The probability of state transition of five diabetic complications and their influencing factors based on multi-state markov models: evidence from a Chinese cohort study

Abstract

Diabetes-related complications remain a leading cause of disability and mortality among individuals with diabetes and substantially impair patients’ health-related quality of life. In this study, we employed a multi-state Markov model (MSM) to characterize transition probabilities across states defined by the cumulative burden of five major diabetes complications in a cohort of 1996 adults aged 20–80 years with type 2 diabetes mellitus (T2DM). We further estimated the effects of covariates on hazard ratios (HRs) for transitions between these complication states. At baseline, 264 participants were free of any complications, of whom 210 (79.55%) developed at least one complication during follow-up. Overall, 1191 (59.67%) patients exhibited changes in their total number of accumulated complications. For participants free of complications at baseline, the probability of remaining complication-free over the follow-up period was 0.615, while the probabilities of progressing to states with 1, 2, or ≥ 3 complications were 0.254, 0.105, and 0.026, respectively. Sedentary lifestyle and elevated glycated hemoglobin (HbA1c) were identified as independent risk factors for complication progression. These results underscore the need for healthcare providers to intensify follow-up care and routine complication surveillance in patients with T2DM.