METHODS
This study was approved by the Institutional Review Board of HealthSouth Medical Center (Birmingham, Alabama). In this study, the 481 participants from a previous study were followed over the course of 10 years (1999- 2008). All were healthy, active youth baseball pitchers (aged 9 to 14 years) at the time that the study was initiated. Although we know of no rule prohibiting girls from playing in the participating leagues,
all participants were boys, who were contacted by telephone each fall and invited to participate in an annual survey.
The participant was asked whether he played baseball during the past 12 months and what positions he played. If he had pitched,
he was asked how many innings he had pitched and for what teams during the past fall, winter, spring, and summer. Whereas the number of pitches thrown would have been preferred instead of the number of innings pitched, most young pitchers and their families did not typically keep track of their annual pitch counts.
Players were also asked what types of pitches they threw in competition (fastball, curveball, etc) and whether they participated in any showcases.
The annual survey included medical questions.
The participant was asked whether he experienced any pain from pitching or other throwing, where the pain was located, and whether he received any medical treatment. Furthermore,
he was asked whether the pain or injury had led to any missed practice, missed games, or surgery.
He was also asked whether he planned on playing organized baseball in the future or he was retired. When a player declared that he was retired,
he was asked if the reason was injury, failure to make the team, lost interest, or another reason. If a player did not play baseball for 2 consecutive years and indicated that he was retired, he was removed from the follow-up list
and called no more.
The number of injuries was determined by identifying all players who had elbow surgery, had shoulder surgery, or had retired owing to throwing injury during the 10-year study. Retirement or surgery resulting from other activities (eg, collisions, weightlifting, batting, fielding, and nonthrowing sports) were not considered an injury in the current study.
The players were then divided into 3 groups: injured, successful, and short-term pitchers. Injured pitchers were those who had elbow surgery, shoulder surgery, or retired owing to injury. The other participants were divided into the remaining 2 groups based on how many years they
pitched in the study; successful pitchers were those who pitched at least 4 years during the study, whereas shortterm pitchers were those who pitched 3 or fewer years. Data for the 3 groups were compared with analyses of variance. Differences were considered significant when P\.05.
Fisher exact test was used to compare the risk of injury between players who pitched 4 or more years during the study and those who pitched less than 4 years. Fisher exact test was then used to investigate the risk of injury for each of the 3 proposed risk factors—high volume of pitching,
throwing curveballs at a young age, and concomitantly playing catcher. To test the risk of injury from high volume, injury rates were compared between pitchers who pitched more than a particular number of innings in at least 1 calendar year and those who did not. Specifically, Fisher exact tests were run using cutoff points of 70, 80, 90, 100, 110, 120, and 130 innings. The cutoff point was used from the Fisher exact test that produced the lowest P value. Similarly, the Fisher exact test with the lowest P value was identified for throwing curveballs before age 10, 11, 12, 13, 14, and 15 years. Participants who played catcher for at least 3 years during the study were considered to be catchers.
Because of the potential that pitchers who simply pitched more years were more likely to experience the proposed risk factors, the data were stratified and reanalyzed to examine this potential exposure bias. Specifically, Fisher exact tests were performed separately for those who pitched
at least 4 years in the study and for those who pitched less than 4 years in the study. Results for each Fisher exact testwere considered significant if P \ .05.
Bookmarks