Short Link: http://j.mp/80VBlx
The key insight I hope to impart in this series of posts that started with 12 quotes is this: Well understood, consistently executed, adaptively resilient, and systematically improvable workflow *between* health care organizations is not possible without well understood, consistently executed, adaptively resilient, and systematically improvable workflow *within* health care organizations. Each post tries to achieve this goal in a different way.
This Week I Play With Words
My plan was to start with authoritative statements from both the primary care and workflow automation domains, and then systematically describe how the ideas they represent fit together. One way to do this is to take a quote from one domain, insert, delete, and substitute some words from the other domain, and see if the quote still makes sense (if this sounds a little flakey, hang in there). For example, “BPM suites coordinate tasks and synchronize data across existing systems,” becomes “EMR BPM suites coordinate clinical tasks and synchronize clinical data across existing systems,” and the phrase “work-in-progress” becomes “care-in-progress,” and so on.
You get the idea. Here’s quote 12 from the BPM domain:
- Papazoglou and Ribbers, “e-Business: Organizational and Technical Foundations”. J. Wiley & Sons, April 2006, quoted at S-Cube, last retrieved 11/10/09:
- “BPM suites coordinate tasks and synchronize data across existing systems. They also help coordinate human process activities, streamlining tasks, triggers, and time lines related to a business process, and assuring they are completed as defined by a process model. A BPM suite makes processes more efficient, compliant, agile, and visible by ensuring that every process step is explicitly defined, monitored over time, and optimized for maximum productivity.
- A true BPMS enables business users to:
- Model and simulate all interaction patterns between workers, systems and information sources to create shared understanding about how to optimize business processes and results.
- Coordinate and manage the handoff of work across boundaries.
- Provide real-time feedback to business managers about work-in-progress to support in-line business process adjustments.
- Monitor process outcomes to performance targets, and continuously refine and adjust process flows and rules.”
- Webster, adapted from Papazoglou and Ribbers:
- EMR BPM suites coordinate clinical tasks and synchronize clinical data across existing pediatric, pediatric subspecialty, and non-pediatric primary care EMRs. They also help coordinate clinical activities, streamlining clinical tasks, triggers, and timelines related to a care coordination process, and assuring they are completed as defined by a care coordination process model. An EHR BPM suite makes care coordination processes more efficient, agile, and visible by ensuring that every care coordination process step is explicitly defined, monitored over time, and optimized for maximum productivity.
- A true EHR BPM suite enables physicians and staff to:
- Model and simulate all interaction patterns between physicians and other clinical and non-clinical staff, systems, and information sources to create shared understanding about how to optimize care coordination processes and results. [CW: “well understood”]
- Coordinate and manage the handoff of patient care tasks within and across organizational boundaries. [CW: “consistently executed”]
- Provide real-time feedback to pediatricians and care coordinators about care-in-progress to support patient care process adjustments. [CW: “adaptively resilient”]
- Monitor care coordination outcomes to performance targets, and continuously refine and adjust care coordination process flows and rules. [CW: “systematically improvable”]
Sounds good! I’d consider such a product if I wanted to build a high-performance medical home system. Actually, I do. And you should too.
First, I emphasized “care coordination” rather than simply “patient care” or “healthcare” because I wanted to stress the relevance of workflow and BPM technology to the care coordination aspect of the health system as described in quote 3.
Second, I emboldened four bracketed phrases of my own contrivance (“well understood,” “consistently executed,” “adaptively resilient,” and “systematically improvable”) because they provide a useful checklist, so to speak, for comparison of traditional EMRs to process-aware EMRs (“process awareness” being a prerequisite for applying BPM to EMR processes).
And third, a caveat—EMR BPM doesn’t exist yet. You can’t simply buy a BPM suite and use it for EMR BPM. Right now the closest you can get to EMR BPM is an EMR workflow system. However EMR workflow systems have the process-aware foundations required for full-blown EMR BPM, that is, well understood, consistently executed, adaptively resilient, and systematically improvable EMR-mediated care processes.
Let’s Try It Again
Recall the topics of pragmatic interoperability and conversations between EMRs in a previous post? Quote 7, adapted from the original, described a conversation between the EMRs representing a community pediatrician and a pediatric subspecialist. Let’s further adapt it to a diagram of an “e-contract” (quotes 10 and 11) between two workflow management systems (circa 2000, though I was not aware of it at the time).
Adapted from Figure 1 in Grefen, Aberer, Hoffner & Ludwig,
CrossFlow: Cross Organizational Workflow
Management inDynamic Virtual Enterprises, 2000.
[CW: you’ll need to copy and paste the link into your browser and
and then replace the single quote with a real one, the one
on your keyboard should work, WordPress does not
appear to handle this character well], last retrieved 11/10/09)
Here’s quote 7 after I’ve inserted “Invoke,” “Monitor,” “Control,” and “Get Result” at the places in text that correspond to the arrows in the diagram:
“EMR workflow systems (WfSs) will need to coordinate execution of workflow processes among separate but interacting EMR WfSs. For example, when a general pediatric (GP) EMR workflow system (GP EMR WfS) forwards (“Invoke”) a clinical document to a pediatric subspecialist (PS) who is also using an EMR workflow system (PS EMR WfS), the GP EMR WfS eventually expects a referral report back from the PS EMR WfS. When the result arrives (“Result”), it needs to be placed in the relevant section in the correct patient chart and the appropriate person needs to be notified (perhaps via an item in a To-Do list). If the expected document does not materialize within a designated interval (“Monitor”), the GP EMR WfS needs to notify the PS EMR WfS that such a document is expected and that the document should be delivered or an explanation provided as to its non-delivery. The PS EMR WfS may react automatically or escalate to a human handler. If the PS EMR WfS does not respond, the GP EMR WfS may cancel its referral (“Control”) and also escalate to a human handler for follow up (find and fix a workflow problem, renegotiate or terminate an “e-Contract”). Interactions among pediatric EMR workflow systems, explicitly defined internal and cross-EMR workflows, hierarchies of automated and human handlers, and rules and schedules for escalation and expiration will be necessary to achieve seamless coordination among pediatric EMR workflow systems.”
The quote above is a “conversation” between two EMRs, one acting for a pediatrician, the other acting for a pediatric subspecialist. Similar sorts of diagrams are possible for a subspecialty-based medical home for pediatric patients that require regular visits to pediatric subspecialists, or for transfer of a patient’s records from a pediatric to an adult medical home. One of the advantages of workflow automation is that process definitions can change without resorting to rewrite of the programming code that makes EMRs possible. It should be up to pediatric, subspecialist, family medicine, obstetrics & gynecology, and other primary care stakeholders, to negotiate cross-practice EMR workflows and e-contracts consistent with their clinical and business objectives. They should not need to take into account the fragile, ambiguous, unscalable, frozen workflows of traditional EMRs (that is, EMRs lacking workflow management system foundations and BPM functionality).
When presenting a new concept (to my intended audience, I certainly didn’t invent the ideas, just this specific example of their application), it can be helpful to present the same basic idea in several different ways. The following is another diagram of a process shared across organizations.
This figure is adapted from the introduction to the 2001 dissertation, The Contracting Agent: concepts and architecture of a generic software component for electronic business based on outsourcing of work. (Andries van Dijk. – Eindhoven : Technische Universiteit Eindhoven). Workflow Engine A kicks off workflow A1 through A5, but between A3 and A4 outsources steps B1 through B3 via Web services.
Rob Allen’s Workflow: An Introduction also covers communicating workflow engines at a basic level (pages 10-24) and is a useful overview of workflow in general.
I can’t resist my little word substitution game again. Dr. van Dijk also notes:
“An influential approach to the modelling of communicating information systems is called the language / action perspective (LAP). The basis for LAP was a growing awareness that linguistic theories are relevant for the design of communicating information systems. A cornerstone of the LAP approach is the linguistic theory of speech acts developed by Searle in 1969.”
You would surely agree that an EMR is an “information system,” and that “interoperating” is an instance of “communicating,” so replace “communicating information systems” with “interoperating EMRs” and you get:
“An influential approach to the modelling of communicating information systems is called the language / action perspective (LAP). The basis for LAP was a growing awareness that linguistic theories are relevant for the design of interoperating EMRs. A cornerstone of the LAP approach is the linguistic theory of speech acts developed by Searle in 1969.”
I previously mentioned relevance of linguistic theory to EMR interoperability–I rest that case.
An Obvious Question
An obvious question occurs. If there is such a great “fit” between what EMRs need and what workflow systems and business process management do, why hasn’t it happened yet? I myself have been puzzled by this. I think there is an element of NIHism (Not Invented Here-ism). The United States is a remarkable generator of new information technologies, from the large high tech companies to the university spin-offs to inventors who start in a garage. Much workflow research took place, and many commercial BPM products created, outside the US (as I noted previously).
However, EMR workflow management systems are not prevalent in Europe either (the Soarian system, initially developed in Germany, being the sole exception of which I am aware). So I have another theory, which I will hold for a later post.
Nonetheless, workflow management systems and business process management technology is diffusing into the healthcare industry at an increasing rate. Some day most pediatric and primary care (and other general-purpose and specialty-specific) EMRs will be EMR workflow systems, although by then I expect the phrase “workflow system” to disappear. It will be the non-workflow system EMRs that will require qualification, much as we use “analog watch” or “silent movie” today (so-called retronyms) to distinguish them from their modern descendants.