Objective: To describe the associations between coffee consumption and glucose tolerance among elderly subjects.
Design: A survey among an unselected non-institutionalized elderly population. Diabetes was assessed on the basis of self-reports and 2-h oral glucose tolerance test for the subjects on diet treatment or with normal glucose tolerance (NGT). Coffee consumption was assessed by a questionnaire.
Setting: Three municipalities in Northern Finland Subjects: All non-institutionalized subjects with baseline normal glucose tolerance or impaired glucose tolerance (IGT) aged 70 years or over.
Main outcome measures: Incidence, persistence or impairment of abnormal glucose tolerance (AGT).
Results: Forty-two percent of those with NGT and 0-5 cups of coffee daily developed AGT, whereas the corresponding figure was 25% for those who drank more than five cups coffee daily. Fifty-five percent of the subjects with IGT or NGT who drank 0-5 cups coffee daily either persisted in IGT or developed AGT. The corresponding figure was 30% for those drinking more than five cups coffee daily. After adjustment for age, body mass index, cardiovascular disease, smoking, alcohol consumption and physical exercise, among those with NGT, the odds ratio (OR) for developing AGT was 2.3 for those drinking 0-5 cups compared to those drinking more than five cups of coffee daily (95% confidence interval (CI) 0.7-7.2). Correspondingly, low coffee consumption was the most powerful predictor associated with the persistence of IGT or the development of AGT (OR 2.9; 95% CI 1.2-6.9).
Conclusions: Low coffee consumption predicts impairment of AGT or persistence of IGT among elderly subjects.