# Estimating cluster-level local average treatment effects in cluster randomised trials with non-adherence

## TL;DR

This paper proposes a method to estimate cluster-level local average treatment effects in cluster randomised trials with non-adherence, using two-stage least squares on cluster summaries, ensuring valid inference with proper adjustments.

## Contribution

It introduces a TSLS approach for cluster-level LATE estimation in CRTs with non-adherence, accounting for heteroscedasticity and small sample corrections.

## Key findings

- TSLS with cluster summaries yields low bias and accurate coverage.
- Proper degrees of freedom correction improves inference accuracy.
- Robust standard errors are essential for valid confidence intervals.

## Abstract

Non-adherence to assigned treatment is a common issue in cluster randomised trials (CRTs). In these settings, the efficacy estimand may be also of interest. Many methodological contributions in recent years have advocated using instrumental variables to identify and estimate the local average treatment effect (LATE). However, the clustered nature of randomisation in CRTs adds to the complexity of such analyses. In this paper, we show that under certain assumptions, the LATE can be estimated via two-stage least squares (TSLS) using cluster-level summaries of outcomes and treatment received. Implementation needs to account for this, as well as the possible heteroscedasticity, to obtain valid inferences. We use simulations to assess the performance of TSLS of cluster-level summaries under cluster-level or individual-level non-adherence, with and without weighting and robust standard errors. We also explore the impact of adjusting for cluster-level covariates and of appropriate degrees of freedom correction for inference. We find that TSLS estimation using cluster-level summaries provides estimates with small to negligible bias and coverage close to nominal level, provided small sample degrees of freedom correction is used for inference, with appropriate use of robust standard errors. We illustrate the methods by re-analysing a CRT in UK primary health settings.