Cardiovascular disease is a leading cause of mortality in people with type 2 diabetes mellitus (T2DM). Beginning in 2015, long-term cardiovascular outcomes trials (CVOTs) have reported cardioprotective benefits for two classes of diabetes drugs. In addition to improving the lives of patients, these health benefits affect relative value (i.e., cost-effectiveness) of these agents compared with each other and especially compared with other agents. While long-term CVOT data on hard outcomes are a great asset, economic modeling of the value of this cardioprotection faces many new empirical challenges. The aim of this study was to identify different approaches used to incorporate drug-mediated cardioprotection into T2DM economic models, to identify pros and cons of these approaches, and to highlight additional considerations.
A review of T2DM modeling applications (manuscript or conference abstracts) that included direct cardioprotective effects was conducted from January 2015 to September 2018. Model applications were classified on the basis of the mechanism used to model cardioprotection [i.e., directly via hazard ratios (HRs) for cardiovascular outcomes or indirectly via biomarker mediation]. Details were extracted and the studies were evaluated.
Five full-length articles and 16 conference abstracts (of which 11 posters were found) qualified for study inclusion. While the approaches used were diverse, the five full-length publications and all but two of the abstracts modeled cardioprotection used direct HRs from the relevant CVOT. The remaining two posters modeled cardioprotection using CVOT HRs in combination with treatment effects mediated through known risk factors.
The classification of empirical methods in cardioprotection was intended to facilitate a better understanding of the pros and cons of different methodologies. A substantial diversity was observed, though most used trial HRs directly. Given the differences observed, we believe that diabetes modelers and other stakeholders can benefit from a formal discussion and evolving consensus.
Diabetes Therapy 2019;10(5):1753-1769