Objective: To evaluate changes in antimicrobial use and expenditures and the rates of selected nosocomial infections due to resistant organisms associated with implementation of an antimicrobial-prescribing improvement program.
Design: Before-after trial comparing 1992 (pre-program), 1993 (a transition year), and 1994 (after full implementation of the program).
Setting and participants: Academic medical center, all patients and physicians.
Intervention: An antimicrobial-prescribing improvement program with prior approval requirement for use of restricted agents.
Main outcome measures: Antimicrobial use and expenditures, rates of selected nosocomial infection marker events.
Results: Between 1992 and 1994, there were substantial decreases in antimicrobial use, from 158,107 to 137,364 defined daily doses, and in expenditures from $2,486,902 ($24.01 per patient day) to $1,701,522 ($18.49 per patient day). After adjusting for changes in purchase prices and census days, we estimated savings attributable to the program of $279,573 in 1993 and $389,814 in 1994. In addition, we found significant decreases between 1992 and 1994 in the rates of enterococcal bacteremia (.34 vs .16 events per 1,000 patient days; P = .016), selected gram-negative bacteremia (.26 vs .11; P = .015), methicillin-resistant Staphylococcus aureus colonization or infection (.66 vs .20; P < .0001), and Stenotrophomonas colonization or infection (.35 vs .17; P = .019). No significant change occurred in rates of nosocomial candidemia or Clostridium difficile toxin-positive diarrhea. Values for 1993 were intermediate between those of 1992 and 1994.
Conclusion: Implementation of an antimicrobial-prescribing improvement program was associated with substantial savings in antimicrobial use and expenditures and significant decreases in rates of selected nosocomial infections due to resistant organisms.