Controlling Use of Antimicrobials in a Community Teaching Hospital
Objective. To test the hypothesis that antibiotic use could be controlled or improved in a community teaching hospital, with improvement defined as reductions in overall use, overall cost, and antimicrobial resistance.
Design. Interventional study with historical comparison.
Setting. A not‐for‐profit, 900‐bed community general hospital with residents in medicine, surgery, obstetrics‐gynecology, and psychiatry.
Participants. Physicians who requested any of the targeted antibiotics.
Interventions. Three categories of inpatient antibiotic orders were monitored beginning in April 2001: conversion from intravenous to oral administration for selected highly bioavailable antimicrobials, cessation of perioperative prophylaxis within 24 hours for patients undergoing clean and clean‐contaminated surgery, and consultation with an infectious diseases physician before continuing administration of selected drugs beyond 48 hours. Data were analyzed after the first 33 months. Patient outcomes were reviewed during the hospital stay and at readmission if it occurred within 30 days after discharge.
Results. From April 2001 through December 2003, a total of 1426 requests for antimicrobial therapy met criteria for intervention. Overall physician compliance with the program was 76%, ranging from 57% for perioperative prophylaxis to 92% for intravenous to oral conversion. Antimicrobial costs per patient‐day decreased by 31%, from $13.67 in 2000 (before program implementation) to $9.41 in 2003. Total savings in acquisition costs were $1,841,203 for the 3‐year period. Resistance to numerous drugs among Klebsiella pneumoniae isolates was also significantly reduced.
Conclusions. A program to improve the use of antibiotics in a community hospital was successful in reducing overall use, overall cost, and antimicrobial resistance.
Received June 9, 2004; accepted August 19, 2005; electronically published February 28, 2006.
Antibiotic‐resistant organisms are a major public health concern and are of particular concern for hospitals.1,2 Although the problem of antimicrobial resistance is complex, experts agree that exposure to antimicrobial drugs exerts selective pressure favoring the emergence of resistance.2 Therefore, use of antibiotics, whether appropriate or not, often leads to microbial resistance. National surveys have suggested that 22%‐65% of antibiotic use in hospitals is inappropriate.3‐6 The effects of inappropriate use of antibiotics on microbial resistance to antibiotic agents, adverse drug events, and healthcare costs have been well described.6‐9 Antimicrobial resistance results in increased morbidity, mortality, and healthcare costs, because multidrug‐resistant bacteria make treatment much more difficult. A pharmacoeconomic study at Duke University Medical Center showed that treating a primary nosocomial bloodstream infection due to methicillin‐resistant Staphylococcus aureus (MRSA) is approximately 3 times as costly as treating a similar infection due to methicillin‐susceptible S. aureus.10 A joint statement by the Society for Healthcare Epidemiology of America and the Infectious Diseases Society of America contends that “appropriate antimicrobial stewardship that includes optimal selection, dose, duration of treatment, as well as control of antibiotic use, will prevent or slow the emergence of resistance among organisms.”11(p584) Several published reports describe implementation of programs to control the use of antibiotics in response to an outbreak of infection with drug‐resistant organisms.12‐14 A proactive approach to antimicrobial resistance was taken at our hospital, even though no major outbreak of infection with drug‐resistant organisms had occurred.
Methods
In April 2001, a multidisciplinary antimicrobial management team was formed at Presbyterian Hospital of Dallas (Dallas, TX) to create a comprehensive antimicrobial management program (CAMP). The team was composed of the Chief of the Department of Infectious Diseases, 2 clinical pharmacists, and a microbiologist. The goals of the team were to improve the overall use of antibiotics, to prevent and/or slow the emergence of resistant organisms, to improve patient outcomes, and to decrease healthcare costs. The design was an interventional study using a historical comparison. CAMP was approved by the pharmacy and therapeutic committee and the medical board of the hospital. The medical staff was notified about the program at section or department meetings, at teaching conferences, and through the pharmacy newsletter.
CAMP Components
The 3 main components of CAMP were as follows: (1) conversion from intravenous to oral administration for highly bioavailable antimicrobials, (2) discontinuation of perioperative antimicrobial prophylaxis at 24 hours for clean and clean‐contaminated surgical procedures, and (3) restricted use of antibiotics that have a high risk for adverse events, have a high cost, have a high potential to promote resistance, or are drugs of last resort. To evaluate compliance with the components described above, a report using the pharmacy computer system was generated daily to identify patients receiving the targeted antimicrobials. Each component is described below.
Intravenous to oral conversion. Certain antimicrobials are so well absorbed that they reach nearly equivalent blood levels whether they are given intravenously or orally. The following antimicrobials were targeted for conversion from intravenous to oral administration: clindamycin, fluconazole, levofloxacin, metronidazole, moxifloxacin, and trimethoprim‐sulfamethoxazole. The criteria for switching from intravenous to oral formulations were ability to take oral medications or food; absence of persistent nausea, vomiting, or diarrhea; and no disorder of the gastrointestinal tract that could decrease drug absorption. When these 3 criteria were met, a clinical pharmacist wrote an order in the patient’s chart to change the antibiotic from the intravenous to the oral formulation. A note was placed on the patient’s chart notifying the physician of the change. Pharmacy records were evaluated to determine whether the order to convert to the PO formulation was countermanded by the physician.
Preoperative prophylaxis. Antimicrobial prophylaxis for certain types of surgery has been shown to decrease postoperative morbidity and infection rates.15‐17 However, inappropriate and extended prophylaxis can lead to selection of resistant organisms.15 The timing of prophylaxis is very important, because adequate tissue concentrations of the antibiotic should be present at the time of incision and should be maintained until the incision is closed.16 This can usually be achieved with a single preoperative dose.17 Our program allowed up to 24 hours of prophylactic antibiotic therapy. If the patient underwent a clean or clean‐contaminated procedure, a note was included in the chart to ask to the physician to discontinue antimicrobial prophylaxis after 24 hours. Compliance with this request was monitored by review of pharmacy records.
Restricted use of antimicrobial agents. The third component of CAMP was the restricted use of certain antimicrobial agents that have a high risk for adverse events, have a high cost, have a high potential to promote resistance, or are drugs of last resort. Quinupristin‐dalfopristin could be prescribed only by an infectious diseases physician. Seven other antimicrobial agents (vancomycin, cefepime, imipenem, ertapenem, linezolid, ceftazidime, and meropenem) could be prescribed only by an infectious diseases physician after 48 hours. A pharmacist left a note on the patient’s chart reminding the treating physician that, after 48 hours, a consultation with an infectious diseases specialist was required to continue treatment with the antimicrobial. The 48‐hour period allowed the physician to review results of culture and susceptibility tests before deciding whether to discontinue antibiotic treatment or consult an infectious diseases physician. It was assumed that infectious diseases physicians would be more judicious regarding the prescription of these drugs, but no formal strategy was used to ensure this. In most cases, the prescribing physician stopped treatment with the drug and substituted another that was not restricted by CAMP criteria; an infectious diseases physician was consulted in only a minority of cases.
Data Collection
Data on the annual antimicrobial doses charged to patient accounts, annual antimicrobial acquisition costs, and patient‐days were obtained from the hospital's Finance Department. Antimicrobial doses in grams were converted to daily defined doses by means of published conversion factors.18 Analysis of variance (ANOVA) was performed to compare the mean daily defined doses per 1,000 patient‐days between 2000 (the year before CAMP implementation), 2001 (the first year of CAMP), 2002 (the second year of CAMP), and 2003 (the third year of CAMP). A P value less than .05 was considered statistically significant. Drug costs in 2001 dollars were used uniformly for all cost calculations, so that inflation or deflation would not affect the analysis.
The Microbiology Department collected data on the susceptibility of major pathogens and published an annual antibiogram, a copy of which was placed in each hospital chart and sent to all physicians. Data from the year before implementation of the program (ie, 2000) were compared with data collected during the second year (2002) and third year (2003) of the program. Student's t test was used to determine statistically significant differences between the years.
Results
Table 1 lists the number of interventions by the Antimicrobial Management Team from April 3, 2001, through December 31, 2003, and the rates of acceptance by treating physicians for each intervention type. During this period, 1,426 requests for antimicrobial therapy resulted in an intervention. Recommendations to switch from intravenous to oral administration were accepted (ie, not countermanded) 92% of the time. Recommendations to discontinue surgical prophylaxis at 24 hours were accepted 57% of the time. Recommendations to either discontinue use of restricted antibiotics or to obtain a consultation with an infectious diseases specialist were accepted 87% of the time. In the majority of cases (54%), use of the drug was discontinued or a nonrestricted antibiotic was substituted by the treating physician; in a minority of cases (46%), a consultation with an infectious diseases specialist was requested. The overall acceptance rate for all interventions was 76%.
CAMP Compliance
Intravenous to oral conversion. Compliance with conversion from intravenous to oral administration has been nearly 100% since July 2001, when pharmacists started automatically converting the route of antimicrobial administration from intravenous to oral. Before this, a note had been left on patient charts asking the physician to make the change. The number of interventions for intravenous to oral conversion has dropped dramatically, indicating that physicians are converting the administration route without intervention. To date, no evidence has suggested that physicians have countermanded requests by the pharmacists to switch from intravenous to oral therapy.
Preoperative prophylaxis. The rate of compliance with surgical prophylaxis interventions has fluctuated. There has been a downward trend in the number of interventions that are necessary, with most physicians now requesting a course of prophylactic antibiotic therapy for 24 hours or less. When noncompliance with the surgical prophylaxis recommendations was analyzed by specialty, it was determined that most instances of noncompliance were associated with 3 surgeons. Operative records and microbiological findings for their patients were reviewed, and no evidence of infection was documented. This information was transmitted to the departmental section chief and to the Quality Resource Management Committee. Monitoring of compliance will continue, and further information about the appropriate use of perioperative prophylaxis will be provided to these surgeons.
Although the rate of compliance with surgical prophylaxis recommendations has been lower than we had hoped, we have had a significant impact on the number of antibiotic doses used for surgical prophylaxis. Use of cefazolin, the primary antibiotic used for surgical prophylaxis at our hospital, decreased significantly between 2000 and 2003 (Figure 1).
Figure 1. Cefazolin mean daily defined dose (DDD) per 1,000 patient‐days. *
versus 2000 by Student's t test. Bars = 95% confidence intervals.
Restricted use of antimicrobials. Compliance with interventions to restrict to use of antimicrobials has remained high, and the number of interventions has remained fairly constant. Figures 2, 3, and 4 show a statistically significant decrease in the daily defined doses per 1,000 patient‐days for cefepime, ceftazidime, and imipenem. Because of supply problems with meropenem, there was insufficient use to draw any conclusions. Use of vancomycin and linezolid increased despite our efforts, most likely because of the increasing prevalence of infection with community‐acquired MRSA (Figures 5 and 6).
Figure 2. Cefepime mean daily defined dose (DDD) per 1,000 patient‐days. *
versus 2000 by Student's t test. #
versus 2001 by Student's t test. Bars = 95% confidence intervals.
Figure 3. Ceftazidime mean daily defined dose (DDD) per 1,000 patient‐days. *
versus 2000 by Student's t test. Bars = 95% confidence intervals.
Figure 4. Imipenem mean daily defined dose (DDD) per 1,000 patient‐days. *
versus 2000 by Student's t test. #
versus 2001 by Student's t test. Bars = 95% confidence intervals.
Figure 5. Vancomycin mean daily defined dose (DDD) per 1,000 patient‐days. All P values >.05. Bars = 95% confidence intervals.
Figure 6. Linezolid mean daily defined dose (DDD) per 1,000 patient‐days. *
versus 2000 by Student's t test. #
versus 2001 by Student's t test. Bars = 95% confidence intervals.
CAMP Effects
Antibiotic costs. Antimicrobial drug costs per patient‐day decreased from $13.67 in 2000 (before CAMP implementation) to $9.41 in 2003 (Figure 7). Total savings in acquisition costs were $1,841,203 for the 3‐year period. This cost savings is based on antimicrobial acquisition cost alone and does not include mixing costs, administration costs, tubing costs, or nursing and pharmacy time.
Figure 7. Antimicrobial cost per patient‐day
Susceptibility. Comparisons between 2000 (the year before program implementation) and 2002 (the first full year of the program) and 2003 (the second full year of the program) revealed no statistically significant change in the rate of susceptibility to ceftazidime and cefepime among Enterobacter cloacae (134 isolates analyzed), Proteus mirabilis (133), nonurinary Pseudomonas aeruginosa (338), and Serratia marcescens (109). Susceptibility to cefepime among E. coli recovered from non–urinary tract sites (548 isolates analyzed) decreased significantly (from 100% in 2000 to 98% in 2003), but the change in the rate of susceptibility to ceftazidime was not statistically significant. Susceptibility to ceftazidime among E. coli recovered from the urinary tract (1,615 isolates analyzed) decreased significantly (from 100% in 2000 to 98% in 2002). Despite decreased use of imipenem, susceptibility among Pseudomonas aeruginosa to this agent decreased significantly between 2000 (93% of isolates) and 2002 (83% of isolates) but improved in 2003 (91% of isolates). Among 218 isolates of Klebsiella pneumoniae, there was a statistically significant increase in susceptibility between 2000 and 2003 to the following antimicrobials: ceftazidime, ceftriaxone, cefuroxime, levofloxacin, gentamicin, tobramycin, and trimethoprim‐sulfamethoxazole (Table 2). Interestingly, only the use of ceftazidime and cefepime was restricted by our program. The change in the rate of susceptibility to cefepime did not reach statistical significance (
).
Discussion
The consequences of overusing or misusing antibiotics are different from those of overusing or misusing other drugs or procedures. Overuse of another type of drug or procedure may harm that specific patient, but overuse of antibiotics and the associated emergence of drug‐resistant organisms can affect the patient in the next bed or even patients that have yet to be admitted to the hospital.
Our program has been successful for many reasons. The components of CAMP were chosen after thorough review of the medical literature. Before program implementation, an entire year was devoted to educating physicians about the importance of the program and obtaining their support. Support from the hospital administration has also been a crucial element in the success of the program, and it could not have been implemented without dedicated personnel from the Departments of Pharmacy and Microbiology. Having an infectious diseases physician as the program champion has also been crucial to its success. Implementation of the program was very labor intensive, especially during the planning and initial implementation stages. Costs of the program included salaries for 2 pharmacists (1.0 full‐time employee), a microbiologist (0.25 full‐time employee), and a physician (prorated at 30 hours per month). Our antibiotic team met 3‐4 times per week for the first 9 months of the program. The program has become less labor intensive with time; we now meet once weekly to review the results of surveillance of antibiotic use. The antibiotic team is working closely with the Departments of Infection Control and Microbiology to monitor microbial resistance patterns and to quickly detect any trends. Overall use of antibiotics is also being monitored to detect any new trends indicating inappropriate use.
Most programs implemented in US hospitals have been created in response to an increase in multidrug‐resistant organisms and have made reduction of resistance one of their major goals. Our program differs from others in that we have not yet experienced an increase in multidrug resistance, with the exception of the dramatic increase in the prevalence of MRSA. We were unable to reduce the use of vancomycin or linezolid because of the increase in the prevalence of community‐acquired MRSA. However, the requirement of a consultation with an infectious diseases physician to continue administration of high‐risk antibiotics after 48 hours of use frequently allowed us to change therapy to drugs with a low risk of adverse events and resistance, such as trimethoprim‐sulfamethoxazole or minocycline. The improvement in susceptibility patterns among K. pneumoniae isolates was unexpected and gratifying. However, we are not certain that the improvement can be entirely explained by our program. For example, increased susceptibility to cefuroxime, ceftriaxone, levofloxacin, and tobramycin cannot be directly attributed to our program, because use of these agents was not restricted. However, it is possible that selection pressures were reduced by the decreased use of imipenem, ceftazidime, and cefepime. We are unable to explain the discordance between reduced imipenem use and increased imipenem resistance among P. aeruginosa.
There are several weaknesses to this study. We failed to include any methodology at the outset to determine improvements in length of hospital stay. Thus, our financial savings may be severely underestimated. Future studies should include length of stay in the analysis. We also did not include a method to formally determine whether any adverse outcomes occurred because of our program. However, given the widespread publicity of our program’s activities, we are confident that any adverse outcomes would have been immediately brought to our attention. To date, no adverse patient outcomes have been attributable to the team’s recommendations.
The major benefit of implementing CAMP is the reduction in the inappropriate use of antibiotics without subsequent adverse outcomes in patients. This reduction can slow or prevent the emergence of drug‐resistant organisms, which, in turn, leads to reduced morbidity, mortality, and healthcare costs.
Acknowledgments
We thank the following individuals for their assistance in preparing and reviewing this manuscript: Richard Gilder, RN, BSN, CNOR, BCNI; Sherry Mercer, CPA; Judith Marshall, RPh; and Mark Feldman, MD.
References
- 1. McGowan JE, Tenover FC. Control of antimicrobial resistance in the health care system. Infect Dis Clin North Am 1997; 11:297‐311.
- 2. Schwartz B, Bell DM, Hughes JM. Preventing the emergence of antimicrobial resistance: a call for action by clinicians, public health officials, and patients. JAMA 1997; 278:944‐945.
- 3. Fraser GL, Stogsdill P, Dickens JD, et al. Antibiotic optimization: an evaluation of patient safety and economic outcomes. Arch Intern Med 1997; 157:1689‐1694.
- 4. Pelletier LL. Hospital usage of parenteral antimicrobial agents: a gradated utilization review and cost containment program. Infect Control 1985; 6:226‐230.
- 5. Marr JJ, Moffet HL, Kunin CM. Guidelines for improving the use of antimicrobial agents in hospitals: a statement by the Infectious Diseases Society of America. J Infect Dis 1988; 157:869‐876.
- 6. Dunagan WC, Woodward RS, Medoff G, et al. Antimicrobial misuse in patients with positive blood cultures. Am J Med 1989; 87:253‐259.
- 7. McGowan JE. Do intensive antibiotic control programs prevent the spread of antibiotic resistance? Infect Control Hosp Epidemiol 1994; 15:478‐483.
- 8. Arbo MD, Snydman DR. Influence of blood culture results on antibiotic choice in the treatment of bacteremia. Arch Intern Med 1994; 154:2641‐2645.
- 9. Goldmann DA, Weinstein RA, Wenzel RP, et al. Strategies to prevent and control the emergence and spread of antimicrobial‐resistant microorganisms in hospitals: a challenge to hospital leadership. JAMA 1996; 275:234‐240.
- 10. Abramson MA, Sexton DJ. Nosocomial methicillin‐resistant and methicillin‐sensitive Staphylococcus aureus primary bacteremia: at what costs? Infect Control Hosp Epidemiol 1999; 20:408‐411.
- 11. Shlaes DM, Gerding DN, John JF, et al. Society for Healthcare Epidemiology of America and Infectious Diseases Society of America Joint Committee on the Prevention of Antimicrobial Resistance: guidelines for the prevention of antimicrobial resistance in hospitals. Clin Infect Dis 1997; 25:584‐599.
- 12. Patterson JE, Hardin TC, Kelly CA, et al. Association of antibiotic utilization measures and control of multiple‐drug resistance in Klebsiella pneumoniae. Infect Control Hosp Epidemiol 2000; 21:455‐458.
- 13. White AC, Atmar RL, Wilson J, et al. Effects of requiring prior authorization for selected antimicrobials: expenditures, susceptibilities, and clinical outcomes. Clin Infect Dis 1997; 25:230‐239.
- 14. Smith DW. Decreased antimicrobial resistance after changes in antibiotic use. Pharmacotherapy 1999; 19(8 pt 2):129S‐132S.
- 15. Antimicrobial prophylaxis in surgery. Med Lett Drugs Ther 2001; 43:92‐97.
- 16. American Society of Health‐System Pharmacists. ASHP Therapeutic Guidelines on Antimicrobial Prophylaxis in Surgery. Am J Health Syst Pharm 1999; 56:1839‐1888.
- 17. Mangram AJ, Horan TC, Pearson ML, et al. Guideline for prevention of surgical site infection, 1999. Hospital Infection Control Practices Advisory Committee. Infect Control Hosp Epidemiol 1999; 20:250‐278.
- 18. NNIS System. National Nosocomial Infections Surveillance (NNIS) System Report, data summary from January 1992 through June 2003, issued August 2003. Am J Infect Control 2003; 31:481‐498.
-
Abstract presented at the 40th Annual Meeting of the Infectious Diseases Society of America; October 24‐27, 2002; Chicago, IL.