Outline of a diabetes disease management model: principles and applications

A complex interactive computer model was developed to determine the health outcomes and economic consequences of different diabetes interventions for user-defined observation periods. The interventions include intensive or co nventional insulin therapy, different oral hypoglycaemic medications, diffe rent screening and treatment strategies for micro-vascular complications, d ifferent treatment strategies for end-stage complications, or multi-factori al interventions. The analyses can be performed on different sub-groups of type 1 and 2 diabetic patients, defined in terms of age, gender, baseline r isk factors and pre-existing complications. The model performs real-time si mulations. Full on-screen documentation of the model structure, logic, calc ulations and data sources is available to maximize the model's transparency . Economic and clinical data used in the disease management model are edita ble by the user, allowing the input of new data as they become available, t he creation of country-specific, HMO-specific, or provider-specific version s of the model, and the exploration of new hypotheses ('what-if analyses). The approach used allows maximum flexibility, adaptability; and transparenc y within the model structure. For the user-defined patient cohorts and inte rvention strategies the diabetes disease management model compares life exp ectancy, expected incidence and prevalence of complications as well as expe cted life-time (or shorter) treatment cost. Diabetes and complication manag ement strategies can be compared in different patient populations in a vari ety of realistic clinical settings. The model allows extrapolation of resul ts obtained from relatively short-term clinical trials to longer-term medic al outcomes, and from trial populations to real-life populations providing a tangible yardstick to judge the quality of diabetes care. The model was u sed to evaluate diabetes care options in Germany, France, Switzerland, UK a nd US. (C) 2000 Elsevier Science Ireland Ltd. All rights reserved.