Group | No. of studies | WMD (95% CI) | P within group | P between group* | Effect model | P heterogeneity | I2, % |
---|---|---|---|---|---|---|---|
Total | 4 | 2.20 (1.21, 3.18) | < 0.001 | < 0.001 | Random | < 0.001 | 99.7 |
Study quality | |||||||
High | 1 | 0.45 (0.34, 0.57) | < 0.001 | Fixed | – | – | |
Moderate | 3 | 2.82 (1.19, 4.46) | 0.001 | Random | < 0.001 | 99.2 | |
Design | < 0.001 | ||||||
Case- control | 3 | 2.90 (−0.39, 6.20) | 0.085 | Random | < 0.001 | 99.7 | |
Nested case-control | 1 | 0.20 (0.15, 0.24) | < 0.001 | – | |||
Sample | < 0.001 | ||||||
Plasma | 2 | 0.49 (−0.08, 1.07) | 0.094 | Random | < 0.001 | 99.4 | |
Serum | 2 | 3.96 (−0.08, 8.01) | 0.05 | Random | < 0.001 | 99.5 | |
Sample size | < 0.001 | ||||||
< 550 | 2 | 3.96 (−0.08, 8.01) | 0.055 | Random | < 0.001 | 99.5 | |
≥ 550 | 2 | 0.49 (− 0.08, 1.07) | 0.094 | Random | < 0.001 | 99.4 | |
Mean/ median age (y) | < 0.001 | ||||||
< 60 | 1 | 0.45 (0.34, 0.57) | < 0.001 | Fixed | – | – | |
60–65 | 3 | 2.82 (1.19, 4.46) | 0.001 | Random | < 0.001 | 99.2 | |
% Male | < 0.001 | ||||||
< 55 | 2 | 1.03 (−0.63, 2.69) | 0.224 | Random | < 0.001 | 98.1 | |
≥ 55 | 2 | 3.40 (−1.72, 8.54) | 0.194 | Random | < 0.001 | 99.9 | |
HTN% | < 0.001 | ||||||
< 70 | 1 | 6.18 (5.47, 6.89) | < 0.001 | Fixed | – | – | |
≥ 70 | 2 | 0.57 (0.32, 0.82) | < 0.001 | Random | 0.018 | 82.2 | |
N/A | 1 | 1.77 (1.57, 1.97) | < 0.001 | Fixed | – | – | |
Diabetes % | < 0.001 | ||||||
< 20 | 2 | 3.31 (−2.30, 8.92) | 0.248 | Random | < 0.001 | 99.6 | |
≥ 20 | 2 | 1.24 (0.20, 2.28) | 0.019 | Random | < 0.001 | 98.4 | |
Current smoking % | < 0.001 | ||||||
< 30 | 2 | 0.57 (0.32, 0.82) | < 0.001 | Random | 0.018 | 82.2 | |
≥ 30 | 1 | 6.18 (5.47, 6.89) | < 0.001 | Fixed | – | – | |
N/A | 1 | 1.77 (1.57, 1.97) | < 0.001 | Fixed | – | – |