Analysis of recurrent events in cluster randomised trials: The PLEASANT trial case study

Grant, Kelly and Julious, Steven A. (2025) Analysis of recurrent events in cluster randomised trials: The PLEASANT trial case study. Statistical Methods in Medical Research. ISSN 0962-2802

Full text not available from this repository. (Request a copy)

Abstract

Recurrent events for many clinical conditions, such as asthma, can indicate poor health outcomes. Recurrent events data are often analysed using statistical methods such as Cox regression or negative binomial regression, suffering event or time information loss. This article re-analyses the preventing and lessening exacerbations of asthma in school-age children associated with a new term (PLEASANT) trial data as a case study, investigating the utility, extending recurrent events survival analysis methods to cluster randomised trials. A conditional frailty model is used, with the frailty term at the general practitioner practice level, accounting for clustering. A rare events bias adjustment is applied if few participants had recurrent events and truncation of small event risk sets is explored, to improve model accuracy. Global and event-specific estimates are presented, alongside a mean cumulative function plot to aid interpretation. The conditional frailty model global results are similar to PLEASANT results, but with greater precision (include time, recurrent events, within-participant dependence, and rare events adjustment). Event-specific results suggest an increasing risk reduction in medical appointments for the intervention group, in September–December 2013, as medical contacts increase over time. The conditional frailty model is recommended when recurrent events are a study outcome for clinical trials, including cluster randomised trials, to help explain changes in event risk over time, assisting clinical interpretation.

Item Type:	Article
Additional Information:	Data availability statement: Access to patient-level data is provided by the CPRD for health research purposes and is dependent on the approval of a study protocol by the MHRA Independent Expert Advisory Committee (ERC). More information on ERC and the protocol submission process can be found at: https://cprd.com/data-access [accessed 23/11/2022]. Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Faculty \ School:	Faculty of Medicine and Health Sciences > Norwich Medical School
UEA Research Groups:	Faculty of Medicine and Health Sciences > Research Groups > Norwich Clinical Trials Unit Faculty of Medicine and Health Sciences > Research Groups > Epidemiology and Public Health Faculty of Medicine and Health Sciences > Research Centres > Population Health
Related URLs:	https://journals.sagepub.com/doi/10.1177...
Depositing User:	LivePure Connector
Date Deposited:	19 May 2025 11:30
Last Modified:	19 May 2025 11:30
URI:	https://ueaeprints.uea.ac.uk/id/eprint/99295
DOI:	10.1177/09622802251316972

Actions (login required)

View Item