Practical estimates of the effect of publication bias in meta- analysis

One of the major drawbacks to using meta-analysis is the possibility of 'publication bias'. The studies for a meta-analysis are usually chosen through a literature review and an inherent selection bias may thereby arise. This paper presents the 'trim and fill' method, a new technique developed for estimating and adjusting for the numbers and outcomes of missing studies. It has the advantage of being computationally simple and, in practical situations, to perform better than existing methods. To illustrate the trim and fill method, the authors apply it to an existing meta-analysis of the effect of using gangliosides in reducing case fatality in acute stroke. (non-author abstract)