- Kenny Lin, MD, MPH
Outside of pregnancy, antibiotics for patients with asymptomatic bacteriuria - that is, a positive urine culture in a patient with no signs or symptoms of a urinary tract infection - do more harm than good. Consequently, comprehensive guidelines and a Choosing Wisely recommendation from the Infectious Diseases Society of America strongly discourage this practice. But just as it is difficult for an interventional cardiologist to not stent a narrowed coronary artery in a patient with stable angina, many physicians have a hard time not treating bacteria that grow in a urine culture, no matter what the science says. To avoid this situation, the U.S. Preventive Services Task Force recommends not screening for asymptomatic bacteriuria in men and nonpregnant women.
Unfortunately, these guidelines are frequently ignored in clinical practice. Perhaps a patient's urine smells funny, or it looks darker or cloudier than usual. Someone (who may not have actually evaluated the patient) obtains a urine culture. The culture grows bacteria, the incorrect diagnosis of "urinary tract infection" makes its way into the patient's chart, and the patient subsequently receives antibiotics that at best do not help, but possibly lead to individual adverse effects and increased antibiotic resistance. Performing urine cultures on patients with urinary catheters is especially problematic, since virtually all of them develop asymptomatic bacteriuria.
A recent study by Dr. Barbara Trautner and colleagues in JAMA Internal Medicine reported on the results of an intervention to reduce treatment of asymptomatic bacteriuria in catheterized inpatients at a Veterans Affairs hospital in Texas. The intervention, part of the aptly named "No Knee-Jerk Antibiotics Campaign," focused on reducing inappropriate urine culture ordering through case audits and feedback and distribution of a guideline-based diagnostic algorithm on a pocket card. When unnecessary screening and/or overtreatment of asymptomatic bacteriuria was judged to have occurred, researchers used a script to provide feedback to teams of internal medicine residents. During the 3-year intervention and maintenance period, the rate of treatment of asymptomatic bacteriuria fell by 75 percent.
"Changing the behavior of clinicians is fraught with challenges, but change is possible," wrote Dr. Manisha Juthani-Mehta in an accompanying editorial. "Some of the components that have been successfully shown to facilitate a change in behavior include education, feedback, participation by clinicians in the change effort, and administrative interventions." The study by Trautner and colleagues demonstrated that for this unnecessary care scenario, it is possible to motivate physicians to Choose Wisely.