Traditional reviews, meta-analyses and pooled analyses in epidemiology

Background The use of review articles and meta-analysis has become an impor tant part of epidemiological research, mainly for reconciling previously co nducted studies that have inconsistent results, Numerous methodologic issue s particularly with respect to biases and the use of meta-analysis are stil l controversial. Methods Four methods summarizing data from epidemiological studies are desc ribed. The rationale for meta-analysis and the statistical methods used are outlined. The strengths and limitations of these methods are compared part icularly with respect to their ability to investigate heterogeneity between studies and to provide quantitative risk estimation. Results Meta-analyses from published data are in general insufficient to ca lculate a pooled estimate since published estimates are based on heterogene ous populations, different study designs and mainly different statistical m odels. More reliable results can be expected if individual data are availab le for a pooled analysis, although some heterogeneity still remains, Large prospective planned meta-analysis of multicentre studies would be preferabl e to investigate small risk factors, however this type of meta-analysis is expensive and rime-consuming. Conclusion For a full assessment of risk factors with a high prevalence in the general population, pooling of data will become increasingly important. Future research needs to focus on the deficiencies of review methods, in p articular, the errors and biases that can be produced when studies are comb ined that have used different designs, methods and analytic models.