PubMed Health⌕ Search
Home/Archive/PubMed

PubMed · 11080563

The Mt. Hood challenge: cross-testing two diabetes simulation models.

J B Brown·A J Palmer·P Bisgaard·W Chan·K Pedula·A Russell
View original source ↗Export metadata ↓

Abstract

Starting from identical patients with type 2 diabetes, we compared the 20-year predictions of two computer simulation models, a 1998 version of the IMIB model and version 2.17 of the Global Diabetes Model (GDM). Primary measures of outcome were 20-year cumulative rates of: survival, first (incident) acute myocardial infarction (AMI), first stroke, proliferative diabetic retinopathy (PDR), macro-albuminuria (gross proteinuria, or GPR), and amputation. Standardized test patients were newly diagnosed males aged 45 or 75, with high and low levels of glycated hemoglobin (HbA(1c)), systolic blood pressure (SBP), and serum lipids. Both models generated realistic results and appropriate responses to changes in risk factors. Compared with the GDM, the IMIB model predicted much higher rates of mortality and AMI, and fewer strokes. These differences can be explained by differences in model architecture (Markov vs. microsimulation), different evidence bases for cardiovascular prediction (Framingham Heart Study cohort vs. Kaiser Permanente patients), and isolated versus interdependent prediction of cardiovascular events. Compared with IMIB, GDM predicted much higher lifetime costs, because of lower mortality and the use of a different costing method. It is feasible to cross-validate and explicate dissimilar diabetes simulation models using standardized patients. The wide differences in the model results that we observed demonstrate the need for cross-validation. We propose to hold a second 'Mt Hood Challenge' in 2001 and invite all diabetes modelers to attend.

Explore related subjects

AlbuminuriaAmputation, SurgicalBlood PressureComputer SimulationDiabetes Mellitus, Type 2Diabetic RetinopathyHumansMarkov ChainsModels, StatisticalMonte Carlo MethodMyocardial InfarctionPrognosisProteinuriaSurvival RateTreatment OutcomeComparative StudyJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

J B Brown, A J Palmer, P Bisgaard, W Chan, K Pedula, A Russell. 2000. The Mt. Hood challenge: cross-testing two diabetes simulation models.. https://doi.org/10.1016/s0168-8227(00)00217-5

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2007

Urinary albumin excretion is associated with impaired flow- and nitroglycerin-mediated brachial artery dilatation in hypertensive adults.

We investigated whether the urinary albumin/creatinine ratio (UACR), a measure of albuminuria, is associated with non-invasive measures of arterial function in hypertensive adults without known coronary heart disease (CHD) or stroke. UACR was measured in the first voided morning urine sample in 469 non-Hispanic white hypertensive individuals (mean age 62.2+/-9.8 years, 41% men) belonging to hypertensive sibships. High-resolution ultrasonography of the brachial artery was used to assess flow-mediated dilatation (FMD)--an endothelium-dependent response--and nitroglycerin-mediated dilatation (NMD)--an endothelium-independent response. Because of skewed distribution, UACR was log transformed after addition of 0.1. The association of log (UACR+0.1) with FMD and NMD, before and after adjustment for CHD risk factors, serum creatinine, and hypertension medication and statin use was assessed using linear regression analyses. In univariable analyses, variables associated with lower FMD were greater age, male sex, history of smoking, lower high-density lipoprotein (HDL) cholesterol, higher serum creatinine and higher log (UACR+0.1); variables associated with lower NMD were greater age, male sex, higher systolic blood pressure, lower HDL cholesterol, higher serum creatinine and higher log (UACR+0.1). In separate stepwise multivariable regression analyses that adjusted for conventional CHD risk factors, serum creatinine and hypertension medication and statin use, higher log (UACR+0.1) was associated with lower brachial artery FMD (P=0.035) and NMD (P=0.0002). These findings highlight the association of increased urinary albumin excretion with impaired vascular reactivity in hypertensive individuals.

Albuminuria↗
PubMed2007

What predicts progression and regression of urinary albumin excretion in the nondiabetic population?

An increase or decrease in urinary albumin excretion (UAE) is associated with, respectively, a higher or lower risk for renal and cardiovascular disease, independent of widely known cardiovascular risk factors. This study aimed to identify factors that are associated with changes in UAE in the nondiabetic population using data of the Prevention of Renal and Vascular End stage Disease (PREVEND) Study, a community-based prospective cohort study. Data of the 6647 nondiabetic participants who completed the first (1997 through 2001) and second (2001 through 2003) screening were used. Change in UAE was categorized as regression (n = 650), stable (n = 5240), or progression (n = 757) on the basis of change in class during follow-up, with classes being a UAE <15, 15 to 30, 30 to 300, and >300 mg/24 h. With the use of stepwise forward multinomial regression analysis changes in BP, fasting glucose concentration, and start of antihypertensive drugs were found to be the most important modifiable variables associated with the risk for progression and regression (P < 0.01 for likelihood ratio test). The odds ratios to develop regression or progression of UAE during follow-up were 0.64 (95% confidence interval [CI] 0.57 to 0.73) and 1.91 (95% CI 1.72 to 2.12), respectively, per 10-mmHg increase in BP during follow-up, 0.89 (95% CI 0.80 to 0.98) and 1.09 (95% CI 1.01 to 1.17), respectively, per 1-mmol/L increase of fasting glucose levels during follow-up, and 1.57 (95% CI 1.21 to 2.06) and 0.70 (95% CI 0.51 to 0.95), respectively, for start of antihypertensive drugs during follow-up. These associations were independent of baseline BP, glucose, body mass index, estimated GFR, and UAE and changes in high-sensitivity C-reactive protein during follow-up. In conclusion, changes in glucose concentration and BP and start of antihypertensive drugs (angiotensin-converting enzyme inhibitors/angiotensin receptor blockers in >50% of cases) are associated with progression and regression of UAE in the nondiabetic population. Although associations do not necessarily suggest causality, it is hypothesized that in the general population, the most important ways to reduce UAE are by lowering glucose concentration and BP (including start of antihypertensive medication), even in normotensive, nondiabetic individuals.

Albuminuria↗