Author Affiliations: Medical Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, Maryland (Dr Prasad); Department of Health Care Policy, Harvard Medical School, and Massachusetts General Hospital, Boston (Dr Jena); and National Bureau of Economic Research, Cambridge, Massachusetts (Dr Jena).
As observational studies have increased in number—fueled by a boom in electronic recordkeeping and the ease with which observational analyses of large databases can be performed—so too have failures to confirm initial research findings.1 Several solutions to the problem of incorrect observational results have been suggested,1,2 emphasizing the importance of a record not only of significant findings but of all analyses conducted.2
Prasad V, Jena AB. Prespecified Falsification End PointsCan They Validate True Observational Associations?. JAMA. 2013;309(3):241–242. doi:10.1001/jama.2012.96867