This dissertation is an epidemiologic query designed to address important questions related to screening and diagnosis of prediabetes and diabetes. We used data from the Atherosclerosis Risk in Communities (ARIC) Study, an ongoing community-based cohort initiated in 1987, to assess four main aims.First, we examined trajectories of kidney function by diabetes status. Persons with diagnosed diabetes declined almost twice as rapidly as persons without diabetes (-2.5 ml/min/1.73 m2 per year [95%CI, -2.6, -2.4]; -1.4 ml/min/1.73 m2 per year [95%CI, -1.5, -1.4], respectively). Those with undiagnosed diabetes, likely early in the disease course, experienced relatively little short-term average decline, but declined in the long-term. This brief period of stagnation suggests potential hyperfiltration.Second, we compared the prognostic performance of five definitions of prediabetes in use by international organizations. Prediabetes prevalence ranged widely depending on the definition (9% to 38%), but all definitions identified persons at high-risk for clinical outcomes. Fasting glucose 100-126 mg/dL was more sensitive for clinical complications, while HbA1c 5.7-6.4% and 6.0-6.4% were more specific and provided modest statistically significant improvements in risk discrimination.Third, we conducted a diagnostic testing study, comparing 1,5-anhydroglucitol (1,5-AG), a less burdensome test, to 2-hour glucose for identification of hyperglycemia. Their concordance was low and, while specific for high glucose levels, 1,5-AG missed many people with elevated 2-hour glucose. Finally, we compared the prospective associations of 1,5-AG and 2-hour glucose with risk of future clinical outcomes. Low levels of 1,5-AG were associated with risk of future outcomes, but not as strongly as 2-hour glucose.Given the proven approaches to prevent progression to diabetes and reduce incidence of its complications, our data reinforced the need for prediabetes and diabetes screening and diagnosis, with consistent identification of individuals to inform public health planning. We proposed further understanding hyperfiltration and exploring new ways to optimize prediabetes definitions. While 1,5-AG does not appear to substitute for 2-hour glucose and may have limited screening utility, additional research is needed to assess it as a measure of glycemic control in persons with diagnosed diabetes. Our research directly informs important next steps to promote prevention of diabetes and its complications.
【 预 览 】
附件列表
Files
Size
Format
View
Screening and diagnosis of prediabetes and diabetes: epidemiologic research to inform prevention