AIM: To develop a novel metabolic computer model of the natural lifetime progression of type 2 diabetes that generates dynamic risk factor trajectories consistent with prespecified lifetime therapeutic strategies, in order to enhance the long-term economic and outcome modelling of type 2 diabetes and its complications. METHODS: The main model drivers of progressive disease were changes in insulin sensitivity and islet beta-cell function derived from an analysis of follow-up results from the Belfast Diet Study. These were related to clinical measures through an adaptation of the homeostasis model assessment. Established causal relationships estimating body mass index, lipids and blood pressure from measures of glycaemia and plasma insulin were calibrated using Third National Health and Nutrition Examination Survey (USA) data, standardizing for age, sex, ethnicity and smoking. The effects of individual interventions were calibrated using published trial evidence, in line with the current understanding of the main modes of action of each agent. RESULTS: A comparison of the effects of common therapies using the model showed both similarities and differences. Large improvements in glycaemic control from lifestyle modifications, further enhanced by oral glucose-lowering drugs or insulin, were reproduced. Projections comparing lifetime therapeutic strategies suggest that simple guidelines may not always be valid. CONCLUSION: This novel mathematical model using evidence from the long-term natural history of type 2 diabetes is able to project the expected effects of various antihyperglycaemic therapies. Coupled with an economic model, this metabolic model may provide a mechanism for healthcare professionals and policymakers to evaluate different long-term strategies for the management of type 2 diabetes.
AIM: To develop a novel metabolic computer model of the natural lifetime progression of type 2 diabetes that generates dynamic risk factor trajectories consistent with prespecified lifetime therapeutic strategies, in order to enhance the long-term economic and outcome modelling of type 2 diabetes and its complications. METHODS: The main model drivers of progressive disease were changes in insulin sensitivity and islet beta-cell function derived from an analysis of follow-up results from the Belfast Diet Study. These were related to clinical measures through an adaptation of the homeostasis model assessment. Established causal relationships estimating body mass index, lipids and blood pressure from measures of glycaemia and plasma insulin were calibrated using Third National Health and Nutrition Examination Survey (USA) data, standardizing for age, sex, ethnicity and smoking. The effects of individual interventions were calibrated using published trial evidence, in line with the current understanding of the main modes of action of each agent. RESULTS: A comparison of the effects of common therapies using the model showed both similarities and differences. Large improvements in glycaemic control from lifestyle modifications, further enhanced by oral glucose-lowering drugs or insulin, were reproduced. Projections comparing lifetime therapeutic strategies suggest that simple guidelines may not always be valid. CONCLUSION: This novel mathematical model using evidence from the long-term natural history of type 2 diabetes is able to project the expected effects of various antihyperglycaemic therapies. Coupled with an economic model, this metabolic model may provide a mechanism for healthcare professionals and policymakers to evaluate different long-term strategies for the management of type 2 diabetes.
Authors: Kathryn E Ferrier; Michael H Muhlmann; Jean Philippe Baguet; James D Cameron; Garry L Jennings; Anthony M Dart; Bronwyn A Kingwell Journal: J Am Coll Cardiol Date: 2002-03-20 Impact factor: 24.094
Authors: P M Clarke; A M Gray; A Briggs; A J Farmer; P Fenn; R J Stevens; D R Matthews; I M Stratton; R R Holman Journal: Diabetologia Date: 2004-10-27 Impact factor: 10.122
Authors: Arran T Shearer; Adrian Bagust; F Javier Ampudia-Blasco; Belén Martínez-Lage Alvarez; Isabel Pérez Escolano; Gonzalo París Journal: Pharmacoeconomics Date: 2006 Impact factor: 4.981