Horse Racing as a Model to Study the Relationship between Air Pollutants and Physical Performance

SIMPLE SUMMARY: Ambient air contains a mixture of pollutants, the effects of which on animal and human health have been widely described. In contrast, the effects on physical performance are poorly understood, largely due to the difficulty of implementing an experimental model to study this problem. Thoroughbred horse racing involves many animal athletes, of similar genetics, environmental exposure, training, and diet, who participate by breathing varying mixtures of ambient air. This paper presents an analysis strategy based on the homogeneity of the races, the distance, and the design of the track. This paper presents a preliminary analysis in which we observe that the level of performance is decreased by concentrations of PM(10), PM(2.5), NO(2), NO, and CO in the air. Thus, we note that this natural experiment may constitute a model of interest to advance the understanding of the problem. ABSTRACT: This study proposes the theoretical principles for the selection of a sample of horse races to study the relationship between air pollutants and performance. These criteria were then applied to an original dataset comparing the correlations between these variables obtained in “Handicap” versus “Conditional” type races. Methods: The mean concentration of pollutants during the six hours prior to the race and the speed of the test were determined in 441 official races at a racecourse in Santiago, Chile, during the summer and winter months of 2012. Using layout, track condition and distance (1000, 1100 and 1200 m) as criteria, a homogeneous group of races (“Handicap”; n = 214) versus a heterogeneous group (“Conditional”; n = 95) were compared using simple correlations (Spearman’s test). Results: Race speed was related to greater levels of PM(10), PM(2.5), NO(2), NO and SO(2) and it was positively related to O(3), a trend that was observed in the 1000, 1100 m races and in the total “Handicap” group. Similar results were observed only in 1000 m for the “Conditional” group with lower Rho, except for PM(10) and PM(2.5). The total races of the conditional group showed lower Rho values and significant associations of the same trend for CO, NO(2), NO and SO(2). Conclusions: Horse races between 1000 and 1200 m of the “Handicap” type appear to be an interesting group to study the relationship between air pollutants and the performance of racehorses. In the future, our observations should be expanded to other distances and other types of races.