Objective To investigate confidence interval estimation for the amount of heterogeneity in meta-analysis.
Methods On the basis of BT’s method, the approximate Q-statistic distribution following linear transformation of Chi-square was applied to improve the accuracy of Q-statistic distribution, and to obtain the confidence interval for the amount of heterogeneity in meta-analysis.
Results In case, the Q1 distribution obtained 95%CI 0.07 to 2.20, while the Q2 distribution obtained 95%CI 0.00 to 1.41; The proposed method Q2 narrowed down the range of confidence interval.
Conclusion On account of improving the accuracy of Q-statistic distribution, the proposed method effectively strengthens the coverage probabilities of the confidence interval for the amount of heterogeneity. And the proposed method can also improve the precision of the confidence interval estimation for the amount of heterogeneity.
Citation: LIU Jintao,QIN Chaoying,DANG Hong. Interval Estimation for the Amount of Heterogeneity in Meta-Analysis Based on Q-Statistic Following Linear Transformation of Chi-Square Distribution. Chinese Journal of Evidence-Based Medicine, 2012, 12(11): 1404-1406. doi: 10.7507/1672-2531.20120219 Copy
Copyright © the editorial department of Chinese Journal of Evidence-Based Medicine of West China Medical Publisher. All rights reserved