Expertise: no longer a sine qua non for guideline authors?