Pancreatic beta-cell function as a predictor of cardiovascular outcomes and costs: findings from the Cardiovascular Health Study.
Curtis LH., Hammill BG., Bethel MA., Anstrom KJ., Liao L., Gottdiener JS., Schulman KA.
OBJECTIVE: To explore relationships between beta-cell function and incident cardiovascular events, death, and medical costs among elderly individuals. RESEARCH DESIGN AND METHODS: In a prospective, population-based cohort of 4555 elderly individuals, we examined the effect of beta-cell function on incident cardiovascular events and mortality. We also examined costs for 3715 of these individuals. We used the computer-based homeostasis model assessment (HOMA) to calculate indices of beta-cell function (HOMA-%B) and insulin sensitivity (HOMA-%S) using baseline fasting glucose and insulin levels. All subjects were followed from 1992/1993 for 6 years or until death. MAIN OUTCOME MEASURES: Discrete-time survival model of the effects of beta-cell function on incident cardiovascular events and all-cause mortality; and semiparametric estimators for calculations of mean 6-year costs. RESULTS: Controlling for HOMA-%S, a 20% decrease in HOMA-%B was associated with increased odds of incident cardiovascular events (odds ratio [OR], 1.09; 95% confidence interval [CI], 1.05-1.14) and death (OR, 1.10; 95% CI, 1.07-1.14). The relationships persisted after controlling for clinical and sociodemographic confounders. A 20% decrease in HOMA-%B was also associated with increased costs (cost ratio, 1.03; 95% CI, 1.01-1.05). The significant association did not persist after controlling for confounders. LIMITATIONS: The sample comprises relatively healthy elderly individuals and is based on data from 1992 through 1999, which may not reflect current experience. The measure of beta-cell function is an estimate generated from single measures of glucose and insulin. CONCLUSIONS: Beta-cell function as measured by HOMA-%B is a significant predictor of incident cardiovascular events and mortality but not of costs, controlling for HOMA-%S and sociodemographic and clinical confounders.