Create a variable block

A variable block groups variables that share a source table, time window, and row-level filters. When passed to omop_recipe via its blocks argument, the block's concept_ids are expanded into individual omop_variable objects that inherit the block's defaults. This is the compact way to add many concepts from one table.

Usage

omop_variable_block(
  id = NULL,
  table,
  concept_ids = integer(0),
  concept_names = NULL,
  time_window = NULL,
  format = "raw",
  value_source = NULL,
  suffix_mode = "index",
  filters = list(),
  population_id = "base",
  expand = FALSE,
  reference_date = NULL,
  unit = NULL
)

Arguments

id

Character or NULL; block ID (auto-generated from table and concept count if NULL).

table

Character; shared source OMOP CDM table (e.g. "condition_occurrence").

concept_ids

Integer vector; concept IDs for all variables in the block.

concept_names

Character vector or NULL; human-readable names matching concept_ids positionally.

time_window

Named list with start/end offsets, or NULL for no window.

format

Character; default output format for variables in this block (e.g. "binary", "count").

value_source

Character or NULL; default value source column (e.g. "value_as_number").

suffix_mode

Character; naming mode for multi-column expansion ("index", "range", or "label").

filters

List of omop_filter objects; row-level filters applied to all variables in the block.

population_id

Character; which population this block belongs to (default "base").

expand

Logical; if TRUE, expand the block's concepts to include vocabulary descendants server-side (default FALSE).

reference_date

Character/Date or NULL; fixed reference date required for a "time_since" block.

unit

Character or NULL; "day" (default for time_since) or complete calendar "month" units.

Value

An omop_variable_block object.

See also

omop_recipe, omop_variable

Examples

if (FALSE) { # \dontrun{
recipe <- omop_recipe(
  blocks = omop_variable_block(
    table = "condition_occurrence",
    concept_ids = c(201820, 320128),
    concept_names = c("Type 2 diabetes", "Essential hypertension"),
    format = "binary"
  ),
  outputs = omop_output(type = "wide")
)
} # }