Table 3.
Stratification analyses for the associations between BMI and risk of ESCC and GC in the Linxian dysplasia population trial cohort.
ESCC |
GC |
|||||
---|---|---|---|---|---|---|
Normal (≥18.5 to <24.0) | Underweight (< 18.5) | Overweight or obese (≥24.0) | Normal (≥18.5 to <24.0) | Underweight (< 18.5) | Overweight or obese (≥24.0) | |
Age at baselinea | ||||||
<55 years | 1.00 | 1.35(0.81–2.24) | 0.67(0.42–1.06) | 1.00 | 1.10(0.76–1.60) | 1.08(0.64–1.81) |
≥55 years | 1.00 | 1.09(0.59–1.86) | 0.72(0.42–1.25) | 1.00 | 0.95(0.69–1.30) | 1.04(0.60–1.81) |
P for interaction | 0.243 | 0.930 | ||||
Genderb | ||||||
Men | 1.00 | 1.13(0.55–2.03) | 0.75(0.38–1.46) | 1.00 | 0.72(0.43–1.32) | 1.58(0.93–2.69) |
Women | 1.00 | 1.30(0.84–2.04) | 0.66(0.44–0.99) | 1.00 | 1.12(0.62–2.01) | 0.75(0.44–1.28) |
P for interaction | 0.020 | 0.047 | ||||
Smokingc | ||||||
Former | 1.00 | 1.05(0.58–2.08) | 0.57(0.14–2.40) | 1.00 | 0.66(0.15–1.79) | 1.35(0.77–2.39) |
Current | 1.00 | 1.08(0.42–1.89) | 0.78(0.25–2.45) | 1.00 | 0.82(0.55–1.23) | 1.19(0.72–1.97) |
Never | 1.00 | 1.27(0.85–1.94) | 0.69(0.47–1.01) | 1.00 | 0.83(0.61–1.15) | 0.96(0.61–1.49) |
P for interaction | 0.545 | 0.235 | ||||
Drinkingd | ||||||
Yes | 1.00 | 1.04(0.42–2.60) | 0.83(0.36–1.91) | 1.00 | 0.81(0.46–1.40) | 0.88(0.35–2.19) |
No | 1.00 | 1.28(0.83–1.94) | 0.65(0.34–1.05) | 1.00 | 0.93(0.58–1.51) | 1.12(0.74–1.69) |
P for interaction | 0.810 | 0.668 | ||||
Family history of cancere | ||||||
Yes | 1.00 | 1.09(0.83–1.44) | 0.58(0.35–0.99) | 1.00 | 0.98(0.68–1.43) | 1.10(0.63–1.91) |
No | 1.00 | 1.25(0.78–2.01) | 0.76(0.48–1.22) | 1.00 | 0.75(0.55–1.68) | 1.03(0.62–1.73) |
P for interaction | 0.003 | 0.992 |
Bold text indicates statistical significance. BMI, body mass index. ESCC, Esophageal squamous cell carcinoma. GC, Gastric cancer, including gastric non-cardia carcinoma (GNCC) and cardia carcinoma (GCC).
Adjusted for gender, smoking, drinking, family history of cancer, education and consumption of fresh fruit.
Adjusted for age at baseline, smoking, drinking, family history of cancer, education and consumption of fresh fruit.
Adjusted for age at baseline, gender, drinking, family history of cancer, education and consumption of fresh fruit.
Adjusted for age at baseline, gender, smoking, family history of cancer, education and consumption of fresh fruit.
Adjusted for age at baseline, gender, smoking, drinking, education and consumption of fresh fruit.