Randomized Controlled Trials without Data Retention

TL;DR

This paper develops algorithms for conducting accurate randomized controlled trials that comply with strict data minimization and anonymization policies, enabling effective treatment effect estimation without retaining individual data.

Contribution

It introduces recursive algorithms and strategies combining bootstrap and federated methods to perform RCT analysis under stringent data retention constraints.

Findings

Effective treatment effect estimates with minimal data retention

Robust inferences from non-i.i.d. data using combined algorithms

Framework supports privacy-preserving RCTs with anonymized data

Abstract

Amidst rising appreciation for privacy and data usage rights, researchers have increasingly acknowledged the principle of data minimization, which holds that the accessibility, collection, and retention of subjects' data should be kept to the bare amount needed to answer focused research questions. Applying this principle to randomized controlled trials (RCTs), this paper presents algorithms for making accurate inferences from RCTs under stringent data retention and anonymization policies. In particular, we show how to use recursive algorithms to construct running estimates of treatment effects in RCTs, which allow individualized records to be deleted or anonymized shortly after collection. Devoting special attention to non-i.i.d. data, we further show how to draw robust inferences from RCTs by combining recursive algorithms with bootstrap and federated strategies.