BackgroundThe association between coffee consumption and incident hypertension remained inconsistent. ObjectiveThis study aimed to examine the association of consumption of coffee, including different types (instant and ground) and with different additives, with new-onset hypertension, and evaluate whether genetic variation in caffeine metabolism and inflammation may modify the association. DesignThis study utilized a prospective cohort design. Participants/settingA total of 98 765 participants free of hypertension enrolled in the UK Biobank between 2006 and 2010 were included. Dietary coffee consumption was collected using 24-hour dietary recall questionnaires. Main outcome measuresThe study outcome was new-onset hypertension, ascertained by self-reported medical conditions, hospital inpatient records, death registers, and primary care records. Statistical analyses performedCox proportional hazards models were used to estimate hazard ratio and 95% CI. Potential modifying effects were assessed by likelihood ratio testing. ResultsDuring a median follow-up of 12.1 years, 7090 (7.2%) new-onset cases of hypertension were documented. Overall, regardless of coffee type (instant or ground) and whether adding milk to coffee, there was a U-shaped association between unsweetened coffee consumption and new-onset hypertension, with a 14% to 18% reduction of hypertension risk at >1 to ≤4 drinks per day, whereas a null association was observed between sweetened coffee consumption and the risk of new-onset hypertension. Relative to coffee nonconsumers, the adjusted hazard ratio (95% CI) of new-onset hypertension for participants who drinking unsweetened coffee 1 or fewer, >1 to 2, >2 to 3, >3 to 4, and >4 drinks/day were 0.92 (95% CI 0.85 to 1.00), 0.82 (95% CI 0.76 to 0.89), 0.86 (95% CI 0.79 to 0.96), 0.86 (95% CI 0.77 to 0.96), and 0.88 (95% CI 0.78 to 0.99), respectively. Moreover, a stronger inverse association between moderate consumption of unsweetened coffee and new-onset hypertension was found in participants with higher C-reactive protein levels (P for interaction =.012), whereas an individual’s genetic variation in caffeine metabolism did not significantly modify the association (P for interaction = .453). ConclusionsRegardless of the type of coffee (instant or ground) or the addition of milk to coffee, moderate consumption of unsweetened coffee (>1 to ≤4 drinks/day), but not sweetened coffee, was associated with a lower risk of new-onset hypertension.