Ishida H, Ichihara K, Matsuda N
Department of Clinical Pathology, Kawasaki Medical School, Kurashiki.
Rinsho Byori. 1998 Oct;46(10):978-86.
To effectively respond to the desire for consultation in clinical practice, we must prepare logical reasoning and evidence which rationally supports laboratory test selection, the interpretation of test results and recommendation of certain tests to physicians. Standard of medical decision making can be used for logic issues such as posttest probability, test characteristics and receiver operating characteristics (ROC) curves and establishing appropriate cut-off points. Although we usually obtain evidence by consulting authorities or the literature, good evidence can also be obtained from meta-analysis. In addition, we can demonstrate the relationship of laboratory tests among several frequently occurring diseases and epidemiological tendencies such as frequency of causative organisms at several infection sites and bacterial sensitivities to antimicrobial agents, because we have access to a large-scale laboratory database. To construct a well-organized knowledge base with explicit evidence, cooperation among many facilities is necessary to develop system, which allows the free exchange of data.
为了在临床实践中有效回应会诊需求，我们必须准备好逻辑推理和证据，以合理支持实验室检查项目的选择、检查结果的解读以及向医生推荐某些检查项目。医学决策标准可用于诸如检验后概率、检查特征和受试者工作特征（ROC）曲线等逻辑问题，并确定合适的临界点。虽然我们通常通过咨询权威人士或文献来获取证据，但也可以从荟萃分析中获得有力证据。此外，由于我们能够访问大规模实验室数据库，所以可以展示实验室检查在几种常见疾病之间的关系以及流行病学趋势，例如多个感染部位的致病微生物频率和细菌对抗菌药物的敏感性。为了构建一个有明确证据支持的条理清晰的知识库，需要多个机构合作来开发系统，以实现数据的自由交换。
