Table 4. The association between tobacco types and between tobacco and opium use, alone or in combination, and ESCC.
| ESCC cases | Matched controls | Unadjusted | Adjusted | |
|---|---|---|---|---|
| N (%) | N (%) | OR (95% CI) | OR (95% CI)a | |
| Tobacco type used | ||||
| Never used | 196 (66) | 432 (76) | Referent | Referent |
| Cigarettes only | 41 (14) | 68 (12) | 1.32 (0.83–2.11) | 1.50 (0.92–2.43) |
| Hookah only | 12 (4) | 18 (3) | 1.66 (0.75–3.67) | 1.69 (0.76–3.77) |
| Nass only | 23 (8) | 20 (4) | 3.03 (1.53–5.96) | 2.91 (1.46–5.77) |
| More than one type | 27 (9) | 32 (6) | 2.12 (1.18–3.83) | 2.11 (1.15–3.86) |
| Used neither tobacco nor opium | 166 (56) | 398 (70) | Referent | Referent |
| Used tobacco but not opium | 43 (14) | 66 (12) | 1.68 (1.05–2.68) | 1.70 (1.05–2.73) |
| Used opium but not tobacco | 30 (10) | 34 (6) | 2.22 (1.27–3.87) | 2.12 (1.21–3.74) |
| Used both tobacco and opium | 60 (20) | 72 (13) | 2.20 (1.42–3.40) | 2.35 (1.50–3.67) |
95% CI=95% confidence interval; ESCC=oesophageal squamous cell carcinoma; OR=odds ratio. ORs were obtained from conditional logistic regression models.
a
Adjusted for education, ethnicity, and total intake of fruit and vegetables.