Trends in the Antibiotic Treatment of Acute Sinusitis: 2003–2020

Eighty-five percent of outpatient pediatric acute sinusitis encounters result in an antibiotic prescription.¹ Evidence-based guidelines from the Infectious Diseases Society of America (IDSA, published 2012)² and the American Academy of Pediatrics (AAP, published 2013)³ for the management of acute sinusitis recommend empirical amoxicillin with (IDSA) or with/without (AAP) clavulanate for 10 to 14 days (IDSA).² Excessively broad or long antibiotic therapy is associated with adverse outcomes.⁴ Nationwide trends in pediatric sinusitis antibiotic treatment have not been assessed since IDSA and AAP guideline publication.⁵ Understanding these guidelines’ impact will inform future efforts to maximize antimicrobial stewardship interventions.

Methods

We evaluated trends in antibiotic dispensations for new acute sinusitis diagnoses among commercially insured subjects aged <18 years between April 1, 2003, and December 31, 2020, using the MarketScan Commercial Claims and Encounters Database. Subjects were identified via an outpatient encounter with an International Classification of Diseases (ICD) code for acute sinusitis (ICD Ninth Edition, 461.×; ICD 10th Edition, J01.×0), along with a same-day oral antibiotic dispensation. Each subject contributed their first qualifying acute sinusitis diagnosis after 6 months of continuous enrollment. Subjects were excluded for an acute sinusitis diagnosis in the 6 months before their qualifying encounter, for cystic fibrosis or chronic sinusitis in the 6 months previous, or for a same-day code for acute otitis media, cellulitis/abscess, community-acquired pneumonia, Streptococcal pharyngitis, or urinary tract infection. We used interrupted time series analyses to evaluate the impact of the IDSA guidelines on antibiotic selection on the basis of quarterly intervals (weighted per quarter by number of subjects), using ordinary least-squares regression with Newey-West standard errors and a one-quarter lag. We performed 2 sensitivity analyses:

1. excluding subjects dispensed an oral antibiotic in the 30 days before the sinusitis diagnosis; and

2. including subjects who filled a prescription on the day of or day after their sinusitis diagnosis.

Statistical analyses were performed using the Aetion Evidence Platform and Stata, Release 17. The Brigham and Women’s Hospital institutional review board waived the need for informed consent.

Results

Among 3.75 million patients with qualifying acute sinusitis diagnoses, 3.19 million (84.9%) dispensed an antibiotic comprised the final cohort (Supplemental Fig 3). Median age was 9 years (interquartile range, 4–14). The most frequently dispensed antibiotic was amoxicillin (36%) (Fig 1, Supplemental Tables 1 and 2). After IDSA guideline publication, there were immediate changes in the proportion of children dispensed amoxicillin (1.56% absolute increase [95% confidence interval (CI), 0.05%–3.1%]), azithromycin (1.31% absolute decrease [95% CI, −2.32% to −0.31%]), and cefdinir (1.93% absolute decrease [95% CI, −2.85% to −1.00%]) (Fig 1, Supplemental Table 3). Comparing postintervention to preintervention dispensation trends, there were positive changes in first-line antibiotic trends (amoxicillin, 0.32% per quarter [95% CI, 0.24%–0.41%]; and amoxicillin-clavulanate, 0.26% per quarter [95% CI, 0.21%–0.31%]) and negative changes in the trends of azithromycin (−0.56% per quarter [95% CI, −0.61% to −0.51%]) and cefdinir (−0.12% per quarter [95% CI, −0.18% to −0.07%]), concordant with the guidelines. Overall, 80% of subjects were dispensed 10 days and 14% were dispensed ≥11 days of antibiotics (Fig 2). The 2 sensitivity analyses revealed no changes in the distribution of antibiotics dispensed.

FIGURE 1.

Nationwide trends in antibiotic selection for pediatric acute sinusitis. Proportions of subjects dispensed a given antibiotic, stratified by quarter. Antibiotics were identified via generic names. Trend lines are interrupted time series analyses, with 2012, quarter 2, as the intervention point, with a one-quarter lag. We opted to only assess the impact of the IDSA guidelines on antibiotic dispensations because the AAP guideline was published in such close succession with the IDSA guideline, we did not feel we could distinguish between the effect of the 2 guidelines. Others: penicillin V, dicloxacillin, oral cephalosporins (except cefdinir), fluoroquinolones, tetracyclines, macrolides, clindamycin, linezolid, and trimethoprim-sulfamethoxazole.

FIGURE 2.

Nationwide trends in antibiotic duration for pediatric acute sinusitis. Proportion of subjects with acute sinusitis who received the noted duration of antibiotics (captured through days’ supply dispensed). Azithromycin was excluded given its routine use as a 5-day prescription. Inset shows all durations except 10 days to visualize trends more easily.

Discussion

In a nationwide cohort of commercially insured children with acute sinusitis, the proportion dispensed amoxicillin with or without clavulanate increased after IDSA guideline publication, although one-third of subjects were still dispensed a non–first-line antibiotic. This suggests that evidence-based guidelines can impact prescribing practices and highlights the importance of rigorous studies evaluating the optimal empirical antibiotic selection and duration to inform updated guidelines.

Although antibiotic dispensations capture what patients received, they may not reflect prescriptions as written, because some may not have been filled. Restricting encounters to the first qualifying diagnosis per patient may have resulted in a younger population, compared with including >1 encounter per patient. This study lacks a negative control and confounding because other events affecting time trends cannot be excluded.

There are an estimated 65 antibiotic prescriptions for sinusitis per 1000 pediatric population in the United States annually.¹ If all US children prescribed an antibiotic for sinusitis were prescribed 5 to 7 days, as per 2021 AAP guidance,⁶ rather than the patterns described herein, 14.7 to 24.4 million fewer days of antibiotics per year would be prescribed.⁷ Nonfirst-line and potentially excessively long antibiotic courses are prescribed to children with acute sinusitis, presenting an opportunity for antimicrobial stewardship interventions.

Glossary

AAP

American Academy of Pediatrics

CI

confidence interval

ICD

International Classification of Diseases

IDSA

Infectious Diseases Society of America