I.  Is the security technology of electronic healthcare records good enough?

The security problem with electronic health records isn't encryption or firewalls, it is nothing that can be fixed with better technology.

The real issue issue is the people who already have access to the records: the 10000 people who work for a university hospital system-- or the 2000 that just got laid off, looking to make a fast buck. 

The real risk is if you are in a divorce proceeding and your spouse wants to use your mental health against you; or your boss/employee wants to know if you're gay; or your insignificant other wants to know etc etc.  So someone makes a call to that guy they know who works there...

How could I not be tempted to look up my nurse-girlfriend's medical record... just to see?  How could my brother not be tempted to ask me to look up his girlfriend at my hospital?  How could janitor #7 not be tempted to ask secretary #895 to look up... for the purpose of...?

All this goes on already, and so far the records are only accessible within the system.  Imagine what will happen when they are nationalized.

The analogy for electronic health records is not financial records.

If you want a working analogy for the world after electronic health records, it's facebook.

II.

I hope I don't need to point out that there's a billion-- 19 billion, actually- dollars behind EHR; about $10k/year for five years goes to the doctors, and the rest, I assume, goes to lobbyists.  No, I haven't done the math, it's an estimate. 

Most of the content in the record is information bias.  It seems important because it exists.  Do you need to know your fiancee's driving record before you finally pop the question?  Of course not.  But if some guy says, "I'll tell it to you for free--" would you say no?  Information bias.

All medical information is redundant.  That means each doctor has to find out for himself.  Sure, we use them as a shortcut, but we would never act on them.  To do so would be negligence.  In other words: trust, but verify.


III. 

Here's a hilarious story about EHR:

At the same time (that Farrah Fawcett was fighting to prove someone at UCLA was selling her medical records to the media) UCLA repeatedly asked her to donate money to the hospital for a foundation to be set up in her name.

The university went so far as to give her a prewritten letter that she could sign and fill in a dollar amount for the foundation, documents show. It also created an official-looking proposed announcement that said, "Ms. Farrah Fawcett has established a fund in the Division of Digestive Diseases with the expansive goal of facilitating prevention and diagnosis in gastrointestinal cancers."

See?  They think that having a wing named after her will make up for the fact that she's dying and that one of their employees sold the records.

But a hospital is a business, they're going to get paid somehow.  Anyhow.

(For a million reasons and this one, you do not want nationalized electronic health records.)

IV.  EHRs Are Really About Pretending To Reduce Costs Without Actually Reducing Costs: That's Why The AMA Is Behind Them 

The point of EHR is, of course, that it will "reduce costs."  Oh my God.

The cost of healthcare has nothing at all to do with the value of healthcare. 

If you cut the reimbursement for tests, procedures and visits, and the cost of medications, to 20% of their current cost-- you will not reduce the total spent on healthcare even one cent. 

Got that?  No?  I will repeat it, for emphasis:

If you cut the reimbursement for tests, procedures and visits, and the cost of medications, to 20% of their current cost-- you will not reduce the total spent on healthcare even one cent.

The simple reason for this is basic economics: supply and demand.  The reason people can't see this is that they have misidentified the supply and the demand.  They think demand for healthcare comes from the patient, and the supply is allocated by insurers via doctors.  Wrong.  Doctors are both the supply and the demand.

Do you think that if reimbursement goes down, doctors will say, "ok, I guess I'm in a lower tax bracket?"  They will simply ramp up their business.  See more patients, order more tests, whatever.  Maybe they'll buy the test machines for their office and bill the insurer for the services (e.g. an x-ray machine.)

They're not unethical, they're not trying to mismanage patients, they are just trying to survive.  Why should a Goldman Sachs trader make three times more than him?   They'll do whatever is necessary to help the patient, and whatever necessary to get paid.

Healthcare is a business.  The business of healthcare is business.

V.  What's the solution?

The solution is nuanced.

Part 1 is that you have to remove the incentive to increase the utilization of healthcare:   doctors have to be salaried.  I can't believe I just said that, but it is true.

The second part is the tricky one: you have to change the culture of medicine, the mindset of doctors.  You have to make the natural reflex of doctors not to try to make more money, but to use less healthcare.  That's nearly impossible, which means that it will fail in this generation.

So, President Obama, take the long term view: doctors become salaried starting 2020.

Face it-- doctors today are smart enough to find a way to beat the system.  I am smart enough to beat the system, and it is impossible to expect me not to try to beat the system because I believe that I am not acting against the interests of society in doing it, even though I am. 

But eventually I will die. The idea of a 2020 start date is that it will change the type of person who decides to become a doctor. Like academic researchers: no one goes into it for the money.  I didn't go into it for the money, either, but when I got here there were opportunities to make money.  How could I not take them?  All of those considerations have to be removed.  Then you will get a new kind of person trying to become a new kind of doctor.

Yes, this new kind will have problems of his own (apathy, institutional bias, etc.)  But it will cost less.  That's what you wanted, right?