boostr

This is a generic implementation of a combined vaccine and vaccine efficacy modelling framework first proposed in White et al (2015). Please cite this work if you use this vaccine model.

The model consists of two parts. The first predicts antibody titres over time following initital dose(s) and subsequent booster doses.

$$ titre(t) = titre_{dose} \left( \rho_{dose} e^{-r_{dose}^s (t - t_{dose})} + (1 - \rho_{dose}) e^{-r_{dose}^l (t - t_{dose})} \right) $$

Where subscript $dose$ indexes for the primary series dose(s) or subsequent booster doses, $titre$ is the maximum titre on receipt of a dose or booster, $rho$ the proportion of the response that is short-lived and $r^s = \frac{\log_e(2)}{d^s}$ and $r^l = \frac{\log_e(2)}{d^l}$ where $d^s$ and $d^l$ are the half-lives of the short and long lived components of the antibody response respectively.

The second translates antibody titre to vaccine efficacy using a parameterised dose response curve.

$$ V(t) = V_{max} \left( 1 - \frac{1}{1 + \left( \frac{titre(t)}{\beta} \right)^\alpha } \right) $$

Where $V_{max}$ is the maximum vaccine efficacy, $alpha$ the shape parameter and $beta$ the scale parameter

Installation

You can install boostr from github with:

pak::pkg_install("mrc-ide/boostr")