Table 4.
Multivariate Odds Ratios and 95% Confidence Intervals for the Joint Effects of Fruit and Vegetable Intake and Tobacco Smoking on Stomach Cancer
| Cases (n = 80) N |
Controls (n = 146) N |
Unadjusted |
Adjustedb |
||
|---|---|---|---|---|---|
| OR (95% CI) | OR (95% CI) | ||||
| Tobacco smoking | Vegetable intakea | ||||
| Ever | ≤ 50% percentile | 37 | 56 | 1.00 (reference) | 1.00 (reference) |
| > 50% percentile | 34 | 52 | 0.99 (0.54–1.80) | 1.23 (0.63–2.40) | |
| Never | ≤ 50% percentile | 6 | 14 | 0.65 (0.23–1.84) | 0.63 (0.21–1.87) |
| > 50% percentile | 3 | 24 | 0.19 (0.05–0.67) | 0.25 (0.06–0.95) | |
| Tobacco smoking | Fruit intakea | ||||
| Ever | ≤ 50% percentile | 42 | 56 | 1.00 (reference) | 1.00 (reference) |
| > 50% percentile | 29 | 52 | 0.74 (0.41–1.36) | 0.81 (0.42–1.57) | |
| Never | ≤ 50% percentile | 5 | 10 | 0.67 (0.21–2.10) | 0.63 (0.19–2.15) |
| > 50% percentile | 4 | 28 | 0.19 (0.06–0.58) | 0.20 (0.06–0.65) | |
aFruit and vegetable intake were categorised into two: low intake (< 50% percentile) and high intake (≥ 50% percentile) based on the distribution of intake among controls; bEstimates were adjusted for age (years, continuous), education levels (primary school, secondary school, high school or higher, unknown), family history of cancer (yes, no), alcohol consumption (ever, never), fruit or vegetable intake (whenever one was not the main exposure), and H. pylori serostatus (negative, equivocal, positive)