BACKGROUND: Since the licensure of varicella vaccine in 1995 and implementation of a routine 2-dose vaccination program in 2006, varicella morbidity and mortality has decreased substantially in the United States. Case-based varicella reporting was implemented in Georgia in July 2011, and since then, varicella clusters continue to be reported among household, childcare, school, and prison settings. The objective of this analysis was to describe varicella cluster epidemiology and inform prevention and control measures by comparing cluster and non-cluster (sporadic) varicella cases occurring in Georgia during 2012-2016.
METHODS: Clinical characteristics, demographics, and vaccination history of varicella cases occurring from January 1, 2012 through November 30, 2016 were captured via case-patient and physician interviews. A cluster was defined as ≥ 3 epidemiologically-linked varicella cases occurring in a 21-day period. Cluster and non-cluster cases were compared using two-sample t-tests and Chi-square testing at a 5% significance level.
RESULTS: Of 715 reported varicella cases, 553 (77%) were non-cluster, 110 (15%) were cluster-associated, and 52 (7%) had unknown cluster status. Twenty-five varicella clusters were reported; four (16%) in schools or daycares, four (16%) in prisons or detention centers, 16 (64%) in households, and one (4%) at a university. Clusters ranged from 3 to 11 cases (mean: 4.20). The mean age among cluster cases (13.20 years) was significantly higher (p<0.001) than non-cluster cases (9.34 years). The proportion of cluster cases reported as White was significantly lower (p <0.04) than among non-cluster cases. The proportion of cases reported to have Hispanic ethnicity was not significantly different based on cluster status. Cluster cases were significantly less likely to have a history of varicella vaccination; the odds of vaccination among cluster cases were 0.39 [95% confidence interval (CI): (0.25, 0.60)] times those of non-cluster cases. Additionally, the odds of having a moderate or severe rash (250-500 lesions or >500 lesions) among cluster cases were 2.75 [95%CI: (1.52, 4.97)] times that of a non-cluster case.
CONCLUSIONS: In Georgia, varicella epidemiology differed between cluster-associated and sporadic cases. Clusters were more likely to involve unvaccinated individuals. Additionally, cluster-associated cases were more likely to be older and more severe than non-cluster cases. More data is needed to fully explain these differences, including the role of transmission setting on varicella epidemiology. Case-based reporting remains an important strategy for identifying varicella clusters and informing prevention and control measures.