Causal Inference & Treatment Effects Analysis

Usage

treatmenteffects(
  data,
  treatment_var,
  outcome_var,
  covariates,
  patient_id,
  time_var,
  event_var,
  cluster_var,
  causal_method = "propensity_score",
  outcome_type = "continuous",
  estimand = "ate",
  ps_method = "logistic",
  ps_specification = "main_effects",
  balance_threshold = 0.1,
  matching_method = "nearest_neighbor",
  matching_ratio = "1to1",
  caliper_width = 0.2,
  replacement_matching = FALSE,
  weight_trimming = TRUE,
  trim_quantiles = 0.05,
  stabilized_weights = TRUE,
  weight_normalization = "sum_to_n",
  outcome_model = "linear",
  include_ps_in_outcome = TRUE,
  balance_assessment = TRUE,
  balance_methods = "standardized_differences",
  overlap_assessment = TRUE,
  sensitivity_analysis = TRUE,
  sensitivity_method = "rosenbaum_bounds",
  gamma_values = "1.1,1.2,1.3,1.5,2.0",
  heterogeneous_effects = FALSE,
  effect_modifiers,
  causal_tree = FALSE,
  causal_forest = FALSE,
  instrument_var,
  iv_method = "two_stage_least_squares",
  model_diagnostics = TRUE,
  bootstrap_inference = TRUE,
  n_bootstrap = 1000,
  cross_validation = FALSE,
  comprehensive_report = TRUE,
  individual_effects = FALSE,
  save_weights = FALSE,
  save_matched_data = FALSE,
  regulatory_documentation = TRUE
)

Arguments

data

the data as a data frame

treatment_var

Binary treatment variable (treated vs. control)

outcome_var

Primary outcome variable of interest

covariates

Confounding variables to adjust for in causal analysis

patient_id

Patient identifier for tracking matched pairs

time_var

Time to event variable for survival outcomes

event_var

Event indicator for survival outcomes

cluster_var

Clustering variable (e.g., hospital, physician) for robust standard errors

causal_method

Primary method for causal inference

outcome_type

Type of outcome variable

estimand

Target causal estimand

ps_method

Method for estimating propensity scores

ps_specification

Complexity of propensity score model

balance_threshold

Maximum standardized mean difference for adequate balance

matching_method

Specific matching algorithm to use

matching_ratio

Ratio of controls to treated units

caliper_width

Maximum distance for matching (in standard deviations)

replacement_matching

Allow controls to be matched to multiple treated units

weight_trimming

Trim extreme propensity score weights

trim_quantiles

Quantiles for weight trimming (e.g., 0.05 = trim top/bottom 5\

stabilized_weightsUse stabilized inverse probability weights

weight_normalizationMethod for normalizing weights

outcome_modelModel for outcome regression

include_ps_in_outcomeInclude propensity score in outcome model (doubly robust)

balance_assessmentAssess covariate balance before and after adjustment

balance_methodsMethods for assessing covariate balance

overlap_assessmentAssess propensity score overlap between treatment groups

sensitivity_analysisPerform sensitivity analysis for unmeasured confounding

sensitivity_methodMethod for sensitivity analysis

gamma_valuesGamma values for Rosenbaum bounds (comma-separated)

heterogeneous_effectsAnalyze treatment effect heterogeneity

effect_modifiersVariables for subgroup analysis and effect modification

causal_treeUse causal trees to identify effect heterogeneity

causal_forestUse causal random forests for individualized treatment effects

instrument_varInstrumental variable for IV analysis

iv_methodMethod for instrumental variable estimation

model_diagnosticsPerform comprehensive model diagnostics

bootstrap_inferenceUse bootstrap for confidence intervals

n_bootstrapNumber of bootstrap samples

cross_validationUse cross-validation for model selection

comprehensive_reportGenerate comprehensive causal inference report

individual_effectsEstimate individual treatment effects

save_weightsSave propensity scores and weights to dataset

save_matched_dataSave matched dataset for further analysis

regulatory_documentationInclude regulatory compliance documentation

A results object containing:

results$overview					a table
results$treatment_effect_estimates					a table
results$balance_assessment					a table
results$propensity_summary					a table
results$matching_summary					a table
results$weight_summary					a table
results$model_diagnostics					a table
results$sensitivity_analysis					a table
results$heterogeneity_analysis					a table
results$individual_effects					a table
results$overlap_plot					an image
results$balance_plot					an image
results$effect_plot					an image
results$sensitivity_plot					an image
results$causal_report					a html

Tables can be converted to data frames with asDF or as.data.frame. For example:results$overview$asDFas.data.frame(results$overview) Comprehensive causal inference analysis for estimating treatment effects in observational studies. Implements propensity score methods, inverse probability of treatment weighting (IPTW), matching techniques, and doubly robust estimation. Includes covariate balance assessment, sensitivity analysis, and causal effect estimation with confidence intervals. Essential for comparative effectiveness research, real-world evidence studies, and observational research where randomized controlled trials are not feasible. data('treatment_data')treatmenteffects( data = treatment_data, treatment_var = "treatment", outcome_var = "outcome", covariates = c("age", "sex", "baseline_severity"), causal_method = "propensity_score", estimand = "ate" )