Maybe the ICD-10 codes do have some unnecessary granularity
I'm a pretty strong advocate of data granularity and diagnosis code specificity. But maybe the ICD-10 code set has gone a code too far.
David Pittman, of MedPage Today, thinks that "assault by drowning while in bathtub" may just be a code too far. It's not that someone who has been killed in a bathtub doesn't deserve to be a statistic. That could be a thing. We don't know until we count it.
But he got an assist from the Skeptical Scalpel who wonders how many times a physician needs to encounter drowning victims. ICD-10 codes include:
- X92.0XXA Assault by drowning and submersion while in bathtub, initial encounter
- X92.0XXD Assault by drowning and submersion while in bathtub, subsequent encounter
- X92.0XXS Assault by drowning and submersion while in bathtub, sequela
And I have to laugh in agreement. Unless we're expecting a zombie apocalypse, clinicians shouldn't need "X92.0XXD". Or is there another reason I can't see?
I mean another reason other than the "A," "D" and "S" are available for every ICD-10 code. But it's not like clinicians and medical coders need to remember three times as many codes.
I hope no one is spending any time memorizing "X92.0XXD."
Why we want to count assaults by drowning and submersion while in bathtub
Consider the Wonk Blog post, "100 Americans die of drug overdoses each day. How do we stop that?" That's shocking. We ought to do something about that.
Or should we?
I'm not taking a stand here. But if you want to argue the point, you need to understand the scope of the problem. If it is a problem. To do that, you need to know how many people are dying of drug overdoses. Then that number can be used to support or reject the premise that we need to do something about fatal drug overdoses.
So should we do more to prevent homicides of people in bathtubs?
I don't know. How may people die each day of assaults by drowning and submersion while in bathtubs?
These kinds of numbers help us define problems and help us focus on where to spend our resources. We can't solve every problem so we need the data to help us prioritize.
No matter where you stand on the issue, you're going to want real numbers so you don't have to trust someone else's assertion that it matters.