Table 1 Accuracy of extracted duration and selected indications
From: Detecting inappropriate total duration of antimicrobial therapy using semi-automated surveillance
Duration of therapy (inaccurate duration/number of screened records) | Indication of therapy (inaccurate selections/number of screened records) | |
|---|---|---|
Error rate location AMC (%) | 10/200 = 5.0% 6 prescriptions not electronically prescribed (all were OPAT prescriptions) 1 prescription of which start fell before inclusion paeriod and therefore not included in the dataset 1 missing non-J01 antibiotic (oral metronidazole) 2 prescriptions were not terminated after discharge or death 18 transfers to other hospital leading to missing data | 31/200 = 15.5% Incorrectly registered indications affecting duration of therapy: 12/31 = 38.7%a |
Error rate location VUMC (%) | 8/200 = 4.0% 8 prescriptions not electronically prescribed (of which 5 were OPAT prescriptions) 14 transfers to other hospital leading to missing data | 37/200 = 18.5% Incorrectly registered indications affecting duration of therapy: 23/37 = 62.0%c |
Total | 18/400 = 4.5% | 68/400 = 17.0% Incorrectly registered indications affecting duration of therapy: 35/400 = 8.8% |