Invited commentary: the 21st century epidemiologist--a need for different training?

The accompanying paper by Ogino et al. (Am J Epidemiol. 2012;176(8):659-667) cogently suggests a need for including modern approaches like molecular pathological epidemiology (MPE) in our research. However, Ogino et al. make an assumption that epidemiology has previously not included pathology or modern technologies in epidemiologic studies and that there is a unique need for the specialty of MPE. The new molecular pathology is yet another technique that can improve epidemiologic investigations. There is a long tradition of combining good pathology with epidemiologic research, especially in studies of cardiovascular disease. Large epidemiologic studies have successfully integrated specialty expertise in a collaborative and mutually beneficial approach to test specific hypotheses. The author is concerned that MPE techniques, whether they involve metabolomics, genomics, proteomics, or microarrays, will come to drive epidemiologic studies without any specific hypothesis-testing or unique population characteristics. The epidemiologist would then become little more than a collector of study subjects and a distributor of the various specimens to the laboratories.