Dietary tendencies as predictors of marathon time in novice marathoners.

Background: The effects of dietary factors such as carbohydrate (CHO) on endurance-running performance have been extensively studied under laboratory-based and simulated field conditions. Evidence from "real-life" events, however, is poorly characterized. The purpose of this observational study was to examine the associations between prerace and in-race nutrition tendencies and performance in a sample of novice marathoners.

Methods: Forty-six college students (36 women and 10 men) age 21.3 ± 3.3 yr recorded diet for 3 d before, the morning of, and during a 26.2-mile marathon. Anthropometric, physiological, and performance measurements were assessed before the marathon so the associations between diet and marathon time could be included as part of a stepwise-regression model.

Results: Mean marathon time was 266 ± 42 min. A pre-marathon 2-mile time trial explained 73% of the variability in marathon time (adjusted R2 = .73, p < .001). Day-before + morning-of CHO (DBMC) was the only other significant predictor of marathon time, explaining an additional 4% of the variability in marathon time (adjusted R2 = .77, p = .006). Other factors such as age, body-mass index, gender, day-before + morning-of energy, and in-race CHO were not significant independent predictors of marathon time.

Conclusions: In this sample of primarily novice marathoners, DBMC intake was associated with faster marathon time, independent of other known predictors. These results suggest that novice and recreational marathoners should consider consuming a moderate to high amount of CHO in the 24-36 hr before a marathon.