Background Previous studies on high level athletes has shown that players with jumper's knee jumps as high and perhaps even higher than the symptom-free athletes. The reason for this difference is unknown, and we know little about when this difference occurs.

Objective Assess the Jumping ability and change of this ability as a risk factor for developing jumper's knee.

Design Prospective cohort study.

Participants Elite volleyball players at junior high school (Top Volley Norway) at Sand, Norway.

Method Standardised jumping on the power platform (Boscotest). Standing jump (SJ) and countermovement jump (CMJ) were tested at baseline and then twice a year. The jumper's knee diagnosis was based on a standardised clinical examination.

Results A total of 113 symptom-free students were included, with a mean observation time of 1.8 years. 19 of these developed jumper's knee (16 boys and three girls). The three symptomatic women these were excluded in the analysis. The groups were equal at baseline; asymptomatic athletes had SJ of mean 27.5 cm and CMJ at 35.0 cm, while the jumper's knee group had SJ at 28.0 cm (p=0.67 vs asymptomatic) and CMJ (37.2 cm (p=0.10). Jumping ability changed in both groups. For the asymptomatic it was SJ: 5.5 cm (95% CI 1.9 to 9.1) for and CMJ: 6.0 cm (4.1–7.9) while the corresponding results in jumper's knee group was SJ: 3.7 cm (−1.8 to 9.3) (ANOVA, p=0.003) and CMJ: 4.5 cm (0.3 to 8.8) (ANOVA, p=0.001).

Conclusion There is no difference at baseline between the men who developed jumper's knee with those who remained asymptomatic. Those who remained asymptomatic, however, increased their jumping ability more than those who developed a finding that conflict with data from previous cross-sectional studies.