Table 2. Results of meta-analysis for KCNJ11 E23K polymorphism and T2D risk.
Sub-group analysis | No. of studies | K allele | Heterozygous | Homozygous | |||||||||
OR (95%CI) | P(Z) | P(Q)a | P(Q)b | OR (95%CI) | P(Z) | P(Q)a | P(Q)b | OR (95%CI) | P(Z) | P(Q)a | P(Q)b | ||
Ethnicity | 0.10 | 0.06 | 0.01 | ||||||||||
Caucasians | 22 | 1.12 (1.08–1.16) | <10−5 | 0.001 | 1.09 (1.06–1.13) | <10−5 | 0.13 | 1.33 (1.18–1.50) | <10−5 | 0.0008 | |||
East Asians | 14 | 1.13 (1.08–1.17) | <10−5 | 0.02 | 1.13 (1.05–1.22) | 0.0009 | 0.02 | 1.30 (1.18–1.42) | <10−5 | 0.06 | |||
Indians | 5 | 1.06 (0.87–1.29) | 0.56 | <10−5 | 1.00 (0.82–1.23) | 0.98 | 0.01 | 0.90 (0.69–1.18) | 0.45 | 0.02 | |||
Others | 7 | 1.09 (0.97–1.23) | 0.15 | 0.01 | 0.98 (0.76–1.28) | 0.91 | <10−4 | 1.19 (1.00–1.43) | 0.05 | 0.31 | |||
Sample size | 0.32 | 0.14 | 0.04 | ||||||||||
Large | 22 | 1.12 (1.08–1.17) | <10−5 | <10−5 | 1.12 (1.10–1.15) | <10−4 | 0.12 | 1.17 (1.07–1.28) | 0.007 | 0.02 | |||
Small | 26 | 1.13 (1.07–1.18) | <10−5 | <10−5 | 1.10 (1.02–1.19) | 0.009 | 0.002 | 1.32 (1.20–1.46) | <10−5 | 0.0007 | |||
Mean BMI of cases | 0.37 | 0.20 | 0.03 | ||||||||||
<25 | 12 | 1.15 (1.11–1.21) | <10−5 | 0.15 | 1.15 (1.08–1.22) | <10−5 | 0.21 | 1.40 (1.28–1.55) | <10−5 | 0.74 | |||
25∼30 | 25 | 1.12 (1.06–1.19) | <10−4 | <10−4 | 1.13 (1.06–1.21) | <10−4 | 0.004 | 1.18 (1.03–1.34) | 0.008 | <10−4 | |||
>30 | 6 | 1.10 (1.03–1.18) | 0.008 | 0.07 | 1.11 (1.04–1.19) | 0.001 | 0.09 | 1.16 (1.01–1.33) | 0.02 | 0.10 | |||
Total | 48 | 1.12 (1.09–1.16) | <10−5 | <10−5 | 1.09 (1.03–1.14) | <10−5 | 0.0001 | 1.26 (1.17–1.35) | <10−5 | <10−5 |
P(Z): Z test used to determine the significance of the overall OR.
P(Q)a: Cochran's chi-square Q statistic test used to assess the heterogeneity in subgroups.
P(Q)b: Cochran's chi-square Q statistic test used to assess the heterogeneity between subgroups.