We sought to assess the effect of coffee consumption on blood pressure in humans. Our data sources included a MEDLINE search of the literature published before December 1997, bibliography review, and expert consultation. We selected controlled trials in which coffee consumption was the only difference between the intervention and control groups, mean blood pressure change was reported for each group or period, and treatment lasted for >24 hours. Of 36 studies initially identified, 11 (522 participants) met these inclusion criteria. Information on sample size, study design, participant characteristics (gender, race, age, baseline blood pressure, and antihypertensive medications), and treatment results were abstracted by 3 reviewers using a standardized protocol. Treatment effect of coffee consumption on blood pressure was estimated with the use of a random-effects model. In the 11 trials, median duration was 56 days (range, 14 to 79 days), and median dose of coffee was 5 cups/d. Systolic and diastolic blood pressure increased by 2.4 (range, 1.0 to 3.7) mm Hg and 1.2 (range, 0.4 to 2.1) mm Hg, respectively, with coffee treatment compared with control. Multiple linear regression analysis identified an independent, positive relationship between cups of coffee consumed and subsequent change in systolic blood pressure, independent of age of study participants and study design characteristics. The effect of coffee drinking on systolic and diastolic blood pressure was greater in trials with younger participants. Our findings provide support for a relationship between coffee consumption and higher blood pressure. Trials of coffee cessation of longer duration and in persons with hypertension should be performed.