This thesis examines several aspects of the debate on the choice of best firstline treatment for infantile spasms. The first part of the thesis describes, analyses and interprets outcomes from the United Kingdom Infantile Spasms Study (UKISS), a multicentre randomised controlled trial that compares treatment with prednisolone or tetracosactide (hormonal treatments) against vigabatrin. This study was coordinated from the Bath Unit for Research in Paediatrics at the Royal United Hospital, Bath, and showed that the early primary clinical response, cessation of spasms, was significantly more likely in infants who were allocated hormonal treatments. However, the overall proportions with relapse-free response at the end of the study period, when the child reached the age of 12 to 14 months, were similar in all three treatment groups. In a group of infants of a priori interest who had no identified underlying aetiology, neurodevelopmental outcomes were significantly better in children who had been allocated hormonal treatments, although this finding was not robust within a sensitivity analysis. The second part of the thesis examines problems with case definitions and outcome measures in epidemiological and interventional studies of infantile spasms, and describes the development of West Delphi, a consensus elicitation process that has produced published proposals for standardised case definitions and outcome measures for such studies. Its most novel proposals were: (1) that the primary clinical outcome cessation of spasms should be defined by the absence of observed spasms for at least 28 days from the time of the last observed spasms, with the last spasms being observed within 14 days of treatment allocation; and (2) that there should be a standardised primary electroclinical response that studies should also report as a primary outcome.