Synopsis: An antibiotic control program was associated with significant reduction in costs in some nosocomial infections.

Source: Frank MO, et al. Decrease in expenditures and selected nosocomial infections following implementation of an antimicrobial-prescribing improvement program. Clinical Performance and Quality Health Care 1997;5(4):180-188.
The infectious diseases and clinical pharmacy departments at Indiana University Medical Center in Indianapolis reported their experience with an "antimicrobial prescribing improvement program." They analyzed financial as well as microbiological data from 1992 to 1994. In 1993, they instituted a program in their 365-bed hospital that consisted of restrictions on some antibiotics-such as amikacin, clarithromycin, azithromycin, cefaclor, cefotaxime, imipenem, oral vancomycin, fluconazole, and the quinolones. Patients could not receive these drugs without the approval of a clinical pharmacist (who handled 81% of the requests) or infectious disease department faculty (not resident or fellow) approval. Some additional drugs such as tobramycin, piperacillin, ceftazidime, and intravenous vancomycin were put in a "controlled" status with automatic review at 72 hours. Other antibiotics such as gentamicin, erythromycin, doxycycline, ceftriaxone (1 g/d), cefazolin, clindamycin, metronidazole, and amphotericin were unrestricted.

The program, which was modeled after that at Barnes Hospital in St. Louis,1 brought an estimated $279,000 savings in 1993 and $390,000 in 1994-even though the number of hospital beds was about the same during the two periods, and the intensivity of care was at least as great if not greater. The program brought the cost of antimicrobial use from $24.01 down to $18.49/patient/day. These costs do not include the benefits of reduced nosocomial infections. With an average cost of a nosocomial bacteremia at about $3,500, the program would have saved an additional $115,500 in 1994.

In addition, indicators of nosocomial infection were studied-with a finding that there was a significant reduction in the rates of enterococcal bacteremia (P = 0.016), selected gram-negative bacteremia (P =0.015), methicillin-resistant Staphylococcus aureus colonization and infection (P < 0.0001), and Stenotrophomonas colonization and infection (P = 0.019). They did not find a significant reduction in nosocomial candidemia or Clostridium difficile toxin diarrhea. The antimicrobial susceptibility patterns remained stable over the review period.


This article provides additional evidence of the value of a hospital program to control antibiotic use. Multiple prior studies have shown a cost savings. Some have shown microbiologic effect with reduced Gram-negative resistance to antibiotics2as well as impact onC. difficile colitis.3White and Atmar recently found they could increase susceptibilities through an antibiotic control program at a hispital in Houston. They have found no change in survival among patients with Gram-negative bacteremia.4

There has, however, been little evidence that the actual incidence of other specific nosocomial infections can be affected by antibiotic restrictions. There is also a growing need to determine the overall clinical impact and outcome on the quality of care that may be affected by inappropriate antibiotic use.

This program is consistent with the recent guidelines developed by the Society for Healthcare Epidemiology of America (SHEA) and the Infectious Diseases Society of America (IDSA) for the prevention of antimicrobial resistance in hospitals.5 It also points out the benefits of a team effort with a microbiologist, a clinical pharmacist, and an infectious disease specialist working together.

The savings seemed to be well worth the cost-which was for the salary of a half-time clinical pharmacist and half time infectious disease faculty position. Their exact salaries were not specified.

These types of programs will, hopefully, become the standard and we will be able to gather more information about the actual clinical impact of changes in antimicrobial prescribing. They certainly seem to be cost effective-although many hospitals are still not willing to accept the value of infectious diseases input.


1. Woodward RS, et al.Am J Med 1987;3:817-823.

2. Bamberger DM, Dahl LS. Arch Intern Med 1992;52: 554-557.

3. Pear SM, et al. Ann Intern Med 1994;20:272-277.

4. White AC, et al. Clin Inf Dis 1997;25(2):230-241.

5. Shales DM, et al. Infect Control Hosp Epidemiol 1997;18:275-291.