Univariate and multivariate regression outputs for models examining the relationship between use of the intervention (arm), the covariates (period and set) and the incidence of erroneous medication orders
| Univariate model | Multivariate model | ||||||
| Erroneous medication orders | n (%)* | OR | 95% CI | P value | OR | 95% CI | P value |
| Arm | |||||||
| Control | 34 (28.3%) | 1 | – | – | 1 | – | – |
| Intervention | 8 (6.6%) | 0.18 | 0.08 to 0.41 | <0.01 | 0.16 | 0.06 to 0.43 | <0.01 |
| Period | |||||||
| Period 1 | 12 (10%) | 0.33 | 0.16 to 0.69 | <0.01 | 1.1 | 0.41 to 3.18 | 0.79 |
| Period 2 | 30 (25%) | 1 | 1 | ||||
| Set | |||||||
| Set 1 | 27 (22.5%) | 1 | – | – | 1 | – | – |
| Set 2 | 15 (12.5%) | 0.49 | 0.25 to 0.98 | 0.04 | 0.43 | 0.17 to 1.04 | 0.06 |
Bold indicates reference category.
*n=number of erroneous medication orders and percentage out of the total 120 medication orders observed per arm.