Abstract:
|
Publication bias (PB) is a serious problem in meta-analysis (MA), affecting the validity and generalization of conclusions. Current approaches to assessing PB can be distinguished into two classes: selection models and funnel-plot-based methods. Selection models use weight functions to adjust the overall effect estimate and are usually treated as sensitivity analyses. Funnel-plot-based methods include visual examination of a funnel plot, regression and rank tests, and the nonparametric trim and fill method. Although these approaches have been widely used, measures for quantifying PB are seldom studied. Such measures can be used as a characteristic of a MA; also, they permit comparisons of PBs between different MAs. Egger's regression intercept may be considered as a candidate measure, but it lacks an intuitive interpretation. We introduce a new measure, the skewness of the standardized deviates, to quantify PB. This measure describes the asymmetry of the collected studies' distribution. In addition, a new test for PB is derived based on the skewness. Large sample properties of the new measure are studied, and its performance is illustrated using simulations and three case studies.
|