counted_prevalence {rprev}

Count prevalence from registry data.

Description

Counts contribution to prevalence at a specific index from each year of a registry. A person is included as contributing to disease prevalence if they are incident within the specified time-span, and are either alive or censored at the index date. The rationale for including censored cases in prevalence estimation is that such cases have typically been lost to follow-up, and are often more likely to have been alive at the index date than not.

Usage

counted_prevalence(formula, index, data, start_date, status_col)

Arguments

formula

A formula of the form <event date column> ~ <entry date column>.

index

The date at which to estimate point prevalence as a string in the format YYYY-MM-DD.

data

A data frame with the corresponding column names provided in form.

start_date

The initial date to start counting prevalence from as a Date object. Typically the index date - (Nyears * 365.25). Allows for non-whole year prevalence estimations.

status_col

The name of the column holding a binary indicator variable of whether the individual experienced an event at their event time or was censored.

Value

The number of prevalent cases at the specified index date as a single integer.