Ear infections (or, what we like to call “acute otitis media”) are one of the staple diagnoses of pediatrics. Most kids have at least one before their third birthday.  And most pediatricians see at least one every day by 11. You would think we would always get it right. But I have a confession — we don’t. In fact, children are misdiagnosed and over-treated at an alarming rate.

There are several reasons for our poor performance:

1. Ear infections are actually pretty complicated. In 2013, the American Academy of Pediatrics (AAP) issued a practice guideline called The Diagnosis and Management of Acute Otitis Media. It’s written for an audience that should already have a solid foundation of knowledge about the topic. It’s 40 pages long, cites 275 references, and the only thing it talks about is ear infections.

2. It’s not always easy to look in a child’s ears. Some kids cooperate. Most don’t — especially when they’re not feeling well. If you’ve ever helped hold down your screaming 18-month-old so the doctor could look in his ears, you have some idea what this is like. But when you’re the one standing there holding the otoscope, trying not to jab a hole through the kid’s eardrum as he flails around, and struggling to see enough to make a diagnosis … well, it’s not much fun. Complicating all this is the fact that there’s almost always a clump of wax right in your line of sight, which must be removed (at the cost of making the child even angrier before you try to take a look).

3. It’s not always easy to interpret what you see. Many children are diagnosed with ear infections based solely on red ear drums. But that’s not good enough. Ear drums, just like your child’s face, turn red with fevers or screaming. That doesn’t mean they are infected. A good diagnosis is quite a bit more complicated (and that’s why there’s a 40-page document to discuss it). This step becomes even more difficult when the person holding the otoscope doesn’t look at a lot of ears. When children get sick on weekends or at night, their parents often take them to an urgent care facility or emergency room. If the facility happens to serve a lot of children, they may have a good chance of getting it right. If not, they tend to overcall — a lot.

(Let me be clear: I’m not bashing doctors or other providers that primarily care for adults. They certainly know far more than I ever will about the clinical problems they most commonly encounter. But if they are used to seeing adults with back pain and heart attacks, they’re probably not very comfortable looking in the ears of a screaming 2-year-old.)

4. Not every misdiagnosis is an accident. It’s disturbing but true. I remember clearly an infant that I saw in the emergency room one evening. She was 6-weeks-old and had been prescribed antibiotics for an ear infection the day before. (This case was horribly mismanaged, because an infant that young is at risk for life-threatening infections and requires a much more thorough workup to rule out more severe illnesses.) I tried to look in her ears, but the canals were so small that there’s no way anyone could have seen enough to make the diagnosis. Sadly, that wasn’t the only diagnosis that has made me doubt a physician’s integrity. If you don’t trust your doctor, find another one.

5. Even with an accurate diagnosis, most ear infections go away on their own. Really. Without antibiotics. Why? Many of them are caused by viruses (so antibiotics wouldn’t help anyway). The majority of the others can be successfully eradicated by the immune system without any help from us. In 2013, the Cochrane Collaboration published a review of 12 studies comparing antibiotics to placebo for treatment of acute otitis media. This level of evidence is as good as it gets. They found that 60 percent of patients had improved 24 hours after diagnosis, whether they were prescribed antibiotics or not.

Antibiotics helped a little on subsequent days, but we still have to treat 20 children with antibiotics to improve the symptoms of one child. This is known as the number needed to treat (NNT): How many patients we have to treat in order to help one of them. In this case, it means that, for every 20 children that get antibiotics, 19 get them unnecessarily. We have to treat 33 children to prevent one ruptured ear drum. And if we look at the long-term outcome of hearing loss, treating with antibiotics doesn’t make a bit of difference.

6. It’s not easy to explain to parents why you’re not treating an ear infection with antibiotics. But based on the evidence we have (which is better in this case than for most illnesses that we treat), for children over 6 months of age without severe symptoms, waiting to treat is a perfectly acceptable decision. It has no risk of side effects, costs less, doesn’t contribute to antibiotic resistance, doesn’t require families to make their kids take medicine and is almost as effective as ten days of antibiotics.

But why does it even matter? It’s just an ear infection, right? What’s the problem with just giving some antibiotics and moving on with life?

1. An incorrect diagnosis can obscure an actual problem. I remember admitting a child to the hospital who had undergone brain surgery a few weeks before. He had a fever, up to 104 degrees, and he was admitted to rule out infectious complications of the surgery. When I examined him, his right ear looked pretty red, but his temperature was also 103 degrees at the time. I decided to reexamine him after his temperature had come down. (Both ears looked fine.) Had I been content to call that an ear infection and send him home with antibiotics, we likely would have missed the abscess in his abdomen that was picked up when he started having pain a couple days later.

2. Antibiotics have side effects. Diarrhea is common and inconvenient. Rashes are also common, and can lead to patients reporting allergies to antibiotics that prevent us from using them when it might really matter. There are other less common — but very severe — side effects, which can include death.

3. Misdiagnosis can lead to unnecessary surgeries. The current recommendation is to consider surgical intervention (ear tubes) for children with three ear infections in 6 months or 4 in 12 months. If some of these were incorrectly diagnosed, we are putting children through unnecessary surgical procedures. While this is not–as surgeries go–a particularly risky procedure, they all carry risks. And when there is no benefit, the only acceptable risk is none at all.

4. We are breeding bacteria that we can’t kill. Antibiotic resistance is a very real problem, and it’s scary. A recent report estimates that by the year 2050, up to 10 million people per year could die because of infections by antibiotic-resistant bacteria. We have created this problem, in large part, by prescribing antibiotics when we’d do just as well without them. If you think Ebola is bad, just wait.

So, there it is. Another one of medicine’s many dirty little secrets. I think you deserve to know. And I think that, armed with the truth, you will be better prepared to make informed decisions about your child’s health. There’s no need to run to the emergency room just because she’s pulling at her ears or has a fever (except in the cases I discuss here). Find a pediatrician that you can trust and call them if you have questions about your child’s health. And if your doctor tries to explain why antibiotics aren’t necessary, listen (and be thankful for a doctor who cares enough to take the time for that conversation).

Chad Hayes is a pediatrician who blogs at his self-titled site, Chad Hayes, MD.