Old Thoughts On Vaccination

Saturday, September 29th 2007
• •

Should We Be Reserving Flu Shots For The Elderly During Shortages?

A review of the studies which have shaped the policy, questions the all-cause mortality benefit of vaccinating everyone over 65 years old.

Influenza vaccination may save many fewer older patients’ lives than generally claimed, according to researchers here.

The reason is that estimates of a 50% or greater reduction in all-cause mortality have emerged from cohort studies fraught with selection bias, asserted a review article in the October issue of The Lancet Infectious Diseases.

The “illusory” estimates arose primarily from methodologically weak cohort studies, the GWU researchers said.

These studies used nonspecific endpoints, typically all-cause mortality and non-laboratory-confirmed influenza outcomes, while attempting to adjust for selection bias in multivariate models with health-status covariates defined by diagnostic codes.

But, in one study, adjustment for diagnostic codes indicating severe illness and frailty was found to increase the mortality difference between vaccinated and unvaccinated groups even before the flu season. This suggested that the method left uncontrolled bias.

Indeed, two studies revealed that most influenza-related deaths occurred in small subsets of older adults with low vaccine coverage who were hospitalized in autumn.

Without cohort studies, “the remaining evidence is not sufficient to show that vaccination substantially reduces the risk of influenza-related mortality among elderly people,” they wrote.

Perhaps the most damning evidence,

Age-adjusted estimates for influenza-related mortality in excess mortality studies showed no reduction in flu-related deaths during a period when vaccine coverage increased by 50%. Nor was there any increase in mortality during the 1997-1998 flu season when the vaccine completely mismatched circulating strains.

When it comes to public health and vaccinations the health community has already thrown out the premiere-ness (and that is a word) of non-maleficence in favor of utilitarian measurements of benefit. The point is even if the benefit of vaccinating all of the elderly is less than expected it is almost certainly still greater, I imagine, than the complications or bad reactions that come with giving the vaccination. This review, even if confirmed by better designed studies in the future, probably contributes little to the public health policy concerning who gets vaccinations.

And it isn’t like there are other subsets of the population who could benefit more from the vaccine during periods of shortage. As long as it is the providers and the markets doing the rationing (and not the government) I have no problem with it, even if the evidence (as above) doesn’t support the rationing as strongly as we might have once believed.