Pneumonia and urinary tract infections are the two most common infections requiring hospitalizations and major reasons for the overuse of broad-spectrum antibiotics.
While UTIs require 480,000 hospitalizations annually in the U.S., pneumonia accounts for more than 1.5 million adult admissions. Up to 40% of these UTI patients unnecessarily receive extended-spectrum antibiotics while that figure rises to approximately 50% of pneumonia patients, according to estimates.
New studies now show that real-time and algorithm-derived computerized alerts about the best antibiotic match for an individual patient can significantly improve the appropriate medication for patients and reduce the possibility of antibiotic resistance, a major public health threat.
The two Journal of the American Medical Association INSPIRE (Intelligent Stewardship Prompts to Improve Real-time Empiric Antibiotic Selection) studies, both carried out in 59 HCA Healthcare hospitals in 12 states, highlighted the problems associated with the over-prescription of broad-spectrum antibiotics where physicians often choose to prescribe such medication out of concern their patients could be sick with antibiotic-resistant bacteria.
The algorithm-derived alerts at the time of prescription showed a better match for 28% of the pneumonia patients and 17% of urinary tract patients when compared with physicians not receiving such alerts, according to the studies. Such patients did not increase intensive care transfers or length of stays.
The alerts used patient characteristics from the electronic medical record as well as hospital and location-specific data to determine the patient’s risk for an antibiotic-resistant infection. Assessment of risk was based on pre-trial data from more than 200,000 HCA Healthcare patients with pneumonia and UTI. Physicians treating patients with a low risk for antibiotic-resistant bacteria were prompted to give standard-spectrum antibiotics.
“The INSPIRE trials have found a highly effective way to help physicians follow treatment recommendations to optimize antibiotic selection for each patient,” said Sujan Reddy, M.D., medical officer in the epidemiology, research and innovations branch of the Centers for Disease Control and Prevention’s (CDC's) division of healthcare quality promotion. “These trials show the value of harnessing electronic health data to improve best practice.”
The trials, which were ongoing when the COVID-19 pandemic began and showed continual benefits when other hospitals in the U.S. were reporting large increases in the use of broad-spectrum antibiotics, identified patients with low risk for antibiotic resistance, prompting physicians to prescribe standard-spectrum antibiotics instead.
“Many different bacteria can cause pneumonia or UTI, and picking the best-matched antibiotic can be a challenge,” said Shruti Gohil, M.D., assistant professor in the division of infectious diseases at the University of California, Irvine School of Medicine. “Results from these trials show that giving physicians an alert informing them of their patient’s actual risk for antibiotic resistance can help them choose the best antibiotic and reduce extended-spectrum antibiotic use.”
HCA Healthcare is now implementing the new algorithm-derived system across its entire network, which includes 186 hospitals and approximately 2,400 ambulatory sites of care.
The studies, led by Harvard Pilgrim Health Care Institute, HCA Healthcare and the University of California, Irvine, were funded by the CDC.