Examining vancomycin use in hospitalized
Medicare patients in the United States
Randolph R Peto, MD, MPH
MassPRO, Waltham, Massachusetts, USA
Is it too late to try to reduce the use of broad-spectrum antibiotics
in American hospitals? A government coordinated project in the eastern portion of the United States is based on
the hypothesis that it is not
too late. The Health Care Financing Administration (HCFA) brought together a group of quality improvement organizations
to examine vancomycin use in hospitalized Medicare patients (the vast majority of whom are 65 years and older).
This is one of many projects in the Health Care Quality Improvement Program (HCQIP), initiated by HCFA in 1993
to apply state-of-the-art tools of quality measurement and quality improvement to services provided to Medicare
patients. Other current HCQIP infectious disease projects include the early diagnosis and treatment of hospitalized
community-acquired pneumonia, diagnosis and treatment of urinary tract infections, and use of pneumococcal and
The project to examine vancomycin use was built around guidelines from the Hospital Infection Control Practices
Advisory Committee (HICPAC)1.
The committee, in conjunction with the US Centers for Disease Control and Prevention, focused particularly on suggestions
to prevent the spread of vancomycin-resistance.
The guidelines summarize situations in which vancomycin use is "prudent," "appropriate" or
"acceptable" and include:
- treatment of serious infections
caused by beta-lactam-resistant gram-positive organisms.
- treatment of infections caused
by gram-positive microorganisms in patients who have serious allergies to beta-lactam antimicrobials.
- when antibiotic-associated colitis
fails to respond to metronidazole therapy or is severe and potentially life-threatening.
- prophylaxis, as recommended by
the American Heart Association, for endocarditis following certain procedures in patients at high risk for endocarditis.
- prophylaxis for major surgical
procedures involving implantation of prosthetic materials or devices at institutions that have a high rate of infections
caused by methicillin-resistant S.epidermidis or methicillin-resistant S. aureus.
Using these guidelines, the HICPAC project is assessing the levels
of "unsupported" vancomycin use among hospitalized Medicare patients. To date, the project has completed
baseline data collection from patients hospitalized in the first half of 1995. The overall results have not yet
been published in the peer-reviewed literature. However, this article presents preliminary information from 550
patients in six participating hospitals in Massachusetts only (one of 14 states and U.S. territories participating
in the overall project). Analysis of the Massachusetts data suggest significant opportunities for improvement.
The range of unsupported use in each hospital varied from about 40% to 60%. Up to 80% of the surgical use did not
meet guideline criteria. It appears that up to one-half of the unsupported use in surgery could be eliminated if
prophylactic coverage was discontinued after a maximum of two doses. Much of the rest of the unsupported use might
be eliminated if vancomycin was restricted to situations in which serious allergy to beta-lactam agents was unmistakable.
For example, the project did not consider gastrointestinal side effects (diarrhea, nausea, etc.) from beta-lactams
to be serious allergies. However, fear of potential medical malpractice liability may account for reluctance of
U.S. physicians to prescribe antibiotics if the patient history suggests any type of side effect, not just a serious
well-documented allergic reaction.
Instances of unsupported use were more numerous with medical therapy (e.g., treatment of suspected serious infections)
than in surgical prophylaxis. Vancomycin was used in about 10% of cases with cultures positive for Gram-positive
organisms and sensitive to beta-lactams. These findings suggest a failure to reevaluate antibiotic therapy when
culture and antimicrobial sensitivity studies are complete.
The guidelines do not support, but also do not refute, the empiric use of vancomycin for suspected serious infections
while awaiting the results of cultures. In the HCFA project analyses however, most empiric use is considered unsupported
if 1) cultures were never obtained or 2) culture results were negative for gram-positive organisms. Responding
to the baseline results of the study, some infectious disease physicians in the participating hospitals appear
unwilling to withhold empiric vancomycin therapy while awaiting culture results. In fact, the study was not designed
to be able to resolve this area of contention fully. The study only examined vancomycin uses in each hospital.
Other antibiotics prescribed for suspected cases of serious gram-positive infections were not examined.
The guideline authors do not suggest ways to enhance compliance with the recommendations, stating, ". . .
Further study is required to determine the most effective methods for influencing the prescribing practices of
physicians. "As part of this HCQIP project, participating quality improvement organizations will remeasure
hospital vancomycin use after individual improvement plans have been carried out. Project leaders plan to share
innovative intervention strategies and lessons learned after the next phase of the project has been completed.
This cooperative effort should help in reducing "unsupported" vancomycin use in the future.
Note: The analyses upon which this publication is based
were performed under Contract number 500-96-P601, entitled, "Utilization and Quality Control Peer Review Organization
for the State of Massachusetts," sponsored by the Health Care Financing Administration, Department of Health
and Human Services. The content of this publication does not necessarily reflect the views or policies of the Department
of Health and Human Services. The author assumes full responsibility for the accuracy and completeness of the ideas