Table 2.
Subgroup analysis of plant-based diets and cardiovascular mortality, cancer mortality, and all-cause mortality.
| Subgroup analysis of the relationship between of PDI, hPDI, uPDI and cancer mortality, CVD mortality and all mortality | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Subgroup | No. of studies | RR (95% CI) | I2% | p for heterogeneity | No. of studies | RR (95% CI) | I2% | p for heterogeneity | No. of studies | RR (95% CI) | I2% | p for heterogeneity |
| PDI | hPDI | uPDI | ||||||||||
| Cancer mortality | 7 | 0.88 (0.79, 0.98) | 84.71% | 0.02 | 8 | 0.91 (0.83, 0.99) | 85.61% | 0.03 | 7 | 1.10 (0.97, 1.26) | 93.11% | 0.14 |
| Location | ||||||||||||
| Korean | 1 | 0.88 (0.75, 1.03) | – | – | 1 | 1.00 (0.84, 1.19) | – | – | 1 | 1.23 (1.02, 1.48) | – | – |
| UK | 1 | 1.04 (0.95, 1.13) | – | – | 1 | 0.90 (0.82, 0.99) | – | – | 1 | 1.16 (1.06, 1.27) | – | – |
| US | 5 | 0.85 (0.75, 0.95) | 83.25% | <0.01 | 6 | 0.90 (0.80, 1.00) | 89.44% | <0.01 | 5 | 1.07 (0.90, 1.28) | 95.38% | <0.01 |
| Age | ||||||||||||
| ≤55 | 2 | 0.78 (0.61, 1.01) | 71.60% | 0.06 | 3 | 0.91 (0.86, 0.95) | 0 | 0.52 | 2 | 1.18 (1.02, 1.36) | 0.00% | 0.46 |
| >55 | 5 | 0.91 (0.81, 1.03) | 87.89% | <0.01 | 5 | 0.90 (0.78, 1.03) | 91.55% | <0.01 | 5 | 1.09 (0.93, 1.27) | 95.37% | <0.01 |
| Total | ||||||||||||
| ≤100,000 | 2 | 0.84 (0.56, 1.26) | 87.46% | <0.01 | 2 | 0.98 (0.86, 1.12) | 11.34% | 0.29 | 2 | 1.01 (0.89, 1.14) | 0.00% | 0.37 |
| >100,000 | 5 | 0.89 (0.79, 1.00) | 87.11% | <0.01 | 6 | 0.89 (0.81, 0.99) | 89.06% | <0.01 | 5 | 1.13 (0.97, 1.32) | 95.10% | <0.01 |
| Female | ||||||||||||
| ≤50% | 2 | 0.80 (0.69, 0.93) | 84.27% | 0.01 | 2 | 0.81 (0.56, 1.18) | – | – | 2 | 1.15 (0.83, 1.60) | 98.40% | <0.01 |
| >50% | 4 | 0.98 (0.91, 1.05) | 32.72% | 0.22 | 5 | 0.93 (0.89, 0.97) | 23.12% | 0.27 | 4 | 1.08 (0.97, 1.19) | 70.20% | 0.02 |
| Follow up time | ||||||||||||
| ≤8 | 2 | 0.73 (0.67, 0.80) | 0.00% | 0.48 | 2 | 0.76 (0.58, 1.00) | 80.87% | 0.02 | 2 | 1.26 (1.04, 1.54) | 67.40% | 0.08 |
| >8 | 4 | 0.96 (0.87, 1.05) | 76.02% | 0.01 | 5 | 0.94 (0.90, 0.98) | 39.58% | 0.16 | 4 | 1.02 (0.94, 1.11) | 71.74% | 0.02 |
| CVD mortality | 7 | 0.81 (0.76, 0.86) | 0.00% | <0.01 | 9 | 0.85 (0.77, 0.94) | 85.13% | <0.01 | 8 | 1.19 (1.07, 1.32) | 80.03% | <0.01 |
| Location | ||||||||||||
| Korean | 1 | 0.77 (0.57, 1.04) | – | – | 1 | 1.23 (0.90, 1.68) | – | – | 1 | 1.55 (1.07, 2.24) | – | – |
| Spain | – | – | – | – | 1 | 0.42 (0.20, 0.88) | – | – | 1 | 1.11 (0.56, 2.19) | – | – |
| UK | 1 | 0.77 (0.66, 0.90) | – | – | 1 | 0.92 (0.79, 1.08) | – | – | 1 | 1.10 (0.93, 1.30) | – | – |
| US | 5 | 0.81 (0.75, 0.87) | 63.50% | 0.03 | 6 | 0.83 (0.74, 0.92) | 88.46% | <0.01 | 5 | 1.18 (1.04, 1.34) | 87.60% | <0.01 |
| Age | ||||||||||||
| ≤55 | 3 | 0.75 (0.65, 0.85) | 19.90% | 0.29 | 5 | 0.94 (0.80, 1.10) | 65.19% | 0.02 | 4 | 1.22 (0.90, 1.65) | 76.41% | 0.01 |
| >55 | 4 | 0.83 (0.78, 0.88) | 49.41% | 0.12 | 4 | 0.79 (0.72, 0.86) | 73.70% | 0.01 | 4 | 1.19 (1.06, 1.34) | 85.81% | <0.01 |
| Total | ||||||||||||
| ≤100,000 | 2 | 0.75 (0.61, 0.92) | 58.62% | 0.12 | 3 | 0.83 (0.59, 1.16) | 73.23% | 0.02 | 3 | 1.13 (0.81, 1.59) | 77.54% | 0.01 |
| >100,000 | 5 | 0.82 (0.78, 0.87) | 35.28% | 0.19 | 5 | 0.85 (0.76, 0.95) | 89.16% | <0.01 | 5 | 1.21 (1.08, 1.36) | 82.83% | <0.01 |
| Female | ||||||||||||
| ≤50% | 2 | 0.81 (0.74, 0.88) | 56.82% | 0.13 | 2 | 0.75 (0.65, 0.86) | 83.96% | 0.01 | 2 | 1.23 (0.95, 1.60) | 95.10% | <0.01 |
| >50% | 4 | 0.78 (0.68, 0.90) | 67.89% | 0.03 | 6 | 0.88 (0.78, 0.99) | 79.72% | <0.01 | 5 | 1.12 (0.97, 1.29) | 64.72% | 0.02 |
| Follow up time | ||||||||||||
| ≤8 | 2 | 0.78 (0.72, 0.85) | 0.00% | 0.46 | 2 | 0.85 (0.55, 1.34) | 91.15% | <0.01 | 2 | 1.41 (1.29, 1.54) | 0 | 0.96 |
| >8 | 4 | 0.81 (0.74, 0.88) | 67.34% | 0.03 | 6 | 0.84 (0.76, 0.92) | 80.38% | <0.01 | 5 | 1.09 (1.00, 1.18) | 58.74% | 0.05 |
| All mortality | 13 | 0.84 (0.79, 0.89) | 81.99% | <0.01 | 15 | 0.85 (0.80, 0.90) | 89.83% | <0.01 | 14 | 1.18 (1.09, 1.27) | 89.97% | <0.01 |
| Location | ||||||||||||
| German | 1 | 0.46 (0.29, 0.74) | – | – | 1 | 0.76 (0.51, 1.14) | – | – | 1 | 1.29 (0.84, 1.98) | – | – |
| China | 2 | 0.94 (0.83, 1.07) | 13.68% | 0.28 | 2 | 0.80 (0.73, 0.88) | 5.63% | 0.3 | 2 | 1.18 (1.09, 1.26) | 0.00% | 0.47 |
| Korean | 1 | 0.76 (0.68, 0.85) | – | – | 1 | 1.05 (0.93, 1.18) | – | – | 1 | 1.30 (1.15, 1.47) | – | – |
| Spain | – | – | – | – | 1 | 0.74 (0.52, 1.05) | – | – | 1 | 1.18 (0.88, 1.58) | – | – |
| UK | 1 | 0.87 (0.81, 0.93) | – | – | 1 | 0.92 (0.86, 0.98) | – | – | 1 | 1.29 (1.20, 1.38) | – | – |
| US | 8 | 0.84 (0.78, 0.90) | 85.09% | <0.01 | 9 | 0.84 (0.78, 0.91) | 93.00% | <0.01 | 8 | 1.14 (1.03, 1.25) | 93.55% | <0.01 |
| Age | ||||||||||||
| ≤55 | 4 | 0.81 (0.73, 0.90) | 69.13% | 0.02 | 6 | 0.90 (0.84, 0.96) | 60.40% | 0.04 | 5 | 1.17 (0.98, 1.39) | 87.36% | <0.01 |
| >55 | 9 | 0.85 (0.80, 0.91) | 83.58% | <0.01 | 9 | 0.82 (0.75, 0.90) | 93.33% | <0.01 | 9 | 1.18 (1.08, 1.29) | 91.80% | <0.01 |
| Total | ||||||||||||
| ≤100,000 | 8 | 0.85 (0.75, 0.96) | 80.60% | <0.01 | 9 | 0.85 (0.81, 0.90) | 23.04% | 0.24 | 9 | 1.14 (1.04, 1.26) | 73.47% | <0.01 |
| >100,000 | 5 | 0.83 (0.77, 0.88) | 86.70% | <0.01 | 6 | 0.85 (0.78, 0.94) | 96.67% | <0.01 | 5 | 1.22 (1.08, 1.37) | 95.92% | <0.01 |
| Female | ||||||||||||
| ≤50% | 4 | 0.79 (0.69, 0.91) | 86.11% | <0.01 | 4 | 0.74 (0.58, 0.94) | 96.87% | <0.01 | 4 | 1.25 (0.97, 1.59) | 96.30% | <0.01 |
| >50% | 8 | 0.86 (0.81, 0.92) | 78.18% | <0.01 | 10 | 0.88 (0.85, 0.92) | 59.46% | 0.01 | 9 | 1.14 (1.06, 1.22) | 80.15% | <0.01 |
| Follow up time | ||||||||||||
| ≤8 | 5 | 0.81 (0.71, 0.93) | 87.07% | <0.01 | 5 | 0.75 (0.64, 0.88) | 90.24% | <0.01 | 5 | 1.30 (1.17, 1.45) | 74.80% | <0.01 |
| >8 | 7 | 0.87 (0.81, 0.92) | 75.59% | <0.01 | 9 | 0.87 (0.86, 0.89) | 7.01% | 0.38 | 8 | 1.10 (1.02, 1.18) | 84.24% | <0.01 |
Heterogeneity analysis was conducted on the basis of region, age, proportion of female participants, total number of participants, and follow-up duration. RR and 95% confidence intervals (CIs) were calculated. I2 ≤ 50% and I2 > 50% indicate significant and nonsignificant heterogeneity, respectively. Statistically significant RR values are bolded in the table.