Cluster randomised trials (CRTs) often use geographical areas as the unit of randomisation. Despite this, explicit consideration of the location and spatial distribution of observations is rare. In many trials, the location of participants will have little importance, however in some, especially against infectious diseases, spillover effects due to participants being located close together may affect trial results. This PhD takes a multidisciplinary approach to apply and evaluate spatial analysis methods in CRTs, furthering understanding of how spatial analysis can complement traditional evaluation of CRTs. I began by conducting a systematic review of CRTs that used spatial analysis techniques. I found only 10 published papers, most of which being supplementary analyses of the main trial. I then conducted a spatial analysis of an Oral Polio Vaccine (OPV) transmission household CRT. This provided additional insights into the underlying mechanism of polio transmission that support the global cessation of OPV and emphasises the difficulties of the global eradication of polio. Following this, I performed a spatial reanalysis of an insecticide-treated bed net CRT, applying approaches from the systematic review and a new method I developed called cluster reallocation to assess the presence and impact of spatial spillover in the trial. This analysis confirmed the previous estimate of intervention effect while showing evidence of a spillover effect. I carried out simulation studies to evaluate the impact of spillover and spatial effects on the standard CRT model and compared spatial regression to non-spatial models. These simulations focus on how to generate spatial spillover effects and the magnitude needed before spatial consideration becomes important to CRTs. I found that non-spatial CRT models are relatively robust to spatial effects and that the use of spatial models does not appear to improve upon the non-spatial model. The collective findings of this thesis highlight that standard CRT approaches are typically robust to small scale spillover effects and consideration of the spatial distribution of observations appears to provide little utility in the main analysis of a trial. Despite this, spatial methods can provide additional insights into the mechanism of interventions and are well suited to secondary analyses of CRTs, especially with the increasing collection of GPS data in CRTs.