Analysis of interrupted time-series relating to statewide sports injury data

This study aims to present a new approach towards the analysis of intervention time-series studies in the context of sports-related injury data. We used Victoria-wide hospital admission injury data associated with the sport of Australian football during the period 2006 to 2013. To estimate the state-wide effect of an implemented exercise training intervention that aimed to reduce the number of football-related injuries, time-series analysis was performed using a generalised least square (GLS) method. We show how the GLS method can be used to evaluate the impact of the intervention. Trend and seasonal patterns time series were also assessed using the 'Seasonal and Trend decomposition using Loess' nonparametric seasonal decomposition procedure. The model identified a decreasing trend in the seasonally adjusted number of injuries after the implementation of the intervention in the hospital admission data. The seasonal decomposition plots also indicate strong seasonal patterns in the injury time series.