Handling of covariates in stepped-wedge cluster randomized trials: Protocol for a methodological review

Abstract

Background: Stepped-wedge cluster randomized trials (SW-CRTs), which randomize clusters to times at which they transition from control to intervention arms, are an increasingly popular design for evaluating health system and policy interventions. Reporting of SW-CRTs should ideally be guided by the 2018 CONSORT extension for SW-CRTs. Although substantial methodological development has taken place in recent years, several gaps remain in our knowledge about the best practices for the design, analysis, and reporting of SW-CRTs. Objectives: We are undertaking a methodological systematic review of SW-CRTs published 2016-2022. Our primary objectives are to describe current methods and practices for handling covariates (number of and types of covariates) in the design and analysis of SW-CRTs; examine adherence to CONSORT guidelines around reporting of covariates including balance at baseline; and identify prevalent issues in SW-CRTs and gaps in the methodological literature. Methods: We will identify primary reports of SW-CRTs published in English 2016-2022 from three sources: a previously published review covering 2020 to 2021, an existing database of pragmatic trials published 2014-2019, and an updated search capturing trials up to 2022. A data extraction form will be used to standardize the extraction of information about the trials. Analyses will be primarily descriptive in nature.