Influenza Vaccination of Healthcare Workers Is an Important Approach for Reducing Transmission of Influenza from Staff to Vulnerable Patients

Cluster randomized trials of staff vaccination in these settings have consistently shown benefit to residents (2–5) and this evidence has been used to support influenza vaccination of healthcare workers more generally. Since part of the rationale for vaccinating healthcare workers is patient protection some employers have chosen to make staff vaccination a condition of employment, which has led to legal challenge. [...]De Serres et al assert that the cluster-randomized trials violate the principle of dilution whereby the greatest reductions should be observed in the most specific outcomes (e.g. greater relative reductions would be expected for laboratory confirmed influenza than for all cause mortality). Since the studies were not adequately powered to assess which outcomes had the greatest reductions and the confidence intervals for all outcomes overlap we do not agree that the studies individually or collectively violate this principle. [...]as discussed above, the statistical uncertainty around estimates of effect mean that whilst we can be confident that there was a highly significant reduction in all cause mortality we do not know whether this represents a greater reduction than that in influenza like illness deaths. [...]De Serres et al’s main criticism is that if the numbers needed to vaccinate (NNV) to prevent one death in our study were extrapolated to all LTCF staff in the US the number of deaths averted would be considerably greater than the annual number of deaths estimated to be due to influenza in the US.