. 2020 Dec 24;6(1):e000892. doi: 10.1136/bmjsem-2020-000892

Table 4.

Binary logistic regression analysing the association of previous traumatic knee injuries and total training load with current knee overuse injuries

	Dependent variable	Predictor	B	e^B	SE_B	P value*
Model 1: χ²=8.215, p=0.004, R²_Nagelkerke=0.361, Cohen f=0.75, n=26	Current knee overuse injury (yes; no)	Previous severe traumatic knee injury (yes; no)	2.516*	12.375	0.976	0.010**
Model 2: χ²=6.033, p=0.014, R²_Nagelkerke=0.276 Cohen f=0.62, n=26	Current knee overuse injury (yes; no)	Total training hours	0.023 (0.006 to 0.093)†	1.023	0.376	0.034*

*Level of significance: p<0.05, **p<0.01.

†Data are expressed as regression coefficients B and the lower and upper bias corrected accelerated bootstrapping-based CIs with 10 000 samples in brackets.