For Clinicians | Antibiotic Eye Drops for Pink Eye

For Clinicians | Antibiotic Eye Drops for Pink Eye

What the Evidence Says and What Schools Still Require

By Dr. Jamie Wilkey, PharmD — Director of Clinical Strategy, Jase
Medically reviewed and edited by Kristen Carpenter, PA-C

“But the school won’t take him back without the drops. I need them!“

That sentence gets said at pharmacy counters and repeated into phone triage every week by the frantic parents of kids with pink eye. A kid woke up with one lid crusted shut, the front office called by 9:15, and now a parent is standing in front of one of us needing an Rx so their child can go back to school or daycare.

So we write the script. Yep. About 69% of the time, going by the claims data.¹

Today we’re talking about what those antibiotic eye drops actually do. Then why the visit is usually about authorization rather than infection, the one presentation where antibiotics make a difference (it isn’t drops), and the short list of findings that should change the plan no matter what the school form says. Let’s jump into what to do about those goopy eyes that look awful.

Do antibiotic eye drops change the course of pink eye?

A little, in the cases that are actually bacterial, and not nearly enough to explain how often we reach for them.

The 2023 Cochrane review pooled 21 trials and 8,805 patients with acute bacterial conjunctivitis. About 55% of the placebo group cleared on their own by day four to nine, compared with 68% on antibiotics.² That works out to a number needed to treat of 7 for clinical cure, against a number needed to harm of 32 for ocular adverse effects with the non-fluoroquinolone drops.³ By day six to ten the two groups have mostly converged.

That is the ceiling on the benefit. It only applies to the bacterial cases. The problem here is that we cannot reliably tell which ones those are. AAFP’s 2024 clinical review states that no single sign or symptom accurately differentiates viral from bacterial conjunctivitis,⁴ and that includes the two findings most of us rely on, purulent discharge and morning matting. So the drops go out to a mixed population in which we cannot cleanly identify the patients who stand to benefit.

The prescribing data shows how that plays out. Across 44,793 pediatric ambulatory encounters, topical antibiotics were dispensed within a day for 72% of office-based visits and 57% of ED visits, but only 34% of eye clinic visits.¹ You could reasonably say that is specialty selection, since eye clinics see referrals rather than the 9:15 phone call. Fair enough. Then look at the encounters where viral conjunctivitis was the documented diagnosis, where there is nothing for an antibiotic to act on: 28% of those children got drops anyway.¹

The 24-hour rule is school policy, not a clinical standard

Nothing in pediatric guidance requires a child with pink eye to be on drops before going back to school or daycare. The AAP position, carried in the Red Book and in Managing Infectious Diseases in Child Care and Schools, is that a child with conjunctivitis and no fever and no change in behavior does not need to be excluded, and that otherwise healthy, well-appearing children with red, watery eyes should not be excluded or isolated at all.⁵ 

Now let’s check out what states actually publish. A 2022 survey of all 50 state policies found:

• 15 states have no conjunctivitis policy at all
• 10 allow students to stay in school
• 5 permit return 24 hours after starting antibiotics
• 5 require a physician’s approval
• 17 states plus DC give inconsistent recommendations, and not one state policy anywhere references the American Academy of Ophthalmology⁶

So the 24-hour rule that sends these families to us is written into state policy in five states.

Before we make the school nurse the villain, though: they are working from guidance that does not agree with itself. AAO’s own patient education page, updated August 2024, advises staying home until symptoms start to improve.⁷ AAP says don’t exclude. A district with no state guidance to point to picks the rule that sounds careful, and 24-hours-on-antibiotics is the one everybody has heard and feels better for parents who don’t want their kids to catch the goop eye from other kids in the class.

This is what gets awkward and that we weren’t really trained for. We were taught to make a clinical decision about an infection. What we are being asked for is authorization to re-enter a building, and it costs the family a visit and a copay to get it.

The drops are not buying a smoother course either. In that same pediatric cohort, ambulatory revisits within 14 days ran 3.1% among children who got antibiotics and 3.6% among those who did not, hospitalizations 0.03%, ED revisits 0.12%, with no differences between the groups.¹

Check out what your own state actually publishes before you write the Rx. In 15 of them, there is nothing to comply with.

Check the ears when you see the eye

There is one presentation where antibiotics clearly do the work, and the treatment is oral, not topical.

Conjunctivitis-otitis syndrome is a good one to remember in the back of your mind. Ear infections turn up alongside bacterial pink eye often. Published estimates run from 32% to 73%, and plenty of those kids never complain about ear pain.9 Haemophilus influenzae is the usual culprit. It was 70% of isolates in a cohort of 67 children with conjunctivitis plus a concurrent ear or sinus infection.⁸ Drops cannot reach the middle ear, so they do not fix the real problem. Close to 30% of US Haemophilus influenzae isolates resist ampicillin, and amoxicillin fails the same way, since the resistance is usually a beta-lactamase.¹⁰ Amox-clav holds up against almost all of them. So amox-clav or cefdinir here, not plain amoxicillin.

Look in the ears of every kid whose parent brings you a red, crusted eye, including the ones with no ear symptoms. If the ear is involved, you treat one infection with one oral antibiotic and the eye clears with it. That is the case where the right answer is an Rx, and it still isn’t drops.

What actually changes the plan

Short list, and none of it depends on sorting viral from bacterial (which is the point, since we cannot do that reliably anyway).

• Contact lenses. Any lens wearer with a red eye gets seen today, not tomorrow. Contact-lens-associated keratitis is most often Pseudomonas, and it can leave permanent central corneal scarring and vision loss even when treatment is prompt and appropriate.¹¹ ¹² Empiric coverage needs a fluoroquinolone, and these patients belong with ophthalmology.
• Moderate to severe pain, any change in vision, or photophobia. Not conjunctivitis territory anymore.¹¹
• Sudden heavy purulent discharge in a sexually active adolescent or adult. Treat as gonococcal until proven otherwise. It needs systemic ceftriaxone with empiric anti-chlamydial coverage and same-day ophthalmology, because untreated it progresses to corneal melting and perforation.¹³
• A baby under a month old.  Different illness, and it needs a clinician today rather than drops.⁷
• Not improving, or keeps coming back. Failure to respond and recurrent episodes both go to ophthalmology.¹¹
• Anything you find yourself wanting a steroid for. Steroids can slow corneal healing and raise intraocular pressure, and ‘needs steroids’ is itself a referral criterion.¹¹

The contact lens question is the one to build into intake. Your MA or tech can ask it before you ever walk in the room. It is most likely to get missed on a busy afternoon.

All other treatment is a warm cloth, hand hygiene, and a few days to pass.

Where preparation actually helps here

Look again at what the parent in the opener was offered. Two options: get an Rx, or keep the kid home. Nobody handed them the third thing, which was a plain statement of what their state and their district actually require and if the drops actually help the condition. That’s similar to most access problems we write about. Care is either fully authorized or fully unavailable, with nothing sanctioned in between. Usually the missing middle is a medication that should have been on the shelf before anyone needed it. Appropriate medical preparation means having the right thing ready before you need it, and sometimes the right thing turns out to be a warm cloth, four days, and knowing you are not cutting a corner.

The bottom line

Most pink eye clears on its own. Antibiotic drops help modestly in the cases that are bacterial, and we cannot reliably tell which ones those are. The 24-hour rule that sends these families to us is school policy, not medicine, and in most states nobody wrote it down at all.

Look in the ears, and know the short list that changes the plan. The rest is a conversation, and that conversation is worth more to the family than the prescription is, even if they really want that Rx when they walk in the door to see you.

Sources

1. Shapiro DJ, Geanacopoulos AT, Subramanian SV, et al. Antibiotic Treatment and Health Care Use in Children and Adolescents With Conjunctivitis. JAMA Ophthalmology. 2024;142(8):779-780. https://jamanetwork.com/journals/jamaophthalmology/fullarticle/2820326
2. Chen YY, Liu SH, Nurmatov U, et al. Antibiotics versus placebo for acute bacterial conjunctivitis. Cochrane Database of Systematic Reviews. 2023;3:CD001211. https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD001211.pub4/full
3. Sonoda K, Takeuchi Y. Antibiotics for acute bacterial conjunctivitis. American Family Physician. 2023;107(6):581A-581B. https://www.aafp.org/pubs/afp/issues/2023/0600/mbtn-acute-bacterial-conjunctivitis.html
4. Winters S, Frazier W, Winters J. Conjunctivitis: Diagnosis and Management. American Family Physician. 2024;110(2):134-144. https://www.aafp.org/pubs/afp/issues/2024/0800/conjunctivitis.html
5. Children’s Mercy Kansas City. Conjunctivitis Care Process Model: Recommendations for Inclusion or Exclusion from Daycare/School, citing the AAP Red Book (2015) and the 2002 AAP / American Public Health Association child care exclusion recommendations. https://www.childrensmercy.org/health-care-providers/evidence-based-practice/cpgs-cpms-and-eras-pathways/conjunctivitis-care-process-model/recommendations-for-inclusion-or-exclusion-from-daycare-school/
6. Lee T, Kuo IC. Survey of state conjunctivitis policies for school-age students. Journal of AAPOS. 2022;26(3):115.e1-115.e5. https://pubmed.ncbi.nlm.nih.gov/35378302/
7. Mukamal R. When Do You Need Antibiotics for Pink Eye (Conjunctivitis)? American Academy of Ophthalmology, reviewed by Rupa K. Wong, MD, August 26, 2024. https://www.aao.org/eye-health/tips-prevention/over-prescription-antibiotics-pink-eye
8. Hu YL, Lee PI, Hsueh PR, et al. Predominant role of Haemophilus influenzae in the association of conjunctivitis, acute otitis media and acute bacterial paranasal sinusitis in children. Scientific Reports. 2021;11. https://pmc.ncbi.nlm.nih.gov/articles/PMC7794412/
9. Conjunctivitis and Conjunctivitis-Otitis Syndrome. Pediatric EM Morsels. https://pedemmorsels.com/conjunctivitis-and-conjunctivitis-otitis-syndrome/
10. Potts CC, Rodriguez-Rivera LD, Retchless AC, et al. Antimicrobial Susceptibility Survey of Invasive Haemophilus influenzae in the United States in 2016. Microbiology Spectrum. 2022;10(3):e02579-21. https://pmc.ncbi.nlm.nih.gov/articles/PMC9241922/
11. Hashmi MF, Gurnani B, Benson S. Conjunctivitis. StatPearls. Updated January 26, 2024. https://www.ncbi.nlm.nih.gov/books/NBK541034/
12. Hatami H, Ghaffari Jolfayi A, Ebrahimi A, et al. Contact Lens Associated Bacterial Keratitis: Common Organisms, Antibiotic Therapy, and Global Resistance Trends: A Systematic Review. Frontiers in Ophthalmology. 2021;1:759271. https://www.frontiersin.org/journals/ophthalmology/articles/10.3389/fopht.2021.759271/full
13. Dvorak SF, Lee C. Hyperacute Gonococcal Conjunctivitis. Brown Hospital Medicine. 2025;4(4). https://bhm.scholasticahq.com/article/144966-hyperacute-gonococcal-conjunctivitis