Search This Blog


Sunday, May 10, 2009


“Meaningful Use”: a pivotal definition for new-wave medical records systems

by e-Patient Dave on May 10, 2009

I’ve struggled with what to say about this subject for two weeks, because I want to “get it right” but it’s vast. So I’m giving up any hope of being comprehensive, and I’m just going to say what little I know, and what I think, and let any discussion happen from there.

The issue

We (the US, and probably the world) are at a pivotal moment in healthcare, a turning point. The direction we choose now can have a lasting effect on how well healthcare works in the coming years – even for the next decade, I suspect. So as we consider this, I suggest that we think not in 2010 terms, but 2020 terms.

The issue at hand is the definition of “meaningful use.” You’ve probably heard of the $19B in the US “ARRA” economic stimulus package that’s set aside to encourage adoption of electronic medical records systems. (A Google search for “HIT stimulus ARRA billion” produces 25,700 hits.) Medical practices that have EMRs and put them to “meaningful use” will get higher reimbursement from the government.

When the stimulus package was passed by Congress, the term “meaningful use” was intentionally left to be clarified later. See David Kibbe’s commenthere for three items that were specified. And here we are: the work is now underway to define and clarify that term. Many meetings are happening in Washington to get this done.

I believe, and many basically agree, that meaningful use should be defined to include full patient access to their data in EMR systems. Here’s why.

First, let’s look at the world for which we’re designing: 2020.

2020: The world for which we’re planning

The systems we design today will be in use a long time from now, so I suggest we look at the world as it will be in 2020, and how we’ll be using these systems then. What does that world look like?

  • We will all be ten years older. You will be ten years older; your children will be ten years old; your parents will be ten years older.
  • The internet will be ten years more advanced. If you think back to the dial-up world of ten years ago (1999), when browsers were just five years old, there’s been a huge amount of progress … and, significantly for this discussion, a lot of it’s related to having more access to our personal data online: banking, plane reservations, shopping, and more. Think how that will change in the next decade.
  • Handheld computers (smartphones) will be ten years more advanced. (More advanced? Heck, the iPhone was only introduced 28 months ago.)
  • Connected e-health devices will be out of their infancy: WiFi blood pressure monitors, bathroom scales, glucose monitors, you name it. It’s fairly certain that by then we’ll be able to use cheap devices that send routine data to some central storage place, where smart software (your choice of smart software) can send out alarms or reminders, your care team can view it … and you should be able to view it, too. And make notes on it.
  • Doctors will be ten years older. Fast Company magazine’s Doctor of the Future Jay Parkinson will be middle aged (Ha!). Our oldest doctors will be retired (or dead). Today’s youngest doctors will be well into their careers and owning practices; a new generation of providers will be practicing who were born after (1994). They have never known a world without household email. They expecteverything to be online.

In this world, what kind of data access do we want? We need to build that into our design choices now. Why?

The product development cycle

It takes time to plan, design, code, test, install, and train on systems. And once that’s done, the decisions that are made will be with us for a long time – until vendors decide to go through that whole process again.

Sure, a lot of code gets written and updated fast. But if I’ve learned one thing about healthcare IT since I got in the e-patient game, it’s that healthcare systems are big, complex, and slow to change. These are not things that get re-engineered frequently.

Docs/nurses etc: how old are the systems you use today? Some are new, but many of you are using systems designed in 2000, right?

And here’s where regulatory requirements come in.

The impact of regulatory requirements

Some regulatory requirements profoundly affect product design. If a law is passed saying you can get a tax credit for a computer where the “Q” key is colored orance, vendors will offer that, because customers will buy it. Similarly, if a law requires that a system have a particular feature, people will only buy that, so vendors put it in the product plan, right from the get-go.

That’s the case with HIPAA: many systems are required by the government to be HIPAA compliant. So vendors design accordingly.

And so it is today with the definition of “meaningful use.” Providers will only receive stimulus money if the system they choose meets the definition of meaningful use. So the definition we choose today can have a profound impact on what you can do with your data far into the future.

My principles

  • Patient is a first-person word. Your time will come: someday it will be you, your child, your mother, your spouse on that hospital bed or at that roadside being tended by an EMT. The way to think about this is in the first person: when that time comes, what should be possible with “my data,” not “patients’ data.”
  • It’s my data. It’s my life that’s at stake. I have a right to seek the best care in the world, and if that means exporting a copy of my data from your system and taking it somewhere else, I have a right to do that..
  • Corollary: No more proprietary captive data. (Edited 11:20am EDT) We must put an end to the era where a system provider thinks the data they collect is their property. Lives are at stake. Whose data is it, anyway?
    Vendors must adapt to a world where they earn their margins by creating on-going value, not by holding data captive. This includes images (CT scans, MRIs, etc, at full diagnostic resolution) as well as lab results and everything else.
    Added 11:40am EDT: My opinion is that “whose data is it” applies to providers as well as system vendors. If you haven’t yet, considerwhat happened to Web guru Doc Searls last year when he couldn’t share his scan data with another MD.
  • Enable participatory medicine – doctor-patient collaboration. Make it possible for each party to view the same data. (Ideally, I’d like to enable collaboration tools such as online discussion of my medical records – but that’s beyond the scope of this post.)
  • Let each constituency say what works for them. Patients shouldn’t say what doctors need, and doctors shouldn’t mandate how patients should and shouldn’t describe things.
    Warning: experts on both sides should be able to comment on / warn the other about apparent errors. Docs must be able to say “Whoops, you overlooked this,” and patients must be able to say “Whoops, you overlooked this.” (See related discussion of “who will vet the vetters” in “Medpedia: who gets to say what info is reliable?”)


These are my personal recommendations, not necessarily those of or the Society for Participatory Medicine. (Hey you out there - what do you think? Participate in the comments! You count!)

  • Grant patients full access to their medical records.
    Exception: the best minds I know have said doctors need the ability to store private notes, and I can grant that. But I want full access to all my data: all the numbers, all the test results, all the radiology reports, everything.
    Added 11:45am: Among other things, the records might contain errors, and we can help clean them up.
  • Let me export my data and take it elsewhere. I have a right to seek the best care anywhere, even if it means going somewhere else. Systems must be designed to allow exporting data (and, of course, importing data). I know there are lots of details to be worked out. Let’s just start with this principle: it’s my data.
  • Enable the most able. Don’t dumb down the data. Allow different levels of viewing – simple, medium, expert. (Hint: let each group participate in creating the language, and let users vote it up or down.)

This is our golden moment. It’s not fundamentally hard to give patients access to their data, not any more than it’s hard for banks to give us access to our banking information. It took banks two or three generations to get it right, but the technology exists, and now it’s widespread. Let’s encourage EMR vendors to get it right on the first try, by defining meaningful use to include giving us full access to all our medical data.

A great organization called the Markle Foundation has recently issued a paper that lays out their thoughts on this issue. I like it, and so do the rest of the people on who’ve voiced an opinion in the back room. More on that Sunday.

No comments:

Post a Comment