STROBE and pre-registration of observational studies

STROBE and pre-registration of observational studies Alain Braillon, 1 Florian Naudet2 Aronson rightly acknowledges that observational studies only generate conjectures and warns that residual confounding can be quite difficult to fix.1 But the issue is deeper than this. The Finnish study that reported elevated suicide rates during a full moon only among premenopausal women and only in winter, which was debunked by Plöderl et al,2 questions the integrity and the professionalism of our system. Hypotheses were not pre-specified and results only identified in subgroups, not to mention the lack of rationale. Why isn’t pre-registration of the hypotheses, designs, and analytical plan—using Open Science Framework, for example3—the mandatory norm in observational research? The fundamental failing of observational studies is that they are subject to vibration of effect caused by the inherently large number of researcher degrees of freedom.4 Such flexibility is a gateway for p hacking and HARking, which stands for “hypotheses after the results are known.”5 Additionally, editors should require the use of STROBE guidelines by researchers for the reporting of observational research (https://www.strobe-statement.org/). STROBE is an international, collaborative initiative of epidemiologists, methodologists, statisticians, researchers, and journal editors involved in the conduct and dissemination of observational studies, with the common aim of strengthening the reporting of observational studies in epidemiology. STROBE is to observational studies as CONSORT is to randomised clinical trials. Better attention to these two quite simple practices might improve the overall credibility of observational research findings.