Some guidelines on guidelines: they should come with expiration dates.

The STROBE guidelines (for Strengthening the Reporting of OBservational Studies in Epidemiology) add to a lengthy catalog of attempts to keep epidemiologists on more or less straight and more or less narrow paths charted by guideline authors. STROBE has an ambitious goal, and may prove highly useful for some. It raises concern, however, about a problem generic to guidelines, namely how long they will be useful. Guidelines may be inevitable, but they can foster ossification and be counterproductive. Who today would be happy with epidemiology guidelines issued in, say, 1960, 1970, 1980, or even, 1990? One solution is to offer each set of guidelines with an expiration date, beyond which it would cease to apply. Such a policy would at least prompt revisions. We propose that the STROBE guidelines might expire on 31 December 2010 or 3 years after any revision.