Overview
The serocalculator R package provides a rapid and computationally simple method for calculating seroconversion rates, as originally published in Simonsen et al. (2009) and Teunis et al. (2012), and further developed in subsequent publications by deGraaf et al. (2014), Teunis et al. (2016), and Teunis and Eijkeren (2020). In short, longitudinal seroresponses from confirmed cases with a known symptom onset date are assumed to represent the time course of human serum antibodies against a specific pathogen. Therefore, by using these longitudinal antibody dynamics with any cross–sectional sample of the same antibodies in a human population, an incidence estimate can be calculated. Further details are below.
A Proxy for Infection
While the exact time of infection is impossible to measure in an individual, antibody levels measured in a cross–sectional population sample can be translated into an estimate of the frequency with which seroconversions (infections) occur in the sampled population. So the presence of many high antibody concentrations indicates that many people in the population likely experienced infection recently, while mostly low concentrations indicate a low frequency of infections in the sampled population.
In order to interpret the measured cross-sectional antibody concentrations in terms of incidence, we must define the antibody dynamic over time to understand the generalized antibody response at different times since infection. This dynamic must be quantified over time to include an initial increase in serum antibody concentration when seroconversion occurs, followed by a gradual decrease as antibodies wane. In published studies, this information on the time course of the serum antibody response has been obtained from longitudinal follow–up data in cases who had a symptomatic episode following infection. In this case, the onset of symptoms then provides a proxy for the time that infection occurred.
The Seroincidence Estimator
The serocalculator package was designed to calculate the incidence of seroconversion by using the longitudinal seroresponse characteristics. The distribution of serum antibody concentrations in a cross–sectional population sample is calculated as a function of the longitudinal seroresponse and the frequency of seroconversion (or seroincidence). Given the seroresponse, this marginal distribution of antibody concentrations can be fitted to the cross-sectional data and thereby providing a means to estimate the seroincidence.