Add a survival (time-to-event) output to the plan

The historical single-outcome call produces one row per cohort episode with an event indicator and time-to-event in days. Advanced calls can retain named endpoints as survival, competing-risk, recurrent-event, or counting-process data. Calendar dates and source event identifiers are never returned. Requires a cohort to be set. Historical plans without an explicit censoring field are censored at the end of the observation period containing the index episode; they never bridge an unobserved gap to a later period.

Usage

ds.omop.plan.survival(
  plan,
  outcome_table = "condition_occurrence",
  outcome_concepts = NULL,
  tar = list(start_offset = 0, end_offset = 730),
  event_order = "first",
  name = "survival",
  outcomes = NULL,
  censoring = NULL,
  format = NULL,
  washout_days = 0L,
  tie_policy = "priority"
)

Arguments

plan

An omop_plan object.

outcome_table

Character; OMOP table containing outcome events (e.g. "condition_occurrence", "procedure_occurrence").

outcome_concepts

Numeric vector; concept IDs defining the historical composite outcome. Omit when using `outcomes`.

tar

Named list; time-at-risk window with start_offset and end_offset (integer days relative to cohort_start_date).

event_order

Character; "first" or "last" to select which event occurrence determines the time-to-event value; advanced recurrent/counting formats also accept `all`.

name

Character; output name used as a key in the plan's outputs list.

outcomes

Named list of endpoint specifications. Each endpoint contains `table`, `concept_set`, and optional safe row `filters`.

censoring

Named list controlling observation-period, death, cohort-end, and optional administrative-date censoring.

format

Character; `survival`, `competing_risk`, `recurrent_events`, or `counting_process`.

washout_days

Non-negative integer washout between events of the same named endpoint.

tie_policy

Character; `priority`, `error`, or `all`. The latter is restricted to recurrent-event output.

Value

The modified omop_plan with the survival output appended.

See also

ds.omop.plan.events, ds.omop.plan.cohort

Examples

if (FALSE) { # \dontrun{
plan <- ds.omop.plan()
plan <- ds.omop.plan.cohort(plan, cohort_definition_id = 1)
plan <- ds.omop.plan.survival(plan,
  outcome_table = "condition_occurrence",
  outcome_concepts = c(201826, 443238),
  tar = list(start_offset = 0, end_offset = 365),
  event_order = "first"
)
} # }