Back in the late 1970’s this author was assembling a large database on patients with Guillain-Barre Sydrome (GBS) treated at the Massachusetts General Hospital (Ropper 1991). Because a small percentage of cases of GBS follow a variety of immunizations this author declined to get vaccinated against the “swine flu” which was touted at the time to be a major healthcare threat. (The controversy over whether swine flu vaccine caused GBS swirled for several years. At most there may have been a very small increase in GBS attributed to swine flu vaccine over that associated with traditional flu vaccine.) But it certainly raised the issue of vaccine efficacy vs. untoward consequences.
Similarly, when this author turned 50 years old and was offered the influenza vaccine, I looked to the literature. There really was no good evidence of a net benefit at a personal level to influenza vaccination of a healthy 50 year-old without chronic diseases.
But in recent years, there has been a broadening of the population for which influenza vaccination is recommended. Basically, it is now recommended for everyone above the age of 6 months who does not have a contraindication. Moreover, for healthcare workers there has been a real push for flu immunization in order to protect the patients whom we care for. The percentage of healthcare workers who get the flu vaccine has become a metric in value-based purchasing programs for hospitals and will become a publicly-reported statistic. In New York State the Department of Health has had a big push to increase the number of healthcare workers immunized against the flu. So the author has acquiesced in recent years and has received influenza immunization, primarily to serve as an example for other healthcare workers. We understand the concept of herd immunization.
But now a researcher from Johns Hopkins has challenged the evidence base for widespread influenza immunization (Doshi 2013). Doshi points out that most of the studies supporting influenza immunization have been retrospective observational studies, not randomized controlled trials. Hence, they may be subject to the “healthy user” bias (i.e. that people getting the flu vaccine may be, in general, more health conscious and more likely to get the vaccine). He also notes that some of the statistics purporting all-cause mortality benefits of 27-30% (one even as high as 48%) seem implausible. He states “If true, these statistics indicate that influenza vaccines can save more lives than any other single licensed medicine on the planet.” Overall, Doshi suggests that the efficacy of the influenza vaccine has probably been oversold and the potential side effects underplayed.
Shortly after Doshi’s article appeared in the British Medical Journal, another article from CDC appeared describing the magnitude of influenza-related illness and hospitalizations averted in the US by influenza vaccination from 2005 to 2011 (Kostova 2013). Applying modeling to statistics from various databases, the authors estimate that from one million to 5 million cases of influenza illnesses are averted annually and from 7700 to 40,400 hospitalizations are averted in the US annually by the influenza vaccine.
So how do we reconcile these disparate pictures of the effectiveness of the influenza vaccine? We don’t purport to know the answer. We’ll just point out that there are numerous examples in healthcare where practice have been adopted on the basis of observational studies, only to be refuted once randomized controlled trials were done. We’ve heard too many times that it would be “unethical” to do a randomized controlled trial for “a treatment we already know works”. We think randomized controlled trials should be the basis for interventions, particularly those impacting such a huge portion of the population.
In the interim, hospitals are pretty much stuck with complying with recommendations of the regulatory bodies. It is possible to get close to 100% compliance with staff immunization in hospitals. An abstract presented at the recent APIC (Association for Professionals in Infection Control and Epidemiology) meeting highlighted Loyola University Medical Center that took its rate from 65% to 99% (APIC 2013). This was achieved without significant loss of the healthcare workforce there.
Many hospitals still deal with significant staff opposition to mandatory immunization. Let’s hope that someday we can reassure them that there is a sound evidence base for the recommendation of mass immunization for influenza.
Ropper AH, Wijdicks EFM, Truax BT: Guillain-Barre Syndrome. FA Davis: Philadelphia 1991
Doshi P. Influenza: marketing vaccine by marketing disease. BMJ 2013; 346: f3037 (Published 16 May 2013)
Kostova D, Reed C, Finelli L, et al. Influenza Illness and Hospitalizations Averted by Influenza Vaccination in the United States, 2005–2011. PLoS ONE 8(6): e66312. doi:10.1371/journal.pone.0066312, published online June 19, 2013
APIC. Mandatory flu vaccination of healthcare personnel does not lead to worker exodus. APIC News Release June 7, 2013 regarding Oral Abstract #012 at the 40th Annual Conference of APIC “Four-Year Experience with Mandatory Seasonal Influenza Immunization for All Personnel in a University Medical Center”, Sunday, June 9, 2013
Print “PDF version”