Thesis_SarahArnup_final.pdf (6.5 MB)
Improving the design, reporting and statistical methods used in cluster randomised crossover trials
thesisposted on 2019-06-28, 02:07 authored by SARAH JANE ARNUP
The cluster randomised crossover (CRXO) design is a type of randomised trial that is used to assess the effects of interventions. The aim of this research was to assess the design, reporting quality, and statistical methods used in CRXO trials, and to facilitate improvement where shortcomings were identified. The research showed that incomplete reporting and the application of inappropriate methods for sample size calculation and analysis were common. This resulted in the development of graphical illustrations that explained the intuition behind the design, a proposal for a reporting guideline, detailed sample size considerations and sample size formulae for stratified trials.