A decision-support algorithm to help primary care physicians assess adolescents and adults with uncomplicated acute bronchitis reduced unnecessary antibiotic use by about 10%, according to a report published online Jan. 14 in JAMA Internal Medicine.
Plus, printed and computer-assisted approaches alike decreased the overuse of antibiotic treatment in primary care practices, said Dr. Ralph Gonzales of the departments of medicine and epidemiology and biostatistics, University of California, San Francisco, and his associates (JAMA Intern. Med. 2013 Jan. 14 [doi:10.1001/jamainternmed.2013.1589]).
Dr. Ralph Gonzalez
Reduced antibiotic use did not result in a significant increase in return visits to either the study’s primary care practices or a hospital, the researchers noted. So it appears that there was no appreciable increase in the adverse clinical consequences of withholding antibiotics, such as a rise in the incidence of pneumonia.
"In aggregate, these findings support the wider dissemination and use of this clinical algorithm to help reduce the overuse of antibiotics for acute bronchitis in primary care," the investigators said.
Dr. Gonzales and his colleagues tested the algorithm in a randomized, controlled trial involving 33 primary care practices from Geisinger Health System in rural and semirural central and northeastern Pennsylvania.
In addition to patient education materials, the decision-support algorithm included clinician education materials, such as:
– Prompts for taking an appropriate history and physical examination of all patients presenting with cough illness.
– A way to calculate a patient’s probability of having pneumonia.
– A list of relevant testing and treatment options for bronchitis.
– Feedback on the clinicians’ performance in appropriately prescribing antibiotics.
Eleven primary care practices were randomly assigned to use a printed version of the decision-support algorithm, 11 to use a computerized version, and 11 to serve as control practices where no decision-support algorithm was implemented.
The study included all of the practices’ board-certified internal medicine and family practice physicians, nurse practitioners, physician assistants, and registered nurses. The patient population comprised all adolescents and adults aged 13-64 years who presented with uncomplicated acute bronchitis during a single winter-season intervention period.
The data from these 6,242 patient cases were compared with those of 9,808 cases that occurred during the three winter seasons preceding implementation of the decision-support algorithm.
The number of visits for acute respiratory infections and the proportion diagnosed as uncomplicated acute bronchitis remained stable over time and across the study sites.
Compared with the preintervention period, the percentage of patients who were prescribed antibiotics during the intervention period decreased by 11.7% (from 80.0% to 68.3%) in practices using the print algorithm, and by 13.3% (from 74.0% to 60.7%) in practices using the computerized algorithm.
Those declines were significantly greater than the change in antibiotic prescribing seen in the control practices, where clinicians actually increased the use of antibiotics by 1.8%, from 72.5% to 74.3%, Dr. Gonzales and his associates said.
The percentage of patients who were not given antibiotics and who subsequently developed pneumonia requiring return visits remained "low" in all practices at all time periods, ranging from 0.5% to 1.5%.
The study results indicate that both conventional (printed) and computerized strategies for decision support are effective at reducing unwarranted use of antibiotics in uncomplicated acute bronchitis, the investigators said.
However, the findings may not be applicable in all settings, the researchers cautioned, because the study included only small- to medium-sized primary care practices within an integrated health care system in a rural and semirural region.
In addition, the study could not establish whether the declines in inappropriate prescription of antibiotics were due to the patient education component, the clinician education component, some other component, or simply to all clinicians’ knowledge that they were being monitored, the researchers said.
The Centers for Disease Control and Prevention supported the study. Dr. Gonzales reported ties to Phreesia, and an associate reported ties to Merck.