What did I do?

I determined statistical approaches for handling complex assumptions in the relationship between training load and injury risk.

Why did I do it?

Training load can potentially be modified to reduce injury risk. Although the amount of research on training load and injury risk has increased in recent years,1 our understanding of this complex relationship remains limited. Previous studies have applied a plethora of statistical approaches with suboptimal or unknown performance.2 The variety of approaches has confused clinicians attempting to apply latest research, and has barred systematic reviews from comparing studies through meta-analyses.3 Despite these challenges, few recommendations exist for statistically handling hypothesised assumptions between training load and injury risk. Consequently, researchers have continued to use suboptimal approaches.

How did I do it?

I first conducted a brief review of the training load and injury risk field to obtain a rough estimate of how missing data were currently handled.4 This information would guide my methodological choices for comparing methods of dealing with missing data. This review encompassed only studies on sport populations published in the last decade (n=108).

In three studies,4–6 I simulated hypothetical relationships between training load and the probability of injury, using 1-season cohort data from Norwegian Men’s Premier League football (n=39 players) and Norwegian academy U-19 football (n=81 players). Ideally, a study uses multiple measures to capture different aspects of training load. Athlete-reported training load was …