Previous: Healthcare Triage News #2: Pot, Foul Language, and Pain
Next: Sex, Cars, and Food: Healthcare Triage News #3



View count:75,274
Last sync:2023-10-26 17:00
You've probably heard of "evidence-based medicine". It's the idea that we practice based on research and data. There's another way of practicing called "eminence-based medicine". It's the idea that we listen to the person who's been around the longest or who has somehow managed to be labelled the expert.
It used to be that such a person would periodically get to write a review article in some journal, and that would be how everyone learned what to do in medicine. That's a problem. We've got a solution. Systematic reviews!

For those of you who want to read more or see references, look here:

John Green -- Executive Producer
Stan Muller -- Director, Producer
Aaron Carroll -- Writer
Mark Olsen -- Graphics

You've probably heard of evidence-based medicine. It's the idea that we practice based on research and data. There's another way of practicing called evidence-based medicine. It's the idea that we listen to the person who has been around the longest or has somehow managed to be labeled an expert. Used to be that such a person would periodically get to write a review article in some journal and that would be how everyone learned what to do in medicine. That's a problem. We've got a solution. That's the topic of this week's Healthcare Triage.

[Healthcare Triage opening theme]

See, the issue with a review article is that it's just an opinion, and just like op eds and actual news are kept in separate parts of a newspaper, so too are review articles and research. "But Aaron," eagle-eyed viewers are shouting, "you've cited review articles many times on Healthcare Triage. Have you been relying on opinion?" No! I've relied on systematic reviews, and those are different.

If you look at an article in a research journal, you'll see that they all have a basic structure. First up is the background. It's a review of what we know about a subject, as well as what we don't know. It sets the stage for the paper you're about to read.

Next comes the objective. It's short, and it tells you what the study you're reading wanted to do. If you've written the background well, the objective makes total sense.

Then comes the methods. That's the section where you explain how you did the study. Depending on the study in question, this can be simple or complex. But the key is that it has to be transparent enough that someone could replicate your study, 'cause good science is science that's been tried repeatedly.

Next are the results. That's where you talk about what you found. Usually, there's a table 1 that describes the group you looked at using your methods. There's also usually other key tables or figures that show what you found. The accompanying text has the rest of the details.

Finally, there's the discussion, where the authors place the results in context. What do they mean in the larger picture? What are the implications? Additionally, this is where you usually talk about the limitations of your research. All studies have them, and good authors acknowledge them and talk about how they could affect the findings.

And that's it! Systematic reviews are research because they follow this structure. They have a method section. They're reproducible. Basically, researchers doing systematic reviews set out to find all the relevant research in a field and then combine it together in one big collection. But by stipulating how they search for studies, how they determine what studies were good enough for inclusion and what they pulled from them, systematic reviews allow others to judge the merits of the work and to test its conclusions if desired.

These are the kinds of studies I bring to Healthcare Triage. They're different. One of the first pieces of research I ever did was a systematic review. It was about gastroesophageal reflux in infants.

You may have heard of gastroesophageal reflux, or GERD. It's when acid in your stomach backs up into your esophagus. It's really bad heartburn. In adults, we usually treat it with dietary changes and medications. In infants, though, it's a totally different animal. About half of all healthy infants will vomit more than twice a day. About ninety-five per cent of them completely get better without treatment. Infants vomit more often because they have an all-liquid diet. They have an immature esophageal sphincter that doesn't close off the stomach from the esophagus. They eat every few hours and they have small stomachs. Tons of infants will have symptoms of gastroesophageal reflux.

This makes parents panic. They worry that something is wrong. They agonize over whether their kids are getting enough food. When I was a resident, the hospital I worked at would construct foam wedges for these infants to sleep on. They thought that infants who were sleeping at an angle would be less likely to have milk come back up. The wedges cost about $150. I was convinced that they didn't work, though, that they cause needless worry, and that it was unfair to make people pay for them.

Other doctors advocated for thickening feeds. They thought this would make it harder for kids to vomit, but this meant moms couldn't breastfeed. It also cost money. I wasn't convinced that that worked, either.

So, I did what any hugely annoying resident would do. I conducted a systematic review. I searched the medical literature for all studies that looked for treatments for GERD that were non-drug and non-surgical. I found more than 2,500 articles that might have been on point. I then excluded any articles that weren't clinical trials or weren't on conservative therapies. That wheedled the list down to thirty-five. Then, along with two other reviewers, we went through them carefully to make sure that they were really good studies. Ten of them made the cut. All of them were randomized control trials.

Two of them were on positioning infants, i.e. wedges. One of them found that putting an infant at sixty degrees in an infant seat made reflux worse. The other found that raising the bed to thirty degrees, like with the wedge, made no difference at all.

One study looked at pacifier use. Didn't help. Some studies looked at thickening feeds with rice flour. Didn't work. And thickening feeds with carob bean gum. Didn't work. And then, one study found that carob bean gum was better than rice flour, which was strange because carob bean gum wasn't better than placebo. Changing formulas didn't work, either.

In other words, none of the conventional therapies used for reflux had supporting evidence behind them. Now, since I was a resident, you might ask, "Who cares what I say?" But this was a systematic review. It had methods. Anyone could check my work, so it was actually published in the medical literature, and it remains one of my most cited studies.

Systematic reviews are research. They're not eminence-based medicine. They're the backbone of evidence-based medicine.

But let me rant about GERD and infants for one more minute. Since I published that paper, treatment with drugs has become much more common. Today, we commonly treat infants with proton pump inhibitors, or PPIs. Between 1999 and 2004, the use of one child-friendly liquid form of PPIs increased more than sixteen-fold. This was in spite of the fact that PPIs have never been approved by the FDA for the treatment of GERD in children.

In 2009, a randomized placebo-controlled trial examining how well a PPI works for infants with symptoms of GERD was published. It found that the drug had no more of an affect than placebo. It also found that children who received the PPI had significantly more serious adverse events, including lower respiratory tract infections. In 2011, someone else published a systematic review of PPIs for GERD in children. Guess what? They don't work, either. The fight continues.

[Healthcare Triage closing theme plays]