Comment by guylhem | Hacker News Reader

Comment by guylhem | Hacker News Reader

The article is quite incomplete. The highest price is the opportunity cost of doctors going into billing instead of doing clinical work - especially for hospitals

I can say first hand that there is an extremely lucrative business-opportunity for physicians to go full time in billing - using domain knowledge and a careful study of the billing schemes to increase financial returns.

For exemple, what if you could identify the patient cases where manual recoding has the highest probable financial return, and compare individual returns to the cost of the human resources you have at hand (from medical coders to physicians) - to assign the best human resource to each case?

In a highly complex case, recoding by a physician who perfectly knows the coding and billing rules can easily double the value billed.

So for a >200k case, it can be worth having a physician spend 20 minutes carefully studying every aspect of the file, aided by specific software modelling the possible billing scenarios.

If you add some programming knowledge and statistical knowledge, all this can be easily automatized, and make even more lucrative.

A full industry is created, based on exploiting the arcane artificial rules imposed on the system - just like for taxes.

The real problem here it that it takes doctors, for which the next best alternative use would have been doing clinical work on patients.

I guess a society only gets what it pays for.

That is partially true, it doesn't have to be a physician, but it does have to be somebody with real knowledge and skill and not just a 3 Month course somewhere.

Someone I knew, would optimize the billing codes so that she could "write-off" non-covered items, and maximize patient care. Her salary was still quite low compared to other doctors in executive roles.

There is no reason that a licensed doctor is required to do what you describe. A software programmer with a one-year billing code training can do it. Doctors are rate limited by med schools, but that only applies to patient care, not support functions like billing.

There is no reason at all - no one is "requiring" doctors to do anything at all. However, the invisible hand of the market do provides incentives for doctors willing to do that job!

The reason is the difference in financial gain if for some reason (domain knowledge) a doctor can perform the same job more efficiently - which is usually the case in medical coding, where a doctor usually know better which complications are linked with which diseases, and which procedures must have been performed and thus, if they don't appear, have usually been done but for some reason the coding didn't happen.

That's just 2 quick examples - a physician can "dig" in the file for supporting evidence to back the claim. Some are quite good at it, and with the numbers at hand, tiny differences in productivity make huge financial gains.

Say that worker 1, anyone with good training, can make 2x returns, while worker 2, a doctor, can make 2.2x returns.

Depending on the volume, you hire both and give the common files to worker 1, and big profile cases to worker 2.

Each worker is usually paid at the marginal value of labor.

When the marginal value of billing improvements become greater than the marginal value of clinical care, labor do arbitrage and moves to the higher paying sector - in this case, doctors are going for clinical practice to billing.

Someone with basic training - or even algorithm - can be efficient, and provide a first line of coding. But domain knowledge still rules and brings in big bucks.

Trust me - my job is to bill patients/insurance as much as we are legally allowed to, using any mean necessary, in a way that will stand in court if challenged.

We're long past data-mining for potential missing codes - we now run simulations to isolate which coding is the most probable and the most lucrative, and divert the file to the right person according to the potential financial gains. Any loophole in the coding rules is exploited big time !

A quick example - several years ago there was no consequence if you "forgot" codes, however simulations figured out in 10% of the case this resulted in greater gains.

Consequently, after a matching legal advice, codes were forgotten 10% of the time - of course, the right times, until forgetting code was disincentivized thought legal changes.

It would be lovely if the market provided incentives to these same experts to spot overly vigorous coding; to spot patterns associated with fraud.

Maybe that's happening?

Vigorous coding is just as legal as paying as little taxes as allowed by the tax code. People are paid to make sure everything done strictly follows the letter of the law, with every t crossed and every i dotted.

OTOH, fraud is easy to catch - if the case repartition suddenly changes, or of there are some mistakes, something must be happening.

But the present incentives are for making rock solid, court-proof claim cases.

It induces all kind of interesting changes- like making sure 5 different health professionals all considered the patient was in end-of-life to qualify for a more lucrative palliative care.

As long as there are the 5 signatures on the paper (and in a 2nd line of defence, proof they saw the patient - like signatures, agendas, etc) it's as good as printing money.

One would have to be really really dumb or overtly greedy to engage in fraud.

Reply on news.ycombinator.com