Care Coordination and EMR Workflow Systems: Key Ideas

Short Link:

As the phrase “clinical groupware” gains currency [UPDATE: well, it gained and then lost that, but what it means is just as important today!], it’s worth considering the history of groupware in general, and workflow in particular, to understand the relationship between EMR workflow systems and clinical groupware. This relationship is at the technological heart of the care coordination problem.

Workflow systems are a form of groupware, and EMR workflow systems are a form of clinical groupware. Jonathan Grudin, in a 1994 Communications of the Association for Computing Machinery article (second most cited for “groupware” in Google Scholar) wrote:

“Desktop conferencing, videoconferencing, co-authoring features and applications, electronic mail and bulletin boards, meeting support systems, voice applications, workflow systems, and group calendars are key examples of groupware.” (Groupware and Social Dynamics: Eight Challenges for Developers, 1994, my emphasis)

Last week I described the landmark 2000 HIMSS presentation and proceedings paper about a workflow-based clinical groupware system installed in ten pediatric practices and one family medicine practice. In it I quoted from two early (1988 and 1992) collections of readings about groupware. I found so much relevant material that I collated, annotated, and published it (see below) so it can become part of a larger conversation about clinical groupware. I’ll refer to this material in future posts.


 From My Bookshelf

[TABLE=3]It is fitting to close this litany of groupware, coordination, and workflow quotes and comments with one more wrinkle, what Frisse, Schnase, and Metcalfe call “The Problem of Language: The efforts to integrate information from disparate sources into a single, unified, computer-based patient record are challenged more by the enormous range of human expression than by technology” (Models of Patient Records,1994). Using the phrase “medical groupware,” not “clinical groupware”, they eloquently describe the importance of medical “conversation” to clinical groupware (see my earlier posts on syntactic, semantic, pragmatic, and “conversational” EMR interoperability):

Models for Patient Records

“When performance is defined as the result of collective efforts rather than as the result of the actions of an individual, software systems supporting these activities may be labeled under the popular rubric groupware….Although it is tempting to think of these activities as “transactions” it is equally valid to consider them “conversations” related to the solution of specific tasks….Using conversations as a central metaphor for handling patients’ records reflects workflow in a clinical setting….the introduction of groupware designed to facilitate conversations will allow for the acknowledgement and representation of the centrality of human conversation rather than force individuals to reconstruct these conversations through examination of data tables and unstructured patient records….medical groupware helps us redefine where our information systems are going and reflect on their origins and true purpose….it should be remembered that the system is nothing more or less than the community of individuals who collectively care for one another.” [CW: my emphasis]

Some workflow systems literally model, execute, and monitor speech acts (proposals, counter-proposals, promises, excuses, and so on). If we are to move from “conversation” as an interesting metaphor, to practical ways to coordinate the “community of individuals who collectively care for one another,” we will need both the informal and spontaneous clinical groupware, and the more formal and prescriptive clinical groupware known as EMR workflow systems. Their strategic combination is at the technological heart of the care coordination opportunity.


  1. Baecker, R. Part I: Introduction, Baecker, R. (Ed.) Readings in Groupware and Computer-Supported Cooperative Work: Assisting Human-Human Collaboration, Morgan Kaufmann, 1992.
  2. Coleman, D. & Khanna, R., Groupware: Technology and Applications, Prentice Hall, 1995.
  3. Ellis, C, Gibbs, S, & Rein, G, Groupware: Some Issues and Experiences, Communications of the ACM, Volume 34, No 1, January, 1991.
  4. Flor, N, & Hutchens, E. Analyzing Distributed Cognition in Software Teams: A Case Study of Team Programming During Perfective Software Maintenance, In Joenemann-Belliveau, T, Moher, T. & Robertson, S. (Eds.) Empirical Studies of Programmers, Fourth Workshop, Ablex, 1991.
  5. Frisse, M, Schnase, J, Metcalfe, E, Models of Patient Records, Vol 69, No 7, July 1994, Academic Medicine.
  6. Grief, I. (Ed.) Computer-Supported Cooperative Work: A Book of Readings, Morgan Kaufmann, 1988.
  7. Grudin, J. Groupware and Cooperative Work: Problems and Prospects, In Laural, B (Ed.), The Art of Human Computer Interface Design, Addison-Wesley, 1990.
  8. Johnson-Lenz, P. & Johnson-Lenz, T. Groupware: The Process and Impacts of Design Choices. In Kerr, E. & Hiltz, S. (Eds.), Computer-Mediated Communication Systems: Status and Evaluation, Academic Press, 1982.
  9. Khoshafian, S. & and Buckiewicz, M., Introduction to Groupware, Workflow, and Workgroup Computing, Wiley, 1995.
  10. Malone, T. & Crowston, K, What is Coordination Theory and How Can It Help Design Cooperative Work Systems, In Halasz, F. (Ed.) CSCW 90: Proceedings of the Conference on Computer-Supported Cooperative Work, Los Angeles, Oct 7-10, 1990, ACM.
  11. Rodden, T. & Blair, G. CSCW and Distributed Systems: The Problem of Control, Bannen, L., Robinson, M, & Schmidt, K, (Eds.) Proceedings of the Second European Conference on Computer-Supported Cooperative Work, Sept 25-27, 1991, Amsterdam.

9 thoughts on “Care Coordination and EMR Workflow Systems: Key Ideas”

  1. Dear Chuck: Great piece! There are really no new ideas, are there? But good ones can be re-cycled as times change. I would suggest you contact some of the people and firms in the Clinical Groupware Collaborative, to lend your expertise and also to learn about what these folks are trying to accomplish. I’d also suggest that the involvement of the patient/consumer in the groupware and workflow model that we’re espousing for EHR technology is new and very important, IT that supports the new Participatory Medicine conceptual framework. It’s not an accident that a pediatrician, Alan Greene, MD, is one of the leaders of that movement.
    Very kind regards, dCK

    1. And thank you David!

      Both for your kind comment and your continuing leadership.

      Your posts and publications are the items I most frequently forward to my colleagues.

      I’ve already spoken with Vince Kuraitis from the CGC, look forward to further exchanges of ideas, and intend for clinical groupware and clinical BPM to realize their full potential.

      There is a great deal to be gained by combining the kind of clinical groupware advocated by the CGC with the process-aware EMR groupware I’ve been promoting for a decade.

      The three technology trends I see coming together are:

      Process-aware EMR workflow/BMP approaches,

      CSCW/Clinical groupware and social informatics approaches, and

      Modular componentized EMR architectures.



  2. David, I see something very different in Chuck’s approach.

    Conventional wisdom: health information exchange first, then maybe get around to thinking about workflow later

    Chuck’s model: understand/define workflow (and EMR usability) FIRST, then define health information exchange (data) requirements.


      1. I received a comment from someone through my contact page that this question deserves a more serious reply, or at least they’d like to hear one. I agree.

        Great clinical groupware across organizational boundaries won’t be possible without great clinical groupware within organizational boundaries. It’s not a question of which to do first but rather whether or not to do both well.

        Effective, efficient, and satisfactory clinical groupware (parallelism with the ISO usability definition is intentional) across organizational boundaries won’t be possible without effective, efficient, and satisfactory clinical groupware within organizational boundaries.

        Without some kind of workflow-based clinical groupware engine, or engines, within and without the physician practice, passing data *and* coordinating the pragmatic effects of that data (that is, supporting and facilitating care coordination) the result will be fragile, ambiguous, unscalable, frozen cross-organizational workflows.

  3. Chuck, your bookshelf looks surprisingly similar to mine. (Seriously, I have those books in almost the same order … are you sure you didn’t sneak in an take the picture in my office? 🙂

    I think you have a really good point that the idea behind “clinical groupware” are very similar to, and very compatible with, the ideas of adaptive case management. Need to get these groups talking.

    Adaptive Case Management Blog Posts

    -Keith Swenson

    1. Hi Keith, was a great conference (I attended specifically for the adaptive case management content).

      Several other folks sent me pictures of similar stacks of books. Maybe I should create a sort of exhibition of pictures of stacks of favorite books.

      Seriously, there are amazing numbers of interesting connections between the workflow management systems/business process management/case management industry on one hand and the EMR/EHR/clinical groupware industry on the other. It’s like standing at an intersection and watching two speeding cars who aren’t aware of each other. The impetus for this blog was an intent to entice and cajole more intellectual commerce between these two complementary and growing software sectors.

      The collision (a good one) is inevitable. It’s just a matter of speeding it up.

      Looking forward to the next Adaptive Case Management conference (with a healthcare track!).


  4. Very interesting article.

    There seems to be a few competing interests here that bring up some challenges. Care coordination and work flow management in EMR’s current state revolves around a single institution organizing its internal teams or individuals. The reality of healthcare today is that care coordination has to happen across multiple systems and disconnected teams or individuals that are acting in silos currently.

    HIEs help with information flow between the parties, but don’t give insights into current actions being taken for true care coordination or a platform for communication between parties. Care coordination must focus on the individual, the patient, and revolve around their ecosystem that involves multiple ‘silos’ of care.

    In practice, we can centralize the health information for a patient into a central record that is very easy for anyone to use. We then empower all of the care stakeholders to see what is happening in real-time, without disrupting their current work flow. Individual providers can see how their care relates to what others are doing and are now incentivized to do so through readmission penalties, patient-centered medical homes, and ACOs. Care managers (professional or informal family ones) can then assign tasks, implement automated workflows (such as reminders or alerts) that work for that individual patient, and can make sure care the overall care is going down the correct path without any unnecessary duplication. We call it CareTree (

  5. Thanks Carl,

    I hope you don’t mind me pulling your self-description (I’ve added my comments too, below)….

    CareTree: A “HIPAA compliant Facebook” for better Care Coordination and Universal HIE

    (From Pioneer Pitch Day, October 16, 2013, New York City)

    “Families are the best care managers, especially with AARP’s projected shortage of caregivers. Families have a patient load of 1 or 2, instead of 100 or 200, and are incentivized to make sure loved ones have the best care, stay healthy, manage their chronic conditions, and stay out of the hospital (or avoid readmissions). Families WANT to be empowered to do this.

    CareTree is the solution. Simply create a patient profile and invite other care stakeholders to access it, controlling who can see what. Users can document care activities, exchange messages, receive text messages for emergencies, track vitals, centralize documents, and keep all of the care activities in one calendar. It’s easy for families to use. Professionals can provide better care, increase patient satisfaction and engagement, and lower readmissions without disrupting their workflow using CareTree’s EMR integration that logs into patient portals and centralizes health records across any EMR nationwide (also creating a self-sustaining national HIE).”

    Thank you for pointing out CareTree to me!

    You make many excellent points: important role of the family, difference between mere data and actionable data, and integration of automated workflows into a health record.

    Not too long ago, EHRs and PHRs were repositories of data. Then data began to flow among them and other systems. This requires syntactic interop (HL7’s pipes “|” and hats “^”) and semantic interop (does this number or code in your system mean the same as in my system). This is all well and good.

    However, there is another level of interop, above syntax and semantics, that is sometimes referred to as workflow or pragmatic (a term from linguistics) interop. A message may be transmitted from system to system due to syntactic interop. The message, should someone happen to view it, has the same meaning across systems, due to semantic interop. However, to ensure that a transmitted message has the intended effect, that is, causes the right thing to happen to the right person at the right time and in the right manner…this requires a layer of intelligent representation of workflow.

    I call this representation an “executable process model.” it can be quite detailed, as in prescribing what happens in what order and which resources are consumed to achieve what goals. Or it can be more proscriptive, simply preventing things from happening, but also allowing users, including patients, to do their job or fulfill their role as they see fit, but more efficiently and effectively.

    I did see suggestions of this workflow level of interoperability in your Pioneer Pitch. I’d love to learn more about your system architecture. Do you represent workflows? If so, at what level of granularity? How easy are they for users and designers to understand and improve? Are you collecting workflow analytics to systematically improve user and patient experience? That sort of thing.

    By the way, I’ll add CareTree to a directory of vendors I maintain. It celebrates People and Organizations improving Workflow with Health IT, or POW!HIT!. If you search Twitter for #POWHIT you’ll see tweets pointing to various POW!HIT! Profiles. I’ll tweet you at @CareTreeMe when it’s up.

    In closing, thank you very much for stopping by. I’m seeing more and more, not just task management (along hardcoded workflows), but true workflow management in healthcare, in which users and designers have the freedom and are empowered to ever more systematically improve healthcare efficiency, effectiveness, and user satisfaction.


Leave a Reply

Your email address will not be published. Required fields are marked *