校准2型糖尿病患者实际治疗效果的估计与应用: 临床试验适用性最大化研究

随机对照试验(RCT)是干预效果评价的金标准，但在结果外推到目标人群时，存在人群代表性不足以及药物的效力-效果差距等问题。校准是提高RCT结果适用性的可行方法。最新校准理论采用多水平网状Meta回归(ML-NMR)，既可保留随机性又可基于目标人群执行校准RCT的网状Meta分析(NMA)，但该理论尚未建立严格的方法学体系，也缺少应用和验证。该理论适合在2型糖尿病(T2DM)新型降糖药评价中开展，已有大量相关RCT，但药物的长期效果、安全性和经济学在实际T2DM中未充分评价。本课题旨在建立基于目标人群校准NMA的评价方法，以三类新型降糖药为例验证该理论。具体目标：建立校准NMA目标人群的确定原则，制定基于目标人群通过ML-NMR执行校准NMA的实际治疗效果的指南和工具；开展校准NMA的卫生经济学评价研究；基于证据质量评价体系，开发校准NMA证据的质量评价模型，确立该方法体系的科学性和适用性。

Randomized controlled trial (RCT) is the gold standard to evaluate the effectiveness of interventions. However, there are some important issues concerning extrapolating the RCT results to the target population, such as underrepresentation of the population and the efficacy-effectiveness gap. Calibration is a feasible method to improve the applicability of RCT results. The latest calibration theory recommends the use of multi-level network Meta regression (ML-NMR), which can both retain randomization and calibrate the network meta-analysis (NMA) of RCT based on the characteristics of the target population. However, no rigorous assessment system of its methodology has been established for this theory, with lack of its application and validation. This theory is suitable for the evaluation of new hypoglycemic agents among type 2 diabetes mellitus (T2DM) because of a large number of relevant RCTs available for these drugs. However, the long-term effectiveness, safety and economics of these drugs have not been fully evaluated in the real-world T2DM population..In response to the above issues, our project aims to establish the assessment system of methodology of calibrating the estimates of NMA based on the target population using three kinds of new hypoglycemic drugs as examples. Specific objectives are: firstly, to establish a set of principles for determining the target population for calibration and develop series of guidelines and assessment tools to calibrate the NMA estimates of real therapeutic effectiveness using ML-NMR based on target populations; secondly, to conduct health economics evaluation research based on calibrated estimates from NMA; thirdly, to develop evidence quality evaluation tool for the calibration with NMA based on current previous assessment system, demonstrating the validity and applicability of the methodology system.