Key Takeaways

Ear infections are close to universal: 62% of children have one by their first birthday, and 83% by age three ¹.

Antibiotics do not reduce pain at 24 hours. By day two or three they help about one child in 20 ².

For every 14 children treated, one gets vomiting, diarrhea, or a rash they would not otherwise have had ².

The diagnosis is a bulging eardrum or new discharge, not a red ear. Bulging is specific (about 97%) but easy to miss (sensitivity about 51%) ³.

Fluid behind the eardrum outlasts the pain. About 75% of cases that follow an ear infection clear within three months ⁴.

The short answer

Most ear infections in otherwise healthy children get better on their own. Antibiotics shorten the tail of the illness a little and prevent a small number of complications. The decision is not “treat or don’t treat”; it is “who benefits enough to justify the side effects.”

What you are looking at	What it usually means	What happens next
Ear pain, fever, miserable at night, eardrum bulging	Acute otitis media	Pain relief now; antibiotics or a 48–72 hour observation plan, decided by age and severity
Pus draining from the ear, pain suddenly better	Perforated eardrum from the infection	Same-day assessment; most heal without surgery
Fluid behind the eardrum, no pain, hearing seems dull	Otitis media with effusion (“glue ear”)	No antibiotics; recheck hearing if it persists
Red eardrum only, no bulge, no fluid, happy baby	Not acute otitis media	Look for another cause; redness alone is not diagnostic
Swelling or tenderness behind the ear, ear pushed forward	Possible mastoiditis	Emergency assessment

The youngest babies are the exception to every “wait and see” rule. Under 6 months, a fever or a suspected ear infection needs a clinician the same day, not a 48-hour experiment.

Why babies get them

The middle ear is a small room behind the eardrum, connected to the back of the throat by the eustachian tube. That tube equalizes pressure and drains fluid. In infants it is short, floppy, and nearly horizontal, so a cold that swells the lining of the nose and throat also blocks the drain ⁵.

Fluid then sits in a warm, closed space. Bacteria that were already living in the nose move in. That is the whole mechanism, and it explains the timing: ear infections follow colds, not teething, not drafts, not water in the ear.

It also explains the age curve. In the Boston cohort that followed 877 children from birth, 62% had at least one episode by age one, 83% by age three, and the peak incidence fell in the second six months of life ¹. Nearly half of the three-year-olds had had three or more episodes.

The diagnosis is not “a red ear”

This is where most of the confusion starts, and it changes what you should do with the information.

The AAP’s diagnostic criteria require either moderate to severe bulging of the eardrum, or new discharge from the ear that is not from the outer ear canal ³. Mild bulging plus ear pain that began within 48 hours also counts. A red eardrum with no fluid and no bulge does not.

The performance numbers are worth knowing, because they explain why a clinician can look and still be unsure. Bulging is highly specific (about 97%), meaning it is rarely wrong when present, but poorly sensitive (about 51%), meaning it is often absent when the diagnosis is real ³. Reduced eardrum mobility does better on both counts, around 95% sensitive and 85% specific.

Two practical consequences:

Ear tugging proves little. Many babies do it to self-soothe. Ear pain is present in only 50% to 60% of confirmed cases ³.
A normal-looking ear does not rule it out, and a red one does not rule it in. If the diagnosis matters, it needs otoscopy by someone trained to do it, and often pneumatic otoscopy or tympanometry.

Does it need antibiotics?

There is no single answer, because the answer depends on four things: age, severity, whether one ear or both, and whether the eardrum has perforated.

The AAP stratifies like this for children 6 months and older ³:

Age	Discharge from the ear	Severe symptoms, one or both ears	Both ears, non-severe	One ear, non-severe
6–23 months	Antibiotics	Antibiotics	Antibiotics	Antibiotics or observation
24 months and older	Antibiotics	Antibiotics	Antibiotics or observation	Antibiotics or observation

“Severe” has a specific meaning here: moderate or severe pain, pain lasting 48 hours or more, or a temperature of 39°C (102.2°F) or higher in the past 48 hours ³. A miserable baby with a 38.5°C fever and a bulging eardrum is not automatically in that category, though a clinician may still treat.

Observation is not the same as doing nothing. It requires a reliable way to be seen again, and a plan to start antibiotics if the child worsens or has not improved within 48 to 72 hours ³.

NICE, the UK guidance body, goes further and frames antibiotics as optional for most children, including those under 2 with infection in both ears. Its reasoning is worth reading, because it is blunt: antibiotics make little difference to pain at 24 hours, make little difference to common complications, and cause diarrhea and nausea ⁶.

The two guidelines disagree because they weigh the same evidence differently. Neither is irrational. If your clinician offers a back-up prescription, that is the NICE approach in practice: fill it if things do not improve in 48 to 72 hours.

What antibiotics do, and what they don’t

The best summary comes from the Cochrane review of 13 randomized trials, covering 3,401 children and 3,938 episodes, all in high-income countries ².

Outcome	What the trial data show
Pain at 24 hours	No meaningful difference (RR 0.89)
Pain at 2–3 days	About one third fewer children in pain; 1 child benefits per 20 treated
Pain at 10–12 days	Two thirds fewer in pain, from one trial
Vomiting, diarrhea, rash	38% more likely; 1 additional child harmed per 14 treated
Abnormal eardrum mobility at 2–4 weeks	Modest reduction
Eardrum perforation	Reduced, about 1 fewer per 33 treated
Recurrence over the next months	No difference
Serious complications	Too rare to detect a difference in these trials

Two numbers capture the trade-off. You treat 20 children to spare one child two days of pain. You treat 14 children to cause one extra episode of vomiting, diarrhea, or rash.

The complication picture is similarly unspectacular. A 2024 meta-analysis that pooled trials and observational studies found acute mastoiditis in 0.02% of children with ear infections, and antibiotics roughly halved that risk. The number needed to treat to prevent one case was 5,368, against a number needed to harm of 23 for adverse effects ⁷.

That does not mean antibiotics are pointless. It means the case for them is strongest in the children most likely to have a complication, and weakest in an older child with mild pain in one ear.

How long does it last?

Pain usually settles within three days, and sometimes takes up to a week ⁶. The infection is not the part that lingers.

Fluid behind the eardrum does. In the AAO-HNS guideline on otitis media with effusion, about 75% of effusions that follow an ear infection resolve within three months ⁴. The rest take longer, and a small number persist.

That fluid is why your baby may seem not to hear you for a few weeks after the fever is gone. It is a conductive loss: sound is muffled on its way to a working inner ear. It is usually temporary, and it is the reason a hearing check is recommended when fluid persists for three months or more, or sooner if the child is at higher risk ⁴.

If hearing is still dull three months after the infection, that is a conversation for the pediatrician, not a reason to buy a supplement.

What actually helps at home

Pain is the thing you can treat, and it is the thing that ruins sleep.

Acetaminophen or ibuprofen, dosed by weight. NICE explicitly recommends regular doses and maximum doses for severe pain in the first days ⁶. Never aspirin in a child.
Fluids. A sore ear plus a sore throat plus fever is a recipe for a dehydrated baby who refuses feeds.
Sleep, however you can get it. Pain is often worse at night, partly because lying flat changes pressure and partly because there are no distractions. Do not prop the mattress or add pillows for a baby under one; that is a suffocation risk, and there is no evidence it helps the ear.
Finish a prescribed course. If antibiotics are started, complete them, even if the child is better in two days.

What does not help:

Decongestants and antihistamines. NICE states the evidence shows they do not help symptoms ⁶.
Cotton buds, oil, garlic, onion poultices, or hydrogen peroxide. None treats a middle-ear infection, and some cause harm.
Ear drops without an examination. If the eardrum is perforated, some drops are the wrong choice. NICE allows anaesthetic drops in specific circumstances only when there is no perforation and no immediate antibiotic ⁶.
Waiting on a young infant. Under 6 months, the calculus changes, and that is a same-day call.

When the eardrum bursts

A perforation sounds worse than it usually is. The pressure behind the eardrum releases, the pain often drops sharply, and pus or bloody fluid drains from the ear.

Most perforations from an infection heal without surgery. Until they do, keep the ear dry, do not insert anything, and do not swim. See a clinician the same day to confirm what happened and to decide whether antibiotics are needed, because a draining ear changes the treatment choice ³.

When infections keep coming back

“Recurrent” has a definition, and it is not “two this winter”: three or more separate, documented episodes in six months, or four in twelve months with at least one in the past six ³.

For that group, the options are not more antibiotics. The AAP recommends against prophylactic antibiotics, because the resistance cost outweighs the benefit ³. Tympanostomy tubes, small vents placed in the eardrum, are an option for recurrent infection and for persistent fluid with hearing loss ³⁴.

The risk factors that are actually modifiable are narrow. A systematic review classified pacifier use and daycare attendance as established, with lack of breastfeeding, siblings, and passive smoke exposure as probable ⁸. That is a useful list because it tells you what not to blame: teething, drafts, and water in the ear are not on it.

Prevention that has evidence behind it

Pneumococcal conjugate vaccine. In the Cochrane review, PCV7 reduced all-cause ear infections by about 6% in low-risk infants and pneumococcal ear infections by 20% to 25% ⁹. That is a real but modest effect, and it is one reason the vaccine is worth the visit.
Annual influenza vaccine. The AAP recommends it for children, partly because flu predisposes to ear infection ³.
Breastfeeding. Exclusive breastfeeding for at least six months is recommended by the AAP for this and other reasons ³.
A smoke-free house and car. Tobacco smoke is one of the few exposures you can remove entirely ³.
Pacifier restriction. In a Finnish trial, clinics that counseled parents to limit pacifier use to sleep time saw 29% fewer ear infections than control clinics. Children who did not use a pacifier continuously had 33% fewer episodes ¹⁰.

When to call

What you see	Why it matters
Baby under 6 months with fever or suspected ear infection	Same-day assessment; this age group is outside the observation guidelines
Swelling, redness, or tenderness behind the ear, or the ear pushed forward	Possible mastoiditis; urgent ⁷
Stiff neck, severe headache, sleepiness, confusion, or a seizure	Possible meningitis; emergency
Drooping on one side of the face	Possible facial nerve involvement; urgent
Fever with a new rash that does not fade when pressed	Emergency, whatever the ears look like
Worsening at any point, or no improvement after 48–72 hours of antibiotics	Reassessment, and possibly a different diagnosis ³
No improvement after 3 days without antibiotics	Reassessment ⁶
Three or more infections in six months	Discuss recurrence and referral ³

The first four rows are the reason “watch and wait” has boundaries. They are rare, and they are the situations where hours matter.

Write down what you saw

An ear infection is a short illness with a long tail, and the tail is what brings families back: a baby who is better but not hearing well, a fever that returns on day four, a third course of antibiotics in a winter.

The details that help a clinician are simple: when the symptoms started, the highest temperature and when it was taken, whether the pain was one ear or both, whether anything drained, what was prescribed and whether it was finished, and how sleep and feeding changed.

Our symptom tracker keeps those entries next to feeding and sleep, and the doctor visit tracker holds the questions you want to ask, so a 10-minute appointment starts with a timeline instead of a guess. If you are still working out whether it is the ear or the teeth, the teething vs ear infection vs cold guide takes the same evidence-first approach to the differential. And if logging every dose is starting to feel like a second job, tracking without data anxiety is the counterweight.

Frequently asked questions

Do all ear infections need antibiotics?

No. Many children, especially over 2 with mild pain in one ear, do well with pain relief and a clear plan to be seen again if things do not improve in 48 to 72 hours ³⁶. Younger children, severe symptoms, both ears, and a discharging ear push the decision toward antibiotics.

How long does a baby ear infection last?

Pain usually improves within three days and can take up to a week ⁶. Fluid behind the eardrum can last weeks to months, with about 75% gone by three months ⁴. Hearing can stay muffled until the fluid clears.

Can I wait and see if my baby has an ear infection?

Only if a clinician has examined the ear and agreed to that plan. Observation requires a reliable follow-up route and a 48 to 72 hour limit ³. It is not appropriate for a baby under 6 months, a child who looks very unwell, or a child with swelling behind the ear.

Is an ear infection contagious?

The infection in the middle ear is not. The cold that set it up is, which is why ear infections cluster in families and daycare rooms ⁸.

Will it damage my baby’s hearing?

Usually not. The common hearing effect is temporary muffling from fluid, and it resolves as the fluid drains ⁴. Persistent fluid for three months or more, or fluid in a child with developmental concerns, needs a hearing assessment and a specialist conversation.

Why is it worse at night?

Lying flat changes pressure in the middle ear, and there are fewer distractions. Pain at night is a common feature and not a sign that the infection has suddenly become more dangerous ¹¹.

Disclaimer: This article is for informational purposes only and does not constitute medical advice. Ear infections can look like other conditions, and young infants need prompt assessment. Always consult your pediatrician about your child’s symptoms, treatment, and follow-up, and seek emergency care for swelling behind the ear, neck stiffness, or a baby who is hard to wake.

Teele DW, Klein JO, Rosner B, “Epidemiology of otitis media during the first seven years of life in children in greater Boston: a prospective, cohort study,” Journal of Infectious Diseases, July 1989. Source ↩ ↩²
Venekamp RP, Sanders SL, Glasziou PP, Rovers MM, “Antibiotics for acute otitis media in children,” Cochrane Database of Systematic Reviews, November 2023. Source ↩ ↩² ↩³
Lieberthal AS, Carroll AE, Chonmaitree T, et al., “The diagnosis and management of acute otitis media,” Pediatrics, March 2013. Source ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸
Rosenfeld RM, Shin JJ, Schwartz SR, et al., “Clinical practice guideline: otitis media with effusion (update),” Otolaryngology–Head and Neck Surgery, February 2016. Source ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶
NHS, “Ear infections,” reviewed January 2025. Source ↩
NICE, “Otitis media (acute): antimicrobial prescribing,” guideline NG91, 2018, updated March 2022. Source ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸
Smolinski NE, Djabali EJ, Al-Bahou J, Pomputius A, Antonelli PJ, Winterstein AG, “Antibiotic treatment to prevent pediatric acute otitis media infectious complications: a meta-analysis,” PLOS ONE, June 2024. Source ↩ ↩²
Lubianca Neto JF, Hemb L, Silva DB, “Systematic literature review of modifiable risk factors for recurrent acute otitis media in childhood,” Jornal de Pediatria, March–April 2006. Source ↩ ↩²
de Sévaux JL, Venekamp RP, Lutje V, et al., “Pneumococcal conjugate vaccines for preventing acute otitis media in children,” Cochrane Database of Systematic Reviews, November 2020. Source ↩
Niemelä M, Pihakari O, Pokka T, Uhari M, “Pacifier as a risk factor for acute otitis media: a randomized, controlled trial of parental counseling,” Pediatrics, September 2000. Source ↩
HealthyChildren.org (AAP), “Ear infections in children.” Source ↩