Making Data Matter

On one of my many road trips recently, I pulled over at a rest area for a brief stretch and caught a glimpse of something I’d never seen before. It almost felt as if I was getting to see the proverbial “man behind the curtain.”  Sitting there in the parking lot next to me was a vehicle with an enormous and complicated camera mounted to the top – the Google maps  Streetview car.  Here it sat, the very low tech way that Google is creating high tech data -putting together a comprehensive map, neighborhood by neighborhood, seemingly one frame at a time by driving across country snapping pictures from the top of this simply modified car.  Even though the data they went out to capture was time consuming and in small bits, their method seemed to be working.

Jumping back in my car and having a bit more time to ponder, I began to think about how we are collecting data in the world of healthcare.  As the industry continues to edge closer and closer to delivery and payment models based on value, alignment, and care coordination, the mad scramble to create and capture truly meaningful data is gaining speed.  The approach to find the “holy grail” of data for most appears to be integration, integration, integration.  Let’s make sure every physician, hospital, payer, and even patient are all on (or have access to) the same system with the same gigantic bucket of data.  If we can just get everyone to push all of their data into the same place, surely we will be able to divine the answers we seek. 

Google’s approach appears to be a bit different. Rather than trying to force everyone to push data into a common place, they are going out and pulling it in, bit by bit and coming up with a very comprehensive, very usable to tool that provides direction and gives meaningful information.  Our approach in healthcare to getting everyone to push their data into a common place would be like Google asking everyone in the country to please send a picture of their home to Google headquarters, hoping we all use the same size, format, color, and resolution. 

How might this model look in healthcare?  Each time a patient refills their blood pressure meds, what if their blood pressure was recorded at the pharmacy and sent directly back to the physician?  Could this allow us to begin to track the outcomes of individual medications in a more meaningful, real time way?  What if our focus was not a common electronic medical record structure that is primarily physician focused, but on creating a common portal for all patients to share data with all physicians?  Would we actually be getting better and more meaningful clinical data from the field as opposed to capturing well coded data that is designed, at least in part, to maximize our reimbursement?

I understand the privacy and operational challenges that something like this would create. It would truly force us to question the model we are currently using, but isn’t that the point?  If we continue to place all of our focus on building the perfect, fully integrated information system, we may be missing the opportunity to capture smaller, discrete pieces of information that may not initially provide us with the “big picture,” but will certainly give us useful direction along the way.

Trackbacks (0) Links to blogs that reference this article Trackback URL
Comments (7) Read through and enter the discussion with the form at the end
Ben Miller - April 6, 2012 10:57 AM

Mark - great post as always. While there are privacy and confidentiality issues to consider, I think the problem with doing what you propose is that there are so many vendors out there competing for healthcare providers to use their product that they have little incentive to work together. The outcome of this is that we have different types of documentation, different places data is stored and therefore difficulty extracting these data in a meaningful way. Third party vendors who could aggregate data from disparate locations are hamstrung by the challenge of creating a standardized template for the data to reside. In essence, capitalism has hurt some of our attempts to integrate information. We must question the current model as it does perpetuate fragmentation in healthcare. So even applying Google's model means that someone somewhere has the "master key" to make sense of all the data as it will not look the same. We will get there, no doubt, but currently many of the vendors putting out great products are not rushing to do standardize their data, where its stored, etc. I think it's going to be all about disruptive innovation moving forward. Thanks again for the great post!

Theresa - April 6, 2012 11:55 AM

This is a great analogy- thanks for sharing! This links back directly to Chris Rangel's recent post: Why Facebook should be a template for electronic medical records Indeed the government and health industries together have failed spectacularly to generate and enforce a timely vision of interchange. Now what we have are poorly motivated vendors offering work-arounds and costly personalized conduits without addressing the big picture. Hopefully, this will be able to shift, but in the meantime, docs and patients are stuck in an inefficient hybrid system.

Dave Chase (@chasedave) - April 6, 2012 4:36 PM

Considering that towns like Portland, Oregon all have health systems on Epic and they can't communicate with each other, even for a congenital optimist like myself, it's easy to be pessimistic about these issues getting solved. I think your approach makes sense. Rather than the be all, end all HIE, it may be a more grassroots "HIE" that bit by bit gets better than trying to solve things top down.

Also questioning integrating billing-centric systems together is wise. A different approach may bear more fruit and be the disruptive innovation that healthcare has long needed.

Paulo Machado @pjmachado - April 10, 2012 10:43 AM

Great post Mark!
I like your suggestion. As long as US Healthcare is for profit, we will have to determine who owns the data & who can monetize it. Right now it is a free for all. I am cautiously optimistic that we will get there but it will require a great deal of 'discomfort' during the transition...

Barry Solomon - April 12, 2012 8:44 PM

Your comments are right on target. It perfectly describes our current situation; trying to force everything into one box, our box, while each competitor has their own box. The providers don't know which way to go for their own practices and the patient is lost in the struggle. No information is shared therefore care not improved.

Ron D Carlson - April 14, 2012 12:44 PM

I am pleased to have just discovered this stimulating blog. Thanks Mark.

Well, where to begin. As a COO/hospital administrator, I can appreciate all of the "whys" proffered as reasons we can not accomplish this data mission.

I find that Mark's Google analogy is most helpful in this regard in a key way. Wise men such as Dr. Einstein remind us that you can never solve a problem from where the problem lies. You must first move in your own framework.

The Google reference is such an example of moving first.

Allow me to run with two more. In another life, I worked in aerospace engineering. Yup, worked on Shuttle, worked on the Viking Martian lander too. You may have heard that the total computer capacity in the Shuttle was less than what is found in today's Smartphone, far less. It should never have worked frankly.

The point being, those missions were considered "impossible" yet accomplished through the same decision stance as exhibited in Apollo 13, namely, failure is not an option.

In other words, "impossible is what we do".

Perhaps you will indulge me as a new contributor here with another analogy. Take the case of 10,000 disparate data collection points using 20 year old technology feeding information hourly into a central repository which then feeds the data outbound through numerous commercial data portals into information available to us all daily. I speak of the weather information we all receive daily. What we view on local TV, etc all is sourced from the US Weather Service which originally gathered, collated and distributed that data throughout the country.

All right, I hear the howling - not applicable. Really? Sure,it's not PHI, sure there are paid data collectors feeding into a government hub with a public face. I've done a tour in government service. Think it's easy using 20 year old equipment to collect and disseminate data in a standard given a bureaucracy? Actually, it's impossible yet it's done daily.

The computational capacity and speed designed into a single MRI speaks to a technological expertise light years beyond Space Shuttle or the weather service.

Which brings this muse back to the health care data sharing questions posed here.

I don't think the root problem with health information is technology frankly, not when we can site an MRI in almost every hospital across the country. I don't think it's who owns the health data or the PHI protection or any of the others usual suspects.

I think it's a problem of willingness. We have not decided that this is what we are going to make happen. We have not decided to do the impossible.

Love 'em or hate 'em, when Google decided to have photos of every street everywhere, that was impossible, right? Well, look what happened by getting started.

I appreciate Mark's fresh approach to change the mindset and get started.

Ron D Carlson
CKO in COO's clothing

Ray Paulk - June 24, 2012 8:13 AM

What a great blog/article. I found it to be very incisive and informational.

Post A Comment / Question Use this form to add a comment to this entry.

Remember personal info?