The Influence of Time of Season on Injury Rates and the Epidemiology of Canadian Football Injuries.

OBJECTIVE: To describe injury rates and injury patterns in the Canadian Football League (CFL) according to time during the season, player position, injury type, and injury location. DESIGN: Prospective, cohort study. SETTING AND PARTICIPANTS: Eight seasons from CFL injury surveillance database. INDEPENDENT VARIABLES: Depending on the analysis, time of season (preseason, regular, and playoffs), player position, injury type, and injury location. MAIN OUTCOME MEASURES: Medical attention and time-loss injury rates per 100 athletes at risk (AAR), and prevalence of time-loss injuries per week. RESULTS: The average game injury rate was 45.2/100 AAR medical attention injuries and 30.7/100 AAR time-loss injuries. Injury rates declined by 1% per week over the season for both medical attention (rate ratio = 0.99) and time-loss (rate ratio = 0.99) injuries, with a substantial decline during the playoffs compared with preseason (rate ratio = 0.70-0.77). The number of ongoing time-loss injuries increased over the course of the regular season. Quarterbacks, offensive backs, and linebackers had the highest game injury rates. Joint/ligament and muscle/tendon injuries were the most common injury types for games and practices, respectively. The lower extremity was the most commonly affected area, specifically the lower leg/ankle/foot and hip/groin/thigh. CONCLUSIONS: There was a 1% decline in injury rate per week during the season and a 30% decline during the playoffs. The number of ongoing time-loss injuries increased over the regular season. Current results can aid league officials and medical staff in making evidence-based decisions concerning player safety and health.

Challenges in interpreting results from 'multiple regression' when there is interaction between covariates.

Properly interpreting research results is the foundation of evidence-based medicine. Most observational studies use multiple regression and report adjusted effects. In randomised trials, adjusted effects are often provided when there are chance baseline imbalances. The estimates for the exposure of interest (eg, treatment) from these adjusted analyses are usually interpreted as population average causal effects (PACEs); for example, what would be the difference in the mean outcome if everyone in the population was treated versus untreated? In this paper, we show this interpretation is incorrect when there is an interaction between treatment and other variables with respect to the outcome. We provide the appropriate methods to calculate the PACE from regression analyses and also introduce alternative methods that have gained popularity over the last 20 years. Finally, we explain why researchers should be cautious when excluding interaction terms based on p values.