TABLE 3.
Summary of results of the associations on tea consumption and risk of cancer outlined by study design, largest meta-analysis of observational studies, and largest meta-analysis of cohort studies1
| Cancer type | Study design | Metrics | Summary effect (random-effects)6 | P value (random-effects) | 95% PI including null | Heterogeneity (I2)2 | Small study effect3 | Concordance | Level of evidence4 |
|---|---|---|---|---|---|---|---|---|---|
| Oral cancer (any tea; any vs. none) | CC | OR | 0.62 (0.55, 0.72) | <0.001 | No | Not large | No | Yes | Convincing |
| Oral cancer (any tea; high vs. low) | Obs | RR | 0.86 (0.79, 0.93) | <0.001 | No | Not large | No | Yes | Suggestive |
| Co | RR | 0.80 (0.67, 0.94) | 0.007 | Yes | Not large | No | Yes | Suggestive | |
| CC | RR | 0.87 (0.79, 0.96) | 0.004 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Obs) | RR | 0.84 (0.75, 0.94) | 0.002 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Co) | NR | ||||||||
| Biliary tract cancer (any tea; any vs. none) | Obs | RR | 0.77 (0.64, 0.92) | 0.004 | Yes | Not large | No | Yes | Suggestive |
| Co | RR | 0.82 (0.70, 0.95) | 0.008 | Yes | Not large | No | Yes | Suggestive | |
| CC | RR | 0.66 (0.42, 1.03) | 0.068 | Yes | Not large | No | No | Nonsignificant | |
| Largest MA (Obs) | RR | 0.77 (0.64, 0.92) | 0.004 | Yes | Not large | No | Yes | Suggestive | |
| Largest MA (Co) | NR | ||||||||
| Endometrial cancer (green tea; high vs. low) | Obs | RR | 0.78 (0.61, 1.00)5 | 0.046 | Yes | Not large | No | No | Suggestive |
| Co | RR | 0.75 (0.44, 1.30) | 0.298 | N/A | N/A | N/A | N/A | Nonsignificant | |
| CC | RR | 0.78 (0.57, 1.06) | 0.108 | Yes | Not large | No | Yes | Nonsignificant | |
| Largest MA (Obs) | RR | 0.78 (0.61, 1.00)5 | 0.046 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Co) | NR | ||||||||
| Liver cancer (green tea; high vs. low) | Obs | RR | 0.87 (0.78, 0.98) | 0.026 | Yes | Not large | No | No | Suggestive |
| Co | RR | 0.87 (0.78, 0.98) | 0.026 | Yes | Not large | No | No | Suggestive | |
| CC | NR | ||||||||
| Largest MA (Obs) | RR | 0.87 (0.78, 0.98) | 0.026 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Co) | RR | 0.87 (0.78, 0.98) | 0.026 | Yes | Not large | No | No | Suggestive | |
| Oral cancer (green tea; high vs. low) | Obs | RR | 0.82 (0.69, 0.96) | 0.015 | Yes | Not large | No | No | Suggestive |
| Co | RR | 0.44 (0.19, 1.04) | 0.058 | N/A | N/A | N/A | N/A | N/A | |
| CC | RR | 0.84 (0.72, 0.98) | 0.030 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Obs) | RR | 0.82 (0.69, 0.96) | 0.015 | Yes | Not large | No | No | Suggestive | |
| Largest MA (Co) | NR | ||||||||
| Breast cancer (green tea; high vs. low) | Obs | RR | 0.82 (0.71, 0.96) | 0.015 | Yes | Large | No | No | Weak |
| Co | RR | 0.99 (0.83, 1.18) | 0.895 | Yes | Not large | No | Yes | Nonsignificant | |
| CC | RR | 0.75 (0.61, 0.92) | 0.006 | Yes | Large | No | Yes | Weak | |
| Largest MA (Obs) | RR | 0.81 (0.67, 0.98) | <0.001 | Yes | Large | No | Yes | Weak | |
| Largest MA (Co) | RR | 0.98 (0.83, 1.16) | 0.821 | Yes | Not large | No | Yes | Nonsignificant | |
| Breast cancer (green tea; any vs. none) | Obs | OR | 0.87 (0.76, 0.99) | 0.040 | Yes | Very large | No | Yes | Weak |
| Co | OR | 0.94 (0.83, 1.05) | 0.278 | Yes | Not large | No | Yes | Nonsignificant | |
| CC | OR | 0.83 (0.62, 1.10) | 0.196 | Yes | Very large | No | No | Nonsignificant | |
| Largest MA (Obs) | OR | 0.87 (0.76, 0.99) | 0.040 | Yes | Very large | No | Yes | Weak | |
| Largest MA (Co) | OR | 0.94 (0.83, 1.05) | 0.278 | Yes | Not large | No | Yes | Nonsignificant | |
| Breast cancer (any tea; any vs. none) | Obs | RR | 0.81 (0.71, 0.94) | <0.001 | Yes | Very large | No | No | Weak |
| Co | RR | 1.06 (0.91, 1.23) | 0.451 | Yes | Not large | No | Yes | Nonsignificant | |
| CC | RR | 0.74 (0.63, 0.88) | <0.001 | Yes | Very large | No | No | Weak | |
| Largest MA (Obs) | RR | 0.79 (0.65, 0.95) | 0.012 | Yes | Very large | No | No | Weak | |
| Largest MA (Co) | NR | ||||||||
| Colorectal cancer (any tea; high vs. low) | Obs | RR | 0.93 (0.87, 0.99) | 0.031 | Yes | Not large | Yes | No | Weak |
| Co | RR | 0.94 (0.88, 1.01) | 0.108 | Yes | Not large | No | Yes | Nonsignificant | |
| CC | RR | 0.91 (0.80, 1.03) | 0.117 | Yes | Large | Yes | Yes | Nonsignificant | |
| Largest MA (Obs) | RR | 0.93 (0.87, 0.99) | 0.031 | Yes | Not large | Yes | No | Weak | |
| Largest MA (Co) | RR | 0.94 (0.88, 1.01) | 0.108 | Yes | Not large | No | Yes | Nonsignificant | |
| Colorectal cancer (green tea; high vs. low) | Obs | RR | 0.87 (0.75, 1.00)5 | 0.050 | Yes | Large | No | No | Weak |
| Co | RR | 0.93 (0.79, 1.10) | 0.408 | Yes | Large | No | Yes | Nonsignificant | |
| CC | RR | 0.73 (0.60, 0.90) | 0.003 | Yes | Not large | No | Yes | Weak | |
| Largest MA (Obs) | RR | 0.95 (0.81, 1.11) | 0.493 | Yes | Very large | Yes | Yes | Nonsignificant | |
| Largest MA (Co) | NR | ||||||||
| Gastric cancer (any tea; any vs. none) | Obs | RR | 0.78 (0.70, 0.86) | <0.001 | Yes | Very large | Yes | No | Weak |
| Co | RR | 1.01 (0.91, 1.13) | 0.791 | Yes | Not large | No | Yes | Nonsignificant | |
| CC | RR | 0.70 (0.62, 0.80) | <0.001 | Yes | Very large | Yes | Yes | Weak | |
| Largest MA (Obs) | RR | 0.76 (0.72, 0.80) | 0.010 | Yes | Very large | Yes | Yes | Weak | |
| Largest MA (Co) | NR | ||||||||
| Leukemia (any tea; high vs. low) | Obs | RR | 0.55 (0.43, 0.72) | <0.001 | No | Not large | No | No | Weak |
| Co | NR | ||||||||
| CC | RR | 0.52 (0.38, 0.72) | <0.001 | No | Not large | No | No | Weak | |
| Largest MA (Obs) | RR | 0.55 (0.43, 0.72) | <0.001 | No | Not large | No | No | Weak | |
| Largest MA (Co) | NR | ||||||||
| Leukemia (any tea; any vs. none) | Obs | RR | 0.76 (0.65, 0.89) | <0.001 | No | Not large | No | No | Weak |
| Co | NR | ||||||||
| CC | RR | 0.69 (0.54, 0.87) | 0.002 | No | Not large | No | No | Weak | |
| Largest MA (Obs) | RR | 0.76 (0.65, 0.89) | <0.001 | No | Not large | No | No | Weak | |
| Largest MA (Co) | NR | ||||||||
| Liver cancer (green tea; any vs. none) | Obs | RR | 0.65 (0.48, 0.88) | 0.004 | Yes | Large | No | No | Weak |
| Co | RR | 0.83 (0.61, 1.11) | 0.205 | Yes | Large | No | Yes | Nonsignificant | |
| CC | RR | 0.46 (0.29, 0.74) | 0.001 | Yes | Large | No | Yes | Weak | |
| Largest MA (Obs) | RR | 0.65 (0.48, 0.88) | 0.004 | Yes | Large | No | No | Weak | |
| Largest MA (Co) | NR | ||||||||
| Lung cancer (any tea; any vs. none) | Obs | RR | 0.76 (0.67, 0.86) | <0.001 | Yes | Very large | No | Yes | Weak |
| Co | RR | 0.91 (0.76, 1.08) | 0.273 | Yes | Very large | No | No | Nonsignificant | |
| CC | RR | 0.69 (0.59, 0.79) | <0.001 | Yes | Very large | No | No | Weak | |
| Largest MA (Obs) | RR | 0.77 (0.68, 0.88) | <0.001 | Yes | Very large | No | No | Weak | |
| Largest MA (Co) | NR | ||||||||
| Ovarian cancer (any tea; any vs. none) | Obs | RR | 0.82 (0.71, 0.94) | 0.006 | Yes | Very large | No | No | Weak |
| Co | RR | 0.92 (0.79, 1.06) | 0.260 | Yes | Not large | Yes | Yes | Nonsignificant | |
| CC | RR | 0.76 (0.61, 0.95) | 0.014 | Yes | Very large | No | No | Weak | |
| Largest MA (Obs) | RR | 0.89 (0.80, 1.00)5 | 0.045 | Yes | Not large | Yes | Yes | Weak | |
| Largest MA (Co) | RR | 0.71 (0.55, 0.93) | 0.013 | Yes | Not large | No | Yes | Suggestive | |
| Thyroid cancer (any tea; high vs. low) | Obs | RR | 0.76 (0.61, 0.96) | 0.024 | Yes | Not large | Yes | No | Weak |
| Co | RR | 0.88 (0.56, 1.37) | 0.568 | N/A | N/A | N/A | Yes | N/A | |
| CC | RR | 0.74 (0.56, 0.97) | 0.029 | Yes | Large | Yes | No | Weak | |
| Largest MA (Obs) | RR | 0.76 (0.61, 0.96) | 0.024 | Yes | Not large | Yes | No | Weak | |
| Largest MA (Co) | NR | ||||||||
1
CC, case-control studies; Co, cohort studies; MA, meta-analysis; N/A, not applicable; NR, not reported; Obs, observational studies; PI, prediction interval.
2
Heterogeneity is defined as “Very large” when I2 > 75%, “Large” when 50% < I2 < 75%, and “Not large” when I2 < 50%.
3
The presence of small study effects is determined if the Egger P value is <0.10.
4
The definition of each category of the level of evidence is presented in Supplemental Table 1.
5
The value is rounded up (to 2 decimal places), and hence is statistically significant.
6
Summary effect with 95% CI value obtained from umbrella review combining meta-analyses of the same comparison.