The NHS has been told to “give over-50s a winter flu jab to cut deaths from heart attack”, the Daily Express reports.
The study included almost 300 people who had had a heart attack, and 300 who hadn’t. It looked at whether they’d had the flu and whether they'd had a flu jab that year.
The results were puzzling. The study found no link between having the flu and having a heart attack, but it did find an association between having the annual flu jab and reduced chances of having a heart attack – around 45% lower odds.
This research doesn’t prove that the flu jab reduces the risk of having a heart attack. It can only establish a link between the two. Further research is necessary to establish whether this link is causal and, if so, exactly how the flu vaccine reduces heart attack risk.
Additionally, the current study only included people who had survived a heart attack, so cannot tell us whether there is any link between the flu jab and fatal heart attacks.
There is not enough evidence about the link between the flu jab and heart attack risk to change the guidelines about who is advised to have a flu vaccine.
Where did the story come from?
The study was carried out by researchers from the University of New South Wales and other organisations in Australia and the United States. The research was funded by GlaxoSmithKline, a pharmaceutical company that produces a flu vaccine, and the Australian National Health and Medical Research Council.
The research was covered widely in the UK and international media. However, no media outlet reported that the study can’t prove that the flu vaccination prevents heart attacks. Nor did any outline the need for further studies to establish whether the vaccine protects against heart attacks.
Newspaper stories reporting that the flu jab could cut deaths due to heart attack are based on speculation. The study did not look at deaths and, because of the way patients were selected, it specifically excluded individuals who had suffered a fatal heart attack.
The Daily Mail implied that the study only involved men. This was not the case (although men did make up the majority of the case group) and male gender was actually associated with an increased chance of heart attack.
What kind of research was this?
This was a case-control study that investigated whether having the flu was associated with an increased risk of heart attack. “Cases” were selected based on hospital admissions for acute myocardial infarction (AMI), the medical term for heart attack. Controls were similarly aged individuals attending outpatient clinics.
Case-control studies are a useful way of investigating the relationship between two or more risk factors and a disease, but they can’t prove that one factor causes the other.
What did the research involve?
The researchers recruited two groups of patients during three winter flu seasons. The first group (the cases) was composed of patients over the age of 40 who were admitted to hospital after suffering a heart attack.
People of the same age who were attending outpatient orthopaedic or ophthalmic clinics during the same winter flu seasons were recruited into the study as controls. Individuals with a history of heart attack and stroke were excluded from the control group.
The primary aim of the study was to determine whether there was an association between having a heart attack and having an underlying influenza infection. The researchers collected nose and throat swabs as well as blood samples to determine if the patient had had the flu that season. Patients who were unable to provide these samples within 72 hours of hospital or clinic attendance and again four to six weeks later were excluded from the study. Importantly, this meant that patients suffering a fatal heart attack were not included in the research.
Additional information was collected and included in the analysis in order to adjust for potential confounding variables. These included:
- smoking status
- alcohol consumption
- flu vaccination status (verified using GP records)
- several other cardiovascular risk factors (high blood pressure, high cholesterol and diabetes)
Standard statistical techniques were used to estimate the association between heart attack and the flu, after adjusting for the variables described above. This analysis also allowed researchers to determine the association between these variables and having a heart attack.
A secondary outcome of interest was the effectiveness of the flu vaccine for reducing the risk of heart attack or flu.
What were the basic results?
There were 826 patients eligible for the study. Of these, 559 (67.7%) agreed to participate and were included in the analysis. Overall, 275 (49.2%) of these participants were cases who had been admitted to hospital after having a heart attack, and 285 (50.8%) were controls attending outpatient clinics.
In all, 276 (49.4%) of the patients were confirmed to have received the seasonal flu jab the year of their hospital or clinic attendance.
There were several significant differences between the groups at the beginning of the study. Compared with controls, patients in the cases group were:
- more likely to be male (78.5% of cases, 45.8% of controls)
- more likely to be under the age of 65 years (64.0% of cases, 25.4% of controls)
- more likely to be married (72.6% of cases, 57.5% of controls)
- more likely to be a current smoker (27.9% cases, 11.2% of controls)
- less likely to live alone (20.1% of cases, 32.6% of controls)
- less likely to report never consuming alcohol (38.4% of cases, 51.8% of controls)
- less likely to report having no chronic diseases (5.5% of cases, 12.0% of controls)
The swabs and blood samples returned evidence of flu infection for 34 (12.4%) of the heart attack patients, compared with 19 (6.7%) in the control group. In unadjusted analysis, heart attack patients were significantly more likely to have had the flu compared with control patients (odds ratio (OR) 1.97, 95% confidence interval (CI) 1.09 to 3.54).
However, when the researchers included other variables in the analysis, the association between flu and heart attack became non-significant (OR 1.07, 95% CI 0.53 to 2.19). This suggests that the previously observed association may have been due to other factors.
Of the confounding variables included in the logistic regression analysis, several were significantly associated with the odds of being admitted to hospital for a heart attack:
- being a man was associated with nearly four-fold increased odds of heart attack (OR 3.83, 95% CI 2.54 to 5.78)
- self-reported high cholesterol was associated with doubled odds of having a heart attack (OR 2.00, 95% CI 1.35 to 2.97)
- being a current smoker was associated with more than doubled odds (OR 2.11, 95% CI 1.25 to 3.56)
Having had the current year’s seasonal flu jab was associated with 45% decreased odds of being admitted to hospital for a heart attack (OR 0.55, 95% CI 0.35 to 0.85).
How did the researchers interpret the results?
The researchers concluded that recently having the flu was not associated with a subsequent heart attack, but that being vaccinated against the flu was protective against heart attack. They recommend further studies looking at the potential protective benefit of flu vaccination for individuals between the ages of 50 and 64 who are at increased risk of having a heart attack, but below the commonly recommended flu jab age of 65 years or over.
This study suggests that there is an association between being vaccinated against the flu and decreased odds of having (and surviving) a heart attack.
The researchers conclude that receiving the flu jab may be protective against heart attack, and report that previous studies have shown a link between flu infection, flu vaccination and heart attack. They further suggest that future policy decisions regarding the age at which the seasonal flu jab is offered take into account its potential effect on cardiovascular events. They say “even a small effect of influenza vaccination in preventing AMI [acute myocardial infarction, or heart attack] may have significant population health gains”.
A key limitation of case-control studies is their vulnerability to selection bias. In the current study, the cases were selected based on their admittance to hospital for a heart attack, and their ability to provide samples 2.5 days as well as 1 to 1.5 months after the cardiac event. This meant that individuals who had a fatal heart attack were excluded. It is unclear based on this study design whether including individuals who suffered and died of a heart attack would alter the observed association between flu vaccination and heart attack.
It’s also problematic that there was a lack of control participants who matched the cases in terms of age. This was because uptake of flu vaccination is higher among those aged 65 and older (in Australia, where the study took place, people aged 65 and over are offered free annual flu jabs). Control group participants were more likely to be over the age of 65. Future studies could either be randomised or match participants on age to account for this confounding variable.
Despite efforts to adjust for key confounding variables, the observed associations may be due to other variables not considered in the analysis. A randomised controlled trial would be required to address this.
There is also the possibility that the results may not be reliable because of small sample sizes. For example, the study’s main objective was to look at whether having had flu was linked with risk of heart attack, yet only 34 cases and 19 controls had evidence of flu. Examining associations where only a small number of people have had the exposure of interest reduces the reliability of risk associations.
Given the limitations in study design and the potential for selection bias, conclusions about the protective effect of the flu jab should be treated with caution. Additional studies (prospective cohort studies or randomised controlled trials) should be conducted to establish whether there is a direct causal link between vaccination and preventing heart attacks, and whether this holds across age groups and severity of cardiac event.
Even if the flu vaccine is found to influence risk of heart attack, it is likely to have far less influence than established risk factors for heart attack such as smoking, high cholesterol, diabetes and high blood pressure.
Read more about reducing your heart attack risk
Analysis by Bazian
Edited by NHS Website
Links to the headlines
Daily Express, 22 August 2013
Daily Mail, 22 August 2013
The Daily Telegraph, 22 August 2013
Metro, 21 August 2013
The Times, 22 August 2013
Links to the science
Heart. Published online August 21 2013