Use of a computerized database to study the effectiveness of an attenuated varicella vaccine

Background. The varicella vaccine Varilrix (GlaxoSmithKline) was introduced in Israel in June 2000 as an optional vaccination for children. Methods. We used the database of a single health maintenance organization that serves 25% of the population in Israel to assess the effectiveness of the vaccine retrospectively. Incidence and complications of varicella were derived from the database from January 1, 1998 until December 31, 2002. Results. Since licensure >30 000 individuals younger than 10 years in this health maintenance organization have been immunized with the vaccine. Annual incidence of disease per 1000 in the study population was 86.6 in 1998, 74.6 in 1999, 74. 0 in 2000, 37.1 in 2001 and 44.6 in 2002. This declining trend in incidence of disease was statistically significant. Complications of varicella occurred in ∼1% of patients throughout the 5-year study period, but there was a parallel decline in the number of patients with complications corresponding to the decline in disease incidence. Vaccine effectiveness for prevention of clinical disease in this population was 92% (95% confidence interval, 91.0 to 92.7). There were varying rates of utilization within communities of varied socioeconomic class, so that in the higher socioeconomic class there was an increased utilization and a corresponding decrease of attack rate; whereas in communities Where there were lower utilization rates, corresponding increased numbers of varicella cases were seen. Conclusion. This database enables long term follow-up of the effectiveness of this vaccine in a large population.