This commentary gives a personal view of drafting the “Strengthening the Reporting of Observational Studies in Epidemiology” (STROBE) statement, by one of its authors. My initial wariness about guidelines for observational research was overcome by focusing on clarity of reporting, rather than on how research should be done. Areas of tension that arose when drafting STROBE include the problem of finding common ground among researchers with different research backgrounds, questions of the intended audience (professional epidemiologists or statisticians vs. all researchers who use epidemiologic study designs), and the fine line between encouraging clarity of reporting vs. prescribing how to do research. STROBE is not an instrument to evaluate the quality of research: research can be reported clearly or not, irrespective of its intrinsic quality. However, the ultimate benefit of STROBE might be that more comprehensive reporting allows for better discussions about published observational research, which may lead to better decisions about what new analyses or new studies are needed to solve a problem.
From the Department of Clinical Epidemiology, Leiden University Medical Center, Leiden, The Netherlands.
Editors' note: Related articles appear on pages 789, 791, 792, 794, 800, and 805.
Correspondence: Jan P. Vandenbroucke, Department of Clinical Epidemiology, Leiden University Medical Center, 1-C9-P, PO Box 9600, 2300 RC Leiden, The Netherlands. E-mail: firstname.lastname@example.org.