Overall, we reviewed four pieces on this study. All the others (CNN, LA Times, NPR) got four star reviews.
We appreciate that journalists sometimes want to bring their own perspective to a story, and we think this can provide valuable context when the reporter approaches the issue carefully. In this blog post about the use of antibiotics to treat children’s ear infections, however, the writer prioritizes her own experience and conclusions while ignoring important evidence and failing to seek out the perspective of any real expert. And so the story provides a lopsided and overly simplified account of an issue that demands more balanced coverage and careful consideration of benefits and harms.
Read the CNN blog post for what we think is a fair comparison – blog to blog, head to head.
This blog piece suggests that antibiotics are more effective than placebos for treating middle ear infections in infants and toddlers — a conclusion supported by two new randomized controlled trials reported in the respected New England Journal of Medicine. While useful to know, this information isn’t enough to tell us how all children with ear infections should be treated. To make these kinds of treatment choices, it’s also important to consider how much more effective antibiotics are than no treatment, and we should be aware that the benefits might depend on how confident the doctor is in the diagnosis (viral colds and other problems often look like ear infections, but won’t benefit from treatment). We also need to weigh the downsides of treatment, which include adverse effects such as diarrhea as well as the potential for increased risk of future infections with antibiotic resistant strains of bacteria. After considering all of these factors, different parents may come to different decisions about whether they want to treat their child immediately with an antiobiotic or reserve treatment for a brief period while waiting to see if he gets better on his own. Reporters don’t do parents any favors when they encourage them to make knee jerk decisions about a child’s care, as this blog did.
Although the cost of antibiotics to treat an ear infection may not be a significant issue for most families, we think costs always should be addressed. We found an estimate on Drugstore.com of $76 for a 10 day course of treatment. There, that wasn’t difficult, was it?
Insufficient data here. The blog says antibiotics are “more effective than placebos in relieving ear-infection symptoms.” But the question is how much more effective are they and does this difference outweigh the downsides of treatment? The story could have helped readers answer this question by providing some data on the absolute rates of improvement reported in the these studies. Had it done so, readers might have observed that in one of the studies, rates of symptom resolution were 80% in the antibiotic group after 7 days compared with 74% in the placebo group. This might have opened the door to a discussion of whether this 6% difference is worth the potential risks of antibiotic therapy. Or, the story could have pointed out in the second study, only 18.6% of antibiotic children failed to get better by day 8, compared to 44.9% of the placebo group — a much bigger benefit for antibiotics. This may have prompted an exploration of the differences between the studies (for one thing, the second study was conducted in Finland, whereas the first was done in Pittsburgh) and consideration of which study may be more representative of the benfits readers can expect for their children.
The author’s reliance on an unrepresentative personal anecdote about her 17-year-old daughter also deserves comment. The story suggests that the study data from infants and toddlers somehow vindicate the author’s decision to procure antibiotics for her daughter. This is an extremely tenuous connection, made worse by the closing suggestion that parents are always justified in asking for antibiotics without thought to creating “superbugs.” Careful consideration of benefits and harms is the cornerstone of good health care decision-making.
To its credit, this blog mentions some of the most common side effects associated antibiotic use such as diarrhea and eczema, and it mentions antibiotic resistance as a problem. However, it failed to put these harms into terms that understandable and useful to readers. Clearly, data about the frequency of adverse effects would help readers weigh these downsides against potential benefits. In addition, while the story says that overuse of antibiotics might be “creating superbugs that may soon conquer the universe” (a characterization that we understand is meant to be humorous), the story should have explained that antibiotic overuse may have real implications for children who get treated unnecessarily. For example, there is some evidence that the use of antibiotics can lead to colonization of the child’s nasal passages with resistant bacteria which can cause future infections that are more difficult to treat. Parents need this kind of information before deciding whether to use drugs on an infection that will probably clear up on its own anyway in most children — and can be effectively treated after a few days with antibiotics if it doesn’t.
Among several deficits in this blog’s characterization of the evidence, the most glaring is its failure to qualify what kind of children were included in the studies being discussed. Both of the trials involved children that were by and large between 6 months and 2 years of age and who doctors were absolutely certain had ear infections. Contrary to the story’s assertions, these are children which current U.S. treatment guidelines say should be treated immediately with antiobiotics. The story’s claim that these studies contradict current practice is simply not true. The “watchful waiting” option mentioned in the story is recommended only for older children (>2 years), or for young children when the doctor can’t be sure if there is an ear infection or not. Young children with an uncertain diagnosis remain a challenging group of kids to treat and there are no clear cut answers provided by these studies. In the everyday world, doctors may have a tougher time identifying ear infections than in these studies because they have less time to get a screaming infant or toddler to sit still for an ear exam. Doctors may also be less experienced than these experts and have a tougher time distinguishing ear infections from other kinds of infections — such as viral colds — that can cause similar symptoms. So, the benefits of antibiotics are still unclear for kids when there is uncertainty about the diagnosis, because kids with other types of infections probably won’t benefit from treatment but will be exposed to potential harms. And these studies tell us very little about the value of antibiotics for treating children older than two (or the author’s 17-year old daughter for that matter), and so the story’s blanket assertion that “antibiotics are in order” and “work better” for ear infections in all children is unfounded and is contradicted by considerable evidence.
Although the blog’s unjustified dismissal of watchful waiting verges on disease-mongering by making ear infections sound worse then they are, this is an issue we’ll deal with elsewhere in the review. The blog didn’t otherwise exaggerate the impact of ear infections.
The story didn’t solicit a comment from any independent expert about the findings. This is a considerable deficit. With a complicated issue such as this, expert feedback is critical to highlighting the nuances in these studies and how they might impact decision-making. That the story takes on an authoritative tone when disseminating incomplete information exacerbates the failure in this area.
Although the story does note that more than half of children got better in one of the studies even though they didn’t receive antibiotics, the story doesn’t give a fair hearing to “watchful waiting” or explain why it is recommended as an option for some children with ear infections. According to data presented in these guidelines, antibiotics shorten ear infections by one day in 5% to 14% of children, but the majority of children will get better without them. Antibiotics also lead to side effects in 5% to 10% of children, infrequent serious side effects, and may contribute to antibacterial resistance. Provided with this full information, different parents may make different choices about whether to treat their children immediately with antibiotics for an ear infection.
CNN, by comparison, ended its piece with a strong take-home message about the shared decision-making conversation parents should now have with their pediatricians.
The availability of antibiotics isn’t in question, so this is not applicable.
As discussed above under the Evidence criterion, the blog fails to adequately describe what is novel about the research and what it means for patients and doctors. The studies confirmed existing guidance in the U.S. to treat young children with antibiotics when there is a certain ear infection. It doesn’t contradict current U.S. guidelines, as the story suggests.
Since the story includes no outside perspective on the research, we can’t be sure to what extent it may have relied on a press release. We’ll call it not applicable.