Fang Jian, Xie Qing-Yu, Meng Qing-Gang, Meng Hong-Yu, Luan Xiang-Jia
Beijing University of Chinese Medicine Beijing 100029, China.
Yinchuan Internet Hospital Yinchuan 750027, China.
Zhongguo Zhong Yao Za Zhi. 2022 Apr;47(8):2237-2243. doi: 10.19540/j.cnki.cjcmm.20220106.502.
Clinical expertise, patient preference, and the best evidence are the three elements of evidence-based medicine. Based on high-level and high-quality evidence, qualitative and quantitative analysis of the prescribing decisions of physicians is beneficial to improving clinical efficacy. A mature methodological system is available for the retrieval, analysis, summary, evaluation, and recommendation of the evidence, but there are still few studies on physicians' prescribing decisions. How to analyze the trend of physicians' prescribing decisions based on the priority ranking in addition and subtraction of prescriptions? Analytic hierarchy process(AHP) is a method for decision making, which arranges the elements of the decision problem into overall goal, criteria, and operational sub-criteria, and uses the matrix eigenvector method to solve the problem. This study aims to analyze the priority of physicians' prescribing decisions for diabetes mellitus with deficiency of both Qi and Yin based on AHP. To be specific, a database of diabetes mellitus cases with deficiency of both Qi and Yin was established and AHP was used to yield the priority ranking of Chinese patent medicine prescriptions in specific clinical scenarios. In the selected cases of diabetes mellitus with deficiency of both Qi and Yin, Xiaoke Pills was the best prescription for the treatment of type 2 diabetes mellitus(deficiency of both Qi and Yin)(normalized=0.388), followed by Liuwei Dihuang Pills(normalized=0.269), Qishen Capsules(normalized=0.230), and Shengmai Injection(normalized=0.113). According to the analysis the available data, for type 2 diabetes mellitus(deficiency of both Qi and Yin), Xiaoke Pills was the most effective prescription in specific scenarios. When the physicians' prescribing decisions are consistent with the evidence, quantitative analysis of physicians' cognition will boost the evidence-based medical decision-making. However, the research results are also affected by the quality of literature, evidence level and priority, which are thus have some limitations. It is recommended that further small data research based on individual cases be carried out to lay a evidence-based basis for the clinical decision-making of type 2 diabetes mellitus.
临床专业知识、患者偏好和最佳证据是循证医学的三个要素。基于高水平和高质量的证据，对医生的处方决策进行定性和定量分析，有利于提高临床疗效。对于证据的检索、分析、总结、评估和推荐，已有成熟的方法体系，但关于医生处方决策的研究仍较少。如何基于方剂的加减用药优先级排序来分析医生的处方决策趋势？层次分析法（AHP）是一种决策方法，它将决策问题的要素分为总体目标、准则和操作子准则，并使用矩阵特征向量法来解决问题。本研究旨在基于层次分析法分析气阴两虚型糖尿病医生处方决策的优先级。具体而言，建立气阴两虚型糖尿病病例数据库，并使用层次分析法得出特定临床场景下中成药处方的优先级排序。在所选的气阴两虚型糖尿病病例中，消渴丸是治疗2型糖尿病（气阴两虚型）的最佳处方（归一化值 = 0.388），其次是六味地黄丸（归一化值 = 0.269）、芪参胶囊（归一化值 = 0.230）和生脉注射液（归一化值 = 0.113）。根据现有数据分析，对于2型糖尿病（气阴两虚型），消渴丸是特定场景下最有效的处方。当医生的处方决策与证据一致时，对医生认知的定量分析将促进循证医学决策。然而，研究结果也受到文献质量、证据水平和优先级的影响，因此存在一定局限性。建议进一步开展基于个体病例的小数据研究，为2型糖尿病的临床决策奠定循证基础。
