Table 1.
Baseline characteristics. Values are numbers (percentages) unless stated otherwise
| Arm 1: education plus support for searching (n=88 practices; n=55 896 patients) | Arm 2: as arm 1 plus feedback (n=87 practices; n=56 194 patients) | Arm 3: as arm 2 plus theory informed behavioural change component (n=87 practices; n=58 569 patients) | |
|---|---|---|---|
| Practices | |||
| Mean (SD) list size | 6360 (3005) | 6538 (3290) | 6893 (3388) |
| Training* | 29 (33) | 31 (36) | 31 (36) |
| Dispensing* | 3 (3) | 6 (7) | 0 (0) |
| Mean (SD) QOF achievement* | 97.5 (4.2) | 98.5 (3.2) | 98.5 (2.7) |
| Median (IQR) No of patients in primary outcome indicator | 584 (429) | 623 (527) | 600 (413) |
| Patients | |||
| Median (IQR) age, years | 77 (10) | 77 (11) | 77 (10) |
| Male sex | 24 625 (44.1) | 24 847 (44.2) | 25 804 (44.1) |
| Scottish Index of Multiple Deprivation fifth: | |||
| 1 (deprived) | 9966 (17.8) | 10 597 (18.9) | 12 102 (20.7) |
| 2 | 14 056 (25.1) | 15 232 (27.1) | 14 204 (24.3) |
| 3 | 9959 (17.8) | 10 746 (19.1) | 9064 (15.5) |
| 4 | 7907 (14.1) | 7749 (13.8) | 8716 (14.9) |
| 5 (affluent) | 11 762 (21.0) | 9620 (17.1) | 11 907 (20.3) |
| Unknown | 2246 (4.0) | 2250 (4.0) | 2576 (4.4) |
| Urban/rural: | |||
| Urban | 47 365 (84.7) | 47 440 (84.4) | 50 107 (85.6) |
| Rural or unknown | 8531 (15.3) | 8754 (15.6) | 8462 (14.4) |
| Health board A | 12 850 (23.0) | 13 505 (24.1) | 13 941 (23.8) |
| Health board B | 16 093 (28.8) | 17 716 (31.5) | 19 239 (32.8) |
| Health board C | 26 953 (48.2) | 24 973 (44.4) | 25 389 (43.3) |
| Mean (SD) No of drugs/patient | 6.1 (3.6) | 6.2 (3.7) | 6.1 (3.7) |
| Baseline performance | |||
| Primary outcome | 3332/55 896 (6.0) | 3341/56 194 (5.9) | 3634/58 569 (6.2) |
| Secondary outcomes†: | |||
| S1: aged ≥75 plus antipsychotic | 691/34 427 (2.0) | 697/33 575 (2.1) | 741/36 113 (2.1) |
| S2: “triple whammy” | 1187/15 632 (7.6) | 1166/15 341 (7.6) | 1326/16 291 (8.1) |
| S3: aged ≥75 plus NSAID | 1016/34 427 (3.0) | 1001/33 570 (3.0) | 1176/36 113 (3.3) |
| S4: antiplatelet plus NSAID | 706/27 519 (3.6) | 768/28 479 (2.7) | 771/28 297 (2.7) |
| S5: oral anticoagulant plus NSAID | 55/5715 (1.0) | 49/5754 (0.9) | 42/5903 (0.7) |
| S6: oral anticoagulant plus antiplatelet | 278/5715 (4.9) | 277/5754 (4.8) | 258/5903 (4.4) |
IQR=interquartile range; NSAID=non-steroidal anti-inflammatory drug.
*Training=accredited for postgraduate general practitioner training; dispensing=dispenses as well as prescribes drugs in rural areas with no pharmacy; QOF achievement=percentage of Quality and Outcomes Framework pay for performance points achieved.
†Outcomes are defined in box 1. Benchmark rates used in feedback (those achieved by 25% of Scottish practices with most optimal prescribing in year before feedback started were S1 1.3% (median practice rate across all arms 2.2%), S2 4.8% (median 7.2%), S3 1.3% (median 2.9%), S4 1.2% (median 2.4%), S5 0% (median 0.5%), and S6 2.2% (median 4.1%).