Abstract P052: Glucometabolic State Transitions

    loading  Checking for direct PDF access through Ovid

Abstract

Background: To assess the frequency and determinants of transitions between normal fasting glucose (NFG), prediabetes and diabetes in an African American cohort.

Methods: We included participants from the Jackson Heart Study who attended the baseline exam (2000-2004) and at least one of two subsequent exams (2005-2008 and 2009-2013), and were not missing data (fasting glucose, hemoglobin A1c, diabetes medication) used classify the glycemic status of participants as NFG, prediabetes, and diabetes at each visit. Transitions were defined as progression (deterioration) or remission (improvement) of glycemic status. Two transitions were possible for those attending all 3 visits. A generalized multinomial logit model, accounting for repeated measures from individuals with 2 transitions as clusters, was used to estimate the odds ratios (OR) for remission versus stable and progression versus stable, adjusting for demographic, anthropometric, behavioral, and biochemical factors.

Results: Among 3,316 participants (baseline mean age 55 years, 36% male, 32% with IFG, 24% with diabetes) with data on at least one consecutive transition, 58.5% remained in their baseline status, 32.5% experienced a progression, 3.0% a remission, and 6.1% both a remission and progression. Among those classified as NFG, prediabetes, and diabetes at baseline: 1) progression occurred among 59%, 39%, and 2%; and 2) remission occurred among 7%, 13%, and 7%, respectively. The average weight change between exams was +1.4 kg for those experiencing only a progression and -4.5 kg for those with only a remission. Subsequent analyses included 2171 with complete data for all covariates which included baseline age, sex, education, BMI, smoking, alcohol use, hypertension medication, systolic blood pressure, log(triglycerides), baseline diabetes status, change in weight and change in log(CRP) between visits. The outcome of remission was associated only with weight change. A 1 kg increment in weight change was associated with a reduced likelihood of remission (OR 0.97, 95%CI: 0.96 - 0.99, p<0.001). The outcome of progression was associated with baseline BMI, log(triglycerides), diabetes status, and weight change (OR for 1 kg increment weight change 1.04, 95%CI: 1.03 - 1.06, p<0.001).

Conclusions: In a contemporary adult African American population, transitions between glucometabolic statuses were frequent and most commonly involved deterioration. Weight changes between examinations were strongly associated with both remission and progression.

Related Topics

    loading  Loading Related Articles