Self-report (SR) has been the primary method used to assess fluid intake during endurance events, but unfortunately, little is known about the validity of SR. The purpose of this study was to compare SR fluid intake with direct measurement (DM) during a 70.3-mile triathlon.
Fifty-three (42 men, 11 women) individuals competing in a 70.3-mile triathlon participated in the study. On the 13.1-mile-run section of the triathlon, 11 research stations provided fluid in bottles filled with 163 mL of water or carbohydrate-electrolyte beverage (CEB). Participants submitted bottles 25 m past aid stations to be reweighed postrace. Participants also answered questions regarding fluid intake postrace. Bland-Altman plots and 95% limits of agreement were used to assess precision of the measures, while least-squares regression assessed linear agreement.
SR intakes during the run ranged from 0–1793, 0–1837, and 0–2628 mL for water, CEB, and total fluid, with corresponding DM intakes of 0–1599, 0–1642, and 0–2250 mL. DM and SR showed strong linear agreement for water, CEB, and total fluid (R2 = .71, .80, and .80). Mean differences between the measures on the Bland-Altman plots were small (13–41 mL), but relatively large differences (±500 mL) between the measures were apparent for some participants.
SR is the predominant methodology used in field studies assessing hydration, despite little to no data confirming its validity. The results herein suggest that fluid-intake-assessment methodology should be chosen on a case-by-case basis and that caution should be used when interpreting data based on SR.
The authors are with the School of Kinesiology, University of Minnesota, Minneapolis, MN.