How Would A “Facebook” Medical Records System Work?

March 26, 2012 in Misc by RangelMD

Last time I went over a few reasons why the current electronic medical records (EMR) industry continues to belch out software that is isolationist and proprietary.  Current EMRs don’t communicate with each other or much outside of the facility or health care system that they serve. Vendors and engineers have literally modeled EMRs to be an exact replacement for paper charts without taking advantage of modern technology beyond that of a fancy word processor.  It’s analogous to transportation technology 100 years ago. The first autos looked like carriages without the horses and very early attempts at flight usually involved a mechanism that flapped like a bird or some ridiculous outfit with wings and feathers.  Modern EMR design has advanced very little beyond being a hard drive replacement for the filing cabinet despite the fact that we have the capability to go well beyond the current technology.

So what does this have to do with Facebook? The important thing about Facebook is that it represents what can be done with “cloud computing” on a massive and efficient scale. Cloud computing means removing data from a local computer or data drive and moving it to a centralized location on the internet where it can be universally accessed. I.e. nobody has to go to a special library or coffee shop in order to access Facebook and nobody has Facebook downloaded and running on their home computer. All that one usually needs for cloud computing is a standard computer and an internet connection thus  minimizing up front costs while allowing simple access.

It is my belief that the current isolationist EMR system is too broken to be fixed in order to create a system where every EMR is capable of communicating with every other EMR. Literally thousands of separate software programs would have to undergo major and expensive revisions in order to meet universal communication standards. Even coming up with industry standards for communication between thousands of different systems would be a massive nightmare (and possibly one of the reasons why it was not addressed in The Health Information Technology for Economic and Clinical Health Act (HITECH) Act of 2009). It would be like trying to get everyone on the planet to agree to a common form of communication (besides soccer). It’s a near impossible task that has been tried in the past with almost zero success.

Instead, what is needed is the standardization of a common computer code or database language for the storage and organization of health care information much like HTML is the universal computer language for internet browsers. Once a standard database language is in place then different EMRs essentially become browsers.  Their selling points then become based on how effectively and clearly they present the information to the user plus whatever additional bells and whistles the user prefers.  And since the free flow of medical information should take priority then why not base these new EMRs on the best communication system ever invented; the internet? Freed of their proprietary shackles, it would no longer be important for electronic medical records to be stored in the physical location of any one practice or hospital. Providers would only need a computer (desktop, laptop, tablet, or smart phone) with an internet connection in order to access a patient’s universal chart from . . . . the “cloud”.

However, there should be several key differences between a social network and a universal electronic health records system. Unlike Facebook, a universal electronic health record (U-EHR) should not be centralized nor dependent upon any one company for development and storage in much the same way that HTML is an industry standard and no one company owns it or the internet. So who is going to store these U-EHRs? The short answer is anyone. Any company that has servers that meet industry standards for safety, efficiency, reliability, and security can store these records.  And who pays for this? Any number of business models can be developed. Like Facebook, servers can come with ads. Or different hospitals and providers can contract with different server companies to provide access to the U-EHR database in much the same way that they contract with an ISP to provide internet access.

Wait a minute. If different companies provide storage of U-EHRs then won’t each version of an individual record be different? Again, communication and flexibility are assets of this system. It should be possible for individual records to compare their current state with records on other servers and update itself if it finds a copy that is more current with recent documentation. This would be similar to how different Usenet servers currently update each other as new posts are added to any one server. This should be done behind the scenes and automatically. illogical conflicts between different copies (i.e. the patient being admitted to different hospitals on the same date) would be tagged to allow users to decide what is the correct or more pertinent information.  Ideally, each record would be like a Wikipedia entry with successive users constantly updating the information. Except that, like current EMRs, older entries and records (office visit notes, lab results, hospital admission records, etc.) would be locked after being electronically signed and changes to THOSE records will only be made with an addendum note.

Such a cloud system has the possibility of allowing for “push” technology for updating U-EHRs. Push technology means actively forwarding information to an application without that application having to call for it (pull) such as sending you email without you having to check it. The push in the case of a U-EHR system would actually be from user to server. Once a new document is entered or changed in the U-EHR that change is communicated or pushed to all known EMR servers and copies of the record. Again, this would be done behind the scenes as different servers receive this new information and records check each other for accuracy.

Such a paradigm shift in how we store and retrieve health care information is going to have a lot of people crying heresy. The idea of decentralizing health care information is going to be alien and threatening to many people. Even though many EMRs currently have the capability for remote access to records over the internet from home or office, many providers are going to oppose such a system because they will feel a loss of control over the records of their patients. They will claim that cloud storage and retrieval of records is not secure or reliable. What happens if the server crashes and it loses all my patient’s records? What happens if the system is hacked and all the records are stolen? What happens if my internet connection goes down? All of these are valid concerns but not absolute contraindications to a cloud based electronic medical records system.

There is no reason to believe that medical information stored in a cloud system would be any less secure than a pile of paper charts crammed into the back closet of an office or an obsolete server running an EMR in a poorly ventilated space.  A single theft, fire, flood, or hard drive failure can take down an entire office EMR but a decentralized U-EHR system would be largely impervious to the destruction of any one server in the highly unlikely event that total disaster ever occurred. And the likelihood of losing your internet connection is about as likely as losing your electricity. The few areas of the country that lack reliable internet access will likely match those areas that lack electricity within a few years.

Security is a bigger concern. Placing very sensitive information on the internet makes this information potentially accessible to anyone on the planet with a computer and a connection. Then again, any EMR with an outside connection for remote access is theoretically vulnerable to attack and professional internet storage providers tend to have far more sophisticated security than Dr. So-and-So running his EMR on a Windows Me server under his desk.  I would think that the loss of one’s bank account would be more of a concern than the loss of one’s medical records but this has not stopped millions of Americans from converting to online banking. In my view, the benefits of online health information far exceeds the risks.

The concerns of individual providers however, will pale in comparison to how much resistance to a cloud system will come from the larger health care industry and EMR vendors in particular. In theory, EMR clients accessing cloud based medical records would be far less expensive (if not free) than current EMR systems which can cost tens of thousands for individual practices or tens of millions for hospital based systems. And incompatible older EMRs will have to be completely redone or scrapped. If your giant hospital conglomerate just spent millions on an old style isolationist EMR just to capture some of the bounty earmarked in the HITECH act, there is going to be a lot of complaining.

Sadly, a universal electronic health records system will not work without substantial participation by the vast majority of health care providers, health care facilities, laboratories, pharmacies, and imaging facilities to ensure that the records are accurate and up to date. The Centers for Medicare and Medicaid Services (CMS) has the power to impose mandates (usually unfunded) and can require all providers and facilities that accept Medicare or Medicaid to participate in such a U-EHR system. But they won’t. CMS would rather mess around with pointless and unproven quality initiatives (read: additional pointless paperwork) than do something that has a significant potential to improve health care safety and quality.

Please share.