"'Plate-throwing rage' raises heart attack risk nearly 10 fold," The Daily Telegraph reports, slightly inaccurately.
This headline reports on a study that found that just seven out of 313 people had felt "very angry" in the two hours before a heart attack – compared to their normal levels of anger. Despite the headline, none of the participants had felt furious or angry to the point of throwing plates or any other objects.
The participants in the study had all been admitted to a cardiac unit following a heart attack. They completed questionnaires to estimate their levels of anger in the 48 hours before the heart attack and their usual levels in the preceding year.
Anger was assessed on a seven-point scale, and seven people reported being "very angry, body tense, maybe fists clenched, ready to burst" (a score of five) in the two hours before the heart attack. This level of anger was associated with 8.5 times the risk of a heart attack in the next two hours than other times.
This type of study alone cannot prove that anger caused the heart attack. And as so few people reported feeling angry before the heart attack, the results are not precise.
The study also relies on accurate recall not only of the period before the heart attack, but also of the usual levels of anger that were reached during the preceding year. People may be more likely to remember anger associated with a dramatic event like a heart attack than anger at other times, and this would affect results.
Where did the story come from?
The study was carried out by researchers from the Royal North Shore Hospital in Sydney and the University of Sydney. There was no external funding.
The media headlines exaggerated the findings of this study, with the Mail Online inaccurately reporting that anger "can cause" a heart attack.
This type of study cannot prove cause and effect. The Daily Mirror failed to explain that the results were only based on seven people.
What kind of research was this?
This was a case crossover study looking at whether episodes of anger are associated with an increased likelihood of heart attack shortly after they occur.
This type of study is uncommon, and aims to assess whether a brief condition or set of circumstances has a temporary impact on risk of an outcome shortly after they occur. It is similar to a case control study, but each case acts as its own control. Case crossover studies then look at how likely the outcome was to occur just after the circumstances occur, compared with any other time.
What did the research involve?
The researchers recruited people admitted to a cardiac unit with suspected heart attack. The participants reported on how angry they were over the previous 48 hours, and also their usual levels of anger. The researchers then looked at whether people were more likely to have had their heart attack in the four hours after an angry episode than at other times in the year.
They also looked at the likelihood of having high levels of anger or anxiety just before the heart attack compared to at other points in the year.
All people admitted to a single cardiac unit (in Sydney, Australia) with suspected heart attack between 2006 and 2012 were eligible for the study. The final 313 participants were those who had evidence of a blockage in an artery supplying the heart (coronary artery) found during angiography (a procedure that looks at the blood flow to the heart).
These participants completed a detailed questionnaire within four days of admission, which included questions about their activities in the 48 hours before the heart attack. They were also asked to rate their level of anger in this period, describe any events that caused it, and estimate how frequently they experienced each level of anger per year, using the following seven-point scale:
- busy, but not hassled
- mildly angry, irritated and hassled, but it does not show
- moderately angry, so hassled it shows in your voice
- very angry, body tense, maybe fists clenched, ready to burst
- furious, forced to show it physically, almost out of control
- enraged, out of control, throwing objects, hurting yourself or others
They also filled out another standard questionnaire about their level of anger and anxiety. They analysed the results comparing the likelihood and level of anger two hours before the heart attack and two to four hours before with their usual yearly estimated likelihood and levels of anger. They also compared the likelihood and level of anger and anxiety in the preceding two hours with the level 24 to 26 hours before.
What were the basic results?
The average age of the participants was 58 and most of them were male (85%). No one reported anger levels of above five (“very angry, body tense, maybe fists clenched, ready to burst”) – so despite the headlines, no-one threw anything.
In the two hours before heart attack:
- seven people reported an anger level of five
- the risk of experiencing a heart attack within two hours of an anger level of five or more was raised by 8.5 times compared with other times (relative risk (RR) 8.5, 95% confidence interval (CI) 4.1 to 176)
- two people had an anger level of four, and this was not associated with a significant increase in risk of heart attack in the next two hours (RR 1.3, 95% CI 0.3 to 5.1)
In the two to four hours before the heart attack:
- one person reported an anger level of five or more, but this was not significantly associated with risk of heart attack
- three people had an anger level of four, which was also not associated with an increased risk
When looking at the levels of anxiety in the two hours before heart attack compared to the level at the same time the previous day, people whose anxiety was in the top 75% of levels (75th percentile) had an increased relative risk of heart attack (RR 2.0, 95% CI 1.1 to 3.8) and this increased for those in the top 90% of levels (RR 9.5, 95% CI 2.2 to 40.8).
How did the researchers interpret the results?
The researchers concluded that episodes of intense anger, defined as being "very angry, body tense, clenching fists or teeth", are associated with increased risk of a heart attack within two hours. Anxiety was also associated with increased risk of heart attack in the next four hours.
This study found that an anger level of five of more (according to their scale) was associated with 8.5 times the risk of a heart attack than at other times. However, there are a number of limitations when considering this result.
Firstly, very few people reported feeling angry shortly before the heart attack – just seven people out of the 313 participants. Therefore the confidence interval for the main result is wide, meaning the results are not particularly precise, and we can’t be certain of the size of the association with risk.
Secondly, the study relies on accurate recall, not only of the period before the heart attack, but also of the usual levels of anger that were reached during the preceding year. As well as the potential for misremembering, it is open to what is called "recall bias". This is where someone is more likely to remember the anger they experienced just before their heart attack if they think it may have contributed to it, than anger at other times in the year.
In conclusion, the study by itself does not prove that increased levels of anger or anxiety directly cause heart attacks. However, a recent systematic review (that we have not assessed) suggests that similar studies also support an increase in risk shortly after outbursts of anger.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 24 February 2015
Daily Mirror, 24 February 2015
Daily Mail, 24 February 2015
Daily Express, 24 February 2015
Links to the science
Acute Cardiovascular Care. Published February 23 2015
European Heart Journal. Published March 4 2014