Risk

Risk Exposure

The vivarium_public_health risk exposure package provides components to assign risk factor exposure values to simulants during a simulation. While the core vivarium framework supplies the values and population machinery, this package uses it to assign each simulant a propensity and translate that propensity into an exposure level through a statistical distribution.

The package is organized around two key concepts:

Exposure — involves initialization of a fixed propensity for each simulant, selection of a distribution type, and registration of an exposure pipeline that converts propensities into exposure values.
Distributions — the statistical models that translate propensities into exposure levels, ranging from simple dichotomous (exposed/unexposed) categories to complex ensemble distributions.

Risk Effect

The vivarium_public_health risk effect package provides components that modify target rates (such as disease incidence or mortality) based on a simulant’s exposure to a risk factor. While the core vivarium framework supplies the values pipeline machinery, this package uses it to translate each simulant’s exposure level into a multiplicative adjustment on one or more target rates.

The package is organized around two key concepts:

Relative Risk — a measure of how much more likely an outcome is for a simulant at a given exposure level compared to a reference level. A relative risk of 1 means the simulant is no more or less likely to experience the outcome than at the reference level. The risk effect component computes a per-simulant relative risk and registers it as a modifier on the target rate pipeline.
Calibration Constant — a value, typically derived from the population attributable fraction, that adjusts the baseline target rate so that, after relative risk multiplication across the population, the overall rate remains consistent with input data.