The aim of this systematic review is to summarise the results of cohort studies that examined the incidence of SCD in marathons and to assess the quality of the methods used. A search of the PROSPERO international database revealed no prospective or published systematic reviews investigating SCD in marathons. The review was conducted using studies that reported and characterised the incidence of SCD in people participating in marathons. Studies were identified via electronic database searches (Medline, CINAHL, SPORTDiscus and Google Scholar) from January 1, 1966 to October 1, 2014 and through manual literature searches. 7 studies met the inclusion criteria and were included in this review. 6 of the studies were conducted in the USA and 1 in the UK. These studies covered a 34-year period involving between 215,413 and 3,949,000 runners. The SCD of between 4 and 28 people are recorded in the papers and the reported estimates of the incidence of SCD in marathons ranged widely from 0.6 to 1.9 per 100,000 runners. The proportion of those suffering SCD who were male ranged from 57.1% to 100% and the mean age reported in the papers, ranged from 37 to 48. This review raises 4 methodological concerns over i) collating reports of SCD in marathons; ii) time of death in relation to the marathon; iii) the use of registrants rather than runners in the estimates of sample size and iv) limited detail on runners exercise history.These four concerns all threaten the reliability and interpretation of any estimate of SCD incidence rates in marathons. This review recommends that the methods used to collect data on SCD in marathons be improved and that a central reporting system be established.3