The Golden Age of crappy medical information is only getting started
When my patients ask questions about health stuff that they read on the internet, many of them sheepishly admit that they “know they shouldn’t be looking stuff up.”
This is partly because the internet has a reputation for unreliability of medical information, and partly because, like the sentiment expressed on the mug above, doctors really condescend about this stuff.
I disagree with the idea that you shouldn’t read about your health conditions, and the internet has increasingly reliable information available on the basics.
Medicine is confusing, and being sick makes us scared. People want information about the things that are going on in their bodies.
A single doctor’s appointment is rarely sufficient to provide information for many people on their diagnosis, workup, and treatment plan.
You only get a few quick minutes with your doctor, and you’re often given a huge amount of information to absorb.
It’s impossible to remember every single part of the discussion - and many doctors do a poor/nonexistent job explaining things - so it’s only natural to seek out additional sources of information.
Unfortunately, a lot of the content out there that’s digestible to non-medical professionals is obsolete, incomplete, misleading, or oversimplified.
And some of the stuff that’s totally spot on doesn’t provide the necessary context to make the information useful.
The information asymmetry in medicine is gigantic
Whenever I speak to someone in a different area of medicine than I am, I’m consistently blown away by the degree to which our skill sets and knowledge bases barely overlap.
If you show an EKG to a head and neck surgeon, he’ll just look at the automatic machine reading to develop an interpretation, just like if you ask me to review a thyroid ultrasound, I will barely be able to identify any of the anatomy.
And that’s a knowledge gap between people who have medical degrees and who have done many years of post-medical school training!
When I make a decision about a diagnostic test or a treatment plan, there are a huge number of layers of information that I’m integrating to come to my perspective.
I’m considering the biology of the disease, the clinical trials looking into treatments, my experience taking care of other patients with the same disorder, and the way that each patient may or may not fit into the data that we have.
Very often, the data that you have doesn’t apply to the patient that you’re seeing. Or you’re making decisions with conflicting clinical data. Or there’s clinical equipoise about the right next step, which basically means that no one knows the right answer.
And then when we discuss the condition, I’m making assumptions and inferences about the level of detail that someone will benefit from while trying to answer questions and efficiently communicate about next steps.
Unfortunately, even the high quality information out there tends to be generic
Have you ever looked up a common medical condition and then tried to deduce the right next step from a reliable website?
Take a condition like atrial fibrillation, for example.
There’s really good info out there from the Mayo Clinic, the CDC, Johns Hopkins, and a ton of other places.
And while these sources tend to be really good at describing what a medical condition is, they give a huge range of different potential treatments.
If you read these websites, you’ll learn that the options for treatment include medications to slow the heart rate, medications to control the heart rhythm, medications to prevent blood clots, or potential surgical procedures to help treat the condition.
So which of these treatments should you get?
The best information out there can’t do anything other than give you a whole bunch of options without really scrutinizing which you’re most likely to benefit from.
When you see a cardiologist to discuss it, that doctor hasn’t just spent time reading and analyzing landmark trials like AFFIRM, CASTLE-AF, EAST-AFNET 4, and CABANA.
They’ve also reviewed your echocardiogram, know your CHADS2VASc score, looked at how much mitral regurgitation you have, seen the size of your left atrium, and considered your blood pressure control, alcohol consumption, and the possibility that you have undiagnosed or untreated sleep apnea.
In other words, the correct and useful information from the Cleveland Clinic doesn’t necessarily provide much insight into your particular case.
All of the official content out there is bland, watered down, and repetitive… and it’s getting easier to produce
When I read these websites, I’m struck by how similar they all are - and how insipid the information is.
Johns Hopkins has the same basic information as Sharecare, which has the same basic information as Medline, which has the same basic information as WebMD.
It’s repetitive to the point that it basically feels regurgitated and repurposed from one site to another.
So you can spend hours reading about this stuff, and end up no more educated than if you had only spent a few minutes looking at one site.
And now, the advent of easy, rapid AI content production is going to make generic, insipid information proliferate in ways that we can’t even imagine.
I played around with ChatGPT3 a bit, asking it a handful of medical questions.
You’re getting the same bland, depersonalized, superficial information that you could get on any reputable website:
Expertise is hard to develop, and more available content makes it both more important and harder to find
As the volume of information sources proliferate, it becomes harder to sort out a finite (and slowly growing) amount of wheat from a readily available, rapidly growing amount of chaff.
So my practical advice for people here is simple:
It’s totally reasonable to read about your medical conditions online
Stick with one or two reliable sources to learn the basics. Recognize that reading more has diminishing returns
There’s nothing wrong with asking your doctor why they’re recommending one treatment over another one that you read about
Keep in mind that these sites aren’t updated often when new data becomes public, so you may have an incomplete picture even when it comes to the basics
Bring a list of questions written down to the doctor’s visit ahead of time so you don’t forget to learn about what you were wondering during your reading.
I’ll readily admit that sometimes questions patients ask either make me think about a condition in a different way or make connections in my mind that I wouldn’t have made otherwise.
So your reading and questioning may actually lead to better medical care.
And now I’ll leave you with this tweet as a reminder that we should all be humble about the limits of our knowledge:
