Set cohort end date to the last of a set of column dates

exitAtLastDate() resets cohort end date based on a set of specified column dates. The last date that occurs is chosen.

Usage

exitAtLastDate(
  cohort,
  dateColumns,
  cohortId = NULL,
  returnReason = FALSE,
  keepDateColumns = TRUE,
  name = tableName(cohort),
  .softValidation = FALSE
)

Arguments

cohort

A cohort table in a cdm reference.

dateColumns

Character vector indicating date columns in the cohort table to consider.

cohortId

Vector identifying which cohorts to modify (cohort_definition_id or cohort_name). If NULL, all cohorts will be used; otherwise, only the specified cohorts will be modified, and the rest will remain unchanged.

returnReason

If TRUE it will return a column indicating which of the dateColumns was used.

keepDateColumns

If TRUE the returned cohort will keep columns in dateColumns.

name

Name of the new cohort table created in the cdm object.

.softValidation

Whether to perform a soft validation of consistency. If set to FALSE four additional checks will be performed: 1) a check that cohort end date is not before cohort start date, 2) a check that there are no missing values in required columns, 3) a check that cohort duration is all within observation period, and 4) that there are no overlapping cohort entries

Value

The cohort table.

Examples

# \donttest{
library(CohortConstructor)
library(PatientProfiles)

cdm <- mockCohortConstructor()

cdm$cohort1 <- cdm$cohort1 |>
  addTableIntersectDate(tableName = "observation", nameStyle = "next_obs", order = "first") |>
  addFutureObservation(futureObservationType = "date", name = "cohort1")

cdm$cohort1 |>
  exitAtLastDate(dateColumns = c("next_obs", "future_observation"))
#> Joining with `by = join_by(cohort_definition_id, subject_id,
#> cohort_start_date)`
#> # A tibble: 56 × 6
#>    cohort_definition_id subject_id cohort_start_date cohort_end_date next_obs  
#>                   <int>      <int> <date>            <date>          <date>    
#>  1                    1          1 2008-06-11        2017-02-26      2008-08-12
#>  2                    1          2 2015-06-19        2015-10-11      2015-08-14
#>  3                    1          3 1993-06-20        1993-07-21      NA        
#>  4                    1          4 2012-01-08        2012-11-30      2012-03-12
#>  5                    1          7 1990-01-30        2010-08-09      1990-12-08
#>  6                    1          9 2003-06-12        2018-04-04      2004-08-28
#>  7                    1         10 2007-11-04        2015-01-10      2008-04-20
#>  8                    1         11 1985-12-18        1994-02-16      1986-12-02
#>  9                    1         13 1989-09-13        2011-02-18      1990-04-18
#> 10                    1         15 2015-07-01        2018-03-17      2015-08-27
#> # ℹ 46 more rows
#> # ℹ 1 more variable: future_observation <date>
# }