Background Shoulder rotation strength deficit measured at one time-point during preseason has been investigated as a risk factor for shoulder problems in sports with conflicting results. However, athletes face changes in physical demands and accumulative training exposure during a season, which likely influence their rotation strength over this period.

Objective We aimed to investigate if shoulder rotation strength changes during a competitive season in a cohort of youth elite handball players.

Design Prospective cohort study.

Setting Danish youth elite handball clubs.

Patients (or Participants) Players (n=292, 45% girls, 14–18 years) without shoulder problems were assessed at preseason and at midseason.

Interventions (or Assessment of Risk Factors) We measured isometric internal rotation (IR) and external rotation (ER) strength in supine with the shoulder abducted 90° in neutral rotation and 30° of IR measured using hand-held dynamometry, and the corresponding ER/IR ratios.

Main Outcome Measurements Changes in shoulder strength and ER/IR strength ratios.

Results From preseason to midseason, ER/IR mean ratios were higher in neutral rotation (boys difference 0.02, 95% CI 0.00 to 0.05; girls difference 0.05, 95% CI 0.01 0.09) and in 30° IR (boys difference 0.15, 95% CI 0.11 to 0.19; girls difference 0.12, 95% CI 0.07–0.17) due to an increase in ER strength and a decrease in IR strength.

Conclusions The present finding supports that shoulder strength ratios change between two time points during a competitive season. Clinicians and researchers should be aware that strength in youth elite handball players may have a normal variation over the course of the season. This warrants ongoing monitoring and should be considered when normative reference values are compared, and when measuring the effect of targeted exercise programmes. Studies investigating if shoulder strength is a causal factor for shoulder problems should consider including the strength variable as a time-varying covariate.