Table 4. Analysis of heterogeneity by “exclusion of one trial each time” method.
| Authors | I2 a | P b | IV 95% CI c | |
|---|---|---|---|---|
| AST;I2 = 76%, p = 0.21 | Byrne, et al.(2014) | 81% | 0.22 | -7.81[-20.30,4.67] |
| Lu-Hua Yang, et al.(2015) | 79% | 0.57 | -4.41[-19.53,10.71] | |
| Purrello, et al.(2007) | 81% | 0.29 | -7.22[-20.68,6.25] | |
| Sofi, et al.(2010) | 74% | 0.64 | -2.88[-14.89,9.12] | |
| Feng-Shang Zhu, et al.(2008) | 47% | 0.002 | -12.13[-19.98,-4.27] | |
| TC;I2 = 56%, p = 0.001 | Byrne, et al.(2014) | 56% | 0.001 | -14.60[-23.43,-5.76] |
| Lu-Hua Yang, et al.(2015) | 30% | 0.01 | -10.48[-18.63,-2.34] | |
| Purrello, et al.(2007) | 58% | 0.001 | -14.46[-23.18,-5.74] | |
| Sofi, et al.(2010) | 48% | <0.001 | -15.78[-23.17,-8.39] | |
| Stephen, et al.(2015) | 62% | 0.002 | -13.96[-22.67,-5.25] | |
| Feng-Shang Zhu, et al.(2008) | 64% | 0.05 | -10.52[-21.24,0.20] | |
| LDL-C;I2 = 70%, p = 0.05 | Byrne, et al.(2014) | 75% | 0.26 | -5.43[-14.93,4.06] |
| Lu-Hua Yang, et al.(2015) | 60% | 0.22 | -4.68[-12.17,2.81] | |
| Sofi, et al.(2010) | 68% | 0.009 | -9.12[-15.98,-2.27] | |
| Stephen, et al.(2015) | 77% | 0.12 | -6.73[-15.26,1.80] | |
| Feng-Shang Zhu, et al.(2008) | 56% | 0.01 | -9.48[-16.76,-2.21] | |
| HDL-C;I2 = 55%, p = 0.06 | Byrne, et al.(2014) | 63% | 0.01 | 8.00[1.66,14.35] |
| Lu-Hua Yang, et al.(2015) | 65% | 0.04 | 6.91[0.24,13.58] | |
| Purrello, et al.(2007) | 65% | 0.008 | 7.64[1.98,13.30] | |
| Sofi, et al.(2010) | 0% | 0.005 | 4.82[1.44,8.19] | |
| Feng-Shang Zhu, et al.(2008) | 60% | 0.01 | 8.16[1.89,14.43] |
a I2
b P
c IV 95% CI were calculated using Cochrane Collaboration’s Review Manager Software RevMan 5.3.