Interventions based on prudent choice guidelines can reduce antibiotic overuse in safety net hospitals

A statewide pay-for-work intervention based on a set of guidelines called Choosing Wisely reduced the number of inappropriate antibiotic prescriptions for acute respiratory infections by an average of 18 percentage points, from 43% to 25% in two large Los Angeles safety regions. in network hospitals.

BACKGROUND

About 25–50% of all antibiotic prescriptions in the US are inappropriate. This can lead to the growth of antibiotic-resistant bacteria and other direct harm that kills more than 35,000 people a year in the United States. Many health systems have engaged in pay-for-performance interventions to reduce inappropriate antibiotic prescriptions, but such interventions have rarely been studied in safety net systems. This is important because performance-based models for improving health care have performed poorly in safety-net systems, resulting in disparate effects for disadvantaged patients.

“Choosing Wisely,” which was launched in 2012 by the Board of the American Internal Medicine Foundation, is an initiative to reduce the use of unnecessary medical services and improve patient outcomes.

In 2015, the Los Angeles County Department of Health Services participated in a statewide pay-for-work program, and one of the targeted areas was inappropriate antibiotic prescriptions for acute bronchitis, according to the Choosing Wisely guidelines.

METHOD

Researchers conducted a nonrandomized trial at LA General Medical Center (formerly Los Angeles County + University of Southern California) and Olive View-UCLA Medical Center, two academic safety-net hospitals in Los Angeles. Based on a state pay-for-performance program, they used interventions that included audit and feedback; doctor’s education; proposed alternatives; procalcitonin, a test to diagnose lower respiratory tract infections; and public commitment. These interventions were based on the Choosing Wisely campaign, which aims to reduce the inappropriate use of antibiotics in respiratory infections, which are usually viral and therefore do not require antibiotics. The researchers also assessed several unintended consequences of the intervention, such as inappropriate antibiotic discontinuation.

EFFECT

These multicomponent interventions, based on wise choice guidelines, can reduce antibiotic use without measurable harm in safety-net hospitals that serve predominantly disadvantaged patients.

COMMENT

“The scientific community has long studied the impact of health care interventions on the medically underserved,” said lead author Dr. Richard Leuchter, clinical instructor of medicine in David Geffen’s Division of General Internal Medicine and Health Services Research at the UCLA School of Medicine. “This study provides real-world evidence that these types of behavioral interventions can reduce antibiotic overuse in under-resourced patients without causing unintended harm, such as reducing appropriate antibiotic use. This has important implications for responsible antibiotic use for the tens of millions of patients served annually in the U.S. network health systems and is an important advance in health equity in ensuring that our latest medical practices actually improve the health of all US communities.

THE AUTHORS

Additional authors of the study are Dr. Catherine Sarkisian, Ph.D. Dr. John Mafi, Carmen Carrillo, Sitaram Vangala, Dr. Oleg Melamed and Dr. Arthur Jeng of UCLA; Dr. AS Rebecca Trotzky-Sirr; Charles Coffey Jr. and Dr. Brad Spellberg of USC; and Dr. Eric Wei of NYC Health + Hospitals. Sarkisian and Mafi are also affiliated with the Greater Los Angeles VA Healthcare System and the RAND Corporation, respectively.

MAGAZINE

The study will be published in a peer-reviewed journal, the American Journal of Managed Care.

FINANCING

The research was funded by the National Heart, Lung, and Blood Institute (R38HL143614), the National Institute on Aging (5K24AG047899-05, P30AG021684-16, and 1K76AG064392-01A1), and the National Center for Translational Advances (UL2TR01, 8KL81TR01, 81).