TTE_pipeline {debiasedTrialEmulation} | R Documentation |
Target Trial Emulation (TTE) Pipeline
Description
Implements a Target Trial Emulation pipeline using propensity score methods, including matching, weighting, and stratification.
Usage
TTE_pipeline(data, xvars, yvars, ncovars = NULL, ps_type, outcome_measure)
Arguments
data |
A dataset containing treatment assignment, covariates, and outcomes. |
xvars |
A character vector of covariate names used for propensity score estimation. |
yvars |
A character vector of primary outcome variable names. |
ncovars |
Optional. A character vector of negative control outcome variable names. |
ps_type |
The propensity score method: "Matching", "Stratification", or "Weighting". |
outcome_measure |
The outcome measure to estimate: "RR" (Risk Ratio), "OR" (Odds Ratio), or "HR" (Hazard Ratio). |
Value
An object of class "TTE" containing the propensity score analysis results
Examples
library("dplyr")
data(demo_data)
xvars <- c("eth_cat", "age_cat", "sex", "cohort_entry_month",
"obese", "pmca_index", "n_ed", "n_inpatient",
"n_tests", "imm_date_diff_grp", "medical_1", "medical_2",
"medical_3", "medical_4", "medical_5")
yvars1 <- colnames(demo_data %>% select(starts_with("visits_")))
yvars2 <- colnames(demo_data %>% select(starts_with("event_")))
# without negative controls
TTE_pipeline(demo_data, xvars=xvars, yvars=yvars1, ps_type="Matching", outcome_measure="RR")
# with negative controls
ncovars1 <- colnames(demo_data %>% select(starts_with("nco_visits_")))
TTE_pipeline(
demo_data,
xvars = xvars,
yvars = yvars1,
ncovars = ncovars1,
ps_type = "Matching",
outcome_measure = "RR"
)
[Package debiasedTrialEmulation version 0.1.0 Index]