Bias in meta-analysis

The objective of the present article is to review the biases in metaanalysi s. This procedure can be included within observational studies, and its bia ses follow the same classification: selection, information and confounding bias. Selection bias can always be present because of the impossibility of enumerating the reference population. More interest deserves publication bi as, influenced by many facts: significant results, design type, language of publication, the existence of conflicts of interest, sample size, institut ional prestige, existence of prejudices, and regression to the mean, There are several graphic and quantitative procedures to detect the error under c ertain assumptions. The correction of this bias is quite unlikely. It is ea sier to guarantee that the selection process of the studies among those loc ated is not subjected to errors, if structured protocols are used. Informat ion bias has been less studied, but it has also been documented. Control of confounding bias depends on the identification of this error in the indivi dual studies, but it can also be introduced through certain analytic techni ques, as the ecologic fallacy, Lastly, it is recommended not to use the pro cedure of Peto in the analysis as this technique provides biased results.