Measurement of adult antibacterial drug use in 130 US hospitals: comparison of defined daily dose and days of therapy

Abstract

Background: Hospitals are advised to measure antibiotic use and monitor its relationship to resistance. The World Health Organization's recommended metric is the defined daily dose (DDD). An alternative measure is the number of days of therapy (DOT). The purpose of this study was to contrast these measures.

Methods: We measured the use of 50 antibacterial drugs that were administered to adults who were discharged from 130 US hospitals during 1 August 2002-31 July 2003.

Results: Of 1,795,504 patients, 1,074,174 received at least 1 dose of an antibacterial drug (59.8%). The mean (+/- standard deviation) of total antibacterial drug use measured by the number of DDDs per 1000 patient-days and the number of DOTs per 1000 patient-days were not significantly different (792+/-147 and 776+/-120, respectively; P=.137), although the correlation was poor (r=0.603). For some individual drugs, such as levofloxacin and linezolid, there was no significant difference between DDDs per 1000 patient-days and DOTs per 1000 patient-days, because the administered daily dosage was nearly equivalent to the DDD. When the administered dosage exceeded the DDD, such as for ampicillin-sulbactam and cefepime, estimates of use based on DDDs per 1000 patient-days significantly exceeded those based on DOTs per 1000 patient-days (P<.001). When the administered dosage was less than the DDD, such as for piperacillin-tazobactam and ceftriaxone, estimates of use based on DDDs per 1000 patient-days were significantly lower than those based on DOTs per 1000 patient-days (P<.001).

Conclusion: The measurement of aggregate hospital antibiotic use by DDDs per 1000 patient-days and DOTs per 1000 patient-days is discordant for many frequently used antibacterial drugs, because the administered dose is dissimilar from the DDD recommended by the World Health Organization. DDD methods are useful for benchmarking purposes but cannot be used to make inferences about the number of DOTs or relative use for many antibacterial drugs.