Guidelines and recommendations in laboratory medicine.

to the probability of discrepancy and thus to the power of this method: the power will be higher if the number of blood groups is higher or if the frequencies are more similar. check method for detecting specimen mix-up. To the Editor: Dr. Keffer (1) has provided a thought-provoking report on the noncompliance of physicians with most clinical practice guidelines (CPGs). He also refers to the Laboratory Medicine Practice Guidelines (LMPGs) produced by the National Academy of Clinical Biochemistry (NACB). As contributors to several LMPGs (2– 4) and active participants in the NACB LMPG program, we offer some insights into the rationale of NACB and how the NACB LMPG process differs from that of CPGs authored by other professional societies. We discuss some reasons for the difficulty in assessing impact and suggest unrecognized impacts of these guidelines. Most CPGs are aimed at physicians and have focused on clinical practices for a disease (e.g., asthma) or symptom (e.g., chest pain). By contrast, the intended audience for LMPGs includes not only physicians, but also clinical laboratorians and manufacturers of clinical assays. These guidelines include recommendations on the appropriateness of offering certain tests for particular clinical situations and for denying or limiting the availability of other assays. These changes and their effects may be difficult to document, but consensus recommendations can prompt manufacturers to construct new assays, such as the urine immu-noassay for methylenedioxymetham-phetamine (Ecstasy) as recommended (3). Because of their potential impact, all NACB LMPGs are presented and thoroughly discussed in sessions at national meetings, and an estimation of the degree of consensus is sought. The proposed guidelines are also presented at other meetings where pertinent disciplines are represented (e.g., cardiologists, emergency medicine physicians, clinical toxicologists, and endocrinologists). This may be unlike some CPGs that are prepared and published by experts, but without open presentation. Although traditional " evidence-based " documentation is noted for many NACB recommendations [e.g., randomized control trials (5, 6)], other recommendations do not lend themselves to support by outcome studies. The NACB guideline for testing of newborns (7), for example, recommended that testing for alkaline phosphatase isoenzymes not be performed. It was impractical to perform a randomized trial to support this recommendation. In the NACB guideline for emergency toxicology testing (3), a 1-h turnaround time was recommended for reporting of test results. There are no outcome studies to show that this improves the clinical management of intoxicated patients …