Table 2.

 Results of multiple regression analysis of factors potentially associated with proportion of matching eligibility criteria

Variable	Odds ratio (95% CI)	P value
Intercept	0.63 (0.24 to 0.91)
Sample size*	1.01 (0.80 to 1.27)	0.93
Multicentre v single centre trial†	0.58 (0.24 to 1.40)	0.22
Industry funding v none†	1.32 (0.65 to 2.71)	0.44
Other funding v none†	1.94 (0.73 to 5.14)	0.18

*Interpreted as relative increase in proportion of matching eligibility criteria per each one unit increase in log transformed sample size.

†Interpreted as relative increase or decrease in proportion of matching eligibility criteria compared with reference category.