Schumacher G E, Barr J T
Clin Pharm. 1984 Sep-Oct;3(5):525-30.
The theory of Bayesian analysis and its application to therapeutic and pharmacokinetic decision making are discussed. Diagnostic and therapeutic decisions are commonly based on institution, experience, and laboratory information; these decisions reflect varying degrees of uncertainty. Bayesian analysis quantifies the decision process by attaching probabilities to the likelihood of accuracy of each of these decision-making factors to achieve an overall estimate of decision quality. Using Bayesian principles to quantify the probability of efficacy and toxicity associated with serum drug concentrations represents one application of Bayesian theory to enhance therapeutic decisions. The Bayesian approach in pharmacokinetics involves the prediction of pharmacokinetic values, dosage regimens, and serum concentrations for drugs. Beginning with mean population pharmacokinetic parameters, one uses observed serum concentrations in individual patients to modify these parameters through Bayesian analysis to improve the accuracy of future serum concentration predictions. As more clinical pharmacokinetic laboratories and consultation services become familiar with the procedure, Bayesian forecasting promises to expand markedly the sophistication of therapeutic drug monitoring.
本文讨论了贝叶斯分析理论及其在治疗和药代动力学决策中的应用。诊断和治疗决策通常基于机构、经验和实验室信息；这些决策反映了不同程度的不确定性。贝叶斯分析通过为这些决策因素的准确性可能性赋予概率来量化决策过程，以实现对决策质量的总体估计。利用贝叶斯原理量化与血清药物浓度相关的疗效和毒性概率是贝叶斯理论在增强治疗决策方面的一个应用。药代动力学中的贝叶斯方法涉及药物的药代动力学值、给药方案和血清浓度的预测。从总体平均药代动力学参数开始，通过贝叶斯分析，利用个体患者观察到的血清浓度来修改这些参数，以提高未来血清浓度预测的准确性。随着越来越多的临床药代动力学实验室和咨询服务熟悉该程序，贝叶斯预测有望显著提高治疗药物监测的精细程度。
