More unscientific claims made for the ‘benefits’ of flu vaccination
Posted Nov 28 2010 7:39pm
No sooner had I written about the fact that the science shows flu vaccination to be a largely pointless exercise, that I came across this on-line piece in the Los Angeles Times telling readers that flu vaccination “help[s] prevent babies from getting sick.” The article goes on to state that the flu vaccination “was effective in preventing influenza in children as young as 9 months old.” You can read a summary of this study (published in the journal Lancet Infectious Diseases) here .
If one wants to gauge the effectiveness of vaccination or say some drug here’s how one should do it: by conducting a randomised, placebo-controlled trial. In the case of flu vaccination, individuals should be randomised to received real flu vaccination or placebo shot, and then the outcomes assessed over a period of time. At the end of such a study, assuming you’ve adhered to good scientific practices, you should have a decent idea whether flu vaccines are effective or not.
The way the Los Angeles Times has reported this study you’d be forgiven for thinking that it was randomised and placebo-controlled in nature. In actuality, it was nothing of the sort.
What the investigators involved in this study was assess the outcomes of young children who developed symptoms of the flu . The investigators asked parents if their children had been vaccinated. Then they looked to see if there was any association between vaccination status and outcome.
Notice I used the word ‘association’ in the last sentence. And that’s because the study was epidemiological in nature, which means all it could ever really do was look at associations. It cannot be used to discern whether flu vaccination in young children improves health outcomes or not. For that, as I stated above, we would require randomised, placebo-controlled studies.
The reason that this has particular relevance is that studies such as the on in Lancet Infectious Diseases is subject to what is known as the ‘healthy user’ effect. It can be that individuals who are health-conscious are more likely to take up vaccination (and more health-conscious parents of children may do the same). What this means is that, all things being equal, users of vaccination (or other treatments) are more likely to be more health-conscious and healthier than non-users. So, any apparent benefit from vaccination may not be due to the vaccine per se, but due to healthier characteristics in those who choose vaccination.
If this sounds a bit far-fetched, consider the findings of some research that concerned flu vaccination in the elderly . I wrote about this research a couple of years ago here . In the study in question, researchers assessed death rates in about 350 vaccine recipients and 350 non-vaccinated individuals (controls) who had been hospitalised with pneumonia. The study was done outside the flu season. Now, with no flu around to be protected from, any difference in mortality would be very unlikely to be due to flu vaccination.
What the researchers found was that those who had had the flu shot were about half as likely to die compared to the non-vaccinated. The authors of this study suggest that the apparent mortality benefit in the elderly (as observed in epidemiological studies) is probably not due to the vaccine per se, but a healthy user effect.
Getting back to the Lancet Infectious Diseases study, you may notice that its authors conclude that their “…findings suggest that influenza vaccine recommendations should be reassessed in most countries.” That’s fine by me, but would it be possible to have some proper science first, please?
1. Heinonen S, et al. Effectiveness of inactivated influenza vaccine in children aged 9 months to 3 years: an observational cohort study. The Lancet Infectious Diseases, Early Online Publication 23 November 2010
2. Dean T., et al. Mortality Reduction with Influenza Vaccine in Patients with Pneumonia Outside “Flu” Season – Pleiotropic Benefits or Residual Confounding? American Journal of Respiratory and Critical Care Medicine 2008;178:527-533