Effect of intervention using relative risk (RR) on the five indicators of general practitioner referral quality
| Effect of intervention | Pooled baseline proportion | Estimated postintervention proportion | Effect of intervention for practices with baseline proportion equal or below the pooled baseline proportion | Effect of intervention for practices with baseline proportion above the pooled baseline proportion | |
| RR (95% CI) | Proportion (%) (95% CI) | Proportion (%) (95% CI) | RR (95% CI) (n=no of practices) | RR (95% CI) (n=no of practices) | |
| Adherence to pathways | 1.57 (1.35 to 1.81) | 42 (35 to 48) | 66 (57 to 76) | 2.09 (1.60 to 2.73) (n=17) | 1.38 (1.16 to 1.64) (n=16) |
| Letter information on suspected diagnosis | 1.31 (1.19 to 1.43) | 59 (53 to 65) | 77 (70 to 84 | 1.74 (1.47 to 2.05) (n=14) | 1.19 (1.09 to 1.29) (n=19) |
| Letter information on previous management | 1.21 (1.11 to 1.31) | 67 (61 to 73) | 81 (74 to 88 | 1.55 (1.35 to 1.78) (n=14) | 1.09 (1.02 to 1.15) (n=19) |
| Conversion of trauma and orthopaedics referrals to surgery | 1.45 (1.05 to 2) | 43 (34 to 52) | 62 (58 to 65 | 2.39 (1.56 to 3.67) (n=17) | 1.13 (0.87 to 1.46) (n=11) |
| Seen once and discharged | 0.51 (1.33 to 1.78) | 13 (10 to 17) | 6 (3 to 9 | 0.50 (0.30 to 0.80) (n=19) | 0.50 (0.16 to 1.52) (n=14) |