filter_relative {admiral}

Filter the Observations Before or After a Condition is Fulfilled

Description

Filters the observations before or after the observation where a specified condition is fulfilled for each by group. For example, the function could be called to select for each subject all observations before the first disease progression.

Usage

filter_relative(
  dataset,
  by_vars,
  order,
  condition,
  mode,
  selection,
  inclusive,
  keep_no_ref_groups = TRUE,
  check_type = "warning"
)

Arguments

dataset

Input dataset

The variables specified by the by_vars and order arguments are expected to be in the dataset.

by_vars

Grouping variables

Permitted Values: list of variables created by exprs() e.g. exprs(USUBJID, VISIT)

order

Sort order

Within each by group the observations are ordered by the specified order.

For handling of NAs in sorting variables see Sort Order.

Permitted Values: list of expressions created by exprs(), e.g., exprs(ADT, desc(AVAL))

condition

Condition for Reference Observation

The specified condition determines the reference observation. The output dataset contains all observations before or after (selection parameter) the reference observation.

mode

Selection mode (first or last)

If "first" is specified, for each by group the observations before or after (selection parameter) the observation where the condition (condition parameter) is fulfilled the first time is included in the output dataset. If "last" is specified, for each by group the observations before or after (selection parameter) the observation where the condition (condition parameter) is fulfilled the last time is included in the output dataset.

Permitted Values: "first", "last"

selection

Select observations before or after the reference observation?

Permitted Values: "before", "after"

inclusive

Include the reference observation?

Permitted Values: TRUE, FALSE

keep_no_ref_groups

Should by groups without reference observation be kept?

Permitted Values: TRUE, FALSE

check_type

Check uniqueness?

If "warning" or "error" is specified, the specified message is issued if the observations of the input dataset are not unique with respect to the by variables and the order.

Permitted Values: "none", "warning", "error"

Details

For each by group ( by_vars parameter) the observations before or after (selection parameter) the observations where the condition (condition parameter) is fulfilled the first or last time (order parameter and mode parameter) is included in the output dataset.

Value

A dataset containing for each by group the observations before or after the observation where the condition was fulfilled the first or last time

See Also

Utilities for Filtering Observations: count_vals(), filter_exist(), filter_extreme(), filter_joined(), filter_not_exist(), max_cond(), min_cond()

Examples

library(tibble)

response <- tribble(
  ~USUBJID, ~AVISITN, ~AVALC,
  "1",      1,        "PR",
  "1",      2,        "CR",
  "1",      3,        "CR",
  "1",      4,        "SD",
  "1",      5,        "NE",
  "2",      1,        "SD",
  "2",      2,        "PD",
  "2",      3,        "PD",
  "3",      1,        "SD",
  "4",      1,        "SD",
  "4",      2,        "PR",
  "4",      3,        "PD",
  "4",      4,        "SD",
  "4",      5,        "PR"
)

# Select observations up to first PD for each patient
response %>%
  filter_relative(
    by_vars = exprs(USUBJID),
    order = exprs(AVISITN),
    condition = AVALC == "PD",
    mode = "first",
    selection = "before",
    inclusive = TRUE
  )

# Select observations after last CR, PR, or SD for each patient
response %>%
  filter_relative(
    by_vars = exprs(USUBJID),
    order = exprs(AVISITN),
    condition = AVALC %in% c("CR", "PR", "SD"),
    mode = "last",
    selection = "after",
    inclusive = FALSE
  )

# Select observations from first response to first PD
response %>%
  filter_relative(
    by_vars = exprs(USUBJID),
    order = exprs(AVISITN),
    condition = AVALC %in% c("CR", "PR"),
    mode = "first",
    selection = "after",
    inclusive = TRUE,
    keep_no_ref_groups = FALSE
  ) %>%
  filter_relative(
    by_vars = exprs(USUBJID),
    order = exprs(AVISITN),
    condition = AVALC == "PD",
    mode = "first",
    selection = "before",
    inclusive = TRUE
  )