Development of a Platform Trial in Paediatric Intensive Care

Background

In recent years, the number of randomised controlled trials (RCTs) in paediatric intensive care has increased substantially. Currently, in the UK, there are 12 RCTs aiming to recruit 14,500 patients from 24 (of the total 28) UK paediatric intensive care units (PICUs). These are conducted as standalone RCTs, competing to randomise patients or run sequentially, meaning it takes a long time to get answers to important clinical questions and it is not possible to measure the interaction with other treatments being trialled.

Conducting standalone, competing, or sequential trials is inefficient and costly, placing a large demand on the PICU research infrastructure and finite resources. In contrast, adaptive platform trials use a shared platform to study multiple interventions simultaneously in a perpetual manner, meaning that results are generated more rapidly and efficiently.

As the UK methods centre for the REMAP-CAP trial, one of the game-changing COVID-19 platform trials, we want to develop an adaptive PICU platform trial to transform PICU research and ensure that patients receive the most up-to-date, evidence-based care more rapidly than the current model of competing or sequential, standalone RCTs.

Aims and Objectives

To complete the targeted preparatory work necessary for the establishment of a large, ambitious, Bayesian, randomised, multifactorial, adaptive, platform trial for paediatric intensive care. Methods

A mixed-methods approach comprising five linked Work Packages: WP1 – exploring parent and young people’s perspectives on the platform trial design; WP2 - engagement with the wider PICU community; WP3 - statistical modelling and methodological work; WP4 - health economics methodological work; and WP5 - data infrastructure and procedures for nesting within routinely collected data.

Timelines for delivery

WP1 and WP2: workshops with YPAGs, parents and PICU community (months 1-3); qualitative analysis (months 4-8). WP3: submission of HQIP application and specification of the Bayesian statistical model (months 1-3); Receipt of data from PICANet and finalise outputs needed for submission of stage 1 Platform trial (months 3-9); Completion of simulation report for platform trial (months 9-11).

WP4: creation of dataset using existing trial data (months 1-2); integrate with modelling work in WP3 (months 4-11); finalise outputs and recommendations needed for submission of stage 1 Platform trial (month 8); Completion of detailed health economic specification (month 11). WP5: Agree data linkage based on outputs of WP3 and WP4 (month 8). Submission of stage 1 funding application (month 9).

Anticipated Impact and Dissemination

The development of a PICU platform trial, in collaboration with parents, young people and the PICU community will ensure that patients receive the most up-to-date, evidence-based care more rapidly than the current model of PICU research of competing or sequential, standalone RCTs. The development of the PICU platform trial will also make evaluating new domains more efficient.

Using the shared platform will minimise set-up costs and shorten the duration of studies, making them less costly. The targeted methodological work will also contribute important generalisable learning around platform study designs and best practice, which will be published in appropriate peer reviewed journals and presented at relevant methodological conferences.