| Histological subtypes of the oesophagus | Adenocarcinoma in subsites of the stomach | |||||||||||
| Squamous cell carcinoma | Adenocarcinoma | Cardia | Distant stomach | NOS | ||||||||
| Cases/person time at risk (years) | RR† (95% CI) | Cases/person time at risk (years) | RR‡ (95% CI) | Cases/person time at risk (years) | RR‡ (95% CI) | Cases/person time at risk (years) | RR§ (95% CI) | Cases/person time at risk (years) | RR†† (95% CI) | |||
| BMI at baseline (kg/m2) | ||||||||||||
| <20.0 | 9/1620 | 2.21 (0.99 to 4.92) | 3/1725 | 1.29 (0.40 to 4.16) | 2/1725 | 0.67 (0.16 to 2.80) | 12/1597 | 1.80 (0.96 to 3.39) | 6/1597 | 0.92 (0.38 to 2.25) | ||
| 20.0–24.9 | 51/24303 | 1.0 (Referent) | 51/25491 | 1.0 (Referent) | 68/25491 | 1.0 (Referent) | 115/24090 | 1.0 (Referent) | 93/24090 | 1.0 (Referent) | ||
| 25.0–29.9 | 26/19175 | 0.63 (0.39 to 1.02) | 60/20172 | 1.40 (0.95 to 2.04) | 76/20172 | 1.32 (0.94 to 1.85) | 99/19029 | 0.97 (0.73 to 1.30) | 67/19029 | 0.85 (0.61 to 1.19) | ||
| ⩾30.0 | 6/3224 | 0.93 (0.38 to 2.26) | 19/3426 | 3.96 (2.27 to 6.88) | 17/3426 | 2.73 (1.56 to 4.79) | 9/3176 | 0.68 (0.34 to 1.35) | 7/3176 | 0.77 (0.35 to 1.68) | ||
| p trend¶ = 0.04 | p trend = 0.001 | p trend = 0.002 | p trend = 0.13 | p trend = 0.33 | ||||||||
| Continuous, 1 kg/m2 increments | 92/48321 | 0.90 (0.82 to 0.98) | 133/50814 | 1.14 (1.08 to 1.21) | 163/50814 | 1.10 (1.04 to 1.16) | 235/47893 | 0.99 (0.94 to 1.04) | 173/47893 | 0.98 (0.93 to 1.04) | ||
| BMI at age 20 years (kg/m2)# | ||||||||||||
| <20.0 | 22/10703 | 1.35 (0.70 to 2.62) | 21/11077 | 1.07 (0.59 to 1.94) | 21/11077 | 0.66 (0.39 to 1.14) | 53/10592 | 1.40 (0.91 to 2.15) | 26/10592 | 0.60 (0.37 to 0.99) | ||
| 20.0–21.4 | 16/10075 | 1.0 (Referent) | 24/10651 | 1.0 (Referent) | 40/10651 | 1.0 (Referent) | 40/10015 | 1.0 (Referent) | 49/10015 | 1.0 (Referent) | ||
| 21.5–22.9 | 11/9886 | 0.72 (0.33 to 1.57) | 37/10289 | 1.61 (0.95 to 2.72) | 39/10289 | 1.02 (0.65 to 1.60) | 49/9813 | 1.24 (0.80 to 1.91) | 40/9813 | 0.92 (0.59 to 1.44) | ||
| 23.0–24.9 | 13/7632 | 1.03 (0.48 to 2.21) | 18/8051 | 1.02 (0.55 to 1.90) | 22/8051 | 0.75 (0.44 to 1.28) | 36/7571 | 1.12 (0.69 to 1.80) | 26/7571 | 0.70 (0.42 to 1.18) | ||
| ⩾25.0 | 12/3152 | 2.49 (1.15 to 5.40) | 13/3287 | 1.97 (0.99 to 3.94) | 16/3287 | 1.47 (0.81 to 2.70) | 20/3133 | 1.60 (0.91 to 2.83) | 12/3133 | 0.82 (0.42 to 1.60) | ||
| p trend = 0.58 | p trend = 0.17 | p trend = 0.17 | p trend = 0.93 | p trend = 0.72 | ||||||||
| Continuous, 1 kg/m2 increments | 74/41449 | 1.07 (0.96 to 1.20) | 113/43356 | 1.04 (0.95 to 1.14) | 138/43356 | 1.07 (1.00 to 1.15) | 198/41123 | 1.01 (0.95 to 1.07) | 153/41123 | 1.01 (0.96 to 1.08) | ||
| Change in BMI after 20 years of age (kg/m2)# | ||||||||||||
| <0 | 18/4119 | 2.57 (1.40 to 4.72) | 8/4308 | 0.75 (0.34 to 1.64) | 10/4308 | 0.68 (0.34 to 1.35) | 17/4097 | 0.77 (0.44 to 1.36) | 16/4097 | 0.85 (0.47 to 1.55) | ||
| 0–3.9 | 32/20134 | 1.0** (Referent ) | 51/21108 | 1.0** (Referent) | 70/21108 | 1.0** (Referent) | 106/20036 | 1.0** (Referent) | 82/20036 | 1.0** (Referent) | ||
| 4.0–7.9 | 16/13529 | 0.73 (0.39 to 1.36) | 37/14152 | 1.34 (0.86 to 2.08) | 45/14152 | 1.22 (0.82 to 1.82) | 61/13350 | 0.85 (0.60 to 1.21) | 45/13350 | 0.85 (0.56 to 1.27) | ||
| ⩾8.0 | 8/3666 | 1.39 (0.62 to 3.15) | 17/3787 | 3.41 (1.88 to 6.18) | 13/3787 | 2.07 (1.08 to 3.97) | 14/3640 | 0.86 (0.46 to 1.59) | 10/3640 | 0.86 (0.41 to 1.80) | ||
| p trend = 0.10 | p trend = 0.001 | p trend = 0.02 | p trend = 0.77 | p trend = 0.70 | ||||||||
| Continuous, 1 kg/m2 increments | 74/41449 | 0.90 (0.81 to 1.00) | 113/43356 | 1.14 (1.06 to 1.23) | 138/43356 | 1.07 (1.00 to 1.14) | 198/41123 | 0.98 (0.93 to 1.04) | 153/41123 | 0.99 (0.93 to 1.05) | ||
*RR, rate ratio; CI, confidence interval; NOS, adenocarcinoma of the stomach, not otherwise specified; BMI, body mass index (kg/m2). The first year of follow-up was excluded from the analyses. Case-cohort analyses were used which mean that the cases were identified from the entire cohort, while the person years were based only on the subcohort.
†Rate ratios adjusted for age; sex; current smoking (yes/no); number of cigarettes/day; and number of years of smoking.
‡Rate ratios adjusted for age and sex.
§Rate ratio adjusted for age; sex; current smoking (yes/no); number of cigarettes/day; number of years of smoking; and highest level of education (primary school, lower vocational school, high school and higher vocational/university).
††Rate ratio adjusted for age; sex; current smoking (yes/no); number of cigarettes/day; number of years of smoking; highest level of education (primary school, lower vocational school, high school and higher vocational/university); and a history of gastric ulcer or bleeding (yes/no).
¶Tests for dose-respond trends were assessed by fitting ordinal variables as continuous terms in the Cox proportional hazard model. p Trends were calculated for all categories of a variable. The reported p values are two sided.
#Only cases and subcohort members with complete data on BMI at age 20 years were included.
**Rate ratios also adjusted for BMI at age 20 years.