Given today’s trends, I’m betting most of us would agree that EMRs need to evolve from transaction-based to intelligence based systems. They need to do better leveraging “big data,” make context-based care recommendations and support smart processes. John likes to call them “Smart EMR,” but what would such an EMR look like?
In a recent issue of Hospitals & Health Networks, Dr. John Glaser, Ph.D., lays out a long– but useful– explanation as to why EMRs are stiffly focused on transactions such as documenting a visit or writing a prescription. (Very short summary: That’s just where they are coming from historically.) Then he offers a take on the “intelligence-based EMR” and what it will take to get there.
Glaser, CEO of the Health Services Business for Siemens Healthcare, was formerly VP and CIO for Partners HealthCare, so he’s got both the vendor and the care provider view, which I think proves very useful for this discussion.
In his article, he argues that the next-gen EMR needs to offer the following:
- foundational sets of templates, guidelines and order sets that reflect the best evidence or established best practice;
- a process-management infrastructure that supports basic transaction checking such as drug-drug interactions, as well as asynchronous alerting like panic lab reporting and process monitoring and guidance;
- team-based care support such as shared work lists, as well as tools for patient engagement and health information exchange;
- novel decision aids like predictive models that can tell us if a particular patient is likely to be readmitted because he or she is fragile or has a substandard social situation at home that may negatively impact healing;
- context-aware order sets and documentation templates that guide the physician and help infer what types of orders should be placed and what types of documentation should be done
- intelligent displays of data, intelligent correction and identification of data, and extraction of structure by going through free text and pulling out quality measures or problems that were not previously in a patient’s problem list, for example.
The question is, are these functions science fiction (i.e. many years away from being standard) or just an evolutionary leap from today’s systems? What are you seeing out there?
Calls for intelligence-based EHRs, intelligent EHRs or smart EHRs are based on well-founded frustration with many current products. Not only am I sympathetic, I too have called for smarter EHRs.
Several years ago I wrote…
Question: Do We Need Smarter Users or Smarter User Interfaces?
Answer: Smarter User Interfaces.
…and suggested how to get to smarter EHRs.
http://chuckwebster.com/2010/07/ehr-workflow/intuitive-vs-intuitable-emrs-ehrs-and-clinical-groupware-do-we-need-smarter-users-or-smarter-user-interfaces-3
However, I’ll offer a devil’s advocate word of caution. (Whip on red cape and horns, grab trident.)
“Smart EHRs”, “intelligence-based EHRs” and “intelligent EHRs” allude (in some cases, in my opinion) to Artificial Intelligence. Don’t fall into the strong AI or AI-complete trap. Strong-AI-complete information systems are equal in intelligence to humans.
http://en.wikipedia.org/wiki/AI-complete
If we only had EHRs as smart as humans, which (who?) could perceive, pattern match, understand context, reason, communicate in natural language, and cooperatively plan, act and improvise. That would solve all our problems! Not gonna happen anytime soon. And even if it did, what other unintended consequences might afflict us. EHRs might go on strike!
OK! That’s enough of that! (Whip off red cape and horns, set aside trident.)
Some EHRs *are* smarter than others, particularly about process and language. There are lots of interesting smart technologies coming down the pipeline. In the near-term, how do we find these smarter-than-the-average EHRs? How do we speed up that pipeline? What are some practical technologies for making EHRs smarter?
Two smart technologies come to mind, both mentioned in “From the Transaction-Based EHR to the Intelligence-Based EHR.”
1. “Process-management infrastructure”
2. “Extraction of structure by going through free text”
The former is is based on process-aware, workflow management, and business process management technology.
The latter is based on natural language processing, an engineering application of the science of computational linguistics.
I’ve written about both…
Clinical NLP at 2012 NAACL Human Language Technology Conference
http://chuckwebster.com/2012/06/natural-language-processing/clinical-nlp-at-2012-naacl-human-language-technology-conference
EMRs and EHRs Need to Solve “The BPM Problem”: Why Not Use BPM to Help Do So?
http://chuckwebster.com/2010/05/ehr-workflow/emrs-ehrs-and-clinical-groupware-need-to-solve-the-bpm-problem-why-not-use-bpm-to-help-do-so
… otherwise I wouldn’t weigh in here.
The way forward is to educate (your blog, thank you), highlight smarter EHRs (separate wheat from chaff), and solicit help from other industries already using smart-tech.
In the meantime, let’s not cement into place dumb EHRs.
Bottom line, it’s all about all that data that’s out there in today’s EHRs (and elsewhere), most of it “frozen”, pretty much inaccessible to docs. If we’re going to fix health care, we’ll need to do exactly as Glaser says, among other things, get to the “intelligent displays of data, intelligent correction and identification of data.” That’s where the money is. And it’s happening now. No, really.
Hi Barry,
I checked out your home page…
http://www.forwardhealthgroup.com
…those are astonishingly cool animations regarding population health concepts!
Seriously, what your “frozen data” and my “frozen workflow” have in common is the “brittleness” of much current HIT software.
http://chuckwebster.com/2009/03/ehr-workflow/litmus-test-for-detecting-frozen-ehr-workflow
Anyway, cool animations!
Cheers
–Chuck
The problems with most EMR documentation is born out of the silly rules insurance companies command that doctors use in order to get paid. I actually write parts of my notes only to prevent insurance companies from denying payment. Those parts have nothing to do with medicine and are a complete waste of time. They are merely there to guarantee financial security (read: “to not get ripped off by insurance company crooks.”)
[…] their own, they sidestep a very important issue: when will EMRs evolve from transaction-based to intelligence-based systems? After all, an intelligence-based EMR can do more to improve healthcare in context than […]