Bayesian inference for causal mechanisms with application to a randomized study for postoperative pain control

Biostatistics. 2017 Oct 1;18(4):605-617. doi: 10.1093/biostatistics/kxx010.

Abstract

We conduct principal stratification and mediation analysis to investigate to what extent the positive overall effect of treatment on postoperative pain control is mediated by postoperative self administration of intra-venous analgesia by patients in a prospective, randomized, double-blind study. Using the Bayesian approach for inference, we estimate both associative and dissociative principal strata effects arising in principal stratification, as well as natural effects from mediation analysis. We highlight that principal stratification and mediation analysis focus on different causal estimands, answer different causal questions, and involve different sets of structural assumptions.

Keywords: Bayesian inference; Causal inference; Mediation analysis; Oral morphine; Postoperative pain; Potential outcomes; Pre-medication; Principal stratification; Randomized Experiments.

Publication types

  • Randomized Controlled Trial

MeSH terms

  • Adolescent
  • Adult
  • Aged
  • Aged, 80 and over
  • Analgesics, Opioid / administration & dosage
  • Analgesics, Opioid / pharmacology*
  • Bayes Theorem
  • Double-Blind Method
  • Female
  • Humans
  • Male
  • Middle Aged
  • Models, Statistical*
  • Morphine / administration & dosage
  • Morphine / pharmacology*
  • Outcome Assessment, Health Care / methods*
  • Pain Measurement
  • Pain, Postoperative / drug therapy*
  • Prospective Studies
  • Self Administration
  • Young Adult

Substances

  • Analgesics, Opioid
  • Morphine