PerspectiveIn the Journals Plus

‘Expected practice’ intervention reduces antibiotic prescriptions

Show Citation

December 5, 2018

Brad Spellberg
Brad Spellberg

An “expected practice” intervention centered around prescribing shorter courses of antibiotics for standard infections substantially reduced antibiotic therapy duration at a downtown Los Angeles hospital with no change in mortality, study findings published in Open Forum Infectious Diseases showed.

According to Brad Spellberg, MD, chief medical officer at the Los Angeles County-University of Southern California Medical Center and professor of clinical medicine and associate dean for clinical affairs at USC’s Keck School of Medicine, and colleagues, while prior studies have found that shorter courses of antibiotics have a similar effect as traditional courses for many infections, many clinicians are unaware of short-course therapy as a stewardship tool and most ID clinicians do not recommend it.

Spellberg explained to Infectious Disease News that “expected practice” interventions — first described in a JAMA article in 2016 — involve institutions setting their own expectations for standards of care among providers that are stronger than guidelines and based on published medical evidence. Primary care and specialty care experts develop the expected practice, which must then be approved by medical staff committees, including the medical executive committee, and hospital leadership, Spellberg said.

“When specific clinical circumstances dictate differing from the expected practice, the clinician is expected to document why in the medical record,” he said. “Thus, the expected practice supports change in practice by helping to inform clinicians about care standards that are expected to be met at the institution.”

“Furthermore,” Spellberg said, “since the institution and its medical staff have set the expected standards, they help shift responsibility for clinical decision-making to the institution itself. Thus, the expected practice provides a measure of psychological reassurance to staff that if they follow [it] and an unfortunate event happens to a patient, the clinician has done the right thing and the institution will bear responsibility for the decision. This is particularly important for antibiotic usage since providers tend to incorrectly and massively overprescribe antibiotics out of fear that not doing so could harm patients.”

Spellberg and colleagues developed an expected practice for antibiotic durations that was implemented at the hospital in October 2016. For the study, they collected all inpatient visits of adults aged 18 years or older for 12 months prior to implementation to establish baseline outcomes. October was treated as a “burn-in period” for the new interventions, and Spellberg and colleagues then collected inpatient visits for 12 months following implementation, from Nov. 1, 2016, through Oct. 31, 2017.

They included any patient with an ICD-10 code for a urinary tract infection (UTI), skin and skin structure infection (SSTI), pneumonia or ventilator-associated pneumonia in the first 20 discharge diagnoses. The primary outcome was antibiotic days of therapy, defined as the “sum total of days of each antibiotic administered as an inpatient plus the outpatient days prescribed upon hospital discharge.” The secondary outcome was total antibiotic exposure, defined as the “sum total of milligrams of antibiotics administered as an inpatient plus the milligrams prescribed as an outpatient upon discharge from hospital.”

After the implementation of the expected practice, Spellberg and colleagues observed a decrease in average antibiotic days of therapy of 10% for UTIs, 11% for both SSTIs and pneumonia, and 27% for ventilator-associated pneumonia. According to the study, the decreases in antibiotic exposures were even larger: 17%, 13%, 29% and 35% for UTIs, SSTIs, pneumonia and ventilator-associated pneumonia, respectively. Moreover, no changes in mortality were observed. According to Spellberg and colleagues, the use of expected practice as a psychological tool may help to improve antimicrobial stewardship.

“We described a no-cost, easy-to-implement, simple intervention that helped our providers significantly improve their antibiotic usage, resulting in a marked decrease in antibiotics prescribed, with no resulting harm to patients,” Spellberg said. – by Marley Ghizzone

Disclosures: Spellberg reports no relevant financial disclosures.

itj+ Infographic

itj+ Perspective

Jasmine Marcelin, MD

Our practice is filled with dogma that suggests that the longer you treat a person’s infection, the more likely you are to eliminate that infection and avoid development of resistance; more and more research has debunked that dogma, clearly demonstrating that shorter courses of antibiotics may be just as efficacious in eliminating the pathogen. Furthermore, shorter durations come with the added benefit of reduced antibiotic exposure and fewer possibilities for antimicrobial resistance or adverse drug effects.

Our colleague Paul E. Sax, MD, once posted a hilarious tongue-in-cheek explanation of how ID doctors decide on duration of antibiotic therapy, with the first cardinal rule being, “Choose a multiple of five (fingers of the hand) or seven (days of the week).” Although the article was written in jest, when we look at current guidelines to treat specific conditions, we often find these seemingly arbitrary duration ranges: 3-5, 7-10, 10-14 days. While making the decision about whether one goes with 3 vs. 5, or 7 vs. 10 days of therapy may seem a bit like voodoo, convincing people that “shorter is better” may be challenging.

Rather than telling people to go look up the guidelines for treatment of each syndrome, the “expected practice” document developed by Yadav, Spellberg and colleagues provides a succinct guide on the expected duration of therapy for 10 common infections treated in the acute-care setting, all of which were treated in 7 days or less. They also provided high-quality evidence to support the use of these expected durations, a short teaching pearl about the syndrome and duration and included an accountable justification requirement for those who deviated from the expected practice. One of the most useful syndromes in their expected practice table was asymptomatic bacteriuria, which of course requires NO antibiotic treatment, but we see treated all the time. Recommendations to NOT treat asymptomatic bacteriuria are in ALL CAPS: Very appropriate. These changes resulted in significant decreases in antibiotic duration of therapy for urinary tract infections, ventilator-associated pneumonia, all pneumonia and skin/soft issue infections.

I love reading antimicrobial stewardship studies demonstrating that the battle against overly broad empiricism and ultra-long duration of therapy can be won with behavioral science. We already know that clinicians respond best to peer comparison and default to societal norms, and simple changes in information delivery and expectations of practice can lead to a significant change in prescribing behavior. The “expected practice” method is a unique approach leveraging our “type A” desires to conform to behavioral norms. Like the peer comparison tool described successfully in outpatient antimicrobial stewardship, “expected practice” is a “Keeping up with the Joneses” meets “Big Brother is watching” method of getting clinicians to do the right thing. One of my favorite (paraphrased) comments in the discussion was that if people prescribed antibiotics for shorter durations because they knew they were being watched, so be it! If that is what it takes to change practice, I am all for it, and perhaps after being watched for long enough, clinicians will believe that it is the right thing to do.

As we teeter near the edge of this cliff overlooking an abyss of antimicrobial resistance, one thing is certain: Although better drugs can help us treat the patients who develop these awful infections, the only thing that can truly stop us from falling off the cliff is a change in prescribing behavior.

Jasmine R. Marcelin, MD

Assistant professor, infectious diseases
Associate medical director, antimicrobial stewardship and infection control
University of Nebraska Medical Center
Omaha, Nebraska

Disclosure: Marcelin reports no relevant financial disclosures.