Reports from the field-Managed care/Primary care

Report cards don't reflect patient variance

Managed care report cards may unfairly penalize providers for patient variances beyond their control, according to a recent study reported in the American Journal of Medicine.

Many managed care organizations impose financial penalties on providers who use excessive medical services in treating their patients. However, because these report cards often fail to take into account differences in the severity of illness, such report cards may be misleading.

For example, a hospital that uses more blood transfusions for patients undergoing hip fracture surgery may have a higher proportion of severely anemic patients than another hospital that uses fewer transfusions for seemingly similar patients, researchers found.

Researchers conducted a retrospective study of 8,776 charts for hip fracture patients 60 and older who underwent surgical hip repair between 1982 and 1993 at one of 19 study hospitals located in four states. They examined transfusion rates among hospitals, patient characteristics associated with transfusion, and whether those characteristics varied among hospitals. Results include:

• Postoperative transfusion rates varied from 31% to 54% of hip fracture patients among the 19 study hospitals.

• Without adjusting for differences in patient severity of illness, four of nine teaching hospitals and two of nine nonteaching hospitals had significantly higher transfusion rates than the reference hospital, while one teaching hospital had a lower rate than the reference hospital.

• After adjusting for patient anemia and other clinical variables, only one of nine teaching hospitals had rates higher than the reference hospital instead of the original four. However, four of nine nonteaching hospitals rather than the original two had higher rates than the reference hospital. In addition, four teaching hospitals and one nonteaching hospital actually had lower transfusion rates than the reference hospital.

[See: Poses RM, Berlin JA, Noveck H, et al. How you look determines what you find: Severity of illness and variation in blood transfusion for hip fracture. Am J Med 1998; 105:198-206.]

Computer use reduces use of vancomycin

The broad-spectrum antibiotic vancomycin is effective against many types of bacteria, but overuse has resulted in vancomycin-resistant bacteria as a major health risk throughout the nation. There is no established antimicrobial therapy for vancomycin-resistant enterococci. The key to reducing physician use of this antibiotic may be clear computerized guidelines for vancomycin use, concludes a recent study in the Journal of the American Medical Informatics Association.

Researchers at Brigham and Women's Hospi tal in Boston tested a computerized order-entry system at the hospital to determine whether it would reduce vancomycin ordering by staff physicians. The system displayed guidelines for appropriate use of intravenous vancomycin each time a physician keyed in an order for the drug. Researchers randomly assigned physicians to the vancomycin guidelines group on the control group and compared the use of the drug in the two groups.

Intervention results in lower costs

Results include:

• Intervention physicians reduced their overall use of vancomycin by 30% compared to the control group.

• Intervention physicians wrote 32% fewer vancomycin orders and had 28% fewer patients for whom they either initiated or renewed an order for vancomycin.

• Intervention physicians prescribed a 36% shorter duration (26.5 days compared to 41.2 days) of vancomycin therapy than did the control physicians.

• Intervention physicians chose less expensive antibiotics in many instances, such as a first- generation cephalosporin such as cefazolin.

At a daily cost of $12 for vancomycin and $9 for cefazolin, if cefazolin were substituted for vancomycin for all cases in the study in which physicians chose not to order vancomycin, the projected savings for Brigham and Women's Hospital would be about $22,500 per year, note the researchers.

[See: Shojania KG, Yokoe D, Platt R, et al. Reducing vancomycin use utilizing a computer guideline. J Am Med Inform Assoc 1998; 5:554-562.]