Validation of the CORE Diabetes Model Against Epidemiological and Clinical Studies

CORE--Center for Outcomes Research, Basel, Switzerland.
Current Medical Research and Opinion (Impact Factor: 2.65). 09/2004; 20 Suppl 1(Suppl 1):S27-40. DOI: 10.1185/030079904X2006
Source: PubMed


The aim of this study was to assess the validity of the CORE Diabetes Model by comparing results from model simulations with observed outcomes from published epidemiological and clinical studies in type 1 and type 2 diabetes.
A total of 66 second- (internal) and third- (external) order validation analyses were performed across a range of complications and outcomes simulated by the CORE Diabetes Model (amputation, cataract, hypoglycaemia, ketoacidosis, macular oedema, myocardial infarction, nephropathy, neuropathy, retinopathy, stroke and mortality). Published studies were reproduced in the model by recreating cohorts in terms of demographics, baseline risk factors and complications, treatment patterns and patient management strategies, and simulating the progress of the cohort to an equivalent time horizon.
Correlation analysis on results from 66 validation simulations produced an R2 value of 0.9224 (perfect fit = 1). A correlation plot of published study data versus values simulated by the CORE Diabetes Model had a trend line with a gradient of 1.0187 (perfect fit = 1). Validation analyses in type 1 and type 2 diabetes were associated with R2 values of 0.9778 and 0.8861 respectively. Correlation of second-order validation analyses (model predictions versus observed outcomes in studies used to construct the model) produced an R2 value of 0.9574, and the value for third-order analyses (model predictions versus observed outcomes in studies not used to construct the model) was 0.9023.
The CORE Diabetes Model provides an accurate representation of patient outcomes when compared to 66 studies of diabetes and its complications. Model flexibility ensures it can be used to compare diabetes management strategies in different cohorts across a variety of clinical settings.

35 Reads
    • "The Economic Assessment of Glycemic control and Long-term Effects of diabetes (EAGLE) model provides clinicians with a flexible and comprehensive tool for the simulation of the long-term effects of Diabetes treatment and related costs in Type 1 Diabetes (T1D) and Type 2 Diabetes (T2D) [2] [12] [13], with risk equations for the probability of complications based on regression analysis. The CORE model [14] [15] shares similarities of objective with EAGLE and is based on interdependent Markov models, which simulate the complications of diabetes and nonspecific mortality using time, state, and diabetes type-dependent probabilities [10]. Other useful instruments are represented by the UK Prospective Diabetes Study (UKPDS) Outcomes Model [16] [17] [18] and by the UKPDS Risk Engine [19], both specific for the progression of T2D. "
    [Show abstract] [Hide abstract]
    ABSTRACT: The increasing prevalence of Diabetes and its related complications is raising the need for effective methods to predict patient evolution and for stratifying cohorts in terms of risk of developing diabetes-related complications. In this paper, we present a novel approach to the simulation of a type 1 diabetes population, based on Dynamic Bayesian Networks, which combines literature knowledge with data mining of a rich longitudinal cohort of type 1 diabetes patients, the DCCT/EDIC study. In particular, in our approach we simulate the patient health state and complications through discretized variables. Two types of models are presented, one entirely learned from the data and the other partially driven by literature derived knowledge. The whole cohort is simulated for fifteen years, and the simulation error (i.e. for each variable, the percentage of patients predicted in the wrong state) is calculated every year on independent test data. For each variable, the population predicted in the wrong state is below 10% on both models over time. Furthermore, the distributions of real vs. simulated patients greatly overlap. Thus, the proposed models are viable tools to support decision making in type 1 diabetes. Copyright © 2015. Published by Elsevier Inc.
    Journal of Biomedical Informatics 08/2015; DOI:10.1016/j.jbi.2015.08.021 · 2.19 Impact Factor
  • Source
    • "The present cost-effectiveness analysis was conducted using the CORE Diabetes Model (CDM) in order to project the long-term clinical and cost outcomes associated with liraglutide and its comparators. A detailed description of the model’s design and operational features has been published elsewhere [9, 10]. The CDM is an Internet-based computer simulation model developed to determine the long-term health outcomes (life expectancy, quality-adjusted life expectancy, cumulative incidences of complications) and economic consequences (annual and cumulative costs per patient, costs associated with complications and treatment), as well as incremental cost-effectiveness ratios (ICER) of interventions in type 1 and type 2 diabetes. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background To evaluate the long-term cost-effectiveness of liraglutide versus sitagliptin or exenatide, added to oral antidiabetic drug mono- or combination therapy respectively, in patients with Type 2 diabetes in Greece. Methods The CORE Diabetes Model, a validated computer simulation model, was adapted to the Greek healthcare setting. Patient and intervention effects data were gathered from a clinical trial comparing liraglutide 1.2 mg once daily vs. sitagliptin 100 mg once daily, both combined with metformin, and a clinical trial comparing liraglutide 1.8 mg once daily vs. exenatide 10 μg twice daily, both as add-on to metformin, glimepiride or both. Direct costs were reported in 2013 Euros and calculated based on published and local sources. All future outcomes were discounted at 3.5% per annum, and the analysis was conducted from the perspective of a third-party payer in Greece. Results Over a patient’s lifetime, treatment with liraglutide 1.2 mg vs. sitagliptin drove a mean increase in discounted life expectancy of 0.13 (SD 0.23) years and in discounted quality-adjusted life expectancy of 0.19 (0.16) quality-adjusted life years (QALYs), whereas therapy with liraglutide 1.8 mg vs. exenatide yielded increases of 0.14 (0.23) years and 0.19 (0.16) QALYs respectively. As regards lifetime direct costs, liraglutide 1.2 mg resulted in greater costs of €2797 (€1468) versus sitagliptin, and so did liraglutide 1.8 mg compared with exenatide (€1302 [€1492]). Liraglutide 1.2 and 1.8 mg doses were associated with incremental cost effectiveness ratios of €15101 and €6818 per QALY gained, respectively. Conclusions Liraglutide is likely to be a cost-effective option for the treatment of Type 2 diabetes in a Greek setting.
    BMC Health Services Research 09/2014; 14(1):419. DOI:10.1186/1472-6963-14-419 · 1.71 Impact Factor
  • Source
    • "Importantly, model validation has been a key part of the CDM's development process. A major validation exercise published in 2004 examined the operational predictive validity of the model against 66 clinical end points from 11 epidemiological and clinical studies [13]. Furthermore , the CDM has been closely associated with the Mount Hood Challenge, which initially contrasted model output from the Global Diabetes Model [14] and the CDM [15] and is now an open 1098-3015$36.00 "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background The IMS CORE Diabetes Model (CDM) is a widely published and validated simulation model applied in both type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) analyses. Validation to external studies is an important part of demonstrating model credibility. Objective Because the CDM is widely used to estimate long-term clinical outcomes in diabetes patients, the objective of this analysis was to validate the CDM to contemporary outcomes studies, including those with long-term follow-up periods. Methods A total of 112 validation simulations were performed, stratified by study follow-up duration. For long-term results (≥15-year follow-up), simulation cohorts representing baseline Diabetes Control and Complications Trial (DCCT) and United Kingdom Prospective Diabetes Study (UKPDS) cohorts were generated and intensive and conventional treatment arms were defined in the CDM. Predicted versus observed macrovascular and microvascular complications and all-cause mortality were assessed using the coefficient of determination (R2) goodness-of-fit measure. Results Across all validation studies, the CDM simulations produced an R2 statistic of 0.90. For validation studies with a follow-up duration of less than 15 years, R2 values of 0.90 and 0.88 were achieved for T1DM and T2DM respectively. In T1DM, validating against 30-year outcomes data (DCCT) resulted in an R2 of 0.72. In T2DM, validating against 20-year outcomes data (UKPDS) resulted in an R2 of 0.92. Conclusions This analysis supports the CDM as a credible tool for predicting the absolute number of clinical events in DCCT- and UKPDS-like populations. With increasing incidence of diabetes worldwide, the CDM is particularly important for health care decision makers, for whom the robust evaluation of health care policies is essential.
    Value in Health 09/2014; 17(6):714–724. DOI:10.1016/j.jval.2014.07.007 · 3.28 Impact Factor
Show more