Site | Descriptor | Number of patients | Total error per patient (median ( range)) | Error rate per order per patient (%) (median (range)) | |||
Pre | Post | Pre | Post | Pre | Post | ||
1 | >700 beds, tertiary teaching hospital, development site for medication chart | 228 | 267 | 2 (0–20) | 2 (0–13) | 14.3 (0–125) | 13.3 (0–175) |
2 | 250 beds, district site, many IMG | 133 | 136 | 2 (0–15) | 2 (0–10) | 20.6 (0–250) | 15.4 (0–75) |
3 | 140 beds, district site, mixture of local trained staff and IMG | 110 | 90 | 2 (0–12) | 2 (0–12) | 18.5 (0–83) | 16.7 (0–81) |
4 | 350 beds, district site, mixture of local trained staff and IMG | 163 | 141 | 2 (0–17) | 1 (0–16) | 20 (0–200) | 20 (0–125) |
5 | Large >800-bed, tertiary teaching hospital | 96 | 117 | 5 (0–19) | 2 (0–17) | 33 (0–157) | 18.2 (0–180) |
All | Representative mix of five sites | 730 | 751 | 2 (0–20) | 2 (0–17) | 20 (0–250) | 15.8 (0–180) |
Kruskal–Wallis test of analysis of variance | p<0.001, NS if site 5 excluded | p = 0.025, NS if site 5 excluded | p<0.001, NS if site 5 excluded | NS difference | |||
IMG, international medical graduates; NS, not statistically significant.