# A hypothesis test of feasibility for external pilot trials assessing recruitment, follow-up and adherence rates

## TL;DR

This paper introduces a hypothesis testing approach to assess the feasibility of large clinical trials based on external pilot data, focusing on recruitment, follow-up, and adherence rates to inform progression decisions.

## Contribution

It proposes a novel hypothesis test framework for external pilot trials, including decision rules and error rate calculations, to improve trial feasibility assessments.

## Key findings

- Developed a hypothesis test for trial feasibility based on pilot data.
- Applied the method to redesign the TIGA-CUB trial.
- Extended the approach to estimate variability of primary endpoints.

## Abstract

The power of a large clinical trial can be adversely affected by low recruitment, follow-up and adherence rates. External pilot trials estimate these rates and use them, via pre-specified decision rules, to determine if the definitive trial is feasible and should go ahead. There is little methodological research underpinning how these decision rules, or the sample size of the pilot, should be chosen. In this paper we propose a hypothesis test of the feasibility of a definitive trial, to be applied to the external pilot data and used to make progression decisions. We quantify feasibility by the power of the planned trial, as a function of recruitment, follow-up and adherence rates. We use this measure to define hypotheses to test in the pilot, propose a test statistic, and show how the error rates of this test can be calculated for the common scenario of a two-arm parallel group definitive trial with a single normally distributed primary endpoint. We use our method to re-design TIGA-CUB, an external pilot trial comparing a psychotherapy with treatment as usual for children with conduct disorders. We then extend our formulation to include using the pilot data to estimate the standard deviation of the primary endpoint. and incorporate this into the progression decision.