Fig. 4: Correlation between the waning effects of the ZVL immunization on HZ and the risk of dementia.

As the effectiveness of ZVL in protecting against HZ waned over up to 15 years after index date (x axis), the magnitude of protection against dementia similarly diminished over the same time frame. Please note that (1) the y axis indicates RR of an outcome (HZ or dementia) in the ZVL cohort, calculated relative to the effect observed in recipients of the PPSV23, which does not confer protection against HZ; (2) y-axis values <1 indicate a lower risk due to exposure; (3) the near horizontal line for HZ protection after ~10 years postindex date indicates that HZ cases were prevented, whereas the return of the dementia risk to an RR of ~1 suggests delayed progression to rather than prevention of dementia diagnoses. The curves represent RR (y axis) for HZ (orange line) and dementia (blue line) in the ZVL cohort calculated relative to the effect observed in PPSV23 cohort, for each month during the follow-up period along with their respective 95% CI bands (shaded areas around each curve) based on the ratio of the Nelson–Aalen estimates of the corresponding cumulative hazard functions. The Person’s R coefficient with a 95% CI shows the level of linear correlation between the curves; RR values <1 indicate a lower risk due to exposure.