Data alone will not save you

“It’s all about the data” was a popular phrase when I came up through the IT ranks. That phrase, though, was used to justify having proprietary (copyrighted by vendors) data storage systems that captured information on individual computers, in databases, and as things grew, in data warehouses.

Communication was point-to-point based on connections between those who created, those who stored and those who consumed data. They were often all within a single system. And we were so happy be computerized that we were just fine with those limitations.

Events

That was well and good in the old world but falls apart in the world in which we find ourselves. Today’s world could be be described as, “It’s all about the event.” An event is a much more complicated thing and relies on data being present or absent. Events need to be watched for or discovered depending on how much can be known in advance. Now for the tricky part: Events depend on data of many types, from many places, with different timing and patterns. It can be requested or simply anticipated. But above all, is must be real-time to be meaningful.

And just to make it more fun, there needs to be a plan for what to do when the moment comes. Events need to have context so that the right information can be acted on by the right person at the right time.

Healthcare

There are few places where this is more important than in healthcare. Medical monitoring is is quickly becoming just one example in the realm of healthcare and complex event processing. Data alone allows us to see vital signs from a patient and understand how their body reacts to their changing environment throughout the day. However…

…what about the 99% of the time that they’re not in a care facility? Did they schedule a physical therapy appointment? Did they show up? Did they pick up their prescriptions? Did they report their blood sugar levels? What were those levels?

By monitoring complex data in combination, real-time, patterns can be discerned that allow medical intervention to take place proactively, avoiding crisis-type intervention that is unscheduled and very costly. We can begin to address what’s important before it becomes urgent, or worse, emergent. Faced with a shortage of doctors, this is the efficient way to use expensive, constrained resources.

Once we start down this road, there are more possibilities than we understand today and there are organizations already moving in this direction.

To get to where this is possible, there needs to be big-picture approach to information and its sources and flows. We need to set healthcare data free to be interpreted and used in ways we haven’t even thought of yet. Data alone will not save you.

Advertisements

Tags: ,

Categories: Data Analytics / Big Data, Healthcare

Author:Chris Taylor

Reimagining the way work is done through big data, analytics, and event processing. There's no end to what we can change and improve. I wear myself out...

Subscribe to the blog

Subscribe and receive an email when new articles are published

6 Comments on “Data alone will not save you”

  1. March 5, 2012 at 1:24 am #

    You are contrasting data-oriented and event-oriented IT approaches. I understand you need a clear theme for this post. Still, all the features you say are important in healthcare IT are crucially dependent on technology that is able to create and guarantee data quality. There is no real point in opposing data-oriented and event-oriented: as you imply, the two need to complement each other. I think what your post is really about is data-driven IT that leads to vendor lock-in versus other data-driven IT.

    • March 29, 2012 at 6:07 pm #

      @Rob: Yes. This article came from a conversation where i made a naive comment, “It’s all about the data.” I was immediately reminded that data by itself isn’t effective unless you can listen for specific things (and things in combination) and then have a way to act upon it.

  2. March 29, 2012 at 1:33 pm #

    Whenever patient “data” changes, a potential “event” occurs. These events can trigger automated workflows, effected by BPM workflow engines executing process definitions. Steps in these workflows may occur automatically (in traditional BPM systems) or generate check lists for use by humans: yes, do it; cancel it, here’s why; remind me later; forward to another user or role; and so on (in some case management systems).

    From the blog post: Clinical Intelligence, Complex Event Processing and Process Mining in Process-Aware EMR / EHR BPM Systems ( http://ehr.bz/x1 )

    “Complex Event Processing, Event-Driven Architectures

    An event is a change in state, such as a patient who gains weight and moves from obese to morbidly obese state categories. A complex patient event is a pattern of detected events amidst a patient event stream (such as moving from obese to morbidly obese combined with being diabetic). Complex event processing, implemented in conjunction with a BPM system, provides means to react to events in real-, or almost real-, time. In the case of a process-aware information systems such as EMR / EHR workflow management (or business process management) systems, patient events can drive automated clinical workflows (such pushing action items to worklists) via workflow engines executing process definitions”

    • March 29, 2012 at 6:06 pm #

      @Charles: Exactly. The biggest challenge I see is that just as we are at a point to have data-influenced, event-driven healthcare, we’re paying for the adoption of EMR’s that silo data and make event detection and response more, not less, difficult. To detect, you need data in motion, and to respond requires BPM capabilities that are mostly still foreign to healthcare.

  3. Ron Webb
    May 11, 2012 at 7:10 am #

    Chris,
    First, I have to admit I’m a data geek. Aside of the occasional WSJ story that touches on healthcare events (a la the Target predicting pregnant customers and the angry father of a sixteen-year old girl), I don’t see a lot of healthcare examples of big data improving efficiency and effectiveness of operations.
    Seeing the results gained by organizations in other industries when they marry data with process is very encouraging. Chris, what do you thing the top barriers are to healthcare adopting some of those best practices?

  4. May 11, 2012 at 3:01 pm #

    Top barriers:
    1) Implementation of EMR’s to collect stimulus money but not to use data in any better way. We have data now, but it isn’t in the places where it can make the difference in outcomes or cost.
    2) Resistance to change by providers far more comfortable with Fee for Service that rewards readmissions, extra tests, etc.
    3) Resistance by physicians who maybe don’t want their results to be compared to others. They might not like what they see, even though every other industry relies on benchmarking.
    4) Defensive medicine that takes a very expensive and often unproven “just in case” approach rather than trusting the proven paths.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: