Table 3.
Association of fish intake with risk of liver cancer and its major histological subtypes.
| Oily fish | Non-oily fish | |||
|---|---|---|---|---|
| No. of cases/ participants | HR (95% CI) | No. of cases/ participants | HR (95% CI) | |
| Liver cancer | ||||
| Never | 38/47,713 | 1.00 (Referent) | 16/20,372 | 1.00 (Referent) |
| <1 serving/ week | 87/144,988 | 0.87 (0.57–1.35) | 75/126,445 | 0.85 (0.46–1.57) |
| 1 serving/week | 93/164,110 | 0.77 (0.49–1.20) | 129/216,732 | 0.88 (0.48–1.61) |
| ≥2 servings/ week | 44/77,773 | 0.68 (0.41–1.13) | 42/71,035 | 0.87 (0.45–1.69) |
| P-trend | 0.087 | 0.83 | ||
| Hepatocellular carcinoma | ||||
| Never | 22/47,713 | 1.00 (Referent) | 8/20,372 | 1.00 (Referent) |
| <1 serving/ week | 41/144,988 | 0.75 (0.42–1.37) | 39/126,445 | 1.01 (0.43–2.38) |
| 1 serving/week | 46/164,110 | 0.70 (0.38–1.28) | 58/216,732 | 0.98 (0.41–2.30) |
| ≥2 servings/ week | 18/77,773 | 0.46 (0.23–0.96) | 22/71,035 | 1.12 (0.44–2.82) |
| P-trend | 0.027 | 0.84 | ||
| Intrahepatic cholangiocarcinoma | ||||
| Never | 13/47,713 | 1.00 (Referent) | 7/20,372 | 1.00 (Referent) |
| <1 serving/ week | 35/144,988 | 1.00 (0.48–2.09) | 27/126,445 | 0.61 (0.23-1.61) |
| 1 serving/week | 39/164,110 | 0.88 (0.41–1.85) | 60/216,732 | 0.76 (0.29–1.95) |
| ≥2 servings/ week | 23/77,773 | 1.06 (0.47–2.36) | 16/71,035 | 0.62 (0.22–1.74) |
| P-trend | 0.96 | 0.71 | ||
Results were adjusted for age (y), sex, ethnic group (White, other), Townsend deprivation index, smoking status (never, former, current), pack-years of smoking (for current smokers), alcohol consumption [never, former, current (<1, 1–2, ≥3 drinks/week)], total physical activity (MET-h/week), dietary factors (consumption of cereal, fresh fruit, fresh vegetables, red meat, processed meat, oily fish or non-oily fish for each other, and coffee), body mass index (kg/m2), diabetes, blood measures of liver function (z scores of alanine aminotransferase and total bilirubin), and fish oil use.