C-Reactive Protein Partially Mediates the Inverse Association Between Coffee Consumption and Risk of Type 2 Diabetes: The UK Biobank and the Rotterdam Study Cohorts

Abstract

ObjectivesGiven its popularity, there is an increasing interest in the study of coffee intake and its effect on health. Previous studies linked coffee consumption to lower type 2 diabetes (T2D) risk. However, potential underlying mechanisms remain unclear. We hypothesized that coffee’s effects on systemic inflammation may play a role. We studied cross sectional and longitudinal associations of habitual coffee consumption with T2D risk and inflammation. MethodsParticipants from UK Biobank (UKB, n = 145,370) and Rotterdam Study (RS, n = 7172) cohorts were included. Coffee intake data were collected through self-administrated food frequency questionnaire or during home interviews. We studied associations of coffee intake with incident T2D using cox proportional hazard models; with longitudinally measured insulin resistance (HOMA IR) through linear mixed effect models; with serum baseline levels of inflammation markers using linear regressions; and the role of inflammation in coffee-T2D associations using mediation analysis. Models were adjusted for sociodemographic, lifestyle and health factors. Results were respectively expressed as hazard ratio (HR); β log transformed HOMA IR level; β log transformed ug/mL; and percentage mediated; and 95% confidence interval [95% CI]. ResultsUKB participants were 58% female and 55.2 years in average; RS were 59.7% female and 65.1 years. The median follow up was 7 (UKB) and 9 (RS) years. The modal coffee consumption was 0.5–2 cups/day (UKB) and 3–4 cups/day (RS). An increase of one coffee cup/day was associated with 4–6% lower T2D risk (RS HR 0.94 [95% CI 0.90; 0.98]; UKB HR 0.96 [0.94; 0.98]); lower HOMA IR (RS β –0.017 [–0.024; –0.010]); lower C reactive protein (CRP, RS β –0.014 [–0.022; –0.005]; UKBB β –0.011 [–0.012; –0.009] and higher adiponectin (RS β 0.025 [0.007; 0.042]. About coffee types, habitual consumers of filtered coffee had the lowest T2D risk (UKB HR 0.88 [0.83; 0.93]), compared to decaffeinated or instantaneous coffee. CRP levels mediated 9.6% (UKB) and 3.4% (RS) of the total effect of coffee on T2D. Adiponectin also showed evidence for mediation. ConclusionsCoffee’s beneficial effects on lower T2D risk may be partially mediated by improvements in systemic inflammation. Among coffee drinkers, filtered coffee may be of preference. Funding SourcesPartially funded by the Institute for Scientific Information on Coffee.