We thank Dr Harkavy for his letter. We agree with Dr Harkavy that the absolute risk of contracting varicella is low. However, we must also acknowledge that the author's attack rate calculations are not appropriate to our data. Attack rates are primarily used to quantify the risk of infection during an outbreak. In contrast, we conducted a population-based observational study. Using well-accepted epidemiologic methods, we showed that children were at a greatly increased risk of medically attended varicella illness compared with vaccinated children, and 5% of all varicella cases in the population were attributed to the parental decision to decline varicella immunizations. Moreover, we do not agree with the author that risk calculations should include children with a history of varicella infection. Healthy children with a history of infection are not at risk of varicella disease and do not receive the vaccine. Including these children could lead to an underestimation of the association between declining immunizations and the risk of varicella infection.
Glanz J, Hambidge SJ. Varicella Vaccine Refusal May Not Be Bad—Reply. Arch Pediatr Adolesc Med. 2010;164(8):780-781. doi:10.1001/archpediatrics.2010.121.1