The challenge of delivering reliable science and guidelines: opportunities for all to participate.

Erroneous research results have potential for harm far beyond the patients recruited into the study, because the leverage of leadership permits them to alter behaviour of thousands of clinicians for many years. Medical research does not yet have as well-developed and rapid-responding safety systems as other spheres where lives are at stake. As doctors we would all leap to assist in a medical emergency for a patient not formally under our care. Yet, when problems occur in clinical research, we do not yet have good mechanisms to all work together quickly and decisively to improve patient safety. In this article we describe how each of us, regardless of our role, can (and perhaps should) take active steps to support delivery of reliable science for patient care.