fit_population: Infer and fit a population model from 'SurvStat' output

In rsurvstat: Download Infectious Disease Data from 'SurvStat' (Robert Koch Institute)

Infer and fit a population model from SurvStat output

Description

SurvStat can be queried for count or incidence. From the combination of these metrics queried across the whole range of disease notifications for any given year we can infer a stratified population size, that SurvStat is using to calculate it's incidence. This is simply modelled with a local polynomial over time to allow us to fill in weekly population denominators.

Usage

fit_population(count_df, .progress = TRUE)

infer_population(
  age_group = NULL,
  geography = NULL,
  years = NULL,
  .progress = TRUE
)

Arguments

count_df	a dataframe from the output of get_timeseries() or get_snapshot()
.progress	by default a progress bar is shown, which may be important if many downloads are needed to fulfil the request. It can be disabled by setting this to FALSE here.
age_group	(optional) the age group of interest as a SurvStat key, see rsurvstat::age_groups for a list of valid options.
geography	(optional) one of "state", "nuts", or "county" to define the resolution of the query. Does not accept a sf map or subset of (unlike get_timeseries()).
years	(optional) a vector of years to limit the response to. This may be useful to limit the size of returned pages in the event the SurvStat service hits a data transfer limit.

Value

the count_df dataframe with an additional population column

a dataframe with geography, age grouping, year and population columns

Functions

• infer_population(): Query SurvStat for data to impute a population denominator

Examples

# snapshot:
get_snapshot(
  disease = diseases$`COVID-19`,
  geography = "state",
  season=2024
) %>%
fit_population() %>%
dplyr::glimpse()

# timeseries
# A weekly population estimate is inferred from the yearly data:
get_timeseries(
  diseases$`COVID-19`,
  measure = "Count",
  age_group = age_groups$children_coarse
) %>%
fit_population() %>%
dplyr::glimpse()



infer_population(years=2020:2025) %>% dplyr::glimpse()

rsurvstat documentation built on Feb. 20, 2026, 5:09 p.m.