Antibiotic Use Climbs for Acute Respiratory Infections

Tara Haelle

July 21, 2015

Antibiotic use, particularly of macrolides, increased among veterans with acute respiratory infections over the course of 6 years, according to a retrospective, cross-sectional study. Moreover, providers accounted for nearly 60% of the variation in prescription rates.

The authors note that their findings, published in the July 21 issue of the Annals of Internal Medicine, mirror similar findings in other national studies, including ones using data from the National Ambulatory Medical Care Survey.

"[A]t the population level, we would expect a much lower overall proportion of antibiotic prescribing based on national treatment guidelines," write Barbara Ellen Jones, MD, from Veterans Affairs Salt Lake City Health Care System at the University of Utah in Salt Lake City, and colleagues.

"The lack of progress in reducing unnecessary antibiotic prescribing for [acute respiratory infections] is a major public health concern."

The researchers analyzed all patient visits to Veterans Affairs (VA) emergency departments and primary and urgent care clinics within 130 VA medical centers between 2005 and 2012 that involved acute respiratory infections with low rates of bacterial infection. These included nasopharyngitis, pharyngitis, sinusitis, acute bronchitis, upper respiratory infection, laryngitis, and tonsillitis. The authors excluded patients with pneumonia, influenza, urinary tract infection, tissue infections, or other infections as well as patients with comorbid conditions that increase the risk for serious bacterial infections. A total of 1,044,523 visits met these criteria, out of more than 2.4 million visits with acute respiratory infection diagnoses.

Among these 1 million visits, 68.4% led to an antibiotic prescription, and 43.4% of these antibiotics were macrolides. The overall rate of antibiotic prescriptions increased from 67.5% in 2005 to 69.2% in 2012 (P < .001). Similarly, macrolide antibiotic use increased from 36.8% to 47.0% during that time. Prescriptions for penicillin, however, decreased, going from 36.0% to 32.1%, as did fluoroquinolone prescriptions, going from 15.0% to 12.7% (P < .001).

With little variation based on fever, age, setting, or comorbidities, 86% of sinusitis cases and 85% of bronchitis cases received antibiotics. In addition, 78% of those with a high fever received antibiotics, and 75% of those in urgent care centers received them.

The providers who fell in the top 10% for prescribing antibiotics, the researchers found, prescribed them in 95% of acute respiratory infection visits compared with 40% of visits (or lower) for the providers in the lowest 10%. Similarly, 89% of visits led to antibiotic prescriptions at the clinics falling in the top 10% for prescribing compared with 41% of visits at the clinics in the lowest 10%.

"[O]ur findings suggest that providers have a strong tendency to choose the same treatment regardless of patient or clinic characteristics, indicating that individual provider preference or 'style' heavily influences the antibiotic decision," the authors write.

Providers accounted for 59% of the variation in prescriptions, whereas clinics accounted for 28% and VA medical centers accounted for 13%. "Antibiotic prescribing was associated with many of the factors we measured, including temperature, distance to clinic, setting type, and geographic region," the authors write. "However, these associations were small and, even when taken together, had limited explanatory power. Antibiotic prescribing was common regardless of the factors we studied."

One limitation in the study was the lack of data regarding symptom duration, physical exams, or provider specialty, any of which might have contributed to the decision to prescribe antibiotics.

"As our understanding of the relationship between providers, patients, settings, and treatment decisions improves, so will our ability to target future information and stewardship efforts," the authors conclude.

The authors have disclosed no relevant financial relationships.

Ann Intern Med. 2015;163:73-80. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.