Invited commentary: Immunization and Bell's palsy in children.

Based on case reports, it has been speculated that certain vaccines may increase the risk of Bell's palsy. In a study using a novel approach (case-centered analysis) among children aged 18 years or less, results of which appear in this issue of the Journal, Rowhani-Rahbar et al. (Am J Epidemiol. 2012;175(9):878-885) found no association between immunization with trivalent influenza vaccine (TIV), hepatitis B virus (HBV) vaccine, or any vaccine (all vaccines combined) and Bell's palsy (definite and probable cases combined). In this commentary, the author evaluates the validity of the case-centered approach. In the study by Rowhani-Rahbar et al., the study population was not representative of the population of children at risk. The analytical approach resulted in the loss of information for 72% of the cases. There were insufficient data to assess TIV and HBV. Only the risk for all vaccines could be assessed, and only if it was assumed that the findings could be generalized. Possible effect modification by antecedent vaccination could not be assessed. The author concludes that a nested case-control study would have been more statistically robust and more informative.