Table 2

Four common questions in sports injury research and how they can be addressed using time-to-event analysis

Time-to-event analysis and time-varying exposures: four common questions in sports injury research
Question 1
What is a time-varying exposure?
Sports injury researchers are addressing the question ‘How much change in training load is too much before injury is sustained, among athletes with different characteristics?’ Does time-to-event modelling allow them to include change in training load as a time-varying exposure to sport injury development?
Key point 1: Time-to-event modelling is well suited to deal with time-varying exposures and its association with sports injury. When using a time-varying training load exposure, the primary exposure of interest must be labelled ‘change in training load’, not ‘training load’. Sudden spikes and reductions in training load are not exposure variables, but exposure levels (known as ‘states’). Consequently, researchers do not examine the association between sudden spikes in training load and injury. They examine the association between changes in training load and injury.
Question 2
Why time-to-event modelling?
Why is time-to-event analysis superior to other analytical concepts when analysing training load-related data that changes status over time?
Key point 2: Unlike logistic regression and Χ2 test, time to event allows flexibility for the sports injury researcher to take into account censoring and compare injury risk across time-varying exposures by using delayed entry functions.
Question 3
Considerations regarding training load-related time-varying exposures
How should sports injury researchers include change in training load in a time-to-event analysis?
Key point 3a: In a time-to-event analysis, change in training load can be included as a categorised variable. This enables for examining non-linear dose–response relationships in the association between changes in training load and sports injury.
Key point 3b: The acute:chronic workload ratio (ACWR) concept is a concrete example of a categorised time-varying training load. In this, the cut-off points of 0.8 and 1.3 have been suggested. It must be stressed that the most suitable value(s) for the cut-off points and whether changes in training load should be based on absolute and/or relative changes remain areas of uncertainty and discussion.
Key point 3c: The sports injury researcher can examine changes in training load using either fixed states and/or transitions between states. Therefore, sports injury researchers are advised to specify up front whether the main objective of the study is to examine injury risk in relation to (1) different (but constant) workload states (eg, low, medium/sweet, high), or (2) the transitions between workload states (eg, moving from high to medium).
Question 4
What about other time-varying exposures?
Many other factors than training load also change status over time (eg, body mass, strength, flexibility). Are sports injury researchers able to include such variables into the time-to-event analysis?
Key point 4: Yes. Many variables change over a player’s season or career that can be important to consider in respect to sports injury. These variables can be included as time-varying effect-measure modifiers and/or time-varying confounders. The challenge in this scenario is to have sufficient data to support inclusion of these variables.