Methods: We used administrative health claims from a 5% sample of Medicare beneficiaries ≥65 years. We defined incident cases as HZ in the first diagnostic position, with no HZ code in the previous year, among beneficiaries enrolled in Part A and B, from 1992-2015. We calculated incidence of first HZ episode by dividing the number of cases by the total number of person-years (p-y). A case was a censoring event. We used Poisson regression to compare HZ incidence by race before and after 2007. We calculated vaccine coverage by dividing the total number of persons with ZVL in Medicare Part D by the number of eligible enrollees.
Results: We identified 266,745 first HZ episodes. Prior to 2007, HZ incidence increased among both blacks and whites. Although incidence was double in whites vs. blacks (10.3 vs 5.0 cases/1000 p-y), the rate of increase was similar (p=0.75). From introduction of ZVL to 2015, HZ incidence decreased 1.8%/year in whites and did not change significantly in blacks (p<0.001) (Figure). By 2015, ZVL uptake in Medicare among blacks was less than half that among whites (7.3% vs.19.9%).
Conclusion: Incidence of HZ increased at a similar rate for black and white Medicare beneficiaries in the pre-vaccine era. In the post-vaccine era, incidence has decreased among whites and plateaued among blacks. Though the reasons for racial difference and temporal changes in HZ incidence remain unknown, ecologic association suggests higher vaccination rates may contribute to steeper HZ declines among whites in the post vaccine era.
A. Guo, None
S. McKay, None
R. Harpaz, None