Du Y, Melchert HU, Knopf H, Braemer-Hauth M, Pabel E (2007): Association of serum caffeine concentrations with serum glucose levels in caffeine-drug users and non-users - results of German National Health Surveys
Diabetes Obesity Metabolism 9 (5): 756-758.

The purpose of this study was to explore the association of serum caffeine concentrations with serum glucose levels in caffeine-drug users and non-users, aiming at the chronic effects of caffeine on glucose metabolism in comparison with known acute effects of caffeine. Eight hundred and fourteen caffeine-drug users and 623 non-users were identified from German National Health Surveys. Their serum caffeine concentrations and glucose levels were measured. The associations of caffeine concentrations with glucose levels were established by correlation analysis and multivariable regression analysis in caffeine-drug users and non-users separately. Antidiabetic therapy was considered. Caffeine concentrations were closely positively correlated to serum glucose levels in caffeine-drug users (Spearman r = 0.117, p = 0.001; partial r = 0.102, p = 0.020) particularly in women (Spearman r = 0.155, p < 0.001; partial r = 0.150, p = 0.005) although the correlation was weak as shown by multivariable regression analysis. The serum glucose levels were significantly higher (5.403 +/- 0.033 vs. 5.306 +/- 0.037 mmol/l) whereas the magnesium level was significantly lower (0.8941 +/- 0.0026 vs. 0.9024 +/- 0.0030 mmol/l) in caffeine-drug users than in non-users. No associations of caffeine concentrations with serum glucose levels were found in any groups of caffeine-drug non-users in our study. Whereas acute intake of caffeine-drugs may impair glucose metabolism, chronic intake of caffeine exclusively from diet has little effects on glucose metabolism and therefore may not contribute to the risk reduction of type 2 diabetes that was found in recent coffee consumption studies.