Friday, November 14, 2008

Sleep on it.......Nawhhh


That's the advice I get......often.....Sometimes I take it, sometimes I don't........

My friend has pointed this out several times.......In fact in my last post, my generalizations lead some people who I wasn't even talking about to get upset. I am sorry about that. But it is no big surprise, the Sherpa has managed to piss off enough people to play out Carly Simon's song...........

You know the one......

You're so vain, you probably think this song is about you

You're so vain, I'll bet you think this song is about you

Dont you? dont you?

But in this case it is a blog post......and the unnamed source is revealed........a little company on the outskirts......To be clear.......it is not about a man......and it is not about someone named Hsu or a company named 23andMe. It is about a company whose marketing tools include fooling the consumer into thinking family history catches less than a SNP scan........I will do what Carly Simon did.....indicate that a letter in the answer is "E"


But I am right and maintain, properly gathered family history has been studied for decades.....GWAS/SNP testing......less than 5 years......


This is why you can't algorithm out genetic counselors, someone needs to track down and verify these histories, and with the geneticists too busy to think straight, someone needs to do it.......until now....it turns out there are many people working on creating social networking of family medical histories.......brilliant idea.....it would absolutely remove the burden of healthcare workers actually doing it....if each patient enters their own history and links to other family members......we then have a self-report of medical history tree.....which is way more accurate than self reported family history....This is the type of thinking that will lead to earlier diagnosis.....through a well studied mechanism.....You may be too late for that one, but let me give you another.....

Rather than invest all that time working on algorithms to pump out reports on barely valid data, we should have more people in silicon valley working on real tools to move personalized medicine forward......yeah I said it.......want to save healthcare? Find a way to auto medical code thus removing billions of dollars spent of healthcare coders each year. What's a coder? Let me explain..........


To get paid by insurance a doctor has to tell the company 4 things.

1. Who the patient is.....the policy number

2. Who the Doctor is.....the National Provider Identification number

3. What the patient has.......the International Statistical Classification of Diseases and Related Health Problems Number (ICD-9)

4. What was done on or to or for the patient...the Current Procedural Terminology Code


With these four things in hand an insurer can approve or reject a claim......pay what is asked or pay what the insurer thinks is fair......


What's the problem? People go to school to learn how to do coding for doctors....why? If coded right, the doctor can collect more money....If coded wrong the doctor makes less......

Doctor's hate menial tasks like this, so they spend their hard earned money on coders, but I am about to let the Valley in on a secret......sorry coders, but your days are numbered....


Coders make between 35 and 50 k per year. Not a lot per person, but if you come up with a program to do the coding, you could save a doctor that much per year.......you could end up saving a hopsital, which often has 10 of these guys on staff 500,000k per year....


There are currently a little under 100,000 of these people at work in the healthcare system.....100,000 at 50,000 USD each.....do you see where I am getting......that is precisely 5 billion dollars A YEAR!!!!!


Betcha not many people know that!!! This could be a 2.5 billion dollar a year company, instantaneously, not ever needing to create a market......Like DTC genetic testing has to do....


I have tech compatriots who say that the DTC genetic testing is the kernel of thought that tech feels is just the start of the healthcare revolution......


I think, man.....why didn't they put Navi or 23andMe's millions into a coding company......

That would be the start of the true healthcare revolution.


....Here's a hint, We ARE an expert society, we outsource things to experts, when we can, we try to replace experts, but that rarely can happen. Instead we should focusing on automating menial tasks....like coding...This empowerment movement would come quicker and the billions in GDP would come sloughing off if we can do one thing.....find a way to remove everyone from healthcare that doesn't talk directly to a patient for the purpose of diagnosing and treating them........If you do that, the GDP spent on healthcare would drop to 8% guaranteed. Here's a hint, don't pick on a field of experts first....all you will get is push back. Which is why ASHG has opened another front on DTC.....


Go take the tech money and pick on the operators, administrators, human relations, custodial, marketing, public relations, coders and everyone else that spawned from providers not wanting to do it......rather than decide to push slightly scientific tests on a scientific field that demands the highest levels of accuracy......Early diagnosis happens more often when a doctor has the time to spend with a patient. Not from a panel of SNPs......most of the time....


The Sherpa Says: True, empowerment can be a good thing....but if you empower healthcare practitioners with a 5 billion dollar bonus for spending 5 more minutes with each patient instead of worrying about coding, then we really could realize personalized medicine.........

7 comments:

Anonymous said...

I disagree with doing away with coders. That is taking jobs away from people who need the jobs. With over 500,000+ jobs lost in the last two months alone, we should not be talking about trying to do away with coders. Instead, we need to talk about preventive measures to help decrease health expenses (not just the cost of healthcare). If you ask me, more money needs to be put into clinical research for preventive care research projects to help learn how to detect diseases at an early state for people who don't have health insurance or are underinsured.

Also, you are missing a few situations where a family history examination can miss cases. I'm not saying SNP testing is better then having a family history done.
If I have some free time this weekend, I'll post a response here about some of the situations you are missing.

Steve Murphy MD said...

Thanks for your thoughts. You are absolutely dead wrong about keeping jobs that could be automated for the sake of the economy......

Healthcare costs are killing the economy and 5 billion is a lot of money....

The best solution is to cut costs on things that won't effect healthcare outcomes while spending in places that improve outcomes....

Coders don't improve outcomes and cost money.....thus wasted expense....in the end. Provided they can be automated...if they can't then it is a justified expense.

-Steve

Anonymous said...

Then outsource all x-rays, MRI's, and other radiology tests so we don't have to pay $50,000+ a year to radiology techs and several hundred thousands of dollars a year to pay for radiologists....we can just pay some person off the street $7 an hour to take the patient from the room to the exam room and have that person hit the "outsource" button after the exam is done.

I disagree with what you are trying to say. We can agree to disagree.

We could almost easily just have every non trauma patient go through an assembly line (x-ray, MRI, echo, lab tests, etc) and just outsource all of the information and have it reported back to a tech and the tech gives the medication or diagnosis to the patient. We wouldn't need all of these doctors, lol.

Andrew said...

*cough* insiders will never cut their own industry *cough*

Also, I think it's a great idea to do away with coders. The problem is that it's the insurance companies who decide how to reimburse and why. I have every reason to believe that they'll try to argue that it's "unethical" to have a machine code insurance, and if that doesn't work, they'll change the standards to make it as hard as possible for the software to work.

"We could almost easily just have every non trauma patient go through an assembly line (x-ray, MRI, echo, lab tests, etc) and just outsource all of the information and have it reported back to a tech and the tech gives the medication or diagnosis to the patient."

We can only hope someday... (assuming it worked!)

Andrew said...

But yes, I very much agree that an automated coding service would be of great benefit to health.

Steve Murphy MD said...

I guess we disagree.....it takes 9 years to learn how to be a
doctor.....it takes 6 months to learn how to code....I hope you
understand the difference....

RT's see patients and help care for them....coders don't....that's
another difference...

BTW.....if all it took was an algorithm of MRIs, XRAYs, CT Scans and labs to make diagnosis, they would have replaced doctors a long time ago.....

-Steve

Anonymous said...

I was being sarcastic Steve (about the assembly line stuff). I wasn't being serious. If you want to make changes, you need to work with the insurance companies.