Skip to contents

District by month panel of non-polio AFP cases

Source: R/spi_prepare.R
spi_case_panel.Rd

Counts the AFP cases in each district and month that are not classified as poliovirus or compatible. Cases still awaiting a laboratory result count as non-polio AFP. Each case is placed in the month of spi_case_date(). The panel is complete: a district and month with no case carries a count of zero, as spi_expected() requires.

Usage

spi_case_panel(
  cases,
  districts,
  months,
  cols = list(),
  polio_pattern = "WPV|VDPV|COMPATIBLE",
  surveillance_types = "AFP"
)

Arguments

cases

Data frame of AFP cases, one row per case, with POLIS column names by default.

districts

Character. District identifiers the panel covers, usually those of the boundaries the model is fitted on.

months

Date. First-of-month dates the panel covers, usually spi_panel_months().

cols

Named list overriding the column names: id (default "adm2_guid"), class (default "classification_all"), surveillance_type (default "surveillance_type_name") and the dates read by spi_case_date().

polio_pattern

Character. Regular expression matched against the upper-cased classification. A match is a poliovirus or compatible case and is not counted. Default: "WPV|VDPV|COMPATIBLE".

surveillance_types

Character. Surveillance types kept. NULL keeps every row. Default: "AFP".

Value

Tibble with one row per district and month: the district column (named as cols$id), month and an integer count, sorted by district and month. This is the cases input of spi_expected().

See also

Other spi preparation functions: spi_case_date(), spi_pop_index(), spi_window()

Examples

cases <- tibble::tibble(
  adm2_guid = c("D1", "D1", "D2"),
  classification_all = c("NPAFP", "WPV1", "PENDING"),
  surveillance_type_name = "AFP",
  paralysis_onset_date = as.Date(c("2024-01-03", "2024-01-09", NA)),
  case_date = as.Date(NA),
  stool1collection_date = as.Date(c(NA, NA, "2024-02-11"))
)
spi_case_panel(
  cases, c("D1", "D2"),
  seq(as.Date("2024-01-01"), by = "month", length.out = 2)
)
#> # A tibble: 4 × 3
#>   adm2_guid month      count
#>   <chr>     <date>     <int>
#> 1 D1        2024-01-01     1
#> 2 D1        2024-02-01     0
#> 3 D2        2024-01-01     0
#> 4 D2        2024-02-01     1