Medication adherence during the run‐in phase of clinical trials: A systematic review of methodological and reporting rigour

TL;DR

This paper reviews how medication adherence is measured and reported in clinical trial run-in phases, finding poor methodological quality and lack of transparency.

Contribution

The study provides the first systematic review of methodological and reporting practices in run-in phases for medication adherence.

Findings

Most run-in studies used pill counts to measure adherence, often with an 80% adherence threshold.

Despite complete reporting of some metrics, no studies publicly reported adherence phase details or data.

All reviewed run-in phases were rated as poor quality and at high risk of bias.

Abstract

Trial run‐in phases are sometimes used to select adherent participants as a strategy to enrich trials, but the methods employed lack clear regulatory guidance and there are concerns about the transparency of reporting. This review aims to characterize the methods used for adherence measurement and public reporting of run‐in studies.

The protocol is published on the Open Science Framework. Clinicaltrials.gov was searched using the terms “run*in”, “lead*in”, “enrichment” or “single‐blind placebo”. The Risk of Bias tool for Observational Adherence Studies (RoBOAS) and the ESPACOMP Medication Adherence Reporting Guideline (EMERGE) were used to assess methodological and reporting quality. A narrative synthesis of the evidence was undertaken.

In total, 249 studies were identified, 34 were included for the analysis, of which 8 specified adherence to be a main purpose of the run‐in. Most (20/34) used pill counts and of those using an explicit threshold for adherence, 13/31 used an 80% cut‐off. The reporting of the adherence measurement method, summary metric and method of data aggregation was complete in 23/34 run‐ins, but none publicly reported the adherence phase considered or presented any adherence data. All run‐ins were judged to be of poor methodological quality, and at critical risk of bias.

The methodological and public reporting quality of medication adherence in run‐in phases of drug trials may compromise their intended purpose of improving the quality of the main trial. These findings have regulatory implications and support the need for guidelines on the measurement, analysis and reporting of adherence during the run‐in phase.

Figures