This lovely article last week, Why Are Medical Bills Full of Mistakes, had plenty of gems. But the answer to the question pretty much lies here:
Common billing errors include being billed for procedures that were cancelled, being billed twice for the same thing, under different names, being billed for a more complex version of what was actually done and being billed for more time than was actually spent. How do these happen? In general, the errors aren’t deliberate fraud.
Frequently staff in the billing office do not talk to doctors but produce bills based on what the doctors write. Doctors don’t document what happened right away if they are busy and so by the time they do make a note, details are often fuzzy. In the case of billing for canceled procedures, the only paper trail available to billers may be the order for the test, and the cancellation may have been communicated by voice, on the fly. When doctors do bill for themselves, it may be difficult to find the correct code, so, in a hurry, we just settle on the first one that resembles what we did. Most of us are not interested in getting better at billing because we hate it. We weren’t trained to do it, and it takes us away from patient care.
It’s estimated that 80% of medical bills have errors. But if we all agree that electronic medical record systems’ were designed primarily and strategically to communicate doctor and hospital bills to insurance companies, you quickly see that your medical records are riddled with errors. Again, it’s not fraud, it’s just juked data so doctors and hospitals can get paid the most money for what they actually did. But when you’re trying to get paid the most, well, it’s like every fishing story that’s ever been told, the fish you caught that nobody saw just keeps getting bigger and bigger.
But the bottom line is almost all structurally coded medical data is fundamentally juked, inflated, and flawed and does not reflect your real life health conditions. And those folks interested in Big Data in healthcare need to understand that much of what they’re dealing with is billing data, not actually real life health data. So can Big Data save healthcare? How can it when the data is so far from real life? It might be able to provide some insight in an alternate, inflated universe of healthcare, but that universe is far from reality.