Evaluation of potential health effects of electromagnetic fields are a time-consuming and complicated procedure which can only be conducted by expert panels (cf. chapter Evaluation). During this procedure all publications of all study types (in vitro, animal experiments and epidemiological studies) on a specific frequency range and particular endpoints are collected and evaluated in terms of their distinct quality (e.g. dosimetry, study design, sample size, statistics), reproducibility of the results and comparability with other studies.
Each of the various study types has its specific advantages and disadvantages which should be considered. The advantage of epidemiological studies and human experimental studies is the direct linkage to humans. However, small sample sizes, an often insufficient retrospective data acquisition and often lack of statistical significance are the disadvantages of these studies. Furthermore it is very difficult to prove the causality of a disease due exposure to electromagnetic fields by epidemiological studies. Therefore experimental in vitro- and animal studies have to be conducted to confirm the observed effects, to uncover mechanisms of action and to test hypotheses. Experimental animal and in-vitro studies provide more reliable results than epidemiological studies due to the possibility of standardized lab techniques with large sample sizes, especially with respect to weak and thereby more difficult verifiable effects. But again, their main disadvantage is the problem of transferring the results to humans. Hence, the results of all study types and the comparison of these results should build the basis to evaluate a consistent effect. Only when the results of each study type (i.e. epidemiological, experimental animal and in-vivo studies) show mainly consistent evidence of the same effect, causality can be concluded. Additionally it should be kept in mind that a biological effect does not inevitably mean an adverse health effect.