The incidence of oesophageal adenocarcinoma has increased over 500% in some countries during the last three decades, but the reasons for this rise are unclear.1 The most intractable problem to date has been explaining why the rates for oesophageal adenocarcinoma vary substantially by race (more common in Caucasians), gender (more common in men), and geography.1^–6 The increasing prevalence of obesity in society may provide a partial explanation for the increases in incidence. The article by Merry et al. in this issue of Gut describes a well-designed cohort study which evaluates whether measures of body size, such as height and the body mass index (BMI), are associated with the subsequent risk of oesophageal and gastric cardia adenocarcinomas (see page 1503).7 Its strengths include its cohort design, high-quality data, the ability to evaluate BMI early in adult life (when carcinogenic processes such as Barrett’s oesophagus may begin) and with weight gain, the completeness of cancer follow-up, the large size (which permitted stratification by gender), and its evaluation of different cancer types within a single population. The study adds knowledge particularly in two areas. First, it found that a high baseline adult BMI and increases in BMI after the age of 20, but not BMI at age 20, were strongly associated with oesophageal adenocarcinoma and cardia adenocarcinoma; this suggests that interventions aimed at preventing weight gain in adulthood may decrease the risk of oesophageal and cardia adenocarcinomas. Second, it provided detailed evaluations of height, which demonstrate no clear association between height and oesophageal or cardia adenocarcinomas. This finding suggests it is excess body weight for any given height, rather than a …