A Latent Principal Stratification Method to Address One-Sided Cluster and Individual Noncompliance in Cluster RCTs

Anthony Sisti, Ellen McCreedy, Roee Gutman, Anthony Sisti, Ellen McCreedy, Roee Gutman · 2025

Read A Latent Principal Stratification Method to Address One-Sided Cluster and Individual Noncompliance in Cluster RCTs by Anthony Sisti, Ellen McCreedy, Roee Gutman, Anthony Sisti, Ellen McCreedy, Roee Gutman (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-15280.txt.

Description

Anthony Sisti, Ellen McCreedy, and Roee Gutman model compliance at two levels inside pragmatic cluster randomized trials in a 2025 biostatistics paper. In such trials, clinics rather than patients are randomized, and both providers and patients can drift from the assigned intervention. Standard analyses of the complier average causal effect track only individual compliance, ignoring how unevenly clusters implement the program.

Their Bayesian framework handles one-sided noncompliance at both levels: binary at the patient level, partial at the cluster level, where implementation is summarized by several continuous metrics. A latent mixture model captures heterogeneity in how clusters take up the intervention. The estimand then shows how treatment effects vary with both sources of compliance, information trialists need when rolling out programs that clinics adopt unevenly.

Catalog record

CREATORAnthony Sisti, Ellen McCreedy, Roee Gutman, Anthony Sisti, Ellen McCreedy, Roee Gutman
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE124 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2509-15280

Browse the full library · About the Tower of Babel