RESUMEN
OBJECTIVES: To assess the effectiveness of live zoster vaccine during more than 10 years after vaccination; and to describe methods for ascertaining vaccine effectiveness in the context of waning. DESIGN: Real world cohort study using electronic health records. SETTING: Kaiser Permanente Northern California, an integrated healthcare delivery system in the US, 1 January 2007 to 31 December 2018. POPULATION: More than 1.5 million people aged 50 years and older followed for almost 9.4 million person years. MAIN OUTCOME MEASURE: Vaccine effectiveness in preventing herpes zoster, postherpetic neuralgia, herpes zoster ophthalmicus, and admission to hospital for herpes zoster was assessed. Change in vaccine effectiveness by time since vaccination was examined using Cox regression with a calendar timeline. Time varying indicators were specified for each interval of time since vaccination (30 days to less than one year, one to less than two years, etc) and adjusted for covariates. RESULTS: Of 1 505 647 people, 507 444 (34%) were vaccinated with live zoster vaccine. Among 75 135 incident herpes zoster cases, 4982 (7%) developed postherpetic neuralgia, 4439 (6%) had herpes zoster ophthalmicus, and 556 (0.7%) were admitted to hospital for herpes zoster. For each outcome, vaccine effectiveness was highest in the first year after vaccination and decreased substantially over time. Against herpes zoster, vaccine effectiveness waned from 67% (95% confidence interval 65% to 69%) in the first year to 15% (5% to 24%) after 10 years. Against postherpetic neuralgia, vaccine effectiveness waned from 83% (78% to 87%) to 41% (17% to 59%) after 10 years. Against herpes zoster ophthalmicus, vaccine effectiveness waned from 71% (63% to 76%) to 29% (18% to 39%) during five to less than eight years. Against admission to hospital for herpes zoster, vaccine effectiveness waned from 90% (67% to 97%) to 53% (25% to 70%) during five to less than eight years. Across all follow-up time, overall vaccine effectiveness was 46% (45% to 47%) against herpes zoster, 62% (59% to 65%) against postherpetic neuralgia, 45% (40% to 49%) against herpes zoster ophthalmicus, and 66% (55% to 74%) against admission to hospital for herpes zoster. CONCLUSIONS: Live zoster vaccine was effective initially. Vaccine effectiveness waned substantially yet some protection remained 10 years after vaccination. After 10 years, protection was low against herpes zoster but higher against postherpetic neuralgia. TRIAL REGISTRATION: ClinicalTrials.gov number NCT01600079; EU PAS register number EUPAS17502.