Consortium Causal Inference Propensity Engine — consortium_propensity

Supports 1:k nearest-neighbor caliper matching and IPTW weighting for observational cohorts.

Usage

consortium_propensity_score(
  con = NULL,
  cohort_table = "cohort",
  exposure_col = "treatment",
  covariate_cols = NULL,
  method = "matching",
  caliper = 0.2,
  ratio = 1L,
  trim = 0.01
)

Arguments

con: Active DBI connection (required if cohort_table is a string table name).
cohort_table: Character table name or data.frame.
exposure_col: Column indicating treatment (1 = target, 0 = comparator).
covariate_cols: Optional character vector of covariate column names to balance.
method: Adjustment method: 'matching' (1:k caliper matching) or 'iptw' (IPTW weighting).
caliper: Caliper distance in units of SD of logit(PS) (default 0.2).
ratio: Ratio for 1:k matching (default 1L).
trim: Trimming threshold for extreme propensity scores (default 0.01).

Value

An object of class c("omop_propensity_result", "list").