Studies of epidemiological and prognostic factors are very important for oncology practice. There is arapidly increasing amount of research and resultant knowledge in the scientific literature. This means thathealth professionals have major challenges in accessing relevant information and they increasingly requirebest available evidence to make their clinical decisions. Meta-analyses of prognostic and other epidemiologicalfactors are very practical statistical approaches to define clinically important parameters. However, they alsofeature many obstacles in terms of data collection, standardization of results from multiple centers, bias, andcommentary for intepretation. In this paper, the obstacles of meta-analysis are briefly reviewed, and potentialproblems with this statistical method are discussed.