Effect of the Low Risk Ankle Rule on ankle radiography, length of stay, patient and physician satisfaction, missed significant fractures and follow-up radiography
| Outcome | % (n/N)* | Between-group differences, % (95% CI)† | ||||||
|---|---|---|---|---|---|---|---|---|
| Phase 1 (wk 1–26) | Phase 2 (wk 27–52) | Phase 3 (wk 53–78) | ||||||
| Intervention | Control | Intervention | Control | Intervention | Control | Phase 1 to 2 | Phase 1 to 3 | |
| Ankle radiography performed | 96.5 (412/427) | 90.2 (415/460) | 73.5 (211/287) | 88.3 (250/283) | 71.3 (243/341) | 87.8 (310/353) | −21.9 (−28.6 to −15.2) | −23.9 (−30.2 to −17.6) |
| Median length of stay, hr (IQR) | 2.0 (1.4–2.9) | 2.9 (2.0–4.3) | 2.0 (2.0–2.8) | 2.5 (1.8–3.8) | 2.1 (1.4–2.9) | 2.5 (2.8–3.6) | 0.4 (−0.2 to 0.9) | 0.4 (−0.1 to 1.0) |
| Patient satisfaction,‡ very satisfied or satisfied | 90.8 (266/293) | 84.4 (217/257) | 86.0 (196/228) | 93.3 (125/134) | 92.1 (232/252) | 94.5 (171/181) | −11.5 (−23.4 to 0.5) | −11.2 (−31.4 to 9.0) |
| Physician satisfaction,§ very satisfied or satisfied | 91.6 (185/202) | NA | 94.5 (120/127) | NA | 98.4 (123/125) | NA | 8.3† (−16.9 to 0.4) | 10.9 (−27.8 to 5.8) |
| Significant fractures, missed | 1.7 (5/301) | 0 (0/344) | 2.5 (5/198) | 1.0 (2/206) | 1.3 (3/236) | 0 (0/283) | 0.008 (−0.004 to 0.02) | −0.002 (−0.02 to 0.01) |
| Radiography performed in follow-up if no ED radiograph | 0 (0/11) | 8.3 (2/24) | 4.5 (3/66) | 7.1 (1/14) | 9.3 (8/86) | 9.5 (2/21) | 0.4 (−0.2 to 1.0) | 0.4 (−0.2 to 1.0) |
Note: CI = confidence interval, ED = emergency department, IQR = interquartile range, NA = not applicable.
↵* Unless stated otherwise.
↵† For the primary outcome (ankle radiographs), differences calculated using general linear model with a compound symmetry error structure to account for the clustering of observations within hospitals; for secondary outcomes interrupted time series with control differences of differences regression model.
↵‡ Patients asked how satisfied they were with care they received in the emergency department (very satisfied, satisfied, neutral, unsatisfied or very unsatisfied).
↵§ Physicians at intervention sites asked how satisfied they were with care they provided in the emergency department ranked as “very satisfied/satisfied/neutral/unsatisfied/very unsatisfied.”