4 ms·
> There are tens of thousands of procedure and drug codes (things that are done or given) and tens of thousands of diagnostic codes (reasons justifying the proc
by some_furry 4y ago
> There are tens of thousands of procedure and drug codes (things that are done or given) and tens of thousands of diagnostic codes (reasons justifying the procedure), creating a space well into the quadrillions of possible routine combinations.
10^5 * 10^5 = 10^10 which is tens of billions, not quadrillions
That said, a combinatorial space can be simply represented as a tuple of two columns (10^5, 10^5) instead of having to map every possible (including nonsensical) permutation.
Increase for three or four dimensions.
- dokein 4y agoPatients have easily 20-30 codes per one hospital visit / one bill and ordering matters so sadly quadrillions may be an underestimate.
- some_furry 4y agoI'm unfortunately aware, but you don't need every possible permutation to have a unique identifier. That's like trying to enumerate all 2^256 or so Bitcoin private keys. It's a stupid way to approach the problem. Enumerate atomic elements, not molecules. Only the most useful or common combinations need naming.
- psychlops 4y agoSo this is assuming that someone would give chemotherapy drugs to a flu patient. The realistic combinations would be much smaller.