Background: Reports on antibiotic use often lack complete definitions of the units of measurement, hampering the comparison of data between hospitals or hospital units.
Patients and methods: To compare methods of measures of in-hospital antimicrobial use, we determined aggregate in-hospital consumption data at a tertiary care university hospital using variations of nominators and denominators. Means of defined daily doses (DDD) of individual antimicrobials per 100 bed-days and per 100 admissions at each hospital and intensive care unit (ICU) were calculated. Furthermore, a literature review was performed for benchmarking purposes.
Results: Antibiotic use in different hospital units ranged from 0.105 to 323.37 DDD/100 bed-days and from 4.23 to 6737.92 DDD/100 admissions, respectively. Including the day of discharge in the denominator 'bed-days' underestimated antibiotic use in various hospital wards by up to 27.7 DDD/100 bed-days (26.0%). Equating 'numbers of patients admitted to the hospital' and 'numbers of admissions' on a hospital level resulted in a difference of 192.6 DDD/100 admissions (64%) because patients transferred between hospital units accounted for multiple admissions. Likewise, reporting antimicrobial (Anatomical Therapeutic Chemical [ATC] group 'J') instead of antibiotic (ATC group 'J01') use led to a difference of 16.5 DDD/100 bed-days (19.3%). The literature review revealed underreporting of complete definitions of antibiotic use measurements.
Conclusions: Data on in-hospital antimicrobial use vary widely not only due to different antibiotic policies at different institutions but also due to different methods of measures. Adherence to the standard of reporting the methods of measurement is warranted for benchmarking and promotion of rational antimicrobial use.