Height and body mass index in relation to esophageal cancer; 23-year follow-up of two million Norwegian men and women Anders Engeland 1, *, Steinar Tretli 2 & Tone Bjørge 3 1 Division of Epidemiology, Norwegian Institute of Public Health, P.O.Box 4404 Nydalen, N-0403 Oslo, Norway; 2 The Cancer Registry of Norway, Institute of population-based cancer research, Montebello, N-0310 Oslo, Norway; 3 Department of Pathology The Norwegian Radium Hospital, Montebello, N-0310 Oslo, Norway Received 08 February 2004; accepted in revised form 13 May 2004 Key words: body mass index, BMI, cohort study, esophageal cancer, height, Norway. Abstract Objective: Associations between body mass index (BMI) and stature and cancers at different sites have been explored in a number of studies. For esophageal cancer there seems to be different effects of BMI for different histological subtypes. We explored these relations in a Norwegian cohort. Material and methods: Height and weight were measured in 2 million Norwegians during 1963–2001. During follow-up, 2245 histologically verified esophageal cancer cases were registered. Relative risks (RR) of esophageal cancer were estimated using proportional Cox regression. Results: Compared with normal weighted (BMI 18.5–24.9 kg/m 2 ) an increased risk of esophageal adenocarcinoma (OA) was observed in overweight men (BMI 25–29 kg/m 2 ): RR ¼ 1.80 (95% CI: 1.48–2.19) and in obese men (BMI ‡30 kg/m 2 ): RR ¼ 2.58 (95% CI: 1.81–3.68). The corresponding risk estimates for women were RR ¼ 1.64 (95% CI: 1.08–2.49) and RR ¼ 2.06 (95% CI: 1.25–3.39). The opposite relation was observed for esophageal squamous cell carcinoma (OSCC). For overweight men the RR of OSCC was 0.72 (95% CI: 0.63–0.82) and 0.68 (95% CI: 0.50– 0.93) for obese. The corresponding RR estimates for women were 0.52 (95% CI: 0.42–0.65) and 0.43 (95% CI: 0.32– 0.59). In addition, the lowest men had the highest risk of esophageal cancer in general. Adjustment for smoking did not change these relations. Conclusion: BMI had opposite relations to the two most common histological groups of esophageal cancer. While low BMI increased the risk of OSCC, high BMI increased the risk of OA. An increased risk of esophageal cancer was found in the lowest men. Introduction Associations between body mass index (BMI) and stature and cancers at different sites have been explored in a large number of studies [1, 2]. The most consistent relations between obesity and risk of cancer have been found for cancers of the colon, breast (in postmeno- pausal women), endometrium, kidney, esophagus, and gastric cardia [1]. Tall persons seem to have a 20–60% increased risk of cancers at many sites (colon, rectum, prostate, breast, endometrium and thyroid) [2]. In the US, it has been estimated that 90,000 cancer deaths could be prevented each year if men and women could maintain ‘normal’ weight [3]. Esophageal cancer is the eighth most common cancer worldwide [4], and is also one of the most lethal cancers, with a 5-year relative survival rate in Norway of less than 10% [5]. The incidence of the most common histologic type of esophageal cancer, squamous cell carcinoma (OSCC), has been relatively stable in western countries [6]. The incidence of esophageal adenocarci- nomas (OA), however, has increased rapidly in the western world during the last 20 years [7, 8]. This pattern has also been observed in Norway (9). Tobacco and alcohol consumption are the primary causes of OSCC [10, 11]. Tobacco smoking is also *Address correspondence to: Anders Engeland, Division of Epide- miology, Norwegian Institute of Public Health, P.O. Box 4404, Nydalen, N-0403 Oslo, Norway; Ph.: +47-23-40 82 13; Fax: +47- 23-40 82 52; E-mail: anders.engeland@fhi.no Cancer Causes and Control 15: 837–843, 2004. 837 Ó 2004 Kluwer Academic Publishers. Printed in the Netherlands.