Quality control in therapeutic drug monitoring: intralaboratory precision and medical requirements.

Formal quality control procedures have become a part of the clinical laboratory. Application of these procedures to therapeutic drug monitoring, which has been introduced into the clinical laboratory relatively recently, is incomplete. Most attention has been directed toward questions of accuracy and laboratory-to-laboratory precision, particularly with respect to antiepileptic agents. Intralaboratory precision must be addressed first, since it is fundamental to accuracy and laboratory-to-laboratory variability and most patients are treated based on results generated by a single laboratory. Intralaboratory precision data were gathered from laboratories already offering therapeutic drug monitoring services. In addition, the intralaboratory precision demands of good medical management was gathered from physicians who use therapeutic drug monitoring services. For 20 specified drugs, most laboratories reported intralaboratory precision (relative standard deviations) of 10% or less. Managing physician responses demonstrated that most require precision of 10 to 15% or 20 to 25%. Individual laboratories should use these data as an indication of the level of intralaboratory precision that may be achieved and that is required. The state of the art of intralaboratory precision is acceptable relative to medical management, but may need improvement as it contributes to laboratory-to-laboratory variability.