Purchase this article with an account.
Nakul Shekhawat, Roni M Shtein, Taylor Blachley, Joshua D Stein; Substantial Over-Prescription of Antibiotics for Acute Conjunctivitis in the United States. Invest. Ophthalmol. Vis. Sci. 2016;57(12):5539.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Acute conjunctivitis is often caused by a virus, thus antibiotic therapy is usually unnecessary. We examined the extent that patients diagnosed with acute conjunctivitis are treated with topical antibiotics & factors associated with antibiotic use.
We analyzed health claims data from a large US managed care plan from 2001-2014. Eligible patients were diagnosed with acute conjunctivitis & continuously enrolled for >90 days after initial diagnosis. We excluded hospitalized patients, those undergoing eye surgery, or those with chronic conjunctivitis. Topical antibiotic use was defined as a prescription fill within 14 days of initial conjunctivitis diagnosis. Patient demographics, diagnosing provider, time to prescription fill, & antibiotic class were studied. Multivariable logistic regression determined factors associated with antibiotic use.
Of 340,630 patients diagnosed with acute conjunctivitis, 198,511 (58%) filled prescriptions for topical antibiotics. 20% of patients (N=38,774) filled prescriptions for combined antibiotic-steroids, which are contraindicated in acute infectious conjunctivitis. Prescription fills differed by age, race, income & diagnosing provider (all p<0.001) but not contact lens wear (p=0.58) or HIV diagnosis (p=0.36) (Table 1). Whites, those with higher incomes, and more educated patients had higher odds of receiving antibiotics for acute conjunctivitis compared with non-whites, less affluent and educated patients (all p<0.0001) (Table 2). Compared to patients diagnosed by ophthalmologists (37% fill), patients had higher percentages & odds of antibiotic fill if diagnosed by urgent care MDs (68% fill; OR 3.02, CI 2.91-3.13), internists (58%; OR 2.64, CI 2.55-2.74), family practice MDs (55%; OR 2.31, CI 2.23-2.40), pediatricians (59%; OR 2.17, CI 2.03-2.32), and optometrists (45%; OR 1.19, CI 1.15-1.24). Patients with HIV were no more likely to receive antibiotics (p=0.81) and those with end-organ damage from diabetes were 18% less likely to get antibiotics (p<0.0001) compared to patients without these conditions.
We identify rampant over-prescription of antibiotics for conjunctivitis in the US among insured patients, including potentially harmful practices that may increase costs, prolong infection duration, & lead to antibiotic resistance. Antibiotic use appears to be driven more by sociodemographic factors & provider type than medical indication.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only