National Trends in Visit Rates and Antibiotic Prescribing for Adults With Acute Sinusitis

Acute sinusitis is diagnosed in over 3 million visits annually among adults and children in the United States.¹ Of these, more than 80% result in an antibiotic prescription¹; however, many of these prescriptions may be unnecessary,^2,3 since sinusitis is most often of viral origin and benefits of antibiotics may be limited.⁴ Prior to 2012, amoxicillin was the recommended empirical treatment for acute bacterial sinusitis; current guidelines now recommend amoxicillin-clavulanate.^5-7 In light of recent studies⁴ and new treatment guidelines,⁷ we sought to examine visit rates and antibiotic prescribing patterns for adults with acute sinusitis in the United States.

We analyzed data from the National Ambulatory Medical Care Survey (NAMCS) and National Hospital Ambulatory Medical Care Survey (NHAMCS) between 2000 and 2009 to estimate visit rates and antibiotic prescribing for acute sinusitis in adults. We estimated the annual number of acute sinusitis visits per 1000 adults in the United States. Acute sinusitis was designated for visits in which any of 3 diagnosis fields contained the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) code for acute sinusitis (461.x). We excluded visits for acute sinusitis that had concomitant diagnoses that could potentially require antibiotics (eg, urinary tract infections) and visits that resulted in hospital admission. We also excluded visits where an aminoglycoside or vancomycin was prescribed, since those patients were presumed to have more serious or atypical infections.

In analyses of antibiotic selection, we restricted our sample to those visits where acute sinusitis was the primary diagnosis. The primary outcome measures were the proportion of visits for acute sinusitis in which any antibiotic was prescribed (antibiotic visits) and the proportion of antibiotic visits in which amoxicillin was prescribed.

All statistical analyses were performed using STATA 11 software (StataCorp) and accounted for the components of the complex survey design. For analysis of time trends, we grouped the survey data and census denominators in five 2-year intervals as recommended by the National Center for Health Statistics.

Between 2000 and 2009, there was a mean of 4.3 million (95% CI, 3.6-4.9 million) outpatient visits per year. Acute sinusitis was diagnosed in 0.5% (95% CI, 0.4%-0.5%) of all outpatient visits among adults during this period. The annual visit rate averaged 19.4 visits (95% CI, 16.5-22.3) per 1000 adults and did not change during the study period.

Antibiotics were prescribed in 83% (95% CI, 78%-86%) of visits for acute sinusitis, and this proportion did not change significantly during the study period (P = .85), which ranged from 74% (95% CI, 62%-83%) to 87% (95% CI, 78%-93%) of visits during this period. In addition, there were no changes in the overall rates of prescribing for any specific antibiotic class or antibiotic agent among visits (n = 1934) where an antibiotic was prescribed. The proportion of visits in which amoxicillin, the recommended agent, was prescribed was 17% (95% CI, 12%-23%) (Figure). Among the antibiotics prescribed other than amoxicillin, the most commonly prescribed were macrolides (29%), quinolones (19%), and amoxicillin-clavulanate (16%).

In the present study, using a nationally representative data set of ambulatory visits, we found that more than 80% of patients diagnosed as having acute sinusitis receive an antibiotic, despite mounting evidence that the benefits of antibiotic treatment for sinusitis are limited. Furthermore, we found that nearly 50% of patients diagnosed as having acute sinusitis received either a macrolide or a quinolone, while fewer than 20% received amoxicillin, the recommended first-line treatment during the study period. The frequent use of macrolides, as shown in this study, is particularly concerning because macrolide use has been associated with treatment failures for respiratory tract infections.⁸ Although likely to be effective, fluoroquinolones are usually unnecessary for sinusitis, and overuse is an important risk factor for colonization and infection with resistant organisms.⁹

We acknowledge limitations to this study. Because sinusitis is diagnosed based on physical examination findings and symptoms, data not captured in NAMCS/NHAMCS, we were unable to determine which patients had acute bacterial sinusitis as strictly defined based on the criteria suggested by recent clinical guidelines.⁷ We were also not able to determine if patients receiving antibiotic treatment had recurrent sinusitis or had previously experienced treatment failure with narrow-spectrum therapy for sinusitis, either of which might have made prescription of a broad-spectrum agent acceptable. Previous studies have suggested, however, that bacterial sinusitis composes a relatively small fraction of acute sinusitis cases seen in primary care, and treatment failure is also uncommon.^3,4 We were also unable to determine which patients had an allergy to recommended agents.

This study highlights that prescribing of broad-spectrum antibiotics for sinusitis, especially quinolones and macrolides, is extremely common. This is an important target for antimicrobial stewardship efforts partially because the benefits of antibiotic therapy are limited. Qualitative research to explore the health care provider and patient attitudes that influence antibiotic selection is a next step to understanding the problem. Also critically important are adoption of clinical guidelines that promote appropriate antibiotic use.⁷ Changes in prescribing behavior of health care providers for sinusitis are urgently needed to improve health care quality and stem the rising tide of antibiotic resistance in the United States.

Correspondence: Dr Fairlie, National Center for Immunization and Respiratory Diseases, Centers for Disease Control & Prevention, 1600 Clifton Rd NE, MS C-25, Atlanta, GA 30333 (iyl9@cdc.gov).

Published Online: September 24, 2012. doi:10.1001/archinternmed.2012.4089

Author Contributions:Study concept and design: Fairlie, Shapiro, Hersh, and Hicks. Acquisition of data: Shapiro. Analysis and interpretation of data: Shapiro, Hersh, and Hicks. Drafting of the manuscript: Fairlie and Hicks. Critical revision of the manuscript for important intellectual content: Fairlie, Shapiro, Hersh, and Hicks. Statistical analysis: Fairlie, Shapiro, Hersh, and Hicks. Obtained funding: Hersh. Study supervision: Hicks.

Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the Centers for Disease Control & Prevention.