In response to our paper showing that the quality of drug therapy guidelines in Canada is quite variable,1 Marc Baltzan has argued that the more fundamental flaw with clinical practice guidelines relates to their potential to be used indiscriminately by some physicians, thereby causing more harm than good. The point is well taken, but it is inappropriate to single out guidelines. Practice guidelines are “systematically developed statements to assist practitioner and patient decisions about appropriate health care for specific clinical circumstances [italics added].”2 All health care technologies, be they drugs, procedures, diagnostic testing, surgical techniques or practice guidelines, must be applied judiciously using clinical judgement that is individualized to the patient's circumstances. Doing anything less constitutes at best negligence and at worst malpractice. Well-developed and valid practice guidelines that are appropriately applied have been shown to improve both the process of care and patient health outcomes.3,4,5 We are unaware of any evidence about the extent to which physicians lacking the necessary knowledge are using guidelines inappropriately and causing harm. However, if this is the real concern then the focus must be on improving medical education and evidence-based practice generally, and not rejecting out of hand evidence-based practice guidelines.
James McCormack and colleagues suggest we should not have evaluated the quality of Therapeutic Initiative letters catalogued in the CMA Infobase because they are systematically developed reviews, not practice guidelines. Our study protocol involved assessing the quality of all drug therapy guidelines listed in the CMA Infobase (mdm .ca/cpgsnew/cpgs/index.asp), which is clearly intended to be a repository of Canadian practice guidelines. We assumed that all documents included in the CMA Infobase were practice guidelines as they had to meet the CMA's criteria for being a guideline, i.e., that the document include information to help patients and physicians make decisions about appropriate health care for specific clinical circumstances. We also understand that the developers of documents considered for inclusion in the CMA Infobase were contacted and asked whether they agreed their document should be included.
We thank Baltzan and McCormack and colleagues for raising these issues and we hope our findings will generate further debate that ultimately will lead to improvements in the quality of practice guidelines produced in Canada.