Predictive validity of an injury score among high school basketball players

GRUBBS, N., R. T. NELSON, and W. D. BANDY. Predictive validity of an injury score among high school basketball players. Med. Sci. Sports Exerc. Vol. 29, No. 10, pp. 1279-1285, 1997. A number of strategies have been investigated in an attempt to identify those individuals most likely to be injured during participation in sports activity. One such strategy identified in the literature involved computing an injury score via a logistic regression equation using measures of structural symmetry to predict the likelihood of athletic injury. The purpose of this study was to investigate the predictive value of the injury score among high school basketball players. Following the establishment of reliability of measures, injury scores were calculated for 62 high school basketball players (34 females, 28 males) before the start of the season. Lower extremity injuries sustained while playing basketball were recorded throughout the season. The predictive value of the injury score equation was determined by calculating sensitivity, specificity, and positive and negative predictive values. The sensitivity and specificity were calculated to be 16.7% and 66.1%, respectively. The positive and negative predictive values were calculated to be 5.90% and 88.1%, respectively. These results indicate that the injury prediction score investigated was not a valid means of predicting injury in high school basketball players. Limitations, possible implications of these findings, and ideas for future related research are presented.