22. An Innovative Approach to Examining the Waning of Vaccine Effectiveness Using Automated Healthcare Data. (4th December 2021)

• Record Type: Journal Article
• Authors: Aukes, Laurie; Bartlett, Joan; Fireman, Bruce; Hansen, John; Lewis, Ned; Marks, Morgan; Saddier, Patricia; Klein, Nicola P
• Is Part Of: Open forum infectious diseases. Volume 8(2021)Supplement 1
• Page Start: S135
• Page End: S135
• Publication Date: 2021-12-04
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofab466.224
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: We conducted a large real-world study of the long-term vaccine effectiveness (VE) of the live attenuated zoster vaccine (Zostavax; ZVL). Using an innovative approach with automated observational data we measured VE for incident herpes zoster (HZ) and severe HZ outcomes including post-herpetic neuralgia (PHN), herpes zoster ophthalmicus (HZO), and hospitalized HZ. This approach could be useful in long-term effectiveness studies of other vaccines. Methods: We assessed VE against HZ, PHN, HZO and hospitalized HZ for up to 10+ years after vaccination at Kaiser Permanente Northern California. We identified incident cases using diagnoses, laboratory tests and prescriptions, and validated a sample by chart review. For each outcome, we used a Cox regression model with a calendar timeline to estimate VE in relation to year since vaccination. The model for HZ included 11 time-varying vaccination status indicators to denote -- at each timepoint during follow-up -- either the number of years since ZVL vaccination (30 days to < 1 year, 1 to < 2 years, . . ., and 10+ years) or that the individual is unvaccinated (reference group). Analyses were adjusted for demographics and time-varying measures of immune compromise status, healthcare use and comorbidities. Results: From 2007-2018, 1.5 million people contributed to analyses; 507, 000 (34%) were vaccinated. During 9 million person-years of follow-up, we observed 75, 135 HZ cases, including 4, 982 (7%) with PHN, 4, 418 (6%) with HZO, and 555 (< 1%) who were hospitalized. VE for HZ was 67% (95% Confidence Interval [CI]: 65-69%) in the first year after vaccination, waned to 50% (CI: 47-52%) in the second year after vaccination, and then waned more gradually to 15% (CI: 5-24%) by 10+ years after vaccination. Initial VE was higher against PHN (83%; CI: 78-87%) and hospitalized HZ (89%; CI: 67-97%) with less waning observed over time (42% by Year 8 for PHN and 53% in Years 5 to < 8 for hospitalized HZ). VE against HZO was 71% in Year 1 and waned to 29% in Years 5 to < 8. Conclusion: Our large population, long follow-up and innovative methods let us estimate VE against HZ, PHN, HZO and hospitalized HZ for 10+ years after vaccination. Our approach could help assess waning and need for boosters for vaccines against other agents including COVID-19. Disclosures: Morgan Marks, PhD, ScM, Merck & Co Inc. (Employee) Patricia Saddier, MD, PhD, Merck & Co Inc (Employee) Nicola P. Klein, MD, PhD, GlaxoSmithKline (Grant/Research Support)MedImmune (Grant/Research Support)Merck & Co Inc (Grant/Research Support)Pfizer (Grant/Research Support)Protein Sciences (now Sanofi Pasteur ) (Grant/Research Support)Sanofi Pasteur (Grant/Research Support)