Vol.:(0123456789) 1 3 Acta Diabetologica https://doi.org/10.1007/s00592-020-01492-x ORIGINAL ARTICLE Analysis via Markov decision process to evaluate glycemic control strategies of a large retrospective cohort with type 2 diabetes: the ameliorate study Fanwen Meng 1  · Yan Sun 1  · Bee Hoon Heng 1  · Melvin Khee Shing Leow 2,3,4,5,6,7 Received: 19 November 2019 / Accepted: 28 January 2020 © Springer-Verlag Italia S.r.l., part of Springer Nature 2020 Abstract Aims Our aim was to explore optimal treatment decisions for HbA 1c control for type 2 diabetes mellitus patients and assess the impact on potential improvements in quality of life compared with current guidelines. Methods We analyzed a large dataset of HbA 1c levels, diabetes-related key risk factors and medication dispensed to 70,069 patients with type 2 diabetes from polyclinics and a large public hospital in Singapore during January 1, 2008, to December 31, 2015. A Markov decision process (MDP) model was developed to determine the optimal treatment policy concerning medication management for glycemic control over a long-term treatment period. We assessed the model performance by comparing quality-adjusted life years (QALYs) gained by the model with those derived by a conventional Markov model informed by current clinical guidelines. Results Numerical results showed that optimal treatment strategies derived by the MDP model could increase the total expected QALYs by as much as 0.27 years for patients at higher risk such as old age, high HbA 1c levels and smokers. In particular, the improvements in QALYs gained for patients with HbA 1c levels of 9% (75 mmol/mol) and above were higher than those with lower HbA 1c levels. However, the potential improvements appeared to be marginal for patients at lower risk compared with current guidelines. Conclusions Use of data-driven prescriptive analytics would help clinicians make evidence-based treatment decisions for HbA 1c control for patients with type 2 diabetes, in particular for those at high risk. Keywords Type 2 diabetes · Glycemic control · Treatment strategy · Markov decision process Introduction Type 2 diabetes mellitus appears large over Asia compared to other regions of the world. Asians, in particular South Asians, are prone to make Type 2 diabetes mellitus to a greater extent than ethnic Europeans [1, 2]. In Singapore, the prevalence of diabetes in adults aged 20–79 years was Managed By Antonio Secchi. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00592-020-01492-x) contains supplementary material, which is available to authorized users. * Fanwen Meng Fanwen_Meng@nhg.com.sg 1 Health Services and Outcomes Research, National Healthcare Group, 3 Fusionopolis Link, #03-08 Nexus@ one-north, Singapore 138543, Singapore 2 Cardiovascular and Metabolic Disorders Programme, Duke-NUS Medical School, Singapore, Singapore 3 Department of Endocrinology, Tan Tock Seng Hospital, Singapore, Singapore 4 Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore, Singapore 5 Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore 6 Clinical Nutrition Research Centre, Singapore Institute for Clinical Sciences, A*STAR, Singapore, Singapore 7 Changi General Hospital, Singapore, Singapore