Logistic regression of predictors of cardiac catheterisation (patients were with treated symptoms of ischaemic heart disease and met appropriateness criteria)
| b3-150 | p Value | Exp(b)3-151 | 95% CI | |
| Age | −0.04 | < 0.01 | 0.96 | 0.93 to 0.99 |
| Sex3-152 | 1.65 | < 0.01 | 5.33 | 1.98 to 14.39 |
| De Bono score | 0.03 | < 0.02 | 1.03 | 1.01 to 1.06 |
| Parsonnet 2 score3-153 | −0.02 | < 0.73 | 0.91 | 0.90 to 1.17 |
| Specialty3-154 | N/A | < 0.89 | N/A | N/A |
| Cardiology | 6.44 | < 0.62 | 623.9 | 0.00 to 9×1013 |
| General medicine | 6.42 | < 0.62 | 615.9 | 0.00–9×1013 |
n = 320; other patients with ischaemic heart disease did not meet criteria or were not coded for relevant variables.
↵3-150 Logistic regression equation is 1/1+e (−b1×1 + b2×2 + b3×3...bixi);
↵3-151 a unit change in an independent variable creates a change in the odds of the dependent variable occurring by a factor of this number—a value of 1 indicates no change in the odds;
↵3-152 coded so that being a man is 1 and being a woman is 0;
↵3-153 Parsonnet 2 score without age;
↵3-154 specialty referent is care of the elderly.