Risk of bias tools for medication adherence research: RoBIAS and RoBOAS

An unbiased means of documenting medication-taking is important to ensure quality evidence about adherence research and to accurately identify individuals at risk of suboptimal adherence for the development of targeted and effective interventions. Guidance to assist researchers in the understanding of risk of bias when conducting or reviewing adherence research is currently not available. To address this gap, tools to identify and gauge the magnitude of important biases that may impact adherence research have been developed.

The Risk of Bias tool for Interventional Adherence Studies (RoBIAS) and the Risk of Bias tool for Observational Adherence Studies (RoBOAS) were constructed from a literature review of key adherence guidelines/frameworks, drafted initially through author consensus. The draft bias tools were piloted and evaluated with expert adherence researchers through an online survey platform to assess the internal consistency and agreement in responses, including gather "free text" feedback to improve the tool's utility.

Of the 121 approached reviewers, only 20 out of the 30 reviewers who consented to participate completed the piloting of the tools. Both tools are structured around four domains relating to: (i) study design, (ii) randomization (RoBIAS tool) and confounding factors (RoBOAS tool), (iii) adherence outcome measurement, and (iv) data analysis. Each domain consists of items/statements, mapped to specific biases relevant to adherence research and study designs, including a domain-based ranking scale to determine the appropriate risk of bias judgement.

The tools are intended to have utility when systematically reviewing adherence research and to inform the design of future adherence studies.