Saturday, March 8, 2014

So, you want to build a website?

In order to explain the problems with Health Information Technology, let's say you want to build a website and you must do it how Health Information Systems do it.

You code up your website with your favorite text editor and code up your website. You already know HTML so that isn't an issue. But now, you need to host it, upload it, etc.

First things first. You need to write a web server. Wait what? Well, HIT doesn't give you anything for free. So, you must now read the IETF standard for TCP/IP. You spend hours reading over it. Bizarre details which no human should ever care about. After hours, you finally have a grasp on how to serve up simple HTML web pages to anyone making a request to your web server.

But, TCP/IP isn't enough, you need to read the HTTP standard, because depending on which HTTP request is made, you need to do something different.

But, now you've run into a problem. In HIT, every system communicates differently. So, each web browser makes requests differently. Not just with TCP/IP, but with HTTP. So, you start with Chrome since it is the top browser and figure out all the nuances. You must then go to Safari, IE, so on and so forth. You'll spend probably half a year just figuring this out.

Six months later, you have your web server up and running.

Now you need to upload your HTML pages. Well, that isn't easy. No FTP clients exist in this world. So, you no must read the IETF standard on FTP and develop your own FTP client. Since you know TCP/IP already, you may not need to read that spec again. So, you spend a month or so developing your own FTP client and finally get it working.

You have now uploaded your website!

But we have a new problem. No DNS servers exist. So, you go and get the IETF spec on DNS and pour over that. Another month of reading and developing, you have finally developed your own DNS server and people can finally start accessing your website.

Does that all sound crazy? Yes.

But that is how Health IT "works" right now.

How do we fix this?

What HIT misses completely is the fact that the rest of the software world revolves around APIs and SDKs. You don't have to reinvent the wheel every time you want to build a piece of software.

Healthcare Leadership, who have built the system the way it is can't seem to come to terms with the fact this isn't how the rest of the software world works.

Yes, there are major concerns over HIPAA, safety, etc. But, neither of those have anything to do with building out SDKs and APIs.

If I want to communicate with EPIC, I shouldn't be required to read through 1000 pages of proprietary HL7 standards (Which were only made "free" in the last year) and hope it works. If I then want to point that solution Cerner, I should have confidence it works.

The way HIT is now, even pointing to another instance of EPIC probably wouldn't even work.

This is entirely unacceptable.

We will not lower costs, improve care the way things are now. This is a tough issue to tackle, but it won't be solved with more standards and certifications. It will be solved by doing things how the rest of software does it.

Heath IT leadership still doesn't get it

I found this comment by Farzad Mostashari to be quite interesting - and telling of what is going on with Health Information Technology at the moment.

And, therein is the problem. Brian Ahier gets what developers and startups needs. APIs and SDKs. For the life if me, I cannot grasp why the former National Coordinator for Health Information Technology cannot see what is important to startups.

Certifications for an EHR?

Why on Earth would a startup care about that? If you are crazy enough and well funded enough to try and start an EHR from scratch in 2014, perhaps.

But, there is so much going on outside of the EHR world (lets call them EHR work arounds) to solve issues that policy has created within EHRs - that one would be remiss to read mundane details about what people sitting in offices in DC think EHR software needs.

HIT is bigger than EHR vendors.

I do find this all disappointing. This comes a week after Mostashari was at HIMSS and got to see all the technology startups are working on. How they are doing things differently.

If the guy that used to "run HIT" can't see - who can on that level?

Tuesday, March 4, 2014

Watch out, horse and carriage makers

At HIMSS this week, my business partner and I had lunch with a gentleman who, I guess you could say does things "the old way" which we do in a technology forward way. He has a successful business in medicine that probably will double in the next year.

Our goal was to talk with him about hopefully augmenting what they do and adding us on as a service.

Well, things didn't go so well.

He explained to use we'd go out of business, that we'd never make it, that we must make serious changes to our idea if we want to succeed.

All quite interesting since we are growing rapidly.

Simply warning to all you old vendors of health information and health information technology. 

There are a lot of us startups entering your space and they plan to steam roll you.

This isn't the 1890's where it took 30 years for cars to replace the horse and carriage.