Posts Tagged ‘ Myocardial infarction ’

A Side Benefit of the ‘Flu Vaccine – Reduction in Heart Attacks

ResearchBlogging.orgIt’s a bold claim, that being vaccinated for Influenza will protect you against having a heart attack or Myocardial Infarction (MI). Well according to a study published last month in the Canadian Medical Association Journal that could well be the case.

The study, “Influenza vaccination, pneumococcal vaccination and risk
of acute myocardial infarction: matched case–control study
“, included 78,706 individuals over the age of 40 in an attempt to determine whether and how much of an effect the vaccine had on MI. The results showed that the vaccine was associated with an almost 20% reduction in risk of MI compared to the unvaccinated population.

My first thought when I read this was that those individuals who get vaccinated might engage in other activities that would lend themselves to reducing risk of MI, healthy eating habits, regular exercise etc. The so called “healthy user” effect. This study attempted to control for this sort of confounding factor by using matched controls with similar risk factors. In particular they performed two further analyses that I think comfortably undermine this interpretation of the results.

First they compared the timing of the receipt of the vaccine, ie early or late in the ‘flu season. Second they looked at subjects who had been vaccinated with the pneumococcal vaccine as a comparison. In the first case there was a greater reduction in risk for those who had received the vaccine early in the season, 21% compared to only 12%. In the second case they found no protective effect for those who had received the pneumococcal vaccine. It seems unlikely that the subjects of the study would vary their healthy habits in precisely the ways they would have to in order to see these results as being independent of the vaccine itself.

An interesting aspect of this kind of epidemiology though is that simply looking at the raw numbers there is a greater incidence of MI in the vaccinated group compared to the unvaccinated group. This is an artefact of how the vaccines are administered clinically. Those patients who have greater cardiovascular risk are also the patients who are more likely to be recommended for vaccination. It’s like noticing that people who buy antiperspirant tend to have sweatier armpits than those who don’t*. Those who are prone to sweaty armpits will likely tend to be those who will buy and use antiperspirant, to accurately gauge effectiveness you would have to control for this factor.

Tying this into the anti-vax focus of this week, one of the claims I’ve seen is that the flu vaccine is useless as it is based on strains that were around the previous season (Here via IAS). This is distressingly simplistic thinking. This paper shows that the vaccine is far from useless, in fact the most benefit was found within ‘flu seasons. If there was no effect of the vaccine on the current ‘flu season strains then the study would not have shown the protective effect that it did. Once again the anti-vax crowd (and by extension the altmed crowd as there tends to be overlap) has shown their inability to grasp the nuance of the situation.

There is a gradation of effect when a vaccine does not exactly match the wild strain, the interaction of antigens and antibodies is more complicated than you might suppose. Changes in the antigen (mutation of the virus over the ‘flu season) mean that antibodies may bind less tightly and therefore have a reduced effect but that is not the same as no effect.

To sum up, health is a complicated thing. Many factors are inter-related and in order to get the whole picture we sometimes have look at things from an unusual perspective. Vaccinating for the ‘flu can reduce your risk of heart attack, who knew? But, we should also be wary that we are keeping things within the realm of plausibility. Infection does cause inflammation which can plausibly effect heart function**. This does not mean it’s valid to blame the Wi-fi at your local school for your vague aches and pains.***


*To make up an example.

** See



Siriwardena AN, Gwini SM, & Coupland CA (2010). Influenza vaccination, pneumococcal vaccination and risk of acute myocardial infarction: matched case-control study. CMAJ : Canadian Medical Association journal = journal de l’Association medicale canadienne, 182 (15), 1617-23 PMID: 20855479

Enhanced by Zemanta

Smoking Bans and the Effect of Health Warnings

In the world today there is an increasing focus on the negative aspects of smoking and a concerted attempt to reduce the presence of smoking in society. Given the harmful effects of this addiction on not only the active smoker but those around them this seems like a prudent move. Two of the approaches with the goal of minimising public exposure to cigarette smoke are the banning of smoking in businesses and public places and the addition of more strenuous warning labels on the cigarettes themselves.

Both of these tactics have been used in New Zealand with varying levels of acceptance (and success). Smoking bans draw the criticism that individual freedoms are being curtailed. This may be a legitimate point but conceptually it is no different than government enforcement of wearing seatbelts while driving on public roads. The aim is to reduce the risk of harm to the public. The real question in each case is whether the intervention is effective in it’s goals.

Addressing this question two studies last year looked at each of these methods, the first I will look at is a meta-analysis (with the concomitant problems those have, that’s another story) of the effect of smoking bans on the hospital admissions of acute myocardial infarction (that’s a heart attack to you and me). The analysis found that smoking bans were associated with an average reduction of heart attacks by 17%.

For each year a ban was in place it was accompanied by a reduction of the incidence rate ratio (the number of new cases per unit of population eg 10 cases per 100,000 people) of 26%. This indicates that the longer a ban is in force the fewer people who will be affected by heart attacks. Looks like an effective strategy to me, 17% is nothing to be sneezed at when it is individual lives you are considering. Depending on individual risk factors the chance of death in the 30 days after a heart attack can be up to 16%.

An editorial discussing these findings in more depth (in the Journal of the American College of Cardiology, the journal this study was published in) can be found Here and is a good read.

The second study focused on the how well explicit (i.e. emphasising death) cigarette pack warnings encouraged smokers to quit. Specifically it looked at smokers for whom the act of smoking formed part of the basis for their self-esteem. Subjects undertook a questionnaire that evaluated whether smoking was tied to their self esteem using statements like ‘‘Smoking allows me to feel valued by others,” and ‘‘Smoking allows me to feel worthy.” (as well as negative versions). The subjects rated how much they agreed with the statements and this was used to determine the smoking-based self esteem for each subject.

Participants were then shown pictures of cigarette packs that either had mortality related warnings (e.g. ‘‘Smoking leads to deadly lung cancer.”) or more moral or self esteem related warnings (e.g. ‘‘Smoking brings you and the people around you severe damage” and ‘‘Smoking makes you unattractive”). After a delay to allow the warnings to be filtered out of conscious awareness the subjects were asked a further series of questions to assess the effect of the warnings (e.g. ‘‘Do you intend to smoke more or less in the future?” ‘‘Do you intend to quit smoking in the future?”).

Subjects for whom smoking formed part of the basis for their self esteem actually increased their likelihood of smoking in response to warnings emphasising mortality. For these people it was the self image warnings that were most effective. Unfortunately is seems that the opposite is true for individuals that do not consider smoking to be an important factor of their self esteem so a one size fits all approach would probably not be effective. The study authors suggest that specific populations could have warnings tailored to be most effective depending on the relevance smoking has to the group identity (e.g. “young smokers who want to impress their peers.”).

This result may be applicable to other areas where minimising harm is the goal, such as drink driving campaigns.

In summary, despite any reservations regarding the form that inducements to stop smoking take it seems that the benefits are indeed worth the attempt. Also, as I often point out, the real world is more nuanced and complicated than we would generally like it to be, more effort may be required to identify sub-groups that respond most to different strategies but this also looks to be worth trying.

Enhanced by Zemanta

Get every new post delivered to your Inbox.

Join 73 other followers