Why doctors give out antibiotics you don't need

medical examiner: Health and medicine explained.

The Pink-Bubble-Gum-Flavored Dilemma

Why doctors give out antibiotics you don't need.

By Zachary Meisel
Posted Wednesday, May 21, 2008, at 3:21 PM ET

ER doctors must sometimes fend off patients' demands for antibiotics

While working a busy night shift in the ER recently, I evaluated a 13-month-old girl. On her chart, the triage nurse had written: "Infant with fever and runny nose. Mother here for antibiotics." The baby was fussy but probably more tired than uncomfortable. Between her squirms, she cooed and smiled at me. Her anxious and upset mother, however, was in far worse shape, repeatedly sticking a rubber bulb syringe up her infant's nostrils in a futile attempt to suck out an endless stream of snot. The mom was also really mad: She had been waiting for more than three hours for a doctor to see her daughter. Now she wanted antibiotics: specifically, a prescription for bubble-gum-flavored amoxicillin.

By my assessment, the child was not acutely ill: She'd had a low-grade fever for two days, her mother said, and a mild cough, but she had clear lungs and appeared well-hydrated. Her eardrum may have had some fluid behind it but wasn't red or bulging. Just as the baby was trying to put my stethoscope in her mouth, paramedics pushed through the ambulance doors with a patient who was having an acute stroke. I had to decide right then if I was going to give this mother the antibiotics she wanted, even though I thought her daughter probably didn't need them.

The profligate prescription of antibiotics—for children and adults with upper respiratory infections, sinus infections, and even middle-ear infections—is a problem because most of these illnesses are caused by viruses, not bacteria, which are what conventional antibiotics attack. Of more concern is the direct connection between antibiotic use and the emergence of drug-resistant "superbugs": As the medicine eliminates germs that are sensitive to it, drug-resistant mutant strains prosper. The result is a major public-health problem. Antibiotic-resistant infections such as methicillin-resistant Staphylococcus aureus may cause more deaths in the United States than AIDS does.

In the doctor's office or the ER, it's hard to tell the difference between bacterial and viral infections, and so doctors are tempted to prescribe antibiotics whenever they're unsure. That's especially true when doctors think that patients expect to take the medicine home, according to a recent study. Investigators interviewed patients with respiratory infections who went to the ER in 10 hospitals affiliated with medical schools, asking whether the patients expected to receive antibiotics and about whether they were satisfied with the care they received when they were discharged. The researchers also asked physicians why they prescribed antibiotics. The main conclusion was that doctors were significantly more likely to prescribe if they believed that patients expected them to—but did a lousy job predicting which patients those actually were. And the patients most satisfied with their care were the ones who left the ER with a better understanding of their condition, antibiotics or no antibiotics. The take-home message for doctors like me: Spend an extra five minutes talking to your patients about their medical problems, and you can send them away happy and without unnecessary medicine.

So once doctors absorb the result of this study and similar investigations, will they write fewer prescriptions? I bet not. To give out fewer antibiotics, the doctors will have to believe that their patients won't benefit from them. If you look closely at the ER study, 73 percent of the patients who received antibiotics for acute bronchitis had illnesses that were either deemed by their doctors to have likely been caused by a bacteria or to have origins that were in that gray toss-up area between a bacteria and a virus. If the doctors were right, and these were bacterial infections, they would, in fact, warrant antibiotics. Also, in many of these cases, the doctors gave other persuasive reasons for choosing antibiotics, including "ill appearance of the patient" and "concern about follow-up."

In my ER world, these factors, if intangible, are understood to be really important in helping us decide how to treat patients. The real dilemma of antibiotic prescriptions is that the most serious consequence for writing them unnecessarily is not a risk to the individual patient but the emergence of the superbugs that pose a risk to public health in general.

Nowhere is this tension between individual care and public health greater than in the ER. Office-based cultures for bacterial infections, which take days to turn around, are not feasible in what we call "the trenches." And because follow-up can never be assured, it's hard to follow recommendations such as those of the American Academy of Pediatrics, which advocates "watch and wait" for 48 to 72 hours for children with middle-ear infections rather than an immediate dose of antibiotics. If we overprescribe antibiotics in the ER, that's because in the trenches the care of one patient often trumps the care of the public. Maybe that's myopic, but there you have it. And it is why efforts to reduce antibiotic use by giving out more information about resistant infections or teaching doctors how to manage patient expectations may ultimately fall flat.

In the end, I did not prescribe antibiotics for the 13-month-old baby. Instead, I took the time to explain thoroughly why I didn't think she needed them (while my colleague took care of the stroke patient). But no matter what that study says, that mother left in a huff— highly dissatisfied, I can assure you. I'm not sure what I'll do the next time I see a similar case. Perhaps I will refuse to write the prescription again, notching another victory for public health. But, for all I know, something intangible will be different: Perhaps the kid just won't look right, or maybe the mother or father will seem too disorganized to be relied on to return if the kid worsens. And that may persuade me to send them home with a bottle of pink-bubble-gum-flavored amoxicillin. It's likely that the fussy kid and his parents won't sleep any better that night. But I will.

Zachary Meisel is an assistant professor of emergency medicine at the University of Pennsylvania. His research interests include injury prevention, patient safety, and prehospital care.