, and stress - the internal response; although the "stress" word is often used to mean both. In general, physical strain is good for you and psychological or social stress is not so good.

This evidence supports the belief that bad relationships increase the risk of certain diseases, and that reducing the risk of disease has to be set in the context of the individual's life and social environment and not just seen as a medical factor which needs treatment by a doctor.

Analysis by Bazian

Edited by NHS Choices

Links to the headlines

Happy relationships good for the heart. The Daily Telegraph, October 9 2007

How a rocky relationship can give your heart a bad beating. Daily Mail, October 9 2007

Relationships may (literally) break your heart. Daily Express, October 9 2007

Links to the science

De Vogli R, Chandola T, Marmot MG. Negative Aspects of Close Relationships and Heart Disease. Arch Intern Med. 2007; 167:1951-1957


How helpful is this page?

Average rating

Based on 0 ratings

All ratings

0  ratings
0  ratings
0  ratings
0  ratings
0  ratings

Add your rating