Relationship between IGA, cortisol, and mood state responses to acute and long-term swim training in elite college swimmers

The prediction of athletic performance is a concern among sport scientists. Within sport psychology, one of the predominant methods for predicting performance is utilizing a trait or state psychological instrument, such as the Profile of Mood States (POMS). However, the concurrent assessment and correlation of other relevant physiological measures with the psychological measures obtained throughout the season of training and competition would appear to improve prediction of athletic performance. The present study examined the effect of acute and long-term swim training on the relationship between the POMS, immunoglobulin A (IGA), and cortisol, and whether these factors taken together could significantly predict athletic performance. Twenty-one Male and eighteen female elite college swimmers were recruited from a major midwestern university for participation. IgA, cortisol, and the POMS taken together accounted for 24% of the variance in the performance criterion variable (p $\u3c$.13). Even though the prediction equation was not significant, it did account for more variance than did the POMS questionnaire alone (5%). The secretion of IgA declined progressively in response to the acute and the long-term swim training suggesting an increased susceptibility to upper respiratory infections. The secretion of cortisol increased in response to the acute and the long-term swim training in only one of the four assessments suggesting that the training had little effect on cortisol. The analysis of the POMS questionnaire revealed that the athletes reported having more vigor, and less tension, depression, anger, fatigue, and confusion during the fourth phase of this study. Weak correlations were documented between IgA and cortisol, and IgA and the POMS suggesting that for this study, the data do not indicate any reliable relationships between these parameters. In conclusion, if relevant parameters could be identified which contribute to the prediction of athletic performance, it might be possible to train athletes more effectively on an individual basis. A biopsychosocial model is recommended in order to account for as much variance as possible in athletic performance