748. Advancing Antimicrobial Resistance Surveillance: An Empirical Approach to Deriving a Composite Drug Resistance Index for Urinary Tract Infections
Session: Poster Abstract Session: Antimicrobials: Resistance Mechanisms
Friday, October 4, 2013
Room: The Moscone Center: Poster Hall C
  • BCCDC_ScientificPoster FINAL 2013_09_23.pdf (515.0 kB)
  • Background: Drug resistance indices (DRIs) quantify the cumulative impact of antimicrobial resistance on treatment outcomes for a bacterial infection. Indices used to date are organism specific and have limited application to empirical therapy for infectious disease syndromes. The purpose of this analysis was to derive a composite DRI for community-acquired urinary tract infections (UTIs) and adopt these methods to account for an empirical treatment regime.

    Methods: Antimicrobial utilization data were obtained from the BC PharmaNet database of outpatient prescriptions. Antimicrobial resistance data were obtained from BC Biomedical Laboratories. For the microbiological approach, DRIs were calculated according to the formula:


    where ρtik is the proportion of urinary isolates for organism i non-susceptible to drug class k at time t, qtik is the proportional utilization rate for drug class k used to treat UTIs caused by organism i at time t, and wi is a weighting factor based on the proportion of urinary isolates tested for each organism i. For the empirical approach, proportional utilization rates were held constant across uropathogens and resistance rates for uropathogens inherently resistant to a given drug class were assumed to be 100%. Trends over time were assessed using the non-parametric Spearman Rank test.

    Results: The DRI calculated using the microbiological approach decreased from 19.1% (95% CI=18.0-20.0%) in 2007 to 17.7% (95% CI=16.9-18.6%) in 2010 (rho=-1.00; p<0.001), while the DRI calculated using the empirical approach decreased from 21.8% (95% CI=20.9-22.7%) in 2007 to 20.8% (95% CI=20.1-21.6%) in 2010 (rho=-0.80; p=0.200). Only the trend calculated using the microbiological approach reached statistical significance. On average, DRIs calculated using the empirical approach were 3% higher than DRIs calculated using the microbiological approach.

    Conclusion: Our results suggest that physicians are adopting their prescribing practices to local resistance patterns, as shown by the favourable DRI trends. An empirical approach to deriving DRIs may more accurately reflect the impact of resistance on the likelihood that an organism will be susceptible to empirically prescribed therapy.

    David M. Patrick, MD, MHSc, FRCPC1,2, Catharine Chambers, MSc1, Mei Chong, MSc1, Dale Purych, MD3,4 and Fawziah Marra, PharmD1,5, (1)British Columbia Centre for Disease Control, Vancouver, BC, Canada, (2)University of British Columbia, Vancouver, BC, Canada, (3)BC Biomedical Laboratories Ltd., Surrey, BC, Canada, (4)Royal Columbian Hospital, Fraser Health Authority, New Westminster, BC, Canada, (5)Pharmaceutical Sciences, University of British Columbia, Vancouver, BC, Canada


    D. M. Patrick, None

    C. Chambers, None

    M. Chong, None

    D. Purych, None

    F. Marra, None

    Findings in the abstracts are embargoed until 12:01 a.m. PST, Oct. 2nd with the exception of research findings presented at the IDWeek press conferences.