Aspirin . . . and other stories

A couple of weeks ago, Minerva mentioned a large trial of low dose aspirin in asymptomatic older people. It turned out that aspirin had no benefit where the primary outcome—disability-free survival—was concerned. Worse, it carried a small increase in the risk of intracranial bleeding. During the five years of the trial, more than a third of participants in the aspirin group stopped taking their allocated treatment. A re-analysis of trial data, adjusting for compliance, still finds no benefit from aspirin on survival but discovers that for haemorrhage, cancer mortality, and all-cause mortality, the risks are greater than estimated by the intention-to-treat analysis (Am J Epidemiol doi:10.1093/aje/kwad168).