The effect of phased caffeine withdrawal and abstention on tinnitus severity was assessed using a pseudo-randomized, double-blinded, placebo-controlled crossover trial of 30 days duration. Sixty-six volunteers who experienced tinnitus and who usually consumed at least 150 mg/day of caffeine participated. The intervention was a direct replacement of usual caffeinated tea/coffee with double-blinded supplies, under one of two conditions. Condition 1: Maintenance followed by phased withdrawal. Condition 2: Phased withdrawal followed by reintroduction and maintenance. Tinnitus severity was measured by the total score of the Tinnitus Questionnaire on Days 1, 15, and 30. Secondary measures included twice daily self-rated symptoms relevant to tinnitus and caffeine withdrawal. Caffeine had no effect on tinnitus severity, the mean difference between caffeinated and decaffeinated days being -0.04 (95% confidence interval -1.99 to 1.93), p=0.97. Significant acute adverse symptoms of caffeine withdrawal were observed. No evidence was found to justify caffeine abstinence as a therapy to alleviate tinnitus, but acute effects of caffeine withdrawal might add to the burden of tinnitus.