Development of a diabetes treatment simulation model: with application to assessing alternative treatment intensification strategies on survival and diabetes-related complications.

摘要

AIMS: The objective of this analysis is to project the long-term impacts on life expectancy and occurrence over 5, 10, and 40 years of microvascular and macrovascular complications of diabetes when using different haemoglobin A1c (HbA1c) thresholds for intensifying treatment of type 2 diabetes. METHODS: A flexible, discrete-event simulation model has been developed to evaluate alternative treatment strategies based on the United Kingdom Prospective Diabetes Study Outcomes Model. In the present analysis, the model is used to investigate the impact of alternative HbA1c thresholds for treatment intensification ranging from 7.0 to 9.0%. For each intensification strategy, the model is run using 80 simulated patients for each of 1224 patient profiles from the Real-Life Effectiveness and Care Patterns of Diabetes Management study (for a total of 97,920 simulated patients) to project the number of patients who will experience diabetes-related complications over time. RESULTS: The use of lower HbA1c thresholds for intensifying treatment is associated with improved long-term outcomes. When the HbA1c threshold for intensifying therapy from oral treatment to basal insulin (T1) is 7.0% and the threshold for intensifying basal insulin to multiple-dose insulin (T2) is 7.0%, simulated patients spend 54% of their time with HbA1c >7.0%, but 95% of their time with HbA1c >7.0% if T1 and T2 are set to 9.0%. More aggressive or proactive treatment postures are projected to reduce clinical events, including diabetes-related deaths and diabetes-related complications, particularly myocardial infarctions (MIs). When T1 and T2 are set to 7.0%, there are 592 fewer diabetes-related deaths in the first 5 years of the simulation and 3740 fewer deaths over 40 years compared with the results when T1 and T2 are set to 9.0%. These decreases in deaths were also associated with a 0.35 year gain in projected life expectancy. Compared with an aggressive strategy with both T1 and T2 being 7%, 644 more patients are projected to experience at leastone episode of MI in the first 5 years if treatment intensification is delayed until HbA1c reaches 9.0%. This number increases over time, reaching 2906 additional patients experiencing at least one MI over a 40-year time period. CONCLUSIONS: We report results from a discrete-event simulation model to explore the impact of alternative treatment strategies for patients with type 2 diabetes. Strategies that intensify therapy (in response to rising HbA1c levels) at lower HbA1c thresholds (e.g. 7.0%) are associated with enhanced projected long-term health outcomes.