Improving quinolone use in hospitals by using a bundle of interventions in an interrupted time series analysis

Abstract

The objectives of the present study were to determine the effects of multiple targeted interventions on the level of use of quinolones and the observed rates of resistance to quinolones in Escherichia coli isolates from hospitalized patients. A bundle consisting of four interventions to improve the use of quinolones was implemented. The outcome was measured from the monthly levels of use of intravenous (i.v.) and oral quinolones and the susceptibility patterns for E. coli isolates from hospitalized patients. Statistical analyses were performed using segmented regression analysis and segmented Poisson regression models. Before the bundle was implemented, the annual use of quinolones was 2.7 defined daily doses (DDDs)/100 patient days. After the interventions, in 2007, this was reduced to 1.7 DDDs/100 patient days. The first intervention, a switch from i.v. to oral medication, was associated with a stepwise reduction in i.v. quinolone use of 71 prescribed daily doses (PDDs) per month (95% confidence interval [CI] = 47 to 95 PDDs/month, P < 0.001). Intervention 2, introduction of a new antibiotic guideline and education program, was associated with a stepwise reduction in the overall use of quinolones (reduction, 107 PDDs/month [95% CI = 58 to 156 PDDs/month). Before the interventions the quinolone resistance rate was increasing, on average, by 4.6% (95% CI = 2.6 to 6.1%) per year. This increase leveled off, which was associated with intervention 2 and intervention 4, active monitoring of prescriptions and feedback. Trends in resistance to other antimicrobial agents did not change. This study showed that the hospital-wide use of quinolones can be significantly reduced by an active policy consisting of multiple interventions. There was also a stepwise reduction in the rate of quinolone resistance associated with the bundle of interventions.

FIG. 1.

Monthly use of intravenous ciprofloxacin from 2005 through 2007, in PDDs. Values predicted by the best-segmented regression model are shown by the solid line. Broken lines show the 95% confidence intervals.

FIG. 2.

Monthly use of ciprofloxacin (intravenous and oral) from 2005 through 2007, in PDDs. Values predicted by the best-segmented regression model are shown by the solid line. Broken lines show the 95% confidence intervals.

FIG. 3.

Annual ciprofloxacin use and overall use.

FIG. 4.

Observed rates of resistance to ciprofloxacin by E. coli isolates from hospitalized patients from 2004 through 2007. Values predicted by the best-segmented regression model are shown by the solid line. Broken lines show the 95% confidence intervals.