Logistic regression analysis of the importance assigned to empirical evidence (median score of quite/very important in comparison to limited importance/not important/did not use)
| Variable | OR (95% CI) | Significance | |
|---|---|---|---|
| Female gender | 1.779 | (1.007 to 3.144) | 0.047 |
| Role | <0.005 | ||
| Role (commissioning and contracts) | 0.320 | (0.180 to 0.570) | <0.005 |
| Role (finance) | 0.194 | (0.048 to 0.779) | 0.21 |
| Role (other) | 0.349 | (0.171 to 0.713) | 0.04 |
| Constant | 1.317 | 0.337 | |
Reference category for role is Public Health. Variable(s) entered on step 1: gender, age, years' experience of NHS commissioning, work role, pay and whether the respondent has any medical qualifications. Likelihood ratio test χ2 (2)=25.3, p<0.0005, Cox and Snell R2=0.09, Nagelkerke R2=0.12, Hosmer and Lemeshow χ2 (4)=1.1, p=0.9, c-statistic=0.65.