This study assessed the knowledge, prevalence, and quantity of caffeine use by athletes competing at the 2005 Ironman Triathlon World Championships. Caf-feine-related questionnaires were self-administered to 140 (105 male and 35 female, 40.3 ± 10.7 y) athletes representing 16 countries. Fifty of these athletes further consented to immediate post-race blood samples for analysis of plasma caffeine and paraxanthine using high-performance liquid chromatography (HPLC). Seventy-two percent of 70 athletes correctly identified caffeine as being an unrestricted substance in triathlon. The majority of athletes [125 (89%)] were planning on using a caffeinated substance immediately prior to or throughout the race. Cola drinks (78%), caffeinated gels (42%), coffee (usually pre-race) (37%), energy drinks (13%), and NoDoz tablets (9%) were the most popular caffeinated choices. Mean ± standard deviation (and range) post race plasma caffeine and paraxanthine levels were 22.3 ± 20 µmol/L (1.7 to 98.4) and 9.4 ± 6 µmol/L (1.8 to 28.9), respectively. Seven athletes (14%) finished with plasma caffeine levels ≥40 µmol/L. Plasma values from elite athletes did not differ from age group competitors. Despite the prevalence of its consumption and the training experience of this athletic group, over one quarter of athletes remained either confused or uninformed about caffeine’s legality. Levels of plasma caffeine taken immediately post race indicated that athletes typically finish with quantities of caffeine that have been shown to improve endurance performance (i.e., ≈ 20 µmol/L or a dose of ≥ 3 mg/kg body weight).